Slope stability probability classification; SSPC

Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SLOPE STABILITY PROBABILITY CL SS FIC TION SSPC 2nd edition Robert Hack Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SLOPE STABILITY PROBABILITY CLASSIFICATION SSPC ITc t·a. "-'~"ary 2nd edition Robert Hack Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ i! PUBI.JSH.ING edition printed in 1996, by "'"'A.wcu edition in by International H1Silti.J:re 90 6164 154 3 rrc puoucanon nun:lber International Institute for 1998 AJ:!:rosoa<~e H.R.G.K. the material orc•tected any means, electronic or mt:~hamcal retl:1.eva1 ""'"'..,_,.,..... without v;;Titten EB The Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 1b Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ iv Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ABSTRACT ,,.,..,,1,.,.,,.., and the poor results of..,,..•.,.,. . . ,"' d~vel;)prnetlt of a rock classification scheme, which has been developed, classities rock mass cm:npe:nsa:ted for and exc:avalciOn dls1::url:~an<~e a for an im:aginaty stability assessment thence mass are calculated. nevv slope a classification pmran1et1~rs Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ vi Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Faust Il Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ CONJ.""EN1S 1~STRACT b V xiv Preface XV A ~'TROmJCTION A. i THE RESEARCH A.l.l Problem definition A.L2 A.2 A.2.1 A.2.2 A.2.3 A.2.4 A.3 A.3.1 A.3.2 A.3.3 3 3 4 INTACT ROCK VERSUS .ROCK MASS Rock mass components Geotechnical units Water Characteristics of intact rock and rock mass THE RESEARCH AREA Climate and vegetation of the Falset research area ue,mo.gtciai and geological characteristics of the FaJset research area ,ithostratigratph:ic units and sub··uruts EXISTING ROCK MASS CHARACTERIZATION & CLASSIFIC..4..TION B.l B.2 B.2.1 B.2.2 B.2.3 INTRODUCTION EXISTING SYSTEMS and characterization systems B.2.3.6 B.2.4 B.3.l B.3.2 B.3.3 19 21 25 Barton'& v·;>v~>>l:ltn Laubscher's MRMR Franklin's Siz.e 26 27 l<':!ck masses B.2.3.7 Rock mass classification systems fur surface emll.n<~er.mg aprl!Jcati()!JS B.2.4J Barton's Bieniawski's RMR smbility B.2.4.2 B.2.4.3 Vecchia - Terrain index for stability of hillsides and scarps B.2.4.4 classification Romana's SMR (modified i'Si:l~nJJ>W!>kil B.2.4.6 Haines (m<}di.lied LalJOSI~ller; B.2.4.7 Shuk - Natural methodology (NSM) B.2.4.8 to assess natural Hudson's RES -rock mass characterization B.2.4.9 .t.x.ca''allil.Oll!ty, rl.ppai;ili.!tv and assessment B.2.4.10 CALCULATION METHODS ,<\ND PARA.l\<1ETERS IN EXISTING CLASSIFICA:r10N SYS·· TEM:S Method of calc-ulation Correlations between different classification systems classification systems Influence of parameters in !8 25 25 R2.4.5 B.3 16 Hi 24 classification systems Classificaticm Modified Hoek-Bmwn failure criterion for NATM- New Austrian Tunnelling Method Hudson's RES - Rock En.gineering 15 22 22 Recent classification systems Bieniawski's RMR B.2.3.1 B.2.3.2. B.2.3.3 B.2.3.4 B.2.3.5 6 6 9 lO l 27 27 27 28 28 28 28 29 29 29 29 30 3l ~' .Ys 32 32 32 33 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B.3.4 Problems with parru.neten; in B.3 .4. l Intact rock B.3.4,2 .13.3.4.3 B.3AA B.3.4.5 B.3.4.6 B.3.4.7 B.3.4.8 B.3.4.9 B.3.4.IO rock mass classification sy&'tems Rock rus·co:nttimJJtY sets 38 Persistem:e of discout:inuities Condition of discontinuities 38 39 39 39 SW~ce:pi;i1)il.i1ijl 40 i:O W;>!;l.th,gt'i•W De.tb.rruation of intact roc.t aoo rock ma.<Js, stress relief Relative orientation of and discontinuities B.3.4.l! B.3.4.12 B.3.4.B Ice and snow influence B.3.4.14 Method of excavation in a discontilmous rock mass u"'"'""·"' e;~pe:ne1nce and with a classification B.3.4.l5 B.4 c .8.3.4.16 SUMMARY 36 36 36 40 40 40 40 42 42 42 43 44 PARA,\IETER DEFINITION A1'\'D I~'ITIAL POINT RATING SYSTEM C.l INTRODUCTION C.Ll Data and srorage C.2 SLOPE GEOMETRY AND STA.N"DARDS FOR VISUAL ASSESSMENT AND CLASSIFICATION OF SLOPE STABILITY C.2.l Ge•cam~try of C.2.2 Visual estimation of C.3 PARAMETERS IN ROCK SLOPE STABILITY 47 49 49 C.3J 54 C.3.2 C.3.2.L4 C.3.2.L5 C.3.2.L6 C.3.2 ..7 C.3.2.2 Shear c.:U.I In la et rock methods means' intact rock field estimates field estimates versus UCS tests i<e1peatabii!It) of intact rock strength estimates of water saturation on intact roek C.3.3.2.2 C.3.3.2.3 C.3.3.2.4 C.3.3.2.5 C.3.3.2.6 C.3.3.2.7 C.3.3.3 C.3.3.4 54 55 55 56 57 58 Conclusions 58 59 Persistence 62 62 60 C.3.3.2 C.3.3.2.I 63 63 64 65 66 Conclusions Alteration of a wall infill material DilSCO.fltilt"I.Ul.!l:y C.3.3.4.I C.3.4 52 .54 54 C.3.2.LI C.3.2.1.2 C.3.2.l.3 C.3.3 5l 51 C.3.3.4.2. of infiil material Conclusions C.3.3.4.3 C.3.3.5 Weathered discontinuities Di,sconti!t1Uity brst features C.3.3.6 C.3.3.7 Effect of water pressure in discontinuities Practical aspects of shear tests on discontinuities C.3.3.8 C.3.3.9 Cmu:lus:ions Sets of discontinuities versus C.3.4J. sets C.3.4.2 disoontinuities and ae1enrunmg parameters C.3.4.2.1 and structural :ama!vr•es. C.3.4.2.2 Scanline method C.3.4.2.3 66 66 67 68 69 69 70 70 71 71 7! 72 73 74 74 74 74 75 75 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ CX!NTE'lv13 C.3.4.2A C.3.4.3 C.3.4.4 C.3.4.5 C.3.5 ~xpos:m:e discmr~ti:rmilty Overall contiition of Conclusions and sets in a rock mass sets in a rock mass 76 77 77 78 78 C.3.5J C. Hi C.3.5.2 Method of excavation External influences C.3.6.l Surfuce run-off water C.3.6.2 Snow and ice Rock rnass creep and stress relief Extemai stresses C.3.6.3 C.3.6.4 78 79 79 79 79 80 80 C.3.6.5 C.3.7 W~uirrnlmy • CA 'INITIAL POR-JT RATING SYSTEM CA.l CA.2 C.4.3 C.4.4 Results Discussion parameters in rock 80 81 81 81 system 83 83 Conclusion 85 SLOPE STABILITY PROBABILITY CLASSIFICATION - SSPC D.l THE DEVELOPMENT OF THE SSPC SYSTEM D.l.l 'Reference Rock Mass' D.LU Detem1ination of parameters & fuctors D.U.2 Mathematical modellling D.Ll.3 'Orientation D.l.2 D.L2.1 D.L2.l.l D.L2.L2 D.L2.L3 D.l.2.1.4 D.L2.I.5 D.l.2.2 D.l.2.2.l D.l.2.2.2 D.L2.2.3 87 88 88 89 90 92 92 92 94 ctiterion' Refinement of initial criterion Correlation of the threshold friGtion values of the friction values Reliability of friction values based on Discussion and conclusion criterion' to test and literature criterion and Di:scontil:miity condition and Conclusions uucoc1mg criterion' Correlation of rock mass parameters with estimated Models D.L3.2 for of rliscontinuities and condition of discontinuities D. .3.3 Linear modei D. .3.4 Discussion and cmtclusions linear model D.L3.4.! Shear model D.l.3.5 The shear plane model and its physical me:aniJ.1g D.L3.5.l D.L3.5.2 Parameters in the shear model D. .3.5.3 Optimiz.ation procedure for the shear model D.1.35.4 Discussion of the shear model D.L3.6 Discussion and conclusions Parameter fur the method of excavation D.1 A .1 Methods of excavation used for in the research area and gecaec:tm:tcal parameters these methods influenced D.1.4.2 Influence of the method of excavation on the discontinuity D. L4.2.1 Interdependency between discontinuity and method of excavation D.L4.22 The values of the paran1eter for the method of excavation D.l.4.3 Reliability of the parameter for the method of excavation Discussion, to literature values and conclusion D. i .4.4 Parameter for the of ., ._,~...,.,~,_.~, mt<m:lA~;:;ena~~nctes between we:amienrtg and iith.ost.ratitgN.tphical unit D.L5.1 Calculation method D.L5.2 Influence of on rock mass parameters used ln the SSPC system D.1.5.3 ~ parameter in SSPC system D.L5.4 0.1.3.1 D.l.5 96 96 97 98 98 99 WO !.00 101 102 D.L3 D.l.4 xi 102 102 103 105 105 106 106 107 108 110 110 113 !!3 ll3 114 l16 117 ll8 !20 120 121 121 123 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D.l.5 .5 Reliability D.l.5 .6 Comparison to literature values D.1.5.7 Conclusions D.1.6 Susceptibility to weathering D. L 7 Water pressures in discontinuities D.2 PROBABILITY ANALYSES D.2.1 Distributions of field dam and derived parameters D.2.2 Probability of orienlation dependent slability D.2.2.1 Probability of 'sliding criterion' D.2.2.2 Probability of 'toppling' criterion D.2.3 Probability of the orientation independent slope slability D.2.3.1 Probability of the linear model for orientation independent slope slability D.2.3 .2 Probability of the shear plane model for orienlation independent slope stability Probability of the cohllllliS and f'IIIIIIS D.2.3.3 D.2.4 Probability of the values for the method of excavation and degree of weathering parameters D.2.4.1 Probability of the values for the parameter of the method of excavation D.2.4.2 Probability of the values fur the parameter of the degree of weathering D.2 .5 Conclusions D.3 THE COMPLETE SSPC SYSTEM D.3 .1 Exposure characterization D.3.2 Reference rock mass Determination of number of geotechnical units in a reference rock mass D.3.2.1 D.3.3 Slope stability probability D.4 RESULTS AND COMPARISON D.4.1 Slope stability D.4 .1.1 Application of SSPC system .D-4 .1.2 AppliCjlfion of H:~s' slope ~:lassificafion 0.4.1.3 Appiication of Romana's SMR system D.4.1.4 Discussion D.4 .2 SSPC system's rock mass 'strength' parameters - rock mass cohesion and friction D.4.2.1 SSPC system versus Bieniawski's RMR system D.4.2.2 SSPC system versus the 'modified Hoek-Brown failure criterion' D.4.2.3 Discussion Conclusions D.4.3 EXAMPLES AND VALIDATION D.5 Example I. Predicting the slabiiity of a slope in Lower Muschelkalk (Tg21) D.5.1 D.5 .1.1 Slope stability by classification Example ll. Plane sliding failure in a 40 year old slope in Upper Muschelkalk (Tg23) D.5.2 D.5.2.1 Slope slability by classification D.5.2.2 Laboratory tests D.5.2.3 Slope stability by limiting-equilibrium back calculation D.5.2.4 Slope stabilitY by numerical analysis - UD£C simulation D.5.2.5 Conclusions example ll D.5 .3 Example m. Non (ijsc~mtinuity related failure in a 4 year old slope in Carboniferous slate D.5 .3 .1 Slope stability probability by SSPC classification D.5 .3 .2 Slope smbility by kinematic analysis D.5.3.3 Laboratory tests D.5.3.4 Slope slability by limiting-equilibrium back calculation D.5.3.5 Slope smbility by numerical analysis- UDEC simulation D.5.3.6 Conclusions example m D.5 .4 Example IV. Inftuence of weathering and method of excavation on the stability of a slope in Upper Muschelkalk (Tg23) D.5 .4 .1 Slope stability by kinematic analysis or calculation D.5.4.2 Slope stability by classification D.5.4.3 Conclusions example IV D.5.5 General conclusions from the examples D.6 CONCLUSIONS TABLES - SLOPE STABILITY PROBABILITY CLASSIFICATION (SSPC) APPENDIX I APPENDIX II STEPS ON DISCONTINUITY PLANES APPENDIX m CORRELATION OF THRESHOLD VALUES OF SLIDING CRITERION 10 TEST AND LITERATURE VALUES INFLUENCE OF WEATHERING ON GEOI'ECHNICAL PARAMETERS APPENDIX IV WEATHERING CLASSIFICATION APPENDIX V EXAMPLES - SSPC FORMS APPENDIX VI BLANK SSPC CLASSIFICATION FORMS APPENDIX VII 125 125 125 126 126 128 128 132 132 133 134 134 134 136 137 137 137 138 139 140 146 147 149 154 154 154 154 156 156 157 158 158 159 159 160 160 161 164 164 165 165 166 166 168 169 169 170 170 171 171 173 174 174 174 174 175 179 185 191 199 205 211 227 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ CONTENJS REFERENCES GLOSSARY SYMBOLS & ABBREVIATIONS INDEX FIGURES TABLES COMPUTER PROGRAMMES CURRICULUM VITAE Figures and tables with numbers starting with 'A' are included in the appendices and starting with 'G' in the glossary. xiii 235 241 247 249 250 252 253 254 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ October 1996 22 May 1998 f!) Centre fur Technical Geoseience, International I11Stitute for Ae!usp,ace -"'-"''""'"'·'""''" 3, 2628 EB Delft, The Netherlands. Sciences (ll 2628 RX De!ft nw13rn11.v of tec:nmHog;y, The Netherlands. of and Earth Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ackruwledgemem :xv Acknowledgement I am sincerely in debt to David Price and Niek Rengers for guiding and helping during the research. The regular discussions during coffee or while having a Spanish lunch in 'El Hostal' with a clear view on a spectacular slope (Fig. 5) will not be easily forgotten. To both I am clearly also in debt for the tedious job of critically reviewing and editing of the manuscript. Johan Kaashoek of the Emsm.us University, Rotterdam, Mathisca de Gunst of the Department of Mathematics and Computer Science of the Free University of Amsterdam and Dieter Genske of the Technical University Delft are acknowledged for their help with and review of the statistical methods used. I also thank my colleagues in I'I'C and in the Technical University Delft for their help during the fieldwork. A special word of thanks should be given to Willem Verwaal and Amo Mulder of the Technical University Delft for getting the samples aad test results used in the research. I thank ITC, being my employer, for giving the opportunity to do the research and providing the financial support. The largest contribution to this research is probably made by the graduate students from ITC and the Technical University Delft who collected the data. They provided the data that allowed me to establish the relations and to develop the classification system. Without knowing it, their ideas and sometimes blunt comments on preliminary versions of the classification system helped me to eliminate ambiguous elements and to improve the system. I like to express my sincere gratitude to the inhabitants of the Falset area in Spain, the City Council of Falset, the Quardia Civil and, in particular, to the staff of the Hostal Sport in Falset. For years they had to put up with students, staff members and me doing 'strange' things to their rocks, hampering traffic, using the swimming pool as site laboratory, being late or too early for dinner and generally being very prominently present. No complaints have ever reached us, on the contrary, they helped us when and wherever possible and provided all filcilities necagary for doing the reeareh. I thank Hmmeke for her assistance and moral support while having suffered my .often irritating moods, and for her loving care provided. Robert Hack 1 October 1996 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ A lN.TRODUCIW.N A INTRODUCTION 1 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 2 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ A INTRODUCflON 3 A.l THE RESEARCH A.l.l Problem definition In the practice of constructing engineering structures, such as buildings, tunnels and slopes, an interaction takes place between the 'ground' and the engineering structure. The knowledge of the consequences of the influence of the 'ground' on the engineering structure and vice versa are often critical for the economic and safe design of an engineering structure. In particular the mechanical response of the 'ground' under influence of the engineering structure should be known before an engineering structure is built. 'Ground' is a very broad term. The 'ground' is any natural material present at the site where the engineering structure is to be built on or in. 'Ground' is normally divided in 'soil' and 'rock'. 'Soil' consists of loose particles not cemented together whereas the particles in rock are cemented together, resulting in a tensile strength. This difference in characteristics between 'soil' and 'rock' haS also re&ulted ·the development ·of dittetenr m:etb.odologtes for the ·catculation of the mechanical behaviour of the 'soil' or 'rock'. Most 'rocks' are not continuous, but contain fractures, faults, bedding planes or more general: 'discontinuity<!> planes' that divide the 'rock' into blocks of rock bounded by discontinuities. The whole array of blocks of rocks and discontinuity planes is then designated the 'rock mass' or 'discontinuous rock mass' . The research described has been done to develop an improved methodology for the assessment of 'rock' slope stability for 'discontinuous rock masses'. m Discontinuous rock masses In the last decades the study of discontinuous rock mechanics has developed tremendously. For constructions, such as slopes, foundations and shallow tunnels it has been recognized that discontinuities have a major influence on the mechanical properties of a rock mass. This perception has major consequences for the assessment of the engineering behaviour of a rock mass. Descriptions and characterizations, engineering geological maps and calculations for engineering muctures.. in. ol'. on a rock .mass have tQ .include dis,cwtin\rity pro~~' Variations in properties, however, can be considerable along the same discontinuity plane. As there may be hundreds of discontinuities in a rock mass, each with its own variable properties, these, taken together with inhomogeneities the rock: material, require that in order to descn'be or calculate the mechanical behaviour ()fthe rock mass accurately, a large amount of data is required. Laboratory and field tests are available to obtain discontinuity properties. Testing in large quantities is, however, time consuming and troublesome. Continuum calculations for engineering muctures in or on a rock mass, whether analytical or numerical, cannot be appropriate, as the simplifications needed to present the rock mass as a continuum are so substantial that it is nearly always highly questionable to what extent the final calculation model still represents reality. Discontinuous 'distinct block' numerical calculations can model the discontinuities and calculate the behaviour of a rock mass in all detail, provided that property data are available. Apart from the need to have powerful computers to do the large number of calculations req\rired by the vast quantity of discontinuities, the test data needed for a detailed numerical discontinuous calculation are never available. An often applied practice to avoid these problems is to simplify the discontinuity model, and estimate or guess the properties or to use literature values. To what extent the result is still representative for the real situation is a question that often remains unanswered. Analytical or numerical calculations should be performed in three dimensions because discontinuities usually make a rock mass m (I) The terms discontinuous rock mass and discontinuity are used in a rock mechanical sense. A discontinuity is a plane that marks an interruption in the continuity and normally has low or zero tensile strength. A discontinuous rock mass is a rock mass containing discontinuities. (see further chapter A.2 and glossary, page 241) Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 4 A. Problem dejirtition 1 development of a new classification system for slope stability. re~•erurch :in the Falset area :in the nn:lvu1ce The data for the reserurch were coJJ:ec:teC! ...,.,,..,.,..,..,.,. of Spain. Within the context groups mountainous typical between 5 and 25 to developed, classifies rock mass parameters We<l,LIJitlfllllg and P"Vf''Mr<>t><'•1> r1Hlh,1"h''>n,••po in an and stabil1tv assessment thence allows assessment of the rw!ut:u~..:,~: of for the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ A ll'.l11?.0DUCTION S necessary tern1lUOJogy and ......,.u ......"'"~ for rock and rock masses .,.,,.!,.,.,,.,... ~t· area used for the research. prucruu1et1ers are evaluated on aeJWlllOC'n of merits for in section C ruud the and the results of the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 6 A.2Intact rock versus rock mass A.2 INTACT ROCK VERSUS ROCK MASS A rock mass may consist of intact rock only, but is more commonly formed from an array of intact rock blocks with boundaries formed by discontinuities (Fig. 1). Within the rock mass the mechanical properties of both the intact rock blocks and the discontinuities may be inhomogeneous and anisotropic. A common relation between rock, rock mass and engineering is (Price, 1984): material properties 1'IU1&f properties * + 'INlSS fabric + environment = 1'IU1&f properties "' * engineering geological engineering geological matrix • • ,__z.._.:_. __ _ _ _ = .. r__ engmeenng utmuvWfl/ir changes produced by engineering work _ * __;.:..:..;;_...;;..;.;!OL,;.;_;;..;;....~-"'--...._ Exact descriptions of rock materiat and follow below. UH~~ matrix .# UH~~ •'-- gi'OIISilU •. -.:~ OJ rock mass are req\iired for understanding the analyses in this research and A.2.1 Rock mass components Intact rock material Intact rock blocks are blocks of rock that do not contain mechanical discontinuities and do have tensile strength. Discontinuities A discontinuity is a plane or surmce that marks a change in physical or chemical characteristics in rock material. A division is made between integral discontinuities and mechanical discontinuities. The latter are planes of physical -weakness. Bedding planes, joints, fractures, mults, etc. are mechanical discontinuities if the tensile strength perpendicular to the discontinuity or the shear strength along the discontinuity are lower than those of the surrounding.rock..material {lSRM, 1978~ 19Ala).• Integral discontinuitiesare discootinuities which are as strong as the surrounding rock material. Integral discontinuities can change into mechanical discontinuities due to weathering or chemical reactions that change the mechanical characteristics. Throughout this book 'discontinuities' denote mechanical discontinuities except where stated otherwise. Discontimdty set Discont:inuities exist as single features (fi.mlt, isolated joint or fracture, etc.) and as discontinuity sets or fumilies (bedding planes, schistosity, cleavage, joints, etc. )<2>. A set denotes a series of discontinuities for which the geological origin (history, etc.), the orientation, spacing and the mechanical characteristics (friction angle, roughness, infill material, etc.) are broadly the same. In some circumstances a discontinuity is treated as a single discontinuity although it belongs to a discontinuity set, in particular if the spacing is very wide compared to the size of the engineering application or to the size of the geotechnical unit (eh. C.3.4.1). <2> ~rious geological processes create discontinuities at a broadly regular spacing. For example, bedding planes are the result of a repeated sedimentation cycle with a change of sedimentation material at regular intervals, folding creates joints at regular separations to allow fur shrinkage or expansion of the rock material, etc.. Normally discontinuities with the same origin have broadly the same characteristics in terms of roughness, infill, etc.. The orientations of discontinuities with the same origin are related to the process that has created them and to the geological history of the rock mass. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ INTRODUCTION 1 1. Intact rock vs rock n:mss. I, case of am.sotrOJ>Y is intersected different from the discontinuities. The are of rock material or of the materiaL A rock mass discontinuities will be more deformable than intact rock. Such deformation will normally take place relative movement discontinuities and be rather than elastic The tensile of a rock mass discontinuities is low and for many rock masses zero. The noros:ttv a discontinuous rock mass is higher due to the storage capacity of the dis,coJatiiuti•ties considerably higher to via the discontinuities. Di:>contiltmities direction normal to the in a the rock mass in the direction of the movement of part complete of the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ !i A.2 Intact rock versus rock mass estimated overbrsak d '·"'4 .p~b~b!e . . '·, void ~ . 6 3. The influence of discontinuities on the of a tunnel in the progress of construction Amold et al., Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ mass as homogeneous geotechnloai 4 4. Rock mass componeni'S. rnass should This would be ''"'·-'·'-·"'"'"""··· various is thus more accuracy obtained a calculation based on more data the economic and environmental value of the structure to be built and envJ.ronmi::nt or human life. For the of a aHowed within a l'.'-'''-'''·"-·'-'•U>''-''"-' upjt nuclear power w-·~''"1 in a calculation for the "-'"""'·'-""·'-'''-·'" (J) A m1it is, in a part of the rock mass in which the mechanical of the inta!;t rock material uniform and the mechanical prcmerne;s of the discontinuities of within each set of discontinuities are the same. In tt'1is re-,search the unit is al.so uniform. Thi;; additional condition is not A<O•i.i!U.HvW.•;;il>C beCilUSe Of the, influence Of am<>UCHiF Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ units in a Greenish and blueish grey consist ofdolomite and Hmeston.::~. No standard rules are available tor the division units and this and El: limestone & dolomite shear zones, etc. often the of a "v'"''•'"u''"""'· unit In 5 is shown in which different units The influence of the different geotc:ctuncal units on the form of the 6. Section of au'-'"'"""'""'"'"' characteristics of a rock mass. Water adds to the of the rock mass, acts as a lubricant in of some illifill materials a.11d water pressure in of a rock mass. discontim.dties reduces the shear and thus also the Therefore it is necessary to of the rock mass and the geme:cnml,caJ units . In tl1is it must be noted that water is often not a continuous feature in time. Water can be ,.,,.,c""'"t and after rain:fuJl Also the en~pm;;erm influence t~J.e presence of water Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ A INI'RODUCIION With time, most rock masses weather, a process strongly infiuenced by the presence of water, which causes the intact rock strength and the discontinuity strength parameters to decrease. To what extent weathering infiuences the mechanical behaviour of a rock mass depends on the type of engineering application, type of intact rock material and discontinuity :infill material, amount and chemistry of percolating water, etc.. Reduction of shear strength of discontinuities due to water Water pressures in a discontinuity reduce the normal stress on the discontinuity and therefore reduce the shear strength along the discontinuity. Sliding over a discontinuity plane is then possible at a lower dip angle than over a discontinuity without water pressure (Fig. 7a, band c). In traditional limiting-equilibrium calculations for slope stability, water pressures in discontinuities are therefore a main reason for slope instability to occur (Hoek et al., 1981, Giani, 1992). In Fig. 7a, b and c it can easily be seen that the discontinuity dip angle for which equilibrium exists decreases (e~e > ~ > y )<4>. 11 ---- equl8brUn: w·•··w·-··tan, ... , .. oqullllbrUn: w•my+p2• (W*OOBy -p1)*tan,. &+p2-p4• (W•-a +pS· p1) •111n., Fig. 7. Block on slope with and without water pressure (W is the weight of the block, cohesion along discontinuities is zero). Accordingly, because both eftects (pressure and weathering) of the presence of water might or might not be present, water is not included in the rock mass or in the geotechnical unit. The influence of water should, however, be included in any calculation of the behaviour of the geotechnical units. A.2.4 Characteristics of intact rock and rock mass A description of some geotechnical properties and characteristics of rock and rock mass is given hereafter. The properties and characteristics are described as far as important for the development of a slope classification system and not in all detail. The underlying mechanisms are only briefly addressed as a full description of all mechanisms in discontinu()us .rock .mechanics would be beyond the scope of this study~ The . reader is referred to the standard literature for further details (Giani, 1992, Good.man, 1989, Hoek et al., 1980, 1981, etc.). Stress tliim1iiitiiiit in a rock mass The stress distribution in a rock mass is strongly infiuenced by the presence of discontinuities. Fig. 8 shows examples of a stress distribution in intact rock and in discontinuous rock masses. The figures clearly show the variation in the stress contours due to the presence and orientation of discontinuities. ~rmation Deformation of intact rock is the change in volume or shape of intact rock under the infiuence of deforming loads. In general, the deformation of intact rock is partly elastic and partly plastic and some rocks also show a time dependent deformation (see also creep, below). Deformation of a rock mass is the change in volume or shape of the rock mass. The deformation is mainly caused by displacements of intact rock blocks along or perpendicular to discontinuities. <4> If rock blocks are completely submerged in water (Fig. 7d) the normal stress on the discontinuity is reduced (pl > p3) causing a reduction in shear strength, but also the driving forces are reduced (p4 > p2). In a completely submerged slope the equilibrium between driving forces and shear strength is, therefore, less disturbed than in a situation with water pressures acting only on bottom and rear sides of the block (Fig. 7b and c). In slopes the rock blocks near the surface of the slope are normally not completely submerged in water and therefore water pressures cause a reduction in normal stress along the discontinuity plane (Fig. 7b) and driving forces may increase if a discontinuity at the rear of the block is filled by water (Fig. 7c). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ A.2lntact roclc versus roclc nuzss 12 The discontinu.ities cause a dramatic change in deformation behaviour of a rock mass in comparison to that of intact rock. The deformation in a rock mass is fur a large part due to shear displacements along discontinu.ities or opening or closure of disoontinu.ities. The shear deformatioD.$ are non-elastic fur larger displacements. Whether the opening or closure of discontinu.ities is elastic or non-elastic depends on the infill ~u.w ICIIId //1'-"'~\\. ,... ' - I _.,./ ,.. l I I I\\ I l \ _Lf_J_ 1 ,/ •R / '"~--"'l. "'... .... Ill lf'-../1\ I \ I \ '\ ~ J!/1 • Ll i • r .. , ... ' ........ , .... "~T no~ I I ~ • I I I I ,, .,;' ·- lnlllned~ llol'lzonCal cllloonllnullll material in the discontinu.ities and the Fig. 8. Stress distribution (bulbs of pressure - lines of equal major principal stress) in discontinuity wall material but a rock mass due to a vertically oriented plane load (after Gaziev et al., 1971). usually the displacements are nonelastic (e. g. fur a common infi1l material such as clay). Therefore, a rock mass shows mostly non-elastic deformation behaviour. Fig. 9 and Fig. 10 illustrate the non-elastic deformation behaviour of rock masses. Rock mass failure In general a rock mass does not full and therefore fuilure of a rock mass is usually defined as the deformation of the rock mass larger than allowed fur a particular engineering construction. U+-------~------~------.-----~ average dl8placement of plate Fig. 9. Example of a cyclic plate-bearing test on fractured rock (after Schneider, 1967). o.oe o.oa 0.84 4L06 O.GII dlaconlnully IIPidnll (Ill) Fig. 10. D~~~~aot mcJD.,. vs discontinuity spacing for plate diameter 8 cm on a model rock mass (after Berkhout, 1985). CompressiWJ, tensile tmd shear strength of intact rock Intact rock material has compressive<S), tensile and shear strength. Rock material consists of mineral grains completely or partially bonded together by cement or another bonding agency. If loaded to fuilure under a compressive, tensile or shear stress, intact rock material will break into smaller pieces of rock when the compressive, tensile or shear strength is reached ('the rock fuils'). Intact rock strength behaviour may be approximated with a 'Mohr-Coulomb fuilure criterion•<6). This allows definition of the intact rock strength in terms of intact rock cohesion and intact rock friction. Strength of a rock mass The 'st:rength' of a rock mass, as often used in the literature or in day-to-day practice, is a confusing and mise expression. A rock mass may be considered to have strength, but, due to the discontinu.ities in a rock mass, this strength is dependent on a variety of:filctors: the shape and size of the rock mass considered, the environment (e.g. <SJ The compressive strength is dependent on the test method, see glossary, page 241. <6> See glossary, page 241. Note: the Mohr-Coulomb firilure criterion does not suit all rocks in all situations and different theoretical or empirical models for whlch the strength of intact rock have been defined. These wiU not be repeated here as these can be found in any standard text book on rock mechanics (e.g. Goodman, 1989, Hoek et al., 1992). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ A INTRODUCI'ION the engineering application, the confining stresses, etc.), the amount and orientation of discontinuities and, although situations of minor importance, the intact rock strength. Consider the sketch m Fig. 11. The rock mass (including the orientation of the discontinuity) and the stresses on the rock mass are in both cases the same. Only the volume of the rock mass is changed. It is easily seen that the rock mass in Fig. lla has a higher 'strength' than in Fig. llb. In Fig. lla intact rock has to be broken, and in Fig. 11 b sliding along the discontinuity is sufficient for 'failure' . 0 mmany 'J.ensile strength of a rock mass 13 1 ! ! ! i a b l l l I t r r The bonding strength between the particles causes the tensile strength of intact rock. A rock mass with discontinuities has only a tensile Fig. 11. Rock mass under stress. strength if the discontinuities have a tensile strength or are filled, coated or cemented with a material that has a gluing or bonding effuct between both sides of the discontinuity. For most rock masses at (near-) surface this is not true and most rock masses have a tensile strength equal to zero. Compressive and shear strength of a rock mass A rock mass consists of rock blocks bounded by discontinuities which have shear strength and may have some tensile strength. The rock mass could thus be considered as a large scale rock material, rock blocks replacing mineral grains. In a rock mass with discontinuities which have a tensile strength, the bonding agent causing the tensile strength may be broken due to compressive or shear loading. This is comparable to the failure of intact rock material and compressive and shear 'strength' may be defined, although these 'strengths' are likely anisotropic and may still depend on the environment. .If. the discontinuities do not have tensile strength the rock mass may be compared to not cemented dense sand, where grains, being the intact rock blocks, fit closely together. The environment (confinement, etc.), the shear strength along the discontinuities, and the intact rock strength determine the maximum com.pressive and shear load that can be sustained(7). Thus 'failure' depends on the configuration of the rock mass and the orientation and variation of the stress fields. Generally valid compressive and shear strength values can therefore not be defined<&>. In some situations where anisotropy is absent or not very important, it is, however, possible to approximate the strength behaviour of a rock mass in models analogous to the methods used for intact rock, but with strongly reduced values for compressive and shear strength. ~ering Weathering is the chemical and physical change in time of intact rock and rock mass material under the influence of the atmosphere and hydrosphere. Two main processes are distinguished: physical and chemical weathering. Physical weathering results in the breakdown of rock material into progressively smaller fragments. The rock and rock mass bieak up due to temperatW:e Variations reswtiiig in differentiiil expatiSioo lltld shriJ.lbge of minerals, freezing and thawing of water, pressures of water in pores and discontinuities, (re-) crystallization pressures, hydration, and frequent swelling and shrinkage of clays due to water absorption, etc.. Chemical weathering results in decomposition of minerals. Water and groundwater with dissolved chemical agents are of major importance as these react with rock and rock mass materiiil. Normally biotic influences, induced by living organisms, plants, bacteria, worms, etc., are included and cause physical as well as chemical weathering. On or near to the surmce the influence of these processes (due to larger temperature variations, influence of vegetation and rain, etc.) is more distinct than deeper below the surface. In this research also the effucts of stress relief, intact rock creep and rock mass creep are included in the definition of weathering as proposed by Price (1995). Intact rock, and rock (7) Comparing a rock mass to intact rock or to an uncemented sand is only partly valid. The elements in a rock mass (rock blocks) fit together like dry masonry, whereas the grains in intact sedimentary rock or in a sand do usually not fit together. The cement in a rock mass is in the discontinuities whereas in intact rock or in a sediment the elements (grains) are bound together by a cement filling the pores between the grains. (S) An altef'lW.tive way to understand rock mass 'strength' is as follows: Ifloaded to fu.ilure under a compressive or shear stress a piece of intact rock will break into smaller pieces of rock when the compressive or shear strength is reached ('the rock tails'). Effectively it then becomes a rock mass (intact pieces of rock with boundaries by fractures = discontinuities). Reversed this leads to the conclusion that a rock mass does not have a compressive or shear strength; it already consists of blocks with boundaries by discontinuities. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 14 A. 2 Intact rock versus rock mass in which are ""' 1"''"'""'" time under a constant solution mass an processes of creep m intact rock may occur, time and perpendicular to discontinuities. the moves downhill in a slow process under to creep if ilie u.u•c;uv "'''"" mechanisms are: deformation intact rock along exJtstJng rock of new mechanical and al.so included. The process is rock and rock mass are over and the dis;coutlnu.ll:y water pressures from vvater Porosity Pnor.r.cil'u is defined as pore space not rock material and is divided in primary and se.c:onaru-y uonJsltv. p,.,,.,,,.,,.,., uor<)Slltv sec:or1,dary ,...,.,..,t,..,:t·ou is the of rock mass Pn•rr.<:inr ilie rock material or mass and uerm.e:ao1m:v of that of the rock mass. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 15 A lNTRODUcriON A.3 THE RESEARCH AREA The research for the development of a slope classification system has been carried out in the area around Falset in northeast Spain, in the province of Th.rmgona (Fig. 12). The area around Falset is particularly suitable for the type of research described because: 1 The variation in geology, lithology and tectonic environment is large, giving different geological environments for the development of the classification system. 2 The topography is mountainous and vegetation is limited, exposing large areas of rock. 3 Access to the area and to rock exposures along existing roads and paths is not difficult. 4 Numerous old roads exist and several new roads have been built in recent years creating large numbers of road cuts, excavated with different excavation techniques. This has allowed for the comparison of stand-up times of slopes, excavation methods and for an assessment of weathering influences. 5 Aerial and satellite images, topographical and geological maps at various scales are available. ,.- .•••.• main road - ·-.. secondary road -river 0 I 5km , N t -·-· I I I ,.' ) I I / . Gratallops.-. . ..... ' '~ .,.,., ,. ~ \, '·""': ' I I I ! Fig. 12. Research area. Colldejoo ~:: ' ... ·-·-·" _.-'-·~., ' '\ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 16 A. 3 The research area Apart from the research for the development of a slope stability classification system also engineering geological mapping has been carried out in the area. The results of this mapping will be reported on in the form of an engineering geological map and accompanying report and legend (Price et al., in preparation). The engineering geological map and report are, however, not part of this study. Detailed descriptions of topography, geology, and engineering geological mapping units are thus omitted. Details of the area and the geology are summarized below as fur as is necessary for understanding the analyses that result in the slope stability classification system. A.3.1 Climate and vegetation of the Falset research area The climate in the Falset area is Mediterranean, characterized by dry and hot summers (temperature ranges from = 15° to 35° C) and moderate winters (10° to 15°). Part of the area is mountainous, ranging up to about 1000 m above sea level. Rivers and streams in the area are mostly dry from Maroh through October/November. It can rain for long periods d.uring the winter and even up to MarohlApril although this is not typical. Sometimes the rain is torrential. Occasionally temperatures below zero do occur. Snow&ll is seldom in the area, but can full in Maroh which is the fieldwork season. Extensive agricultural use is made of the soft soils and weathered rocks in the valleys. The more mountainous areas are covered with forests or are barren rock. A.3.2 Geological and engineering geological characteristics of the Falset research area In the Falset area the strat:igmphy is composed of sediments of Devonian through Quaternary age and intrusive rooks &em CaRK>nifei'Ol:l&tilrougfr~~. ·A -generalized geological table with the lithology and the main engineering characteristics is given in Thble 1. The table only presents a broad impression of the engineering geological mapping units found in the area and is in no way complete in all details. Sedimentary rocks The Palaeozoic consists predominantly of slates interbedded with micro-conglomerates, sand- and siltstones. A low degree of regional metamorphism developed cleavage in the slates. Contact metamorphism has affucted the Carboniferous rooks near granodiorite intrusions. The Triassic corresponds with the Germanic mcies type for Triassic sediments. It is characterized by massive or very thick bedded sandstones with some conglomerate beds at the base (Buntsandstone), followed by thick bedded limestones and dolomites (Lower Muschelkalk), intensely folded and deformed sandy clayey siltstone with gypsum (Middle Muschelkalk) and limestones and dolomites of the Upper Muschelkalk. The youngest formation in the Triassic (Keuper) is a sequence of shales, in t!,te lower part interbedded with Uinestones and dolomites. The Jurassic consists of a series of formations of limestones and dolomites, with broadly similar engineering characteristics. The Cretaceous is represented from the Albian upwards. The Albian consists of (cemented) sands and clays. The remaining Upper Cretaceous consists of limestones and dolomites, with broadly similar engineering characteristics. The Tertiary is mainly marly-arenitic, with an alternation of cemented conglomerates and (not or very weakly cemented) sand and clay layers. The upper part contains limestones and marls. The Quaternary is widespread, mainly as superficial gravelly and sandy slope deposits, fine grained sand and silt deposits on flat areas which are likely of aeolian origin (loess), and gravel in river beds and as terraces. Intrusive rocks Extensive bodies of igneous rocks occur intruded into the Carboniferous formations as granodiorite bodies and aplitic dykes. The intrusions are from Carboniferous through to Permian age and are probably associated with the Hercynian orogeny. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ~ """ GENERALIZED GEOLOGICAL TABLE & PESCRIPTION AND MAIN ENGINEERING CHARACTERISTICS OF THE LITHOLOGY IN THE FALSET AREA(11 i[ Tertiary(3) Cretaceous f ~ ft g-=:: l'< !a. ft :31 [ ~ Upper Cretaceous(4) c Off-white/l,grey, argillaceous to arenaceous, medium bedded to massive, medium to v.large blocky, jointed, slightly weathered, LIMESTONE ANI) DOLOMITE, strong. Alblan c 16 Muschelkalk J Tg3 Red/green/greenish blue/brown/yellow/off-white, argillaceous to fine arenaceous, thinly laminatad to v.thin bedded, v.small blocky, jointad, I folded and deformed, slightly to completely weathered, calcareous sandy silty SHALES, v. to mod. weak, with (small) quantities of gyp$um. Bottom: Interbedded with layers (20 - 100 cm) off-whlte/l.grey, argillaceous to fine arer,~aceous, v.thin bedded, v.small blocky, 1 jointed, LIMESTONE AND DOLOMITE, mod. weak to mod. strong. upper Tg23 Off-white/l.grey/yellowish grey, argillaceous to fine arenaceous, thick laminated to massive, v.small to v.large blocky, jointed, slightly weathered, LIMESTONE AND DOLOMITE, mod.strorlg to strong. middle Tg22 Buntaandstone. Tg21 TIJ1 Carboniferous(2l Hs, H Devonian(2) D --------- ------------- ---lntrusives in Carboniferous ottefl Red:(occasionally greenish grey), argiHaceous to fine arenaceous, thinly laminated to v.thin bedded, jointed, oftan folded and deformed, slightly to c!)mpletely weathered, gypslferous clayey sandy SILTSTONE, v. to mod. weak; large quantities of gypsum up to occasionally more than 80 %. lower PALAEOZOIC Red/ochre, SANDS AND CLAYS, at some locations Weakly cemented. Off-White/l.grey, argillaceous to arenaceous, medium bedded to massive, medium to v.large blocky, jointad, slightly weathered, LIMESTONE AND DOLOMITE, strong. Triassic f ~· T Keuper J !a. Miocene, Oligocene, Eocene MESOZOIC (3) 8' l ~ i g GRAVEL terraces along and in rivers; SAND/SILT/CLAY often on flat agricultural areas (also eolian origin). Brown/yellowish, cemented, CONGLOMERATE layel'l! (massive up to metres thickness) interbedded with brown/yellow, clayey SILT AND SAND layers, in top: LIMESTONE and calcareous sllty SANp layers. Jurassic(4) 8. t 0 Quaternary CENOZOIC late Carboniferous through Perm y1'j/ F02 Off-white/l.grey, arenaceous, medium to thick bedded, medium to large blocky, jointed, slightly weathered, LIMESTONE AND DOLOMITE ICALCARENITEI, strong. Red/brown, rudaceous (bottom) to fine arenaceous !top), v.thlck bedded to massive, slightly weathered, SANDSTONE, mod.strong. Thil)k sequences ( > 100 m) of d. grey, argillaceous, .thinly spaced cleavage, thinly bedded, small to medium blocky or tabular, jointed, folded, sligfttly to mod. weathered, SLATE, mod. strong to t~trong, interbedded with sequences (5- 100 m thick) of grey/brown, thin to thick bedded, medium to large blocky, jointed, folded, slightly weathered, MICRO CONGLOMERATES, SANOSTONES AND SILTSTONES, mod. strong to extr. stropg; folding 1 to > 10 m in slate and > 10 m in other. At two locations 10 to 50 m thick layer of black (with white 5- 10 mm bands), med(ium grained, massive, fresh, GNEISS, v. strong to extr. strong. Layers (6 cmJ of black, argillaceous, thinly laminated, schlstose, folded, slightly to mod.weathered, ORGANIC SHALE, v.weak, interbedded with layers 110 cm) of off-white/brown, amorphous,. v.small to small blocky, jointed, folded, RADIOLARIAN CHERT, v.strong; intensive multiple ~~~~-~~~~~-~~!~?_!~~--------------------------------------------------------------------------L. to d. grey, fine to coarse grained, small to medium blocky, jointed, slightly to highly weathered (also residual soil), GRANODIORITE ~~~!~~~~~~~~~~~-~~!~~~-~X!~~~~~~---------------------------------------------------------------- D. grey, v.fine to fine grained, v.small to small block.y, jointed, slightly to mod. weathered, APLITIC DYKES, mod. strong to v.strong (intrusive In carboniferous sediments and granodiorite). -------- Codes (Q, T, C16, etc.) refer to codes used on the geological map sheets, no. 444, 445, and 471, of the area prepared at a 1 :50 000 scale by the lnstituto Geologico y Minero de Espaiia. Only main codes are included. Notes: 1 Description for rock units according to BS 5930 (1981). (1. = light; d. = dark; v. = very; mod. "" moderately; extr. = extremely) 2 Palaeozoic sedimentary rocks intansively folded Ullder Hercynian orogeny (Carboniferous through Perm); folded on a scale of metres to 1O's of metres. 3 Mesozoic folded under Influence of the Alpine orogeny {late Cretaceous through Miocenel; folding on a scale of 100 m to km's, Tertiary faulted and tilted. 4 Jurassic and Upper Cretaceous consist out of va~lous formations with similar engineering characteristics. 5 Weathering indication characterizes the degree of weathering typically found in surface exposures. ~ ~ 9 0 ~ ~ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ orogeny orogeny several deg;re;a.ses. ,. , . . ,.,,.h...... depressions that were the sea in the southeast the marine shorter and of decreasing age, not younger than Eocene. age The and syn- Quaternary "'"'"'"~'V"-' that cover the area as a a thh::lmess of up of up to 6 m that are found on most of the natural. rrm~x11:ess in the area . .3 a 1.mit or a n"'"""'"'" of the research area can be found in Table 1 A sub-division into Httmstta1tlg:;::apl:nc ""'~'-"''""" is based on I, page 1 or A Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B .EYJSflNG ROCK !.1ASS CHAR.ACTER.lZAllON & CLASSIFICATION B EXISTING ROCK MASS CHARACTERIZATION & CLASSIFICATION 19 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B .EXISTING ROCK lYfliSS CHAR..4Cl'ERJZATJON & a.ASSIF!CATION 21 1 to combine the have (e.g. some extensions to B.2.4). The New Austrian Tunnelling Method (NATM) system includes and contractual not found tum1elling. to the but is to .-....c.,,,,,..T B.2.3.7 the Rock hnl:!;ID(eerJ cmrelatu)ns between the dtti:ere;ot """"""'"-'-'-'"" in the .._.~.._., •...,,!'; classr!lc~ttlcm :result. A summary of basis fur the of a new classification et al., 1974) is also discussed in eh. B.2.3.6. This many of the other s:ystems, and is re!ated to systems and is this are discussed as well as ""'·"'"'"'"''"u and the influence of these t.l:ie final me:tnCICl.S, Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 22 B. 2 Existmg systems B.2 EXISTING SYSTEMS The review of existing systems covers the characterization and classification systems, which are the main and (in the opinion of the author) most interesting systems with good published documentation. Most of these systems have been used in different geological and geotechnical environments for different projects. In many civil engineering or mining projects systems have been developed or existing systems have been modified. Often these have been modified to the particular needs of a project and might not be applicable to other projects or other geological or geotechnical environments. Sometimes parameters or factors of different systems are combined (Japan, 1992). This review only describes the main parameters and characteristics of the systems. All characterization and classification systems are accompanied by (euensive) tables for descriptions of parameters and, if appropriate, by tables with recommendations for civil or mining engineering applications. These tables have not been copied and the reader is referred for the details to the cited literature. B.2.1 Descriptive and characterization systems Two standard systems that characterize a rock mass and express rock mass characteristics in standard terms are those in BS 5930 (1981) and the ISRM Basic Geotechnical Description (ISRM, 198lb). A third, mainly used in the USA, is the Unified Rock Classification System (URCS) (Wtlliamson, 1980, 1984). The systems do not result in a numerical value or direct design recommendation. The systems facilitate communication on rock mass characteristics and are widely used for various purposes. Borehole core mul exposure logging The work by Deere et al. (1964, 1967) and Moye (1967), who published detailed instructions and recommendations for the description of. rock masses and the presentation of rock mass..data in the form of borehole core logs, has been adapted by the working party of the Geological Society Engineering Group in the report 'The logging of rock cores for engineering purposes' (Anon., 1970). British Stmulard BS 5930 The present version BS 5930 (1981) gives recommendations for a standard description of a rock mass. The characteristics are described according to a series of standard terms and phrases and lead to an extensive rock mass name. The geological units of the research area for this study are described according to BS 5930 (Thble 1, page 17). An interesting feature of the British Standard is the recognition of the importance of intact rock block size and form (Fig. 13). Rock blocks are described as very large blocky, very small columnar, etc.. Although not quantified, the descriptive terms relating to block form are very useful in engineering geology. ISRM Basic Geotechnical Description ISRM (1981b) recommends the following geotechnical rock mass parameters to be described or measured: 1) Rock lithology, with geological description Discontinuity spacing (bedding or layer thickness and joint/fracture spacing) 2) 3) Unconfined compressive strength (UCS) 4) The friction angle of the discontinuities The far more extensive ISRM 'Suggested methods for rock and discontinuity characterization, testing and monitoring' (1978b, 1981a) recommends the quantitative description of a very extensive and complete set of rock mass parameters for the characterization of a rock mass. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EX15.71NG ROCK 1\IASS CHAJ?.."CrERlZlil.WN & CLitSSIFlCATION :23 n.,.,,,,.,.,.hrm 1u:cord:it1g to British with n~tios tor block form ISRlVI 100 • 200 i v~ry > 15,000 MOII!j 1-=-::_:..:..::_-l-__str__o_n.::g---t~-:-:--t---:--:---f-_:1!1,;_000----1~,000 i 55 - 103 20. ao ~te a.ooo- s,ooo 21 -ss is ..u,.....~""" because a main user 'Thbie 2. Characterization of intact rock and URCS is the ConDepartment Agriculture. Apart from applications the author is not aware of any application for which :is of major importance. An method of describing intact rock strength is included in the URCS The intact strength in the is related to the deformation of intact rather than to the unconfined compressive strength intact as used in BS 5930 and ISR.l\1. A similar intact rock strength by Bumett was later used for the British with has been servation Service of Ol Soil (Wl This method of e.strtblishi11g intact rock C3.2.1 size intervals: 0.002, 2, 60 mm, etc. 5930, is included in the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 24 B.2 Existing systems B.2.2 Early classification systems Te~TJ~ghi - rock load cl~Daijktltien system K. Terzaghi (1946) classified rock masses with the objective of I I predicting the load on steel arch support sets in tunnelling. The ---r------t-~~. parameters taken into account are the 'rock condition', the I I dimensions of the tunnel, the depth below the terrain surmce and ' ' ' ' below the water table (Fxg. 14). The rock volume supposed to ! w ! ' ' be supported by the steel arch set is hatched in Fig. 14. ! Bl ! I I H The assumption that the steel arch set has to support a certain ' ' volume of rock above the tunnel, implies that the rock is allowed to deform until it can exert a mrce on the support. Terzaghi modelled deformation zones (a crack or shear zone) starting at the toes of the steel arch set in upward direction to allow the volume of rock above the tunnel to rest on the set. The load on the set is assumed to be the weight of the rock volume rock mass in-between the deformation zones up to a certain height above supported the tunnel ~ and the water load (W) (Fig. 14). steel set The 'rock condition' parameter describes the rock mass in steel set B .. various classes such as 'hard and intact', 'hard stratified or Fig. 14. Terzaghi - rock load classification (after K. schistose', etc. . Also classes fOr crushed and swelling rock are Terzaghi, 1946). distinguished. A table is provided which, based on the 'rock condition'> gives the rock load (Hp) parameter as a mctor of the width and height of.tho ~. The fable. -also ineludes estimates of the variation in pxessure on the support (e.g. the presence or absence of side-pressure on the steel arch sets)(!!). - 1 I I I I .. I I Laufjer - sttmd-up time cl~Dsifica,tion Lauffur (1958) related the stand-up time of an un-supported. span to standard rock mass types. Compared to the Terzaghi approach this was a major improvement as disconti.nuities (structural defects) were considered. The characterization of the rock mass was, however, not done by describing different rock mass parameters but had to be selected from a number of characterizations of standard type rock masses prescribed by Lauffer. Later the Lauffer system became the basis mr the New Austrian 1\mnelling Method (eh. B.2.3.6). Deere - RQD i1Ulex cl~Dsification Deere et al. (1967, 1988, 1989) introduced the Rock Quality Designation (RQD). The RQD index is measured on borehole cores, full~, e<!,· [1). RQD = I; lenp& pi«es of mtact core with length·> 10 cm * 100 % ,totQl.ZatP tb-iliMt. . . [1] The intact pieces of core (highly weathered pieces of rock or infill material should not be included) should be measured along the centre line of the core and the RQD values should be calculated separately for each lithostrati.graphic unit. Core runs should preferably be not longer than 1 .or 1.5 m. The RQD values provide a measure of the brokenness of the rock mass. Deere et al. (1%7) related the RQD index to support types for tunnels. It is therefore the first classification system incorporating an index for the amount and quality of disconti.nuities in a rock mass. Recently 'rock quality charts (RQC)' have been based on RQD measurements by ~en et al. (1991, 1992). (Ill Severe doubt has been expressed about the concept of a deformation zone S1arting at the toe of the steel support and developing in upward direction. The development of deformation zones as indicated is only likely in a massive, not jointed (thus continuous), rock mass. In a discontinuous rock mass the deformations will follow existing discontinuities and may well lead to a totally different volume of rock to be supported. Secondly. the deformation zones. will develop in upward direction only under low horizontal stress. With a higher horizontal stress the normal stress on the proposed deformation zones will be too high to allow shearing or tension cracking, thus preventing the development of deformation zones, whereas if the horizontal stresses are considerably larger than the vertical stresses the deformation zones may well develop horizontally rather than vertically. The assumption that the water load has to be supported by the steel set over the full height up to the water table is also unlikely as this would only be the case for a tunnel with impermeable lining capped by a fully permeable waterlogged rock mass. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ll EXISTING ROCK MASS CHARACI'ERI7A'!ON & CLASSIFlC4TION Structure The 25 18 B.2.3.1 The Rock Mass IRS + RQD + (!{fJCf.Cing + condition + groundwater) + reduction = RMR = Rc:ck MIZS'S Rating Intact Rock Strength RQD = Rock Quality De~r£1f11iiltilm spaciwtg .. discontinuity one set (see cm'ldition = expression con.dmon strengtli) of cme set groundwater = for groundwater i;!lftcw (pressure) reduction foetor = f.m orientation of engineering structure relative to main diswn:M1dty set n::U"l'lrt!P.tl~:r haS been to the span and stand- In the extended and has been more up spe~rl:ited The spa,;mg Vlrith the most ad'v'C:rse influence on the RMR and :results in five difJterent 0 ''"""""''"'. B.2.3.2 The 11-,•;,vs1·em of Barron et al. The expresses the with eq. [4]. The first term RQD to the size t."'le intact rock J. term Jw (joint v.rater pru:an::tete:r) environment for the around the tunnel planes and the OJ}(~ll.Utg Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 26 R 2 Existing systems Q set 'fll4mber alteration number can range tJe1tw<~en rock mass. The ......,.......a..,,"', Q range on a ! .....,,..,.;tJ....,.,,;,.. £.2.3.3 page MRMR .,.,...,,,,..... ""'~""'" of Bieniaw:ski. In his system the "'""''""'"''J' pa:ran:tehl~ts methe same as the Bieniawski corKiltmn pruranteJX:r. The number of for the extensive thm in the RMR ,,.,.,,..,.,,•. "'~,..,.,.,..."'""'""'"' RMR "' IRS + RQD + R.MR "' Rock Ma!s ~·~ ., expressir.m. for spacing discontinuitie.s condition = caNiition of discMtinuities (partmUeter also aeJ-Jent:tem tlrGftmi:twll!ter presence or ~ty of grour.dwater i~w loor~Jn!l!t.e·rs RQD and spacing can be repln.ced by the fracture fre~fflel:cvl applied a bolts whatever the MRlv!R value reinforcement for a rock mass with a lOIN RMR the MRMR is not much the R..\1R Laubscher graph for the determine the maxim.um of three "'-''""''·w.v·u is determined The Sltl.llatl1)nS with "'"'""•TI"'F't"'r eh. B.3.4. pnJp()Sed. e:ICcavarion or vice versa. Also methods and the mn:ue1r1ce Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EX1Sl7NG ROCK M4SS CBARACTEPJ7AION & CLASSIFICATION 27 rock masses B.2.3.5 principal elfoctiw stress at o1 a~ [6] miner principal m-ess at oc "' intact rock strength mb and a are parameters describing tlMJ rock maJtS stnu:mre and "' for estimation B.2.3.6 contract aspects and the construction a tunnel. Various circumstances, been developed worldwide, noticeably in Japan (Japan, for tunnelling and a total description of the system is beyond the scope this R2.3.7 the interaction of as external influences on geomo:rphological processes, etc. . The which the """'"'"""'""'n rmlr>ln""t••r Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ stability of a tunnel, is de1ermined. Quantification of the interactions or influences between parameters and between parameters and engineering structure can have any form. These can be, for example, difterential equations, binary operations (0 or 1, for example, for features that are either present or not present), classifications or numerical calculations. How these relations are established (e.g. by engineering judgement or actually proved by testing) is of no importance. The reliability and accuracy of the fin.a1 result depend, however, on the reliability and accuracy of the relations (and obviously of the input data). The methodology resembles the working of a neural network as also pointed out by Hudson, however, the relations between in- and output parameters in a neural network are normally of a simpler form. The methodology is not a classification system, but rather a methodology of thinking for engineering in or on discontinuous rock masses. Hudson gives no detailed applications nor relations between parameters, however, suggestions are given for implementation of the methodology in various forms of engineering in or on discontinuous rock masses. B.2.4 Rock mass classification systems for surface engineering applications Some rock mass classification systems developed for underground excavations have been used for surmce engineering structures such as slopes directly (Bieniawski, 1976, 1989, Barton et al., 1974) or in a modified form (Haines et al., 1991, Robertson, 1988, Romana, 1985, 1991, Selby, 1980, 1982). The system developed by Shuk (1994) is specially designed for slope stability. Also systems have been designed specially for excavation, rippability, etc.. Barton et al. ( 1974) included in his system an estimate of the friction angle for the shear strength of discontinuities. This friction angle can be used in, for example, slope stability calculations. B.2.4.2 Bieniawski' s RMR applied to slope stability Bieniawski (1976, 1989) included not only recommendations for underground excavations but also for foundations and slope stability. The author is not aware whether the system has actually been used for slope stability analyses in the form as presented by Bieniawski. B.2.4.3 Vecchia - Terrain index for stability of hillsides and scarps Vecchia (1978) designed· a classification system to quantify the stability of a hillside or scarp, e.g. natural slopes, based on parameters for 'lithology' and 'attitude', and a 'friction' parameter which is depending on the 'lithology' and 'attitude' parameters. The 'lithology' parameter is determined by the presence of clay and shale in the rock mass and by characteristics of the rock mass such as loose, coherent or massive rock masses. This, combined with interbedded lithologies, results in a series of different standard classes for the lithology, e.g. from shale with a fe\.v coherent beds (rating 10 points) to massive rocks with fe\.v or no discontinuities (rating 90 points). The rock mass in the field is visually compared to the standard classes provided by Vecchia (1978), classified and rated. The 'attitude' parameter assigns a rating ranging from 0 (unmvourable) to 12 (mvoura.ble) to the orientation of discontinuities with respect to the orientation of slope or scarp. The 'friction' parameter is a rating for the friction along the main discontinuity (set) allowing sliding. The 'friction' parameter with a rating between 2 and 10, is assigned on the bases of the classes determined for the 'lithology' and 'attitude' parameters. The 'friction' parameter is thus not a separate parameter established in the field. A terrain index (IT) is calculated as follows: IT = terrain index = lithology + attitutk -. jrld:ion, f7l The simplicity of the system and the limited number of parameters, eftectively only two, which have to be assessed in the field, are very attractive. This simplicity, however, may also be its largest drawback. The quantity of standard lithologies given is limited, will not always fit a rock mass in the field and the visual comparison may be ambiguous. The definition of standard lithologies resembles the approach of standard rock mass classes as used by Lauffer (1958, eh. B.2.2) for underground excavations. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EXISI'lNG ROCK MASS CHA.R/Cl'ERlZAl'lON & CLASSIFICATION 29 Other drawbacks are that there are no provisions for more than one discontinuity set and the limited options for the friction along the discontinuities. An interesting observation (Vecchla, 1978) is made that water in surfilce hillsides or scarps is generally limited to surmce water. Water pressures in the rock mass are therefore not considered. B.2.4.4 Selby - Geomorphic rock mass strength classification Selby ( 1980, 1982) designed the Geomorphic Rock Mass Strength classification. The classification is designed with emphasis on geomorphology rather than engineering. The system resembles the Bieniawski system (eh. B.2.3.1) and includes for a large part the same parameters. Parameters assessed and rated are: intact rock strength (which can also be assessed by Schmidt hammer, eh. C.3.2.1.1), degree of vveathering, spacing of joints, joint orientations, widths (aperture) of joints, continuity (persistence) of joints combined with joint infill, and outflow of water (ratings are given in Thble 4, page 35). The ratings obtained for each parameter are added and the total rating is an expression for the rock mass strength. The rock mass strength is divided in :five classes ranging from very strong to very weak. The total rating is not directly related to slope stability but is used in the qualification and quantification of geomorphologic processes. B.2.4.5 Robertson' RMR (modified Bieniawski) Robertson ( 1988) modified the Bieniawski (RMR) system for use in slope stability analyses. The main distinction with the original system is that for RMR > 40 the stability of the slope is fully governed by the discontinuities whereas for an RMR < 40 the slope stability can be assessed by a modified Bieniawski system. In Thble 4 (page 35) the parameters are listed that are used for determining the slope stability for an RMR < 40. B.2.4.6 Romana' s SMR (modified Bieniawski) Romana (1985, 1991) extended the RMR classification system to slope stability problems expressed in the slope mass rating (SMR). SMR RMR F1 F2 Fs F4 = Slope MtiSS Rllting = Rock MtiSS Rating ~ liS Bieniawsld 1s RMR) = foetor for ptl1'li!Jelism of tM strilr.es of disc~ tl1ld slope face "' j'rJdor for disccmtlmdty. tlip. angle =foetor for relation between slope face tl1ld disco1ltilluity dip = foetor for llldbod. oj.ext:IZWition [8] The parameters F 1, F 2 and F 3 are fur one discontinuity only and therefore the SMR should be calculated for each discontinuity set and the lowest resulting SMR value gives an indication for the stability of the slope. The SMR value predicts the possibility of a 'soil-type' :fiillure (normally for low values) and the amount of plane and wedge :fiillures (normally for higher SMR values). The SMR value is also used to indicate the support measures to be taken for (partially) unstable slopes. B.2.4.7 Haines (modified Laubscher) The Laubscher (eh. B.2.3.3) system is used to forecast rock slope stability in open pits in South Africa (Haines et al., 1991). The adjustment ratings incorporated in the Laubscher system are reported to be of great benefit for slope stability estimation. The design chart to determine the slope dip related to slope height and fuctor of safety using the MRMR of the Laubscher classification is shown in Fig. 15. Haines et al. point out that the system is designed in a mining environment where safety requirements are generally lower than in civil engineering. However, they also incorporated slope dips for slopes with a factor of safety equal to 1.5. These might be suitable for civil engineering. The system has been designed empirically based on existing slopes in open pit mines and analytical calculations. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 30 B. 2 Existing systems The intact rock strength value necess280 ary in the Laubscher system can be replaced by an estimate with Schmidt hammer values for soil and 'softer' 240 rocks and by the density of the. material for 'harder' rocks (Haines et 200 al., 1991). The orientation of the slope with respect to the discontinuity 'E orientations is incorporated in an 2.!ill:180 adjustment percentage. ! - 1120 eo 40 0 0 10 40 50 60 70 MRMR 100 Fig. 15. Design chart to determine slope dip and height using MRMR classification data (after Haines et al., 1991). B.2.4.8 Shuk - Natural slope methodology (NSM) Designing the inclination of a new slope based on slope dips measured on existing natural and artificial slopes is often used in the design of new slopes to be excavated. Normally no formal characterization or classification of the rock mass is applied. The Natural Slope Methodology (NSM) (Shuk, 1994a, 1994b, 1994c, 1994d) is based on this principle. This method uses a statistical analysis of existing natural slopes to predict rock mass and soil parameters, and the probability of slope stability. The method is based on a presumed relation (eq. [9]) between the height and length of a natural slope. [9] P. = non-diltJmsiontlt ~ ~ (rdt#Ml. to tecttmic.s. lW.ItCl' F~ -..) ,,. c = resitlual friction anale, resitlud. cohesion of rock mtZS.V or soil y = unit 'Weight of rock 1IUJS8 or soil a tmd b .. weighting .(tu:tors Equation [9] is only one of the possible relations. Other more complicated relations have not been investigated in depth by Shuk at present. Back analyses of a large number of natural slopes and optimization of eq. [9] result in estimates for different rock (mass) or soil parameters. The method can also be combined with anisotropic behaviour of rock masses and soils. The methodology is very attractive as it does not require extensive field investigations. A problem with the methodology as reported, is that not all relations, parameters and especially the methods used to opti.mize the non-linear relations on the data are clear from the articles published. It is thus impossible to perceive the methodology, or comment on it in detail at present<12>. It is understood that the methodology has been still :further developed and future versions and publications may show the full potential. (Ill Therefore this system has not been included in 'Thble 4. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EXJSJ.'JNG ROCK MASS CHARK:1'ElUZAJ.10N & CLASSIFICATION B.2.4.9 31 Hudson's RES- rock mass characterization applied to assess natural slope instability Mazzoccola et al. ( 1996) presented an example for determining natural slope instability following the Rock Engineering Systems (RES) methodology (eh. B.2.3. 7, Hudson, 1992). The rock mass characterization evaluates the interactions between and the inftuence of all parameters that may be of inftuence on slope stability. Twenty parameters are evaluated ranging from parameters as the geology, folding, etc. to parameters describing the rock mass such as weathering, the number of discontinuity sets, slope orientation, etc.. Also external infiuences are included such as climatological influences, as rainfall, freeze and thaw, etc. . The instability of the slopes is determined following the Rock Engineering Systems (RES) methodology. The publication shows that a good correlation is obtained with a predictability rating for slope instability based on indicators of potential instability of the natural slopes (NatbanaiJ et al., 1992). B.2.4.10 Excavatability, rippability and blasting assessment Various classifications have been developed to assess the excavatability and rippability of rock masses at terrain surfilce (Franklin. et al. 1971, Weaver, 1975, Kirsten, 1982). Franklin. et al. based the excavatability on strength (unconfined compressive or point load strength) and discontinuity spacing in accordance with the Fmnklin size strength classification (B.2.3.4). Weaver based his rippability assessment on the Bieniawski classification for underground excavations (B.2.3.1) while the approach of Kirsten is based on the Barton classification (B.2.3.2). Most exca.vatability or rippability assessment systems are equipment specific, e.g. give recommendations for a particular type of excavation or ripping equipment. Some systems also include seismic velocities to assess rippability ~r, 1975). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 3 CALCULATION METHODS 1 In a newly e. g. a rock mass with a low intact roc:k has often also a small discoliltilmities or both. A correlation different paJran:le~erl_s) to -v-d.!ues which the if the user knows from exJJeriertce 2 or <m It should be noted !hat the correlate lines in and me scatter allows for one to two classes difference between the two systems This may be due to ilie definition of the classes. A more correct ''"'"""'~'"'"rm between the two systems should be base.d on the recomrnended for excavations. The recon:unended types of support are, different for the tvm systems and a made. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EXJSI'lNG ROCK MASS CHARACrEIUZifl'lON & CLtiSSIFlCATION 33 taking lower values for the individual parameters of the system he uses (see also eh. B.4). Because of this, systems should be designed to be operator~independent. RMR .. 91n Q + 44 .. . ---- ·:·------- ·:·------ r--.---t--7-------+--l 80 : :• • 0.01 0.1 1 10 100 1000 Bartoo (Q..value) Fig. 16. Correlation between Bieniawski (RMR) and Bart.on (Q). Data from case histories with RMR and Q-system (after Bieniawski, 1989). (Continuous lines indicate correlating classes of rock mass quality.) B.3.3 Influence of parameters in existing classification systems An inventory .of .the most. important .rock mass parameters of intet:est fur. engineering structures in or on. a rock mass is presented in 'Illble 3. This table is based on the experience and intuition of the author and on the literature. The parameters listed are, in part, those occurring in some of the existing chamcterization and classification systems previously diseussed (eh. B.2). Many systems do, h<>Wever, nofcontain. one or more of the parameters from 'Illble 3 and also the influence of parameters in the existing classification systems is not for all classification systems the same. 'Illble 4 presents the various parameters used in the existing rock mass classification systems and gives a crude indication of the maximum influence of each parameter on the final rating or recommendations for tunnel support or slope geometry. It is impossible for all systems to indicate the influence per parameter exactly because in some systems parameters are not independent or parameters are not linear. The percentages indicate the reduction of the final rating when that parameter is given its minimum value and all other parameters have their maximum value, compared to the rating based on the maximum value of all parameters. If a parameter is linked to another parameter then the other parameter is also changed as required< 14>. Noteworthy differences in the influence of parameters (Table 4) are: The absence of the intact rock strength (except for a low intact rock strength/environment stress ratio), in the Barton system. The absence of discontinuity spacing in the Barton system. <14> Thke for example, the link between Jr and J. in the Barton system; the lowest value for J. is 20 but this cannot be combined with the maximum value (5) for Jr but only with Jr = l. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 34 intact rock'"'"'"!:'" Ui '"" lltHIU !---- \with respect to engineering structure) amount of sets rock block size and form spacing per set ~-'"'""'""'"'"'"' per set Discontinuities material friction Rock mass shear strength along discontinuity !condition of dis· continuity) surface characteristics of discontinuity wall roughness !diiatancy) IHI<>!l\,jll deformation infi!! material Susceptibility to weathering Deformation parameters of intact rock/roe!< mass Engineering Gvv •v , y of engineering structure !size and orientation of a tunnel, height and orientation structure of a slope, ate,) External Water pressure/flow, snow and ice, stress relief, external stress, etc. influences Type of 11able 3. Rock mass parame:lers of interest for (lSJ A reduced HTI'DOI:iai:ce of v.-arer pressures in enJ~In:e:erm.!! structures in or on rock. assessments is also fbund in this research D. l. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ l MAXIMUM NEGATIVE !NFLUENCE Of PARAMETERS (in ,m•v~""'"c from final maximum rating)ll H2l Discontinuities ,.. ' "'"''"'"'""""" systam(2) j range Intact Rock ROD persist- of sets aperture encl.! excaveit~on pressure or load roughness (scale) large I water lOCk ing m«l'lhod of excavation orienta· dimension tkm smal! I I I EARLY SYSTEMS {for .Jn""-'"'_"u"u excawtions) Deere IROD) 0. 100 • [ Wickhiim iRSRl 19 . 120 l 100 35 24 !general area geology '""""""'" 7 RECENT SYSTEMS \for unde;g,vv•ld 0- 100 Bieniawskl ' 0.00008 Barton(3! . iOl 2666 0. 120 laubscher 15 I 20 I a i I 20 I 61 e {reductions are with rock load 90 tar(3) oxtr. good 17 13(5) !no change of class) Js! notE!~()tJ!lh 99 good extr. good very 90od 5 21(5)(61 I sj for a change of class) 90 97 9 15 -T 5 (no change in dassl Q(JOd 11 7 11 "~·H•c"""" 70 poor T I 15 97 95 very axtr. good good 40 3!7) good 100(4} 12 1 I '100(4) 000{4) 20 i ch~~~ei 'changei SLOPE SYSTEMS ISelby IRMRI i 0. 100 20 0- 100 15 20 7 30 7 e 20 6 i ;~~~)~~"~) 0. 100 i Romarm iSMR) 0- 115 0' 100 iL._.. 6 6 10{9) 6 15 6 20 6 13 17 5 11 13(5) 21{5)(6) 17 ~ 20 ~5;J 60 88 0- 100 Vecch!a tm •(8) 12 6 6 6 6 5 5 6 6 15 5 5 I 9 G:l (1001 (10) '13 70 40 5:2 3!7) {note 11) 13 ~~0 ~ ~ ~ ~ V:i Notes: 1 2 3 4 5 6 7 e 9 10 1 influence percentages are only an approximate indication, Some systems are combinations of addi!'lg/subtracting, mtdtiplier/!livider, and/or logarithmic pammaters, not independent and/or non-linear parameters (see ta)(t). lr\fluence percentage ~ (maximum final mting • rating with t!1a ram meter minimum am! all othar parnmetero maximum) 1 maximum final r-~ting x 100 % . For the recent classification sy~tems ai!i'.o the class is indicated that results if the particular pammatar has its minimum value. Thi!lal!ows comp~rison of classes betw<Hm the logarithmic scale of the a-system and the linear scales of the Blaniawsld and Laubscher systems. Terzaghi, lauffer and NATM systems ;,m~ ,not included !IS they do not use a rating for different parameters. Intersections and portals are not considered. Intact rock strength is only of lnf!uence if low comp,.red to stress environment. Graphical {apptoximately logarithmic) rehltions betwot.'fl roof span or hydraulic radius, final rating and stand-up time. Laubscher's system. Parameters for RQI} and d!scnntinulty spacing can be replaced by discontinuity frequency. Amount of discontinuity sets, spacing aM persismnca combined in logarithmic relation iFig. 33 & eq. !13l). Water influence combined with discontinuity ratings, !nfH! combined with persistence. Selby rates present degree of weathering (thus nut future wea!heringl for the whole rock mass fn!lowing as 5930 10981}. flobsr>..son: If RMR < 40 points slope stability govamed by the RMR mting; if RMR > 40 points the stability is ful!y governed by the or!entaticm and stwngtil of thl'! discontinuities. Haines: Final result fmm graph roiating slope heigh1, dip, safety factor and lMRMRl rating. Adjustment parameter for slope orientation in relation with orientatkJr• ut dis•::mtinuities with max!mum of 100\'t•. n ~ ~ ~ ~ ~ ~ Q ~ ~ ~ Q IS""~ ''~"" Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B.3.4 Problems with parameters in existing rock mass classification systems In the previous chapter it is shown that not all systems use the same parameters, that not all systems include all parameters thought to be important for geotechnical purposes and that the infiuence of a parameter on the final classification result is not the same for all systems. Apart from these diffurences the implementation of some parameters can also be questioned. A further discussion of the parameters thought to be important for a classification system for geotechnical engineering is therefore necessary. B.3.4.1 Intact rock strength Intact rock strength is, in most classification systems, defined as the strength of the rock material between the discontinuities. Strength values used are often from laboratory unconfined compressive strength (UCS) tests. Problems caused by the definition of intact rock strength and using strength values based on UCS laboratory tests are: 1 The UCS includes discontinuity strength for rock masses with a small discontinuity spacing. The UCS test sample is most often about 10 cm long and if the discontinuity spacing is less than 10 cm the core may include discontinuities<16>. 2 Samples tested in the laboratory tend to be of better quality than the average rock because poor rock is often disregarded when drill cores or samples break (Laubscher, 1990), and cannot be tested. 3 The intact rock strength measured depends on the sample orientation if the intact rock exhibits anisotropy. 4 UCS is not a valid parameter because, in reality, most rock will be stressed under circumstances resembling conditions of triaxial tests rather than UCS test conditions. Some classification systems (Fra.nklin et al., eh. B.2.3.4) use the Point Load Test solely or as alternative for UCS or hammer tests as the intact rock strength index test. The same problems applying to using the UCS test also apply to the PLS test. The inclusion of discontinuities in the rock will cause a PLS value tested parallel to this discontinuity to be considerably lc:J\1Ver than if tested perpendicular. This efi.ect is stronger for the PLS test than for a UCS test, as the PLS test is basically a splitting test. The size-strength system of Fra.nklin et al. (eh. B.2.3.4), the Unified Rock mass Classification System (URCS, eh. B.2.1), the slope stability system of Haines et al. (eh. B.2.4. 7), the geomorphic rock mass strength classification of Selby (eh. B.2.4.4), and the modified Hoek-Brown failure criterion (Hoek et al., 1992, eh. B.2.3.5) allow for an estimate or 'engineering guess' of intact rock strength using 'simple means' (geological hammer, Schmidt hammer, scratching, breaking by hand, etc.). Although Laubscher (eh. B.2.3.3) also recognises the problems inherent to testing of intact rock strength he actually does not explicitly allow for an 'engineering guess' with 'simple means'. The disadvantage of using a Schmidt hammer for estimation of intact rock strength is the influence of diseontinuitiesbebind the tested smmce. Schmidt hammer values may be inftuenced by a huge and un-qoantifiable loss of rebound if a discontinuity is present inside the rock behind the tested surface (eh. C.3.3.3). B.3.4.2 Rock Quality Designation (RQD) Rock quality designation (RQD)<17) is defined as eq. [10] (Deere et al., 1967). RQD = L kngth pieces of intact core with kngth > 10 total length drilled cm * 100 % [10] The RQD is measured on the borebole core. Normally the RQD is determined for every metre length of borehole core per lithostra.tigraphic unit. The length of unbroken pieces of sound core that are of more than 10 cm (4 (16) With discontinuities are denoted mechanical discontinuities, see glossary, page 241. (I?J RQD is used as an indicator fur rock mass quality directly (eh. B.2.2), but also it is a parameter that is included in many classification systems together with other rock mass parameters. The discussion in this chapter considers the RQD only as a parameter in a rock mass classification system and not as an indicator fur rock mass quality itself. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EX1811NG ROCK MASS CHAR.ICl'ERU.AION &: CLASSJFlCATION 37 inches) length along the centre line of the core (ISRM, 1978b, 1981a), are added and the ratio, as percentage, to the length drilled is the RQD. Recommended is a drilled length of 1 or 1.5 m. In principle the RQD is a very simple test and used worldwide. However, the definition of the RQD and the day-to-day practice of determining the RQD introduces several severe disadvantages that cause the RQD often to be inaccurate or to result in totally misleading values. Many authors have commented on the disadvantages of RQD measurements (R. D. Terzaghi, 1965). Some major problems with RQD measurements are: 1 2 3 The value of 10 cm (4 inches) unbroken rock is arbitrary. The value of 10 cm for unbrobn pieces of rock core is an abrupt boundary. A rock mass with a discontinuity spacing of 9 cm perpendicular to the borehole axis will result in an RQD value of 0 % while a discontinuity spacing of 11 cm will result in an RQD of 100 %. Although a (small) quality dif~rence might result from the difterence in spacings, this is certainly not such a large difimmce that it should result in a difterence between minimum and maximum of the quality assignment Obviously in a real rock mass the spacings between discootinuities are not all the same and therei>re the 10 cm boundary eftect is more or less abrupt depending on the distribution of the spacings. The RQD is biased through orientation with respect to ~ dltleontlnullle 0.09 m discontinuity orientation (Fig. 17 - compare vertical \~ ~ :---.. boreb.ole to horizontal borehole A). If a discontinuity is ~~or~zon~~~~ \ ...---in the borehole core parallel to the borehole (borehole ~~ L ~..r~~~~~~~~;~~~~~~=£._______::;: 4 5 6 B) then ISRM (1978b, 1981a) recommends measuring horlzonlal the length of the core oflSet from the centre line if ~·\, sound pieces of > 10 cm length are prese111: in that / stretch of the core. Depending on the infill thickness of the discontinuity, this might solve the problem of Fig. 17. Bias of RQD due to orientation of borehole. boreb.ole B (RQD = 0 %) in Fig. 17. Weak rock pieces (weathered pieces of rock or infill material) that are not sound should not be considered for determin:ing the RQD (Deere et al., 1967, 1988). 1b exclude infill material will usually not be too difficult; however, excluding pieces of weathered, not sound rock is fuirly arbitrary. The RQD value is influenced by drilling equipment, drilling operators and core handling. Especially RQD values of weak rocks can be considerably reduced due to inexperienced operators or poor drilling equipment. The equipment and especially the core barrels used for geotechnical rock drilling are not standard. It is obvious that the number of breaks caused by the drilling process will be strongly dependent on whether single-, double- or triple-tube core barrels are used. ISRM recommends measuring RQD on cores drilled with. a double-tube core barrel only. The borehole is, however, normally not only made to determine the R.Qii ~Often core are use& Weaker rockor"fracture(( roclimasses to ootain a decent··· core for test samples. The RQD measured on this core is overrated but the amount of overrating is not mown. Alternatively two boreholes should be drilled; one for the RQD with. a double~tube core barrel and one for the samples with. a triple-tube core barrel. The author does not know of any site where this has been the case. On the contrary the author has noticed many sites were the RQD was determined and compared from borehole to borehole irrespective of the core barrels used. The diameter of the borehole core is not standard in geotechnical drilling. A core diameter of not less than 70 mm (H size) is recommended for geotechnical drilling. In massive rocks, however, a reduction is allowed to 55 mm (N size) and in very weak or fractured rock the diameter should be increased between 100 and 150 mm (BS 5930, 1981). The author has noticed tb.a.t in practice very often N or NQ sized boreholes (approximately 47 to 55 mm core diameter) are used independent of the quality of the rock. Bieniawski (1989) allows borehole diameters from BQ to PQ (36.5 to 85 mm) for RQD determination. A larger diameter will result in: 1) fewer breaks during drilling and core handling after drilling, 2) a larger chance tb.a.t a parallel discontinuity is intersected and 3) a larger chance. that pieces of sound rock will be present in the core if a (near-) parallel discontinuity is intersected. In general, smaller core diameters lead to lower values for the RQD and larger diameters to higher values for the RQD. Pieces of rock that are clearly broken through drilling or transport are supposed to be fitted together and the length. should be measured as unbroken (ISRM, 1978b, 1981a). If this is done properly it partly solves the problems mentioned in points 5, 6 and 7, however it is not always easy to distinguish between natural discontinuities and breaks from drilling or core handling. In particular in a fresh rock mass this distinction tripfe::tu.t>e 7 8 barrelS fOr Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ master is often rumost tm:.possi,tne aud a less "'"'.,"'.,.,'"'"''"'0 elll;meer or errors. xrn;snectrve of the lli'lcerta.in. This ""'·"'".,..,,,.,, where the weaker ntt!tOIC)!!:V is not 9 contains of miJtUeJrtce RQD values determined without a bo:renote been nrcmosed IF J, :<: 4.5 IF < 4.5 "'"'"> "'"''',::> RQD "" (1 15 - 3.3 RQD = 100 % discontiw.uities per * J,) % discow.timdties per metre dis1::ominui'ty sets) A more sophisticated approach is a three-dunensionru model to calculate the because 1) the relations etaL, 1991, etru., d.i.s.col:tfuntitjes than a in the same rock mass 3) weak rock pieces (highly weathered :rock ae1ten:nu1atmn of cannot be in these theoretical exciudted, whereas the RQD measured in be caused ori.cmtatlton of the measurement A ooJ:·enme As classification the onentatJ.<m this is not qu<mtille:d B.3.4.3 In many clacSslilciatum "1'"""''.u" of The in such rock masses, ..,...."""""''""•'""'. . . '"'"""'""' do not describe what be done if '"'""'<.A.Uf""' if more R3.4.4 Non-rler~asl!ent UISCOlJ.trrlUI1cy sets do not have the same m11rueJt1ce on the page 241, and eh. C.3.3. How to deal with~-'"''""''-"'"'"" and the ge()ffi.!JfiJ>mc Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EX!.Sl'lNG ROCK MA.'\S CHARACTliJU2'AJON & CJ~4SS1FlCI(l'l0N 39 or exposure, or l or 2 are not co.nmae.rea diSOOfl'linull:y m with good l}l)!iditiort dllloor.tloolty H! wllh v&r;< poor oomll!ion clear which set has the worst influence on of the tunneL B.3.4.6 .._,._.,,...u,,..,"., of a discontimrity can be e. g. ripple an:tsotromc d:~~;co:ntiltlmcy roughness will also be anisotropic. Thus :rmxgiltne:ss used in a M.i:t!!J!HL)I Of 0. HJI.i.IHJIHc!i:s should be assessed in relation should be the rOl!gb:ne:ss can occur. can obviously also not include arutsot:ronk ."'"'1..,,.,..._.,.,. B.3.4.7 Di:scouti:rmity karst features The opt>.n holes coJmmJeraoJlV weaken the from solution Solution leaves oo:em:d disc~~mtinuitie:s. The shear a diminished of contact may break due to The presence d"'~•~<J-.r Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ of karst holes during excavation has also an adverse effect on the slope stability. During blasting the blasting gasses will force their way out of the rock mass via the karstic discontinuities mtb.er than by breaking intact rock or by following discontinuities in the direction of the next borehole. None of the existing systems incorporates a parameter that allows for an inft.uence of karst features. 13.3.4.8 Susceptibility to weathering Susceptibility to weathering is only considered, to a certain extent, in the classification system by Laubscher (1990) and in the modifications of this classification system. Susceptibility to weathering is an important fuctor in slope stability. Within the life span of a civil engineering structure future weathering of discontinuities and rock material may well lead to instability. 13.3.4.9 Deformation of intact rock and rock mass, stress relief Deformation of intact rock is not considered in any of the existing systems, however, it is used for an indirect estimation of the intact rock strength by impact methods (eh. B.2.1). Deformation of intact rock is likely not important for engineering structures which cause low stresses on the rock, e.g. slopes of relatively small heights. Deformation of a rock mass is considered in the Q-system (e.g. Barton et al., 1974, 1976a, 1988, eh. 13.2.3.2) in relation to stress relief due to weak or sheared zones in the rock mass. Deformation of a rock mass in relation to stress relief, not particularly related to weak or sheared zones, may, however, be of importance for slopes. Stress relief and related deformation may cause movements along discontinuities, increase of slope dips, etc. , which inftuencethe stability of a slope. . Aproblemwitluieformation4.a.~ mass.andwith stress· relief is that these cannot be tested, otherwise than with costly tests. B.3.4.10 Relative orientation of slope and discontinuities The orientation of discontinuities in relation with the orientation of the slope has a marked and often decisive efrect on the stability of a slope (sliding, toppling failure, etc.) but not all classification systems used for slope stability assessment incorporate a parameter that allows for this inft.uence (fOr example, Robertson, 1988 for an RMR of less than 40). In the other systems the parameter is fairly crude or not fully decisive or both. For example 13ieniawski allows for a reduction of the final RMR rating by 60 % if the slope is Ullfavourably oriented, and Romana allows a reduction of 52 % ('Th.ble 4). In some systems (m example, Bien.iawski and Romana) only the major discontinuity set or the discontinuity set with the most adverse intluence on the slope stability has an infiuence on the final ratings, with respect to orientation of discontinuities and slope. This results in the same problem as outlined above for the condition of the discontinuity (eh. B.3.4.5). B.3.4.11 Slope height The height of the slope has a direct influence on the stress levels in the rock mass of the slope. High stress levels, comparatively to the intact rock strength, may cause failure of the slope due to intact rock :failure (Gama, 1989). A high slope may also present more opportunities for discontinuity related :failure as the quantity of discontinuities intersected by the slope is huger. Hence, although slope height is likely to be of importance in a slope stability system, none of the existing rock mass sudilce classification systems for slopes incorporates the slope height, except Haines (eh. B.2.4.7) and Shuk (eh. B.2.4.8). B.3.4.12 Water The presence, or the pressure of water in discontinuities, is a parameter incorporated in most systems. Water pressures and water flow in discontinuities may exercise pressures on rock blocks. The shear strength along discontinuities is Ullfavourably influenced because water pressure reduces the normal pressure on the discontinuity and therefore reduces the shear strength, while the presence of water gives a lubricating etrect and may lower the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EXlSllNO ROCK Mr1SS CHARACTERIZ4TlON & Cl.A'\SIF!C4TION 41 5 is not to water seepage. and shortly after rain water no vvater at all discontinuities in 6 7 8 9 section can be simply with, for example, a weir. flowing in and out the section is the amount of water discharged by rock mass surrounding n1P·~l<<lnrw· the of water will tunnel. however, will usually not have a drain at the toe In existing classification expressed classes such as: 10 for underground excavations the , 'moist', - water; the water pressures of static water are of the storage The face is covered by an such as shotcrei.e, wi.th.out except if a mostly, 1.n:1d thus there is ~- flow uf '\Vater in the direction of the face or (l 9) Water flow may be restric!ed to charineis whil.e ilie whole dls,con.timuty is filled pressure still acts over the whol.e surfuce of tl-)e rock masses the of the discontinuities is not \"Vater In static, not water, then the water excavations has, however, been found that in som.e whiie the rock mass is water et al., Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ amount of water flowing out of the rock mass into the excavation. Classes such as 'dry' and 'moist' are not very difficult to establish but classes such as 'dripping' or 'wet' are subjective. The above leads to the conclusion that the methodology used in the existing classification systems that incorporate the influence of water pressures on the mechanical behaviour of a rock mass, should be reconsidered. B.3.4.13 Ice and snow influence Ice and snow can have a severe influence on the stability of a slope. Freezing of water leads to an expansion in volume. Water frozen in a discontinuity will exert a very high pressure on the discontinuity walls. In underground applications this virtually will never be a problem as temperatures underground are normally not below zero. In surfi1ce applications and certainly in slope stability applications freezing of water in discontinuities can, however, be a major factor fur the stability of a slope. Freezing of water may lead to opening and widening of discontinuities, displacements of rock blocks out of the slope :&ce, but also to closure of discontinuities, blocking the discharge of seepage water that may lead to water pressure build-up in the slope. Snow may cause a problem for slope stability because of the additional weight of snow on the slope :face. The influence of ice and snow is also dependent on the orientation of the slope with respect to the direction of the sun as daily temperature changes, especially a regular variation between freezing and thawing, has a negative influence on the quality of the rock mass. The problem of ice and snow influence is not addressed in any of the existing systems for slope stability. B.3.4.14 w Method of excavation Th&way tn.exposure has·boon~ishe&ha& a considerable influeneetmthe ·parameters measured or 'Observed in the exposure. For example, an exposure in a river bed created by slow scouring of the river over probably hundreds to thousands of years creates an exposure with a relatively small amount of visible discontinuities. Stress concentrations have not occurred or were minimal during the creation of the exposure due to the slow process. The tendency for discontinuities to open is minimal and therefore a larger part of the discontinuities is not clearly visible. Contrariwise a blasted excavation shows considerably more discontinuities because partly intact rock has been cracked due to the blasting but also, and often more important, existing internal planes of incipient weakness, which before blasting were not visible, have opened or widened due to the pressure of the blasting gasses and the shock wave, and therefore become visible and thus will be measured as mechanical discontinuities. Some existing classification systems take this effect into account (Haines, eh. B.2.4.7, Laubscher, eh. B.2.3.3, Romana, eh. B.2.4.6, Wickham, eh. B.2.2). These systems reduce the rock mass rating with a parameter to compensate for the damage that will be caused by the method of excavation. B.3.4.15 Seismic velocity in a discontinuous rock mass Some systems include seismic parameters, usually the velocity or apparent velocity of the wave, to assess the quality of the rock or rock mass (Japan, 1992, Weaver, 1975). For rippability, excavation and blasting assessment this is a :fairly standard procedure, but assessments are often specific for types and brands of (excavation) equipment, for blasting procedures or for types and brands of explosives. In excavation or blasting assessment the interpretation is in general simpler than for other applications. The influence of intact rock strength and spacing and orientation of discontinuities (the main rock mass parameters defining excavatability) on seismic waves is comparatively straightforward. To relate seismic velocities to other rock mass or discontinuity parameters (for example, shear strength) is :far more complicated. The behaviour of a seismic wave in a rock mass and the relationships between the rock mass parameters and the seismic parameters are not known in all details and consequently the interpretation is often ambiguous (Cervantes, 1995, Hack et al., 1982, 1990)(20>. A research project has recently been started at ITC and TU Delft to further investigate relations between seismic waves and de1ailed rock mass classification in near surface rocks. <20> Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EXISI'lNG ROCK MASS CHAJ.UCJ.'ERlZAION & CLASSIFICATION B.3.4.16 43 Operator experience and fiuniliarity with a classification Assigning values to some of the parameters in the systems discussed is often subjective and depends upon the operator's experience and the famiJiarity of the operator with the system. Examples for which this is of major importance are: 'the discontinuity set with the most adverse influence on the rock mass or for the engineering application' (B.3 .4.5) and classes such as 'vvet', 'dripping' for water influence (B. 3 .4.12). The merits of a system are clearly reduced if a system depends on the operator's experience or familiarity with the system. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 44 B. 4 Summary B.4 SUMMARY The review of existing characterization and classification systems leads to a series of conclusions and provides some directions for further improvement of parameters and calculation methods for slope stability assessment. These conclusions will be used to develop a new classification system for slopes (SSPC) which is the main topic of this research (section D). The conclusions derived from the review of existing classifications systems are: Method of calculation and parameter type 1 2 3 4 5 6 7 Difterent systems with difterent parameters lead sometimes to approximately the same outcome for the description of the same rock mass, e.g. Bieniawski compared with Barton. These two systems have been used extensively by difterent users, so it is unlikely that the outcome of the systems is totally wrong, however, operator bias may be present. · Intlicrlitemmre omytlie final roc1fmass Classification systemS a:reaescn"bed and not the underlying data analyses that resulted in the choice of weighting :fitctors in the systems. In general, back analysis by linear regression has been used to :fit the weighting :fitctors for most systems. Addition, subtraction, multiplication and division of logarithmic, linear and non-linear parameters are used. No clear advantage from one type of calculation or numeric representation of parameters above another seems to exist. Methods of calculation which combine different parameters in one rating number may not express properly the slope stability because parameters will have an influence on the rating that may not be important for the stability of the slope. The concept of a rock mass quality assessment before and after excavation should be considered as this concept seems logical and has been reported to be beneficial for slope stability assessment (Haines' slope stability assessment, eh. B.2.4.7). Parameters with fixed class boundaries but also with gradational boundaries are used. No specific preference can be found in the literature. Intuitively a scale with gradational boundaries seems to be more appropriate for a real rock mass. Most classification systems have changed during the years of application. This is logical for all systems are empirical. The number of case histories used determines the quality of the system.· The use of any empirical relation is restricted to the geological and engineering conditions of the case histories on which the system was developed. Extensive new data may stimulate an update of the system. No system is 1 :final 1 for there will always be new case histories to either expand its range of use or to improve its quality. Parameters 8 9 10 Parameters that need revision or should not be used at all in a new system are: Intact rock strength, Rock Quality Designation, Spacing of discontinuities, Persistence of discontinuities, Condition of discontinuities, Presence of water, Defonnation of the rock mass in relation to stress relief. Parameters that should be included are: Susceptibility to weathering, Method of excavation. Parameters not used in existing systems but may be considered necessary are: Sur:fitce run-off of water over slopes, Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ B EXlsriNG ROCK MASS C1IAJUCTER1ZAl'ION d: CLASSlFlOfl'ION 45 Ice and snow influence - freezing of water in discontinuities and weight of snow on a slope :tace, Karstic features. 11 Wzter 12 13 14 15 No new terms or definitions should be introduced unless absolutely necessary because this might result in confusion. Water pressures in discontinuities will generally decrease in the direction of the slope :filce, due to stress relief and consequent opening of discontinuities. This is different from the situation around tunnels where, generally, water pressures in discontinuities are present directly behind the tunnel wall. Consequently the influence of water pressures in discontinuities on the final rating of a classification system for slope stability assessment should be smaller than on the final rating of a classification system for the stability assessment of underground excavations. Water ftow and water pressures may be restricted to channels in discontinuities only. The tendency to reduce the inftuence of water, water ftow or water pressure in some of the more recent classifications systems for slope stability may suggest that water has a less strong influence on slope stability than often assumed in the past The influence of water on infill material in discontinuities, the effuct: of lubrication of discontinuities and the influence of water on weathering of the rock mass is likely to be important. Expressions for spacing and condition of a number of discontinuity sets in a rock mass 16 Parameters for spacing and condition of discontinuity sets should be revised so that multiple sets with different discontinuity spacings and conditions can be accounted for. Parameter determination 17 18 19 Determination of parameters should be possible using the simplest means. Any form of (complex:) testing should be avoided where possible. If any test is incorporated then the benefits of this test should be clear. Certainly it should be recognized that the need to do a field or laboratory test will reduce, for economic reasons, the amount of data available. Less data of probably better quality might not be preferable to more data of lower quality. Characterization and classification should be operator independent. Different users of the system should come to the same result Classification systems should be accompanied by exact and detailed descriptions of how to obtain the parameters. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C C 1~4RAtviETERDEFINITIONA.ND INITIAL POINT RATiNG SY:ST£",.11 PARAMETER DEFINITION AND INITIAL POINT RATING SYSTEM 47 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARA.METERDEFINITlON AND lNlTIAL POINT RATING SYS1'EM 49 C.l INTRODUCTION The review of existing classification systems (section B) shows that classification of a rock mass is generally accepted as a useful tool to estimate the influence of the mechanical behaviour of a rock mass on an engineering structure. However, the methodologies and parameters applied in the existing systems may not be appropriate or have to be adjusted to be fully effective in a classification system for slope stability assessment. In this section C parameters are defined such that these are suitable for slope stability assessment. These parameters, and more, were measured in the early stages of this research, which began in 1990. Slope stability was analysed by a point rating system which was modified and developed as the research progressed to give the 'initial point rating' system. It was eventually concluded that a point rating system is not a suitable approach to slope stability classification. Therefore in section D the approach is changed and the final result - a slope stability classification system based on probabilities; the SSPC system- is developed. The outline of section C is as follows: chapter C. 2 - Slope geometry and standards for visual assessment of slope stability The slope stability classification system developed is designed by describing and analysing existing slopes. The standards for measuring the geometry of the slopes and standards for the visual assessment of the stability of these slopes are defined and described in this chapter. chapter C. 3 - Parameters in rock slope stability Parameters of importance in slope stability and possibilities to measure these in the :field, are defined. chapter C. 4 - 'initial point rating' system Based on the results of the parameter analyses an 'initial point rating' system was developed. This 'initial point ratm&' system and the tesUlts .obtained. with the initialdsystem are briefly discussed. C.1.1· Data quality and storage Students and staff of ITC and the Technical University Delft characterized slopes according to standard procedures outlined in the following chapters and produced reports with photographs and descriptions of the slopes. The four years of data collection resulted in 286 characterizations of slopes in the Falset area. Obviously not all data were of high quality as students were in a learning process. This was, however, anticipated, for the involvement of a large number of different persons, not all experienced specialists in rock mechanics, was a preset requirement to avoid operator bias in the development of the system. Nonetheless some of the data received were incomplete, obviously erroneous or inconsequential and could not be used for the research. Because of this all described slopes have also been visited by the author and one or more staff members of ITC or the Technical University Delft. Incomplete data have been completed during these visits. Changing inconsequent or erroneous data incorporated, however, the risk of introducing operator bias from the author or from other staff members. Therefore it was decided that rather than changing the erroneous or inconsequent data these characterizations were altogether Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 50 C.l Introduction disregarded. This resulted in abandoning 36 characterizati.ons<21>, so that 250 acceptable characterizations resulted. Appendix I, 'Iable A 17 shows the number of slope assessments per lithostmt.igraphlc (sub-) unit. Each characterization consists of a maximum of 35 parameters. For 250 characterizations this results in a maximum of 8750 da1a items. This quantity of data can obviously not be handled manually to develop a classification system. Therefore all da1a have been introduced into a database (Dbasem Plus and IV). A programme in the programming language Clipper has been made for the necessary calculations (SSPCCLAS). (ll) From which 20 had been made by one group of students. The work of this group was abandoned altogether because the sites where they reported to have made the characterizations could not be precisely located. These were thus not abandoned because of the characterizations or the slope assessments itself. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARAMETERDEFlNlTION AND lNlTlAL POINT IUI'lNG SYSTEM 51 C.2 SLOPE GEOMETRY AND STANDARDS FOR VISUAL ASSESSMENT AND CLASSIFICATION OF SLOPE STABILITY The development of the classification system was based on existing slopes. The geometry and the stability of the existing slopes had therefore to be properly described and assessed. C.2.1 Geometry of slopes The orientation of a slope (dip and dip-direction) and the height of a slope assessed should be uniform and the stability assessments, whether visual or established by classification, should be made per individual geot:echnical unit. However, not all slopes comply to these requirements and rules have been set up how to describe the geometry of a slope. Latemlly curved slopes If a slope is curved laterally, the slope has to be subdivided in different sections where in each section the dipdirection is broadly uniform. The same applies if a slope dip or slope height changes along a slope laterally. The visually estimated stability (eh. C.2.2) and the stability assessment by classification are also established per section. • Slope height and dip Slope height and dip can be difficult to establish, for the slope is almost never a straight plane. Most slopes tend to become less steep towards the top and often flatten out. In this research the height and dip of the slope have been measured from the toe to the point where curvature indicates a flattening of the slope (Jiijg. 19). If, in vertical direction, a slope consists of different sections with different slope dips, the dip of each section is measured and the visually estimated slope stability (eh. C.2.2) is assessed in each section separately. A classification of the stability of the slope is done for each section individually. In each section the height is taken as the height from the bottom of the section to the top of the slope because the weight of the material above the section will have an influence on the stability of the section. ••• slope : bt!Jndl: : height: ''' btJnch ·.. --~lp Fig. 19. Standards for the geometry of a slope. Stepped or benched slopes Steps and benches on slopes have been measured because the stability of a stepped or benched slope is determined either by the dip and height of the bench or by the dip and height of the total slope (Fig. 19). If the width of the step or bench is large compared to the height of the slope and the rock mass is not prone to large deformations, the influence of the rock mass weight above the bench will, in general, not have a large influence on the outer layers of the rock mass forming the slope below the bench and its stability is governed by the bench dip and height. However, if the width of the bench is small or if the rock mass is prone to large deformations, the stability is governed by the dip and height of the whole slope. Classification of slope stability is done for the sections in-between benches and for the whole slope and the lowest result is assumed to be valid for the whole slope. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C. 2 Slope geol'lllttry and stmulalds for visual assessl'llltnt and classification if slope stability 52 m Multiple geotechmct.d waits one slope If a rock mass in a slope consists of a number of geotechnical units with approximately horizontal boundaries, the visually estimated stability is established per geotechnical unit. Slope stability classification is also done for each geotechnical unit independently. Slope dips can be difreren.t for each geotechnical unit and in each classification the slope dip is used that is characteristic for that geotechnical unit. The slope height used in the classification is the height from the bottom of the geotechnical unit assessed to the top of the slope. If the rock mass in a slope consists of multiple geotechnical units with vertical or inclined boundaries the visually estimated slope stability is established per geotechnical unit and also the classification is done per geotechnical unit. The height used in the calculations is again the height to the top of the slope. In some slopes a slope stability classification per geotechnical unit is not possible, for example, because the geotechnical units are folded. In such a slope, the slope stability classification is done as if the whole slope consists of the geotechnical unit that has the most adverse influence on slope stability. The visually estimated stability is established for the whole slope. C.2.2 Visual estimation of slope stability The research described was directed towards designing a slope stability classification system incorporating all possible mechanisms and modes of fuil.ure. To be able to reference such a newly designed slope stability classification system the stability of the slopes classified in the field has been assessed visually. The stability has been classified in five classes depending upon the absence, presence or impending presence of stability problems. These problems may be 'small' or 'large' depending on the size of the potential or actual rock fiills. Thble 5 gives the five stability classes and the number of slopes assessed in each stability class. • • • of.,l,.....,. ...n.ht1tt.r • b" • "udgement. This •V.!---1 il'>UaiA estJ.matJon ~-J' lS...& Sllljective j mm q.,.....,• .lUV.u.tVDlon L-o..--UGLWGGU·,_5" '1"'1..- ..I!. • • • --...1 •~--1}' ~!1 i!WU · il:l.i.tllreS is particularly sensitive to the experience of the observer. In principle 'large' implies that the unstable rock mass is in the order of tonnes weight while 'small' implies that the unstable rock mass is in the order of kilograms weight. Number of slopes Description Class 109 No signs of present or future slope failures 1 Stable 2 Small problems in near future The slope shows all the signs of impending small failures but no failure has taken place 48 3 Large problems in near future The slope shows all the signs of impending large failures but no failure has taken place 18 4 Small problems The slope presently shows signs of active small failures and has the potential for future small failures 20 5 large problems The slope presently shows signs of active large failures and has the potential for future large failures 55 Total: Note: 250 • The description large or small is independent of slope size. - 'Near future' implies within the engineering lifetime of the slope. Thble 5. Standards for the visual estimation of slope stability and the number of slopes per stability class. The problem of estimating the degree of stability for referencing a classification system is, however, a problem for all classification systems, whether for slopes or for tunnels. For most systems this estimation has been made by a group of observers. For the slope stability classification system described, estimates have been made over a period of four years using at least sixty observers from staff and students of ITC and Delft University of Technology working on 250 slopes. The large number of observers and observations must have significantly reduced the effects of individual observer bias. The purpose of visually assessing slope stability was to compare this with the stability of the slope as assessed by one or another form of classification. However, it should be noted the classification measurement is for a uniform Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARA.METElWEF.lNlTlON AN'l) IN!TL4L POINT RAI'lNG SYSTEM ov~~rrumgs, in etc. which may allow s:~ n""'ri:u"'ll~r was res1erurcn. area, are unstable Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C.3 PARAMETERS IN ROCK SLOPE STABILITY C.3 .1 Introduction The results of the review of the existing classification systems in section B showed that parameters to be used for rock slope classification should be carefully reconsidered and defined to be most efrective in slope stability classification. The following parameters are discussed in this chapter: parameters determining the mechanical behaviour of the rock mass material: intact rock strength and susceptibility to weathering (material properties, eh. C.3.2), shear strength along a discontinuity (eh. C.3.3), sets of discontinuities versus single discontinuities, concept of discontinuity spacing (eh. C.3.4), parameters that are specific to the rock mass at the location of an exposure or slope (exposure and slope spe:cifi:C"'PJrametet'S';ch: C. 3. S) met , parameters that have an infiuence on slope stability, but are not directly related to the rock mass or the slope (external infiuences, eh. C.3.6). The results of the evaluation are summarized in eh. C. 3. 7. C.3.2 Material properties Material properties include the intact rock strength and the susceptibility to weathering of the rock mass. C.3.2.1 Intact rock strength (irs) In most existing classification systems for slope stability assessment intact rock strength is a parameter and is it necessary to obtain the characteristic or mean. value of the intact rock strength of the g~technical unit in which the slope is· made or to be made. To assess whether and how intact rock strength should be a parameter in a rock slope stability classification system, the following should be considered: 1 Intact rock strength is not always included in existing underground or surface classification systems as a (main) parameter. 2 In existing underground excavation and slope stability classification systems (those which include intact rock strength) the contribution of intact rock strength to the final rating is considerably less than other parameters such as discontinuity spacing or condition of discontinuities. 3 Stresses in slopes will be nearly always considerably less than in underground excavation work so that it is unlikely that the infiuence of intact rock strength is as important in slope stability. 4 Failure in slopes is often associated with the shear strength of discontinuities(22)(23). <22> Some of the existing classification systems for slopes attribute slope failure fully to discontinuity failure if the rock mass rating is higher than a certain preset value, e.g. if the rock mass is of a certain quality. For example, the RMR modification by Robertson (1988, eh. B.2.4.5) assumes that slope failure is influenced by a number of parameters, including intact rock strength, for rock masses with a low rating (RMR < 40), but for a high rating (RMR > 40) the stability is dependent on discontinuity shear strength only. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PAJUMETERDEFINlTION AND lNI'I'lAL POINT Rlrl'ING SfSl'EM 55 An analysis of the imluence of steps on discontinuity planes prohibiting sliding along a discontinuity plane (appendix ll) shows that the intact rock strength will not be very critical for most slopes with dimensions 5 as in the research area. Summarized, this leads to the conclusions that the importance of intact rock strength in governing the stability of a slope diminishes with increasing intact rock strength and that a high accuracy in establishing intact rock strength is not necessary. A cut-off value for intact rock strength is used to incorporate the decrease of importance of intact rock strength. Above the cut-off value the contnbution of the intact rock stmlgth to the stability assessment of a slope remains constant. The limited importance of intact rock strength<24> does not require that sophisticated tests are done to establish the intact rock strength. Relatively easy to execute field tests with an impact method (eh. C.3.2.1.1) or with a 'simple means' field test (hammer, scratching, moulding, breaking by hand, etc., eh. C.3.2.1.2) lead to intact rock strength values adequate for slope stability assessment. C.3.2.1.1 Impact methods The Schmidt hammer determines the rebound of a piston activated by a spring. The rebound values measured on rock surfaces have been correlated to intact rock strength. Schmidt hammer values are, however, influenced by the material to a firirly large depth behind the surface. If a discontinuity lies within the influence sphere the Schmidt hammer values will be affected. The Schmidt hammer is thus not considered suitable to measure rock material strength in the field. The same applies to any other impact/rebound devices whose released energy per surface unit area is of the same order of magnitude as the Schmidt hammer of L or N design (eh. C.3.3.3). Equotip or other rebound impact devices (eh. C.3.3.3) might be suitable, but as these devices are only recently applied to rock m:ecnan1cs it· is not yet· certam.· whether the relationships ··befween rebound values and mtacttock ··· strength are correct. C.3.2.1.2 'Simple means' intact rock strength field estimates 'Simple means' field tests that make use of hand pressure, geological hammer, etc. (Bumett, 1975), are used to determine intact rock strength classes in the British Standard (BS 5930, 1981) (the test classes are listed in Thble 6). The 'simple means' field tests to estimate intact rock strength following Thble 6 have been extensively used throughout the research. For all classifications multiple estimates of the intact rock strength, often more than ten, have been made per geotechnical unit and per exposure. The values obtained were averaged. Additional to these estimates also laxge amounts of unconfined compressive strength (UCS) tests(25) have been done in the same geotechnical wits and in the same exposures to establish the reliability .Qf. the strength estimat'e$. If possible, estimates and UCS tests were done both perpendicular and parallel to the bedding or cleavage(26). <23> Sometimes a rock mass with a low intact rock strength (based on unconfined compressive strength- UCS tests) appears to have failed through inmct rock failure, but, on closer examination, the low inmct (UCS) strength is a consequence of a large number of (mechanical) discontinuities in the rock test specimen. Thus a shale may have a very low inmct rock strength as determined by conventional UCS testing (eh. B.3.4.l), but this is not caused by the low strength of the intact material but by the numerous closely spaced bedding planes. For very high slopes, as in deep open pit mines, stresses can become so high that inlact rock failure and shearing through asperities can occur also fur high inmct rock strengths. The inmct rock strength may then be more important. The slope smbility classification system developed in this research is, however, not designed for very high slopes. <24> <2SJ 14 UCS tests (one test from slope 92/5/3004 and all tests of student group 93/4) out of a tom! of 955 UCS tests were clearly outliers with values from 2 to 10 times higher than those measured by other groups in the same area and unit. These UCS tests have been excluded from the analysis. <26> 'Simple means' field tests and UCS tests have also been used for the engineering geological mapping research (see preface), which dam is included in the analyses of 'simple means' testing in this and fullowing chapters. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 56 C. 3 Parameters in rock slope stability The extensive quantity of tests allowed a thorough analysis of the accuracy and reliability of the 'simple means' field tests for estimating the intact rock. strength. This analysis is presented in the following chapters. The estimatedst:rength values in the graphs in this chapter are plotted as the mid values of the ranges of Thble 6. If the strength was estimated to be on the boundary between two classes the boundary value is used. C.3.2.1.3 intact rock strength 'simple means' test (standard geological hammer of about 1 kg) < 1.25 MPa Crumbles in hand 1.25-5 MPa Thin slaba break. easily in hand 5- 12.5 MPa Thin slabs break by heavy hand pressure 12.5- 50 MPa lumps broken by light hammer blows 50- 100 MPa Lumps broken by heavy hammer blows 100- 200 MPa Lumps only chip by heavy hammer blows > 200 MPa Rocks ring on hemmer blows. Sparks fly. 'Dlble 6. Estimation of intact rock strength. Intact rock. streo.gth field estimates versus UCS tests In Fig. 20a the estimated values of intact rock streo.gth by 'simple means' field tests are plotted versus UCS test values for all locations for which both were available, in Fig. 20b(27) the difteren.ces between the UCS test values and the estimated values as percentage of the estimated values are plotted, and in Fig. 20c the averages of estimated and UCS values per unit. In Fig. 20 no difteren.tiation is made for the direction of the measurements. Fig. 20ashewsthat tb.eseatB' iswid& an&~·only low· or·noeorrelationcan be seen; In Fig;· 20bis clearly visible that the differences between UCS and estimated values do not show a normal distribution for lower strength values. The distribution is skewed to higher values, e.g. the UCS values are higher than the estimated values. For high strength values the distribution of the differences is more normal but the average values of the UCS tests per estimated streo.gth class are lower than the averages of the estimated values. A quite good correlation is found for the averages per unit (Fig. 20c). The standard deviation of the UCS values per unit is for most units considerably higher than the standard deviation for the estimated strength value per unit (Fig. 20d). If is assumed that a unit has a characteristic streo.gth distribution with a characteristic mean strength value, which is very likely for the units assessed in the research area, then the estimated value will be nearer the mean value of the distribution because it is an average of more tests. The UCS test value is, however, only a single value or the average of few test values (normally less than three or four) and is likely to differ more from the mean value. This leads to the conclusion, as expected, that the characteristic mean strength value of a unit is better determined by a large quantity of estimated values than by. few UCS tests. The skew of the. distribution of the differences betWeeti"UCS'and estimated values for low "stiengtb.(Fig. 20b) is probably caused the &i that sampi~ are not taken randomly. Samples are very seldom taken from the worst parts of a rock exposure. This is also confirmed by an analysis of the results of intact rock strength estimation and UCS tests for granodiorite with various degrees of rock mass weathering in the same exposure (description rock mass weathering: appendix V, Table A 20). bY In Fig. 21 UCS values are considerably higher than the estimates of intact rock strength for the higher degrees of weathering of the rock mass. The granodiorite has weathered starting from the discontinuities and often a complete sequence of weathering is found. The weathered material and certainly the highly weathered parts, will break from the sample during transport and sawing of the sample. The UCS test is thus done on pieces of rock material less weathered than the average degree of weathering in the unit and therefore leads to a too high strength value. The difference between UCS test values and estimated values for high intact rock strength might be due to a similar, but reversed effect. For high intact rock streo.gth ( > 100 MPa) it is often difficult to get sample blocks out of an exposure without equipment (saw, blasting, etc.) and a tendency exists to do tests on loose blocks that are more easily obtained. These may, however, have a lower strength. This effect is also observed in the granodiorite for which the estimated strength of the fresh exposures is higher than the UCS strength values (27l The averages of UCS values are the averages of all UCS values belonging to the range of estimated strength. A grouping of the UCS values in the same classes as used for the estimate, before averaging leads to about the same values. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C f'M.R/u'vlETER DEF'!NlTlON AND LNffiAL POll'n' RATING :'ITSTEM * -K- -lii-l!-* 'Jf-il!* * 57 * * UCS and estimate * * * * difference b~en UCS and Mtimate as l'liAr'f';MitArm estimate - ~e) /I!!Stlmate Tg!.S Tg22 Tg21 c. averages per unit Tg1 fit Tg1 H H COYII,ll. H Slit. HgM!ss !1f31!odi~ (astlmmed ... estimated ... 1.19 + 0.910*UCS R2 .. 0.78 d. standard deviation of values in graph c (stdev. of estimated intact rock strength is zero forTg1 sst. and Hgneiss) UCS tests,, dashed lines in A and C indicate u'1e relation if estimated rock are obvious in 22, which shows the UCS estimate of intact rock different the estimated range value. For tests m:IJing in the ranges for lower intact values the UCS values are higher than the estimated values for the intact rock UCS is lower than the estimated C.3.2.L4 of the intact rock strength is fuirly good. In field intact rock has been estlt.macted by different students members the same exposure the same geotecl:mical unit. The show that the strength to be the same and a the <>trF•:najrh ;'jtn~:nrom estimates more man one class different often be attributed to real in intact rock within a unit An. Table 6, is that it would be deiJen.deJ1t or person estimates the lower than on the person who e. g. a been observed. The class :ranges are a small or person. Tills has not or Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 58 C. 3 ltuameters in rock slope stability ....... deg!Mof rock Mill 0 1 ~-58S0;1981) • 4 + fmlh slightly 11'110de1llltely • highly 0 lllD >250 100·200 60-100 12.5-50 5·12.5 1.2!1. 5 + 0 50 100 IIV8I1Ig8 ucs (MPa) 150 ·1 0 ~...._ofUCSte8tll(-) Fig. 21. Average estimated intact rock strength vs average Fig. 22. Percen1age of UCS test values fillling in a range diffurent from the estimated range value. UCS for granodiorite units with various degrees of rock mass weathering. to accommodate for most physical strength difte:rences. The possible error made by using estimation by 'simple means.'. . .of intact. mck strength is . discussed.in. .moredetail ch. . D.2.l(lable lS,note 2, ·MO 130). m C.3.2.1.5 Influence of degree of water saturation on intact rock strength Some porous rocks exhibit a difference in intact rock strength depending on the degree of water saturation when tested by UCS tests (Bekendam et al., 1993). The permeability and porosity of the intact rocks in the research area is generally low (the porosity is generally less than a percent) and the difterences in strength due to the degree of water saturation are therefure likely also very small and less than the scatter of the test results for most units. Only the Tgl sandstone unit (Tgl sst.) exhibits a larger porosity, is permeable, and could have shown a strength difte:rence similar to that found in the literature. However, the quantity of tests done on this single unit does :oot allow for conclusive statements. Therefure it is not known whether a strength estimate is inftuenced in the same way by the degree of water saturation as the strength value obtained by a UCS test. rew ucs C.3.2.1.6 The correlation of the estimated value of intact rock strength with the UCS tested in a particular direction could not be proven. Only in strongly anisotropic rocks (e.g. slate) the estimate is in agreement with the results from UCS tests. The highest strength is expected perpendicular to the cleavage direction. For the other rocks the estimation of intact rock strength results in higher values parallel to the bedding direction. In Fig. 23 are shown, per unit, the ratios of the strength perpendicular over the strength parallel for average UCS test values and for average field estimated values. Although this effect has not been studied in detail a possible (and tentative) explanation could be as follows. All rocks included in Fig. 23 have intact rock strengths that are in 'intact rock strength estimate' classes established by hammer blows(~ 12.5 MPa). The field estimate by hammer blows is a form of impact (dynamic) testing by which the rock breaks due to the impact energy (e.g. hammer blow). The impact energy is a limited quantity of enexgy induced into the rock in a small amount of time. Energy induced per time unit is thus high. The UCS test is a static test by which an unlimited. ~ount of energy is induced into the rock until firilure in a relatively large time span. The enexgy induced per time unit is low. Deformation of rock is a time dependent phenomenon. It requires a certain amount of time before a stress is converted into a deformation and vice versa. Stress and deformation are linked and it requires time to transfer stress and deformation throughout a test specimen. In an impact test part of the enexgy dissipates due to crack Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARAMETERDEFINlTION AND lNlTIAL POINI' RATING S'lSl'EM 59 forming directly at the impact point. The remaining energy travels through the rock as a stress/defOrmation wave (e.g. ::C shock or seismic wave). This wave is ' ' 1·6 retlected at layer boundaries and at the .f(27,13) ' end of the sample. When the incident ' ' and retlected waves are at the same "*"(23,27) (11,8) ' ' location and have the same phase, the ' ' ' ' stresses (and defOrmations) are added (1.2) ' .(11,11) : and may cause the rock layer to break. .(1.2) '' a layered sample the distance between fM layers is smaller than the length of the j 0.6 ' ' sample. The wave will loose energy (due ' ' to spherical dispersion, non-elastic defOr' ' mation, absorption, etc.) during 0+-----T-----r-----r'-----r----,-----~----4 travelling through the rock. A wave Tg1 sat. retlected against the end of the sample with a longer travel distance, has thus :Fig. 23. Ratio of average intact rock strength perpendicular over average intact less energy than a wave retlected against rock strength parallel fur UCS and field intact rock strength estimate per unit (values in brackets are the numbers of UCS tests respectively estimates). a layer boWldary. The concentration of energy at a certain point due to the coincidence of direct and retlected waves will also be less. This may be the explanation why a rock sample when tested (by hammer blows) breaks more easily perpendicular than pamllel to the layering and thus that the strength estimate for a sample tested perpendicular is lower than tested parallel. It is likely that this mechanism is less (or does not occur) in very thin spaced layered material (e.g. slate) because the rock at the impact point is easily fractured and broken whatever the orientation. I t m 1 In a UCS test the induction of energy in the sample is so slow that a stress/defOrmation wave will not occur. The whole sample will be stressed and deformed. The tensile strength perpendicular to the layer boWldary planes in a layered material is normally less than the tensile strength of the material. m a UCS test of layered material tested parallel to the layering, :fiillure will occur due to bending and separation of the individual layers, resulting in breaking of layers (starting with the layers at the rim of the sample). Perpendicular to the layering :fiillure occurs due to stress concentrations in the intact rock of individual layers. Bending of the layers and consequent cracking/:fiillure requires mostly less stress/deformation than breaking the rock due to stress concentrations and thus is the measured strength perpendicular larger than parallel to the layering. C.3.2.1.7 Conclusions The estimate of the characteristic strength of intact rock in a geotechnical unit with a 'simple means' test, following Thble 6, is equally good as executing a limited number of UCS tests. Therefore, intact rock strength (irs) in the classification system for slope stability ('initial point rating' system, eh. C.4, and SSPC, section D) has been taken as the intact rock strength established with a 'simple means' test, following Thble 6. The higher accuracy that might be obtained by using UCS tests exists often only in theory. m practice the number of strength tests is so limited in comparison to the variations in strength in the rock mass that very many simple field tests will give a better estimate of the intact rock strength at various locations in the rock mass than a limited number of more complex tests. A cut-off value is used above which the intluence of intact rock strength on the estimated slope stability is constant. For the initial slope stability point rating classification system (eh. C.4) the cut-off value was set at 100 MPa. This was an engineering guess. In the SSPC system (section D) the cut-off value is optimized based on data from existing slopes and results in a value of 132 MPa. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 60 C. 3 Parameters in rock of rock mass for s rock mass weathered from fresh degree of the amount processes, in local circumstances, the necessary size of and. rests n,.,,,......,,'"'"' rock materials to erosive or climatological influence o:r to a use as constmction material (Fookes et aL, Nota: The adjustment is applied to the rat!ng for the stabi!it\1 of the excavation of lauhscher's rock mass classification to predict the future stability. The degrees of rock mass weathering follow BS 5930 i1981 ). values for stability of un·der·R;rcmntd excavations in Thhl.e 7. Selby, 1982). Mostly these tests are done on small """"'"P"·" to weathering has been correlated with rock existing e. g. buildings, _gn1:ves:toraes, etc.. Tests for "'"""""'"''"'"'""' S1tlS(;epitibllH1tv to we:aU!er:m!! of discontinuities in a rock mass are not for the same reasons. Generally, it is assumed that an m<:re.;:ase in the degree of weathering causes a decrease of the shear we::tth~er:ir.t~ on 7 are multiplied. about 50 % ifa mass is exJJected is """'"u·-'>n weathered. within a year. It be noted that !,.,UAll;!\Jii;Ol.:!.I.MY difrerent "'""''"'"''''"' with less variation than the conllJ!U<:ms .., .............. v 1s in a way we:atl:JterJrng is defined as the time necessary to weather a rock del!lm:tion for rock mass we<u~:ueru1~ \ 28l The fuctors used in the system C.4) as in the SSPC system one or more of the mechanisw.s the rock mass and a separate parameter for the ''""'.n""'"'m'cr as a parameter for correction for the influence parameters C.3.5.n. (29) however, w,.,,n.,,.,.;,,., is not necessary in fue classification systems. The and future i.s discussed under exposure and stuoe--soe:cmtc This test is, Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PA.lfAMETERDEFINITION AND lNlTIAL POINT RA11NG SYSTEM 61 no significant weathering is expected. In the SSPC system (section D) susceptibility to weathering is incorporated by establishing the expected degree of weathering at the end of the engineering lifetime of the slope (eh. 0.1.6). The amount of time is established by comparing exposures with a knOWil time of existence within the same geotechnical unit. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 62 C3 Parameters in rock slope stability C.3 .3 Shear strength along a discontinuity The orientation of discontinuities in combination with the shear strength along discontinuities determines the possibility of movement along discontinuities. The influence of discontinuities on various engineering and mining structures and on slope stability is extensively described in the literature (Bamn et al., 1990a, Goodm.an, 1989, Hoek et al., 1980, 1981, etc.). In the literature review (section B) is shown that virtually all rock mass classification systems do include parameters that describe the shear strength along discontinuities in a rock mass. A new-to-develop slope classification system should thus also include one or more parameters describing the shear strength of discontinuities. Considerable differences exist in the methodologies used to incorporate shear strength of discontinuities in the existing classification systems. A basic problem is that shear strength along discontinuities is not fully understood. Some deterministic and empirical models do exist to calculate shear strength from discontinuity characteristics (fOrm of discontinuity, type of infi11 material, etc.), however, most of these methods are not without criticism and do not always work in all circumstances. The literature describing shear strength of discontinuities is extensive and often contradictory. The discussion in this chapter covers only those aspects necessary to illustrate the problems involved in defining a relation tOr shear strength along discontinuities in a slope stability classification system. The emphasis is therefore on parameters that can be determined in the field without extensive testing. The shear strength of a discontinuity is influenced by a number of discontinuity parameters. The discussion of the different parameters leads to a preliminary description of discontinuity parameters determining the shear strength of a discontinuity tOr implementation in a classification system. This was used in the 'initial point rating' system (eh. C.4) and further developed and adjusted tOr the SSPC system (section D). C.3.3.1 Persistence(30> determines the possibilities of relative movement along a discontinuity. Discontinuities are usually difrerentiated in:<3ll 1) persistent discontinuities; the discontinuity is a continuous plane in the geotechnical unit, 2) abutting discontinuities; the discontinuities abut against other discontinuities, or 3) non-persistent discontinuities; the discontinuities end in intact rock (ISRM, 1978b, 1981a). This definition does not consider ditrerences in persistence in d.ifterent directions. It is assumed that the discontinuity is persistent in any direction for the same length. This is not necessarily true. A discontinuity might be persistent in dip direction but not persistent perpendicular to the dip direction or vice versa (ISRM, 1978b, 1981a). The literature review shO\Ved that diffurent classification systems treat persistence in diffurent ways. Some systems (Barton et al., 1974, 1976a, 1988) treat persistence combined with roughness of the discontinuity walls while Selby ( 1980, 1982) combines persistence with the classification of infi11 material. In his classification Laubscher (1990) includes only those discontinuities which are larger than visible, thus those extending tOr a length larger than the exposure or tunnel, or those abutting againSt another discontiritiity: Fuitlfer qUan.titafive~descripfioiiS ofpersisteiice are few and probably riot fully satismctory (Bandis, 1990). The ditrerences in the methodology to incorporate persistence in a classification system were the reason to try to define a new implementation of persistence in the new slope classification system. In the 'initial point rating' system (eh. C.4) the persistence is related to the height of the slope. A non-persistent discontinUity can only move along the discontinuity if the intact rock pieces are broken through. This is dependent on the level of the shear stresses along the discontinuity and hence related to the height of the slope02>. <30> Persistence is treated as a discontinuity property in many of the existing classification systems and often also in the literature (e.g. Barton, 1974, 1976a, 1988, Selby, 1980, 1982, eh. B.3.4.4). <31 > See also glossary, page 241. The number of non-persistent discontinuities in the rock masses that were used fur the design of the new classification system were, however, few and this methodology to incorporate persistence could not be tested. Therefore in the SSPC system (section D) this approach is abandoned and the persistence is incorporated in the characterization of the condition of a discontinuity. <32> Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C BU«MEE'ERDEFINlTIONAND lNlTIAL POINT RAI'lNG SYSTEM C.3.3.2 63 Discontinuity roughness The contribution of discontinuity roughness to the shear strength of a discontinuity can directly be measured with, for example, a shearbox test (eh. C.3.3.8), but only for relatively small surfaces. In theory the contribution of roughness to the shear strength of a laige surfilce can be determined from other easily determined discontinuity parameters, such as the friction of the material (9) and the measurement of roughness profiles (Patron, 1966)(33>. This is, however, too simple fur natural irregular discontinuity surfilces. More complicated theories about roughness profiles, methods to characterize roughness profiles and relations between roughness profiles and shear strength can be found in the literature (Bandis et al., 1981, Barton et al., 1977, Pecker et al., 1971, Grima, 1994, Hsein et al., 1993, ISRM, 1981a, Rengers, 1970, 1971, etc.). However, :many of these relations between roughness and shear strength are hampered by scale effects (Cunha, 1990, 1993) or do not consider all discontinuity properties that are important. In filet the determination of the contribution of roughness to the shear strength is so complicated that exact methods ror laige planes can probably not exist other than by full scale shear tests. Variation of roughness properties throughout a rock mass and the impossibility to establish the roughness properties fur discontinuity surfilces that are not exposed, complicate the matter even further. Ob1:a:ining the properties in the required detail to make it worthwhile to apply a sophisticated methodology, is therefore mostly impossible or impractical. The conclusion is that a relatively simple method to describe the roughness that has a relation with the shear strength, based on as :many as possible simple assessments of outcropping discontinuities, is the only feasible method in a classification system. C.3.3.2.1 Roughness parameters important in slope stability The importance~ of the roughness of a discontinuity partly depends upon the stress configuration on the discontinuity plane in relation with the strength and deformation characteristics of the discontinuity wall material and asperities. Th clearly understand the mechanisms involved, the three fullowing theoretical situations are distinguished. These situations apply to a discontinuity without infill (discontinuities with infill are discussed in eh. C.3.3.4). 1 Overriding of asperities ~ the rock blocks on both sides of the discontinuity are not confined<34> and no shearing through asperities occurs. 2 Deformation of asperities - the rock blocks on both sides of the discontinuity are confined<34> and no shearing through asperities occurs. 3 Shearing through asperities - the rock blocks on both sides of the discontinuity can be confined<34> or not be confined, but shearing through asperities occurs. 1) Overriding qf asperities For a plane sliding situation the normal stress ( = the weight of the block under gravity) on the shear plane is constant in time (influences that can change the stress, such as snow, water, etc. are not considered for this theoreticalsittmtion);ffis assumed thatnoasperities can be shearedoff;because; for example, thesttengdristoo high, the asperities have to be overridden ibr movement along the discontinuity to be possible. Then the most important roughness parameters are the friction of the discontinuity wall material ( t'»btutc) and the maximum roughness angle (illlll%) from the datum reference plane (Fig. 24 left). The deformation characteristics of the rock material adjacent to the discontinuity and the geometry of the asperities at other locations along the shear plane are of no or minor importance. If fPbtulc + illlll% is equal to or larger than 90° movement becomes impossible. Deformation of asperities 2) If a discontinuity is confined and no shearing through asperities can occur, then the angle of the roughness is less important but the geometry (in particular the maximum height) of the asperities, the amount of asperities and the defurmation characteristics will mainly determine the shear strength (Fig. 24 right, deformation is hatched). <33> Formulated in the 'bi-linear shear criterion', see glossary, page 241. <34> Confined denotes here that the rock blocks on both sides of the discontinuity are not free to move in the direction perpendicular to the discontinuity. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ M 3) C3 Parameters in rock slope stabiliJy Shearing asperities through If shearing through asperities can take place then all parameters are of importance, e.g. the strength of the asperity material, the geometry and the deformation characteristics (Fig. 25). Not only all parameters are of importance but also all variations of these parameters everywhere along the plane where contact between the walls will occur during displacement. l~-~---~-- IIO~It~lD lhl~pclllllille A complicating factor is that a Fig. 24. Influence of roughness on displacement without shearing through asperities (left piece of intact rock will often figure: unconfined; right figure: confined). break under stress. Where and when a block of rock breaks is virtually impossible to establish by analytical calculations and highly complicated in a numerical analysis (Baaldman, 1993). Situation 3) is the common situation and nearly all shear disp~ent.AWmg. ~lliti~Jl~gQVe~by a combination. . . . of overriding of asperities, deformation of asperities and shearing through asperities. In slopes, however, the stresses perpendicular to the discontinuities are normally low which reduce the importance of shearing through asperities and the deformation of asperities. Fig. 25. Displacement of block (shearing through asperities and deformation). C.3.3.2.2 Measuring roughness Measuring a roughness profile on an exposed plane is theoretically simple. All that is necessary is to measure the height of the surfilce above and below a certain datum plane at regular intervals. There are, however, practical problems with regard to the datum plane, th~ measuring interval and tlie three-dimensionality of roughness. Datum plane Fig. 26 (left) shows a single block on a slope with the datum plane for this particular block. The datum plane is established by a least squares regression analysis of the profile. The roughness profile can be determined by sampling at a regular interval, measuring the distance below and above the datum plane. Fig. 26 (right) shows the same Fig. 26. Roughness datum plane for single block (left) and same block intersected by vertical block but the block contains a discontinuity (right). (vertical) not-cemented mechanical discontinuity across which no tensile stresses can be transmitted. Thus block B can move while block C remains stable, then datum planes have to be established for both blocks and do not have the same orientation. In a discontinuous rock mass each independent block of rock material has therefore its own datum plane. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C R4RAl-,!ETERDEFINrllON AND lN!TlAL POl!Yl' RA.TING SYSTEM ~S '"'"'"'"J..'-'" to certain ranges. the shear fl ,_ e- <If- r~~~ ;'!!=... 1- . !:-- ~ 4:- 1!11 !)l'<!lillllf! i!i. E !i r,. ~l!i- ~ 2 t- i•t proliia !:1 is ..1:!4 3 l! pralll®o surfaces me:asu1:m!! of C.3.3.2.3 the naked eye to which can be estimated visual ~n'"~'~"''t"'c"" of this method is that it does not need a:u enou~Zn to see traces the m directions. A.n example of this approach are the standard roughness profiles and the relation relates the to shear values that have been developed Barton et al. Barton introduced the JRC as a means to able to describe roughness 1-'"Jll"'"'"' <35l Fraclal repres·en!l>tlo'n is as a solution tor this"''-''"'""' 1989, Lee et al., Research showed, however, that the results pu<w~l'tcumay be accideniaL Fractal rep,resentaticm is therefure not suiiable without further rese--arch and a proper definition of the used (Den Outer et (36) reflection characteristics is that which can be seen. and are not included in visible rottghJness, Iv1easuring rml!lhneilS can be done meter, or, for scale with rulers, tlleod,olltes, iWII.lgl'lme!;s on a limited to a maximum. For the SSPC classification system the maximum is and hence a new geotechni.ca! tm.iL than the maximum, iiJr waviness in bedde-d mul!Jll!lt:'>s Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 66 C. 3 Parameters in rock slope stobility ~::: 'C. . . . a'. JCS • ten rRC . k)g10 ( J~~) + ~,J peak sMor ~ = ~ M1'1lftJl SI1'U8 on ~ pltme [12] =~wall~ stmtgth JRC .. -~ ~, rou.giBMas ctJejJfciMt "' resillual.friction Ollfk A problem with the JRC roughness profiles is that they do not include stepped surfaces and require measurement of the residual friction angle. Also, in the author's experience it is often very difficult to establish the proper JRC number visually. Laubscher (1990) developed a thorough set of descriptive terms for roughness of discontinuities with filctors rating the inftuence on the stability of underground excavations. The descriptions used by Laubscher are partly based on the profiles published by ISRM (1978b, 198la). The roughness is divided in roughness that cannot be seen, but can be felt by using fingers (tactile roughness), and roughness that can be seen, which is described visually at c:lifkent scales. This set of descriptions is used in Laubscher' s classification system for underground excavations (eh. B.2.3.3). Drawbacks are that dimensions for the roughness profiles are not given, the profiles are partly ambiguous, representative profiles for large scale roughness have not been published, and in particular the combination of tactile roughness and small scale roughness is not clearly defined. C.3.3.2.4 Stepped roughness planes Steppedi'Ouglmess·planesfire·planeson·wbicb. aspetitieswitb.sides occor·for which appliesthat tp + t-angle ~ 90°. These asperities are normally denoted as steps on the discontinuity plane, although the i-angle does not have to be 90°. If a step is present perpendicular to the direction of sliding then either the step has to be sheared off bei>re the block can slide or so much dilatancy deformation has to be possible that the block can slide over the asperity. Steps on surfaces often prohibit sliding (appendix ll). None of the empirical relations take this into account. The standard profiles by ISRM (1978b, 1981a) and Laubscher (1990) do, however, include stepped planes (Fig. 69, page 142, and Fig. 70, page 143). C.3.3.2.5 Anisotropic roughness Roughness of a surfilce can be anisotropic (e.g. ripple marks, striation, etc.), and thus the shear strength will be ~on dependent. The<>reticallY .!!I~ rous!l!!~ss should ther~ be m~ in ~ dU'ections, Th~ number of different directions that should be measured depends on the type of the roughness. For example, it is sufficient to measure the roughness in one direction only for a regular striation; perpendicular to the striation the contribution to the shear strength of the roughness due to the striation is Jila.Ximum while parallel to the striation no influence of the striation is present. For less regular surfilces the number of directions in which the roughness has to be measured increases, but roughness in all directions will be again about equal for a fully irregular surface and one measurement will be sufficient. Alternatively the roughness can be measured only in the direction in which shear displacement over the discontinuity is expected (this direction will often be known in slope instabilities). In practice it will mostly be sufficient to determine the roughness in one direction or in two perpendicular directions only; parallel and perpendicular to the maximum roughness. The accuracy of roughness determination and subsequent translation into friction angles is, in general, not high enough to justify the determination of roughness in more than two directions. C.3.3.2.6 Discontinuity history The history and origin of a discontinuity have an influence on the shearing characteristics of the discontinuity. If movement along a.discontinuity has taken place in the past·then two situations are possible: 1 Due to the movement asperities have sheared off completely and the roughness of the discontinuity is nil. The roughness of the discontinuity is determined as found and thus the history is included in the assessment of the roughness. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARAMETERDEFINlTION AND lNlTIAL POINT RATING SYSTEM 67 The movement happened without shearing off the asperities or the asperities are only partially sheared off. The resulting discontinuity has then a non-fitting roughness profile and the dilatancy necessary to allow further displacement is lower (Rengers, 1971). In this situation testing might help to guess an accurate value of the shear strength or an estimate can be made by which amount the necessary dilatancy (or iangle) is reduced due to the displacement For example, the shear strength of a discontinuity that is not fitting at all, is governed by the material friction only. 2 C.3.3.2.7 Conclusions A summary of the diftmm.t ranges for roughness with wavelengths and amplitudes for regular forms of roughness is shown in Fig. 28. The boundary lines are dashed as these are not exact. The wavelengths and amplitudes for the roughness profiles are an indication only. The figure is an attempt to combine normally occurring different types, scales, and measuring methods of roughness and is not expected to cover all forms of rougbness(37). In this research, a new empirical relation between tactile and visible roughness based on the ISRM (1978b, 1981a) profiles, and the friction along a discontinuity plane resulting from roughness, is developed because of the problems with existing shear strength theory and roughness as described above. For this purpose the roughness profiles of ISRM (and Laubscher) have been Jn~!'4:. J~~. .~ess. is .!0 -'E' o.ooot 0.001 ~ (m) wavelenglha===la 0.1 For small wnplll.udes and of 8 triangular/8awt form wherBaa wllh larger amplitudes and the roughness changes to a more alnusolclatform;~ is notinduci•Hn 1he ~visible raughneas. The~ in the graph are cluhecl u these are not exact be Fig; 28; ~nterpremtion of regular forms of roughness as function of"SC~tie and angle. distinguished by feeling with fingers and described in three classes: rough, smooth and polished. The small scale rougtmess·determined on an area of20x26 cm2 of the discontinuity surface, should be visible and is described in three classes: stepped, undulating and planar. Representative example profiles including scales are provided in Fig. 70(38> (page 143). The vertical scale of these profiles is based on the minimum step height requirecl to prohibit crushing efi:cts in steps (eh. C.3.2.1). The large scale roughness determined on an area larger than 20 x 20 cm2 but smaller than 1 x 1 m2 , is described in five classes: wavy, slightly wavy, curved, slightly curved and straight. For large scale roughness examples of profiles with scales and i-angles(39> are presented in Fig. 69(38> (page 142). The roughness profiles are included in Fig. 28. Values for each roughness description that rate the influence on slope stability, have been copied from Laubscher for the 'initial point rating' system. In the SSPC system (section D) the values have been adjusted based on the data obtained in this research. (37) For example, stylolites in limestones or very coarse grained rocks (e.g. porphyritic granites) could plot in the region which is indicated as 'do not normally exist'. Changes between roughness profiles for the 'initial point rating' and SSPC system are only minor. Therefore the profiles are not repeated in this chapter. <38> <39> The i-angles were not included in the 'initial point rating' system but have been derived from dam gathered during the fieldwork for this research (Fig. A 98, eh. D.l.2.1). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C. 3 Parrmreters in rock slope stability 68 If the roughness is direction-dependent the rouglmess should be assessed in two perpendicular directions. If movement along a discontinuity has taken place in the past then the influence of this movement on the shear strength along the discontinuity should be quantified by estimating the remaining i-angle or the discontinuity has to be tested. Alteration of a discontinuity wall C.3.3.3 The discontinuity wall is the rock material directly adjacent to the discontinuity. It is the material which, if in contact, will determine the shear strength along the discontinuity. Determining the shear strength characteristics of discontinuities requires that the joint wall condition or joint wall strength should be established. Various authors have commented on the influence of the strength of the discontinuity wall on shear strength (Bandis, 1990, Barton et al., 1973a, 1973b, 1976b, 1977, 1985, Laubscher, 1990, Fishman, 1990, Rcngers, 1970, 1971, Rode et al., 1990). Often the 'quality' (strength) of the discontinuity wall is lower than the intact rock strength (also eh. C.3.2.1). The decrease in strength may have been caused by weathering features, brought about by chemically charged water percolating through discontinuities that reacted with the wall, etc.. The thickness of the layer having a lower strength may range from microscopic thickness up to many centimetres or more. In shearbox tests the discontinuity wall strength is incorporated in the results, however, shearbox tests can only be done on samples of limited size. Strength and thickness of the joint wall must be known to understand the shear strength test results. 200 maximum. Influence l I I I ~ itfntstt······· !rock I I 400 ~+--.--r--.-,--.--,--.-,--.-. 5 0 10 15 20 depth below surface (mm) I I ~ I -~100 I I I 800 I I 50 I 600 / 2 400 / o~~=r~-r~rT~~-r~~~~_, ~+-----~--~-----.----.----. 0;0 2.5 · 5.0 'Us 10.0 12.5 depth below surface (mm) Fig. 29. Equotip rebound values on weathered discontinuity walls progressively ground down to fresh rock (after Hack et al., 1993a). 0 200 400 800 800 1000 Equotip (l) Fig. 30. UCS vs Equotip (after Verwaal et al., 1993). Rebound tests are a method which may be suitable to assess discontinuity wall strength. The best-known rebound test is the Schmidt hammer<40> (ISRM, 1978a, 1981a, Rode et al., 1990, Stimpson, 1965). Other rebound measurements are based on a hammer, ball or piston which drops from a certain height on to the surfilce to be measured (Equotip, 1977, Hack et al., 1993a, ISRM, 1978a, 1981a, Pool, 1981, Price et al., 1978, Stimpson, 1965, Verwaal et al., 1993). The rebound of the piston, hammer or ball after hitting the surfilce is dependent on the elastic parameters of the tested material and on the strength of the material at the surfilce of the discontinuity. This latter effuct is caused by the crushing of surfilce asperities and surfuce material, which dissipates energy. Most of the rebound tests reported in the literature are not developed to measure the discontinuity wall strength but to measure the intact rock strength. The standard form of the Schmidt hammer releases so much energy over such a laxge area that in most rocks a layer of up to centimetres depth influences the measurement. The ball rebound device (Pool, 1981, Price et al., 1978) and the Equotip device (Equotip, 1977, Hack et al., 1993a, Verwaal et al .• 1993) release considerably less <40> the field. Ditrerent designs of Schmidt hammers for different impact energies exist. 'L' and 'N' design are most commonly used in Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARAlvfETERDEFll>llT!ONAND IN!TMLPO!NT Rlll'ING SYSTEl'tf 69 re.!Jta.Dlle :aneans .to. test shear smmwtn strength is the the number classes necessary to be done with a simple and can C.3.3.4.1 """"''"""'."" or of the . In these descriptions an ...............u ... over a ...il!C·JI>.i.•"'"'~' can be in three ranges et If there are points of contact between discontinuity walls, the the properties of the discontinuity -walls. If iufill tl:w~lrntess the shear srnmg•th for :irregular tliSCOJlltltlW1ty to the amplitude of material. ,..,.,,,..;n,,...,. can be important ''""''"~'"'"" or is strength is mainly determined by dls.coJtlti.tlmity is less than about the grain size of the intact :rock or of the size of the infiU the shear '"""'"'"''l·n but wall material. dlSICOlltlntlity 'IN"afi and the or the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C.3.3.4.2 Origin of a discontinuity or origin of infill material Some classification systems describe discontinuity in1ill material based on the origin of the discontinuity (Brekke et al. , 1972) because the origin of a discontinuity can have a relation with the shear strength characteristics of the discontinuity. For example: bedding planes will often be a potential discontinuity because the plane is formed by more softer or easier weathered materials (e.g. clay) than the rest of the rock mass, whereas tectonic joints will normally have an infill material consisting of weathered intact rock material. This method of description implies the risk of totally wrong assessments. The author has often observed bedding planes that did not contain any clay infill material and observed tectonic joints filled with clay material that was not weathered intact rock. It could even have been that the clay material of the bedding plane had been washed out of the bedding plane and accumulated in the joints. In this situation it would be obviously totally wrong to determine the shear strength parameters based on the origin of the discontinuity. Origin of infill material is obviously a better means of describing the discontinuity characteristics (V'klsh, 1994). However, often the origin of the infill material can only be established by a detailed description of the infill material. Therefore it seems more logical to relate the infill material to shear characteristics directly than by first establishing the origin of the infill. C.3.3.4.3 Conclusions The classes used in this study (and described in this chapter), roughly follow those established by Laubscher (1990). The system is relatively simple and no expert knowledge of geology is necessary. An additional class for 'cemented/cemented infill' discontinuities has been included. lit?jnfill, t:e!!!!'!fed or no~~ed The first distinction to be made is between: no infill, cemented, cemented infill or non-cemented infill. 'No infill' describes a discontinuity that may have coated walls but no other infill. For most discontinuity surmces friction is virtually independent of the minerals of the intact rock. This has been established by many researchers doing tests on smooth, planar surmces to obtain 'Pbo:l~c and is also confirmed by tests done for this research (Hack et al., 1995, appendix HI). Apparent cohesion of the discontinuity walls does depend on the type of mineral but at low levels of low normal stress apparent cohesion is less important (eh. C.3.3.2.1). For mineral coatings on discontinuity walls the same applies (Welsh, 1994), also confirmed by tests done for this research (Hack et al., 1995, appendix Ill). Therefore one class describing the shear strength of a non-cemented, non-filled discontinuity is sufficient. A cemented discontinuity or a discontinuity with cemented. infill has a higher shear strength than a non-cemented discontinuity if the cement or cemented infill is bonded to both discontinuity walls. If there is no cement bond between the discontinuity walls or between the cemented infill and one or both discontinuity walls the discontinuity behaves as a non'"Cemented, non-filled discontinuity. 1\vo classes should be distinguished for discontinuities with a cement bond or with cemented infill bonded to both discontinuity walls: 1) the cement or cemented infill and bonding to both discontinuit.)! walls are .sttonger than the..SUITOunding .intact rock (milure will be m intact rock), and 2) the cement or cemented infill and bonding are weaker than the surrounding rock but still stronger than a non-filled discontinuity. Those that are stronger than the surrounding rock do not need to be considered as a discontinuity, those weaker are described with the class 'cemented/cemented infill'. Non-softening and softening infill A major distinction should be made between non-softening and softening material for discontinuities without cement but with infill material (Barton, 1974, 1980, Laubscher, 1990, Thlinov et al., 1971). Non-softening infill material is material that does not change in shear characteristics under the influence of water nor under the influence of shear displacement. The material may break but no greasing effuct will occur. The material particles can roll but this is considered to be of minor influence because, after small displacements, the material particles will generally still be very angular. Softening infill material will under the influence of water or displacements, attain in a lower shear strength and will act as a lubricating agent. Both classes of softening and non-softening infill material can be further sub-divided in classes according to the size of the grains in the infill material or the size of the grains or minerals in the discontinuity wall. The larger of the two should be used for the description (Thlinov et al., 1971, Laubscher, 1990). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PA.RA.METERDEFINITION AND lNlTIAL POINT RATING SYS1'EM 71 Gouge The so-called 'googe<41> filled' discontinuities are a special case. Gouge filled discontinuities are often the larger discontinuities in a rock mass such as fi:rults. Gouge layers are relatively thick and continuous layers of infill material, mainly consisting of clay but oftlm also containing rock fragments. The common feature of gouge is the presence of clay material that surrounds the rock fragments in the clay completely or partly, so that these are not in contact with both discontinuity walls. The initial shear strength of such a discontinuity will be that of the clay. If the gouge is thicker than the amplitude of the roughness of the discontinuity 'Walls, shear movement is governed by the clay material. If the thickness is less than the amplitude of the roughness the shear strength will be influenced by the wall material and the discontinuity walls will be in contact after a certain displacement; for further displacement the shear strength is governed by the friction along the discontinuity walls in combination with the clay infill and the friction of the rock f:m.gments in the gouge. Flowing material Very weak and not compacted infill in discontinuit.ies flows out of the discontinuities under its own weight or as a consequence of a very small trigger force (such as water pressure, vibrations due to traffic or the excavation process, etc.). For the 'initial point rating' system (eh. C.4) values that rate the influence of the ~rent infill materials on slope stability ha.ve been copied from Laubscher ( 1990) or are studied guesses from the author. The values have been adjusted for the SSPC system (section D) based on the data obtained during this research. C.3.3.5 Weathered discontinuities Weathering of discontinuities results, in most rock material, in weakening of the discontinuity wall and in the formation of infill material in the discontinuity. The shear strength of such a weathered discontinuity is determined more by the presence of infill material than by the reduction of the shear strength due to the weakening of the discontinuity walls<42>. Reduction of the shear strength of the discontinuity walls become important only if the weathered material is flushed out of the discontinuity completely. However, usually a thin layer or coating of weathered material stays behind in the discontinuity. For example, in carbonate rock masses containing some clay, it is oftlm found that the discontinuity walls are slightly weathered and that a thin clay infill is found in the discontinuit.ies, this being all that remains of the weathered rock material. The remaining infill significantly determines the shear strength of the discontinuity. A separate parameter for weathered discontinuities is therefore not necessary. C.3.3.6 Discontinuity karst features Karst features have been found to be of importance in slope stability. The open holes considerably weaken the rock mass (eh. B.3.4.7). In the 'initial point rating' system (eh. C.4) karst was described per discontinuity set in terms of occurrence and size of the karst holes. The values used for the karst parameter (Fig. 36, page 84) in the 'initial point rating' system are studied guesses of the author as no literature references were found. In the SSPC system the values are calculated from the influence of karst on the stability of existing slopes (eh. D.l. 2. 1. 2). C.3.3.7 Effect of water pressure in discont.inuities Water pressures in discontinuities reduce the shear strength of the discontinuities (eh. A.2.3), which is the reason that many classification systems for underground excavations include a separate parameter quantifying this influence. In eh. B.3.4.12 is already discussed that the influence of water pressures on slope stability may be less important than often assumed. The methodology used in this research to develop the classification system for slope <41 > 'Gouge' is an ancient mining term which implies soft, easily extracted material (see glossary, page 241). <42> This has been confirmed during this research fur slope stability assessment where was fol.l.lld that ilie reduction of discontinuity wall strength. is not important if even small quantities of infill material are present (eh. D.1.2.1.2). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 72 C. 3 ltm:uneters in rock slope stability stability, may even further reduce the need for a separate paameter for water pressures (whether swiilce runoff water should be included as a separate parameter is discussed in eh. C.3.6.1). Consider the two following situations: 1) a new slope is made in similar conditions, with respect to water, as the exposures used for the classification (Fig. 31, cut A), and 2) a new slope is made which is totally different with respect to water conditions (Ftg. 31, cut B). ... .. / ·.... ·. '·i ·IS.. • possible new'~ water table after excavation I ' •• i •• •• ·:,_.-.-.-.-.- .l..-. (/.,topsoil ?·· .... ~water tabl8 before excawdioin Fig. 31. New slopes in different conditions with water table. New slope in similar conditions with respect to water 1 All slopes used for the development of the slope stability classification system in this research are situated in a temperate (Mediterranean) climate (eh. A. 3 .1) and with a recurrence period of a few years heavy rainfiill takes place. Therefore, all slopes being used for referencing the slope stability classification system have been subject to rainfiill appropriate to the climate, leading to the presence of water and probably to water pressures in the discontinuities (Fig. 31, cut A). This influence of water in discontinuities is thus likely already incorporated in the weighting of the parameters in the slope stability system. 2 . New slup!! in di/Je:r!!nt C!!1!Jli1ions~wi1h resp!!ct. 10 wate.z:. . . .. If a new slope is made which iJ;ttersects the permanent water table (Fig. 31, cut B) and the classification for the reference rock mass has been made on exposures not intersecting a permanent water table a correction on the classification system may have to be applied to allow for the unfi:tvourable water condition<43>. A correction to the classification for slope stability is thus likely only necessary in those very few occasions where a new slope intersects a permanent water table with water pressures in the rock mass directly behind the slope :fu.ce<44>. In the 'initial point rating' system (eh. C.4) a parameter was incorporated that corrects the stability assessment in case the slope shows permanent water seepage, thus for a slope intersecting a permanent water table. The correction values used for this parameter were the same as those used by Laubscher (eh. B.2.3.3). The quantities of water flowing out of the rock mass as used by Laubscher, have, however, been reduced by the author to be feasible for slopes. In the SSPC system (section D) a parameter for the influence of water has been omitted on the basis of the results of the analysis of the data of the existing slopes (eh. D.l. 7). C.3.3~8 Testing the shear strength of a discontinuity can be done by field and laboratory tests. In practice the various tests contain serious shortcomings and will only give erode estimates of reality. All non-in-situ field and laboratory tests on discontinuities are hampered by difficulties in sampling and executing of the tests. Therefore, no testing of discontinuities is required for the slope stability classification system developed in this research. The problems involved in testing of shear strength have been commented on by many authors (Goodman, 1989, Cunha, 1990, 1993). Example n (eh. D.5.2) illustrates the problems encountered with testing shear strength carried out for this research. <43> It should, however, be considered that: i) a new slope cut will be unlikely to allow free drainage of a pemument water table and artificial measures would be taken to lower the water table behind the slope (drains above the slope, drainage holes in the slope, etc.) and, ii) often exposures used for the classification of the reference rock mass, not intersecting a permanent water table, intersect the increased water table during rain (Fig. 31, cut A). In both situations a correction is not necessary. <44> In the research area this situation does not occur and, in the author's experience, also in other areas this rarely happens. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARAMETER DEFINITION AND INITIAL POINT RATING SYSTEM C.3.3.9 73 Conclusions The evaluation. of discontinuity shear strength properties and the possibilities to measure parameters to describe these properties lead to the conclusion. that a simple classification. of parameters based on. tactile and visual observations of outcroppiug discon.tinuities is the ooly feasible possibility to include discontinuity properties in. a classification. system. M~ sophisticated measurmg methodologies are oot necessarily better, mostly high.l.y un.practical and not sUable· i>r fi.eld use for a classification. system.. The various simple methodologies for the d.itken.t parameters as described in. this chapter, are implemented in. the 'initial point rat:ing' system and are shown. on. the field exposure clwacterization. form in. Fig. 36 (page 84). Some methodologies are modified for the SSPC system in. section. D based on. the data obtained in. this research. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 7JI. C3 Parame1ers in rock or mscm1rnnmtJ.es "'"'''"'"""'<; a form of '""''"'"'''"' n'l.et!il0(1oiog:tes. These are C.3.4.1 A ge{llo;glcru ~malysis of consists in exposures, between the diffurent discontinuities and strucrurai to obtain cmnp.:tete discussion on hO'N to deterroJne d1s:co:nttnu:tt '"'"''"""·'"' the scope of this research and can be found in books Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C R4.ll.AMETERDEFlN!TlON AND INmAL POINTR.4'l'ING SYSTEM 75 C.3A2.2 a in u:nmv,omrao.te discontinuity may wen have been missed if it happened not to cross the selected ""~"""'....,"'· pn:sel!lt in the area axposure where the measurements are or oriented with to the exposure, e.g. discontinuities near to face of the exposure. The errors which may the results of to determine sets and orientations are, in '4"''-''-'"''"'• ""''"'"'"'"'"'"'"'RD. on and character method. A mean orientation value for a disco11ti!mi1ties that to the set in each discontinuity set can be oriented C.3.4.2A In a assessment to determine discontinuity in an exposlll"e, t..l-10se discontinuities that are most unfavourable for the engineering structure or if that is not a priori known, the discontinuities that are rer;:re:senttatwe the set are visually In selection is incorporated the whole area selection is done visually it does not matter whether the discontinuity is accessible or and the character of the discontinuity roughness, the the properties of the selected are measured in detail in 'V}"·"""''... of author based on former and (45l If selected discontinuities not to be accessible the orientation can often be measured from a distance by means, such as cl.inometer and compass or However, of a not-accessible have to be estimated. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 76 C.3 lbrame~n in rock slope stability this research this method gives an equal or better result than the results of extensive measurements of discontinuities for a statistical analysis. If extensive amounts of measurements of discontinuity properties and parameters have to be done, they are always done on a part of the exposure that is (easily) accessible whether representative for the rock mass or not. The same observations have been made by other researchers (Gabrielsen, 1990). It may be thought that a studied assessment for the determination of discontinuity properties would not be accurate enough, but it should be kept in mind that the variation of discontinuity properties in one discontinuity set is often so large that a high accuracy is not very important (ISRM, 1978b, 1981a). C.3.4.2.5 Borehole cores Grouping the discontinuities in sets and determining the mean or characteristic discontinuity properties and parameters of the sets can be done by the methods discussed for exposures<47). It should be noted that borehole cores show only a very small part of a discontinuity surface and that consequently the determination of properties may be less accurate. C.3.4.3 Overall spacing of discontinuity sets in a rock mass Various expressions have been defined to quantify in a single qualitative or quantitative expression the spacings of a number of discontinuity sets in a rock mass. One of the simplest expressions is the RQD (Deere, 1967, 1988, 1989, eh. B.2.2), more detailed expressions, which describe block size and block form in a rock mass, can be found in BS 5930 (1981, eh. B.2.1), Price (1992, eh. B.2.1). Taylor (1980) developed eq. [13] for the description of the...spacing . for. a maximum of three. disconti:miliy sets. For a rock num with OM ~ set: * log1gX factor1 = 0.46 + 0.264 (.x = ~ spacing in cm) with two discontintl.ity sets: factor1 = 0.38 factors • 1 + 0.269 * log10 .x,._ [13] with three discontintl.ity sets: factor1 =0.30 + 0.269 * log10 .xllliliiaalo factors =0.10 + 0.333 * log10 .x.....,.,._ ~ " - < • > ' SJIQCing factor for rock mass "'factor1 * foctor2 *factors (min.imum, intermediate and maximum refer to tire spacing of the discontinuity sets) The graphical representation is shown in Fig. 33. The parameter is calculated for a maximum of three discontinuity sets with the lowest spacings. The method according to Th.ylor is used in Laubscher' s classification system <46J Experiments (unpublished) done by the author while employed in an underground mine showed that scanline analyses compared to studied assessments of the orientation and spacing of various discontinuity sets resulted in nearly the same values if the discontinuity sets were clearly distinct and if done in small (maximum 2 x 2 m) tunnels with crosscuts allowing fur scanlines in all directions (also along the roof). The studied assessments and statistical analyses were done by different qualified engineers who also incorporated discontinuity type and properties in the analyses. The statistical analyses often, however, missed discontinuity sets if the same comparison was done in large tunnels or in tunnels without crosscuts (thus not allowing fur scanlines in all directions), or if the sets were not clearly distinct or had a (very) large spacing. <47l In borehoie cores spacing is often measured irrespective of the discontinuity sets, as, for example, in measuring the RQD. This is often inevitable because the borehole cores are drilled without marking the orientation. The orientation of discontinuities is, however, a main factor in determining the stability of a slope and boreholes drilled for slope stability assessments should thus always be drilled to produce orientated cores. The discontinuity spacing measured in borehole cores may be effected by new discontinuities formed due to the stress relief as a consequence of drilling. The measured discontinuity spacing is then lower than in-situ. This effect is more severe fur borehole cores from a large depth than fur cores from a relatively shallow depth as would be drilled fur the type of slopes for which the classification system is developed and is therefore not further discussed. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARAMETER. DEFINITION AND 1Nl1.'1AL POINT lUl'lNG SfSI'EM for underground excavations (1990, eh. B.2.3.3). Many engineers, illcluding the author, have extensively used, with success, the Laubscb.er system for classification of the stability of underground excavations in a mining environment. The good results obtained with Laubscher' s classification system i>r undeJ:ground excavations is the reason to investigate the possibility to include a parameter describing the spacings of a number of discontinuity sets in a rock mass, calculated analogous to 'Th.ylor (1980), in the classification system for slope stability developed in this research<41). C.3.4.4 Overall condition of discontinuity sets in a rock mass Several options exist to describe the overall properties describing the shear strength of discontinuity sets in a rock mass. In most existing classification systems only that discontinuity set is considered that has the most 71 . . . .. .... . . . .. ... / . "" ... : f diiCOiitint.iltY seF :: ~ : ,;:-: ::::: .. U+-----~~-·~·-·-·-"~"'-+-·~·~·~·~"~~·~"~'·~·~"~"· M+-~~~~~~~~~~7.*~~~~ 1~.~~~~.~.. ~.~.+.,.~.~ ~~.~~ :::::::: . /':/'::·-::/: /: Y: ::.': : ~ U+-~777'~-":7~~..-- .. .... 3 ciscon1lnuil:y se ::: Minimum spacing : :: inUrmedlate spac~ns;:: 11.3·+-......,..--;<-~~__.._,...........::.."""'"'-maximum spacing : :: u~~~.~.,~.~ ..~..~.~ .. -.-.-... -...--------------~ /: . . ~t::;;:: ,': :::::::: ;~" t: :,~::::/ : : ::::::: 0.1+--~--r-------r----.,-------1 0.1 1 10 100 dlscon1!nl.llt.v SPaCing Ccm) 1GOO Fig. 33. Discontinuity spacing iilctors (after 'Thylor, 1980). adverse condition. This can lead to problems as discussed in eh. B.3.4.5. A solution to these problems is to use an average or a weighted mean of the condition of the c:lifrerent discontinuity sets. In the 'initial point rating' system (eh. C.4) the parameter describing the overall · condition of thedisoontinuities is the mean: value of the three discontinuity sets with the lowest condition ratings, weighted inversely against the spacing. For the SSPC system (section D) difmrent methods to quantify an overall condition have been investigated. C.3.4.5 Conclusions The distinction of ditmrent discontinuities or discontinuity sets and the determination of the characteristic orientation, spacing and parameters describing the shear strength can be best done by a studied assessment. Discontinuities within an exposure and within a geotecbnical unit should first be grouped visually into sets. The discontinuity properties and parameters of each set can then be measured at an easy accessible location. Geological and structuml geological approaches can be used to determine these properties and parameters at locations where t;!].e rock Jl!-~~,js not ~· It sh~~t~~, ~,~,that these meth~. do not "·~~t. in highl~ ac~curate values because the variation of properties and parameters in most rock masses is large. This implies that a very high accuracy in determining parameters in an exposure is mostly not necessary. In the 'initial point rating' system (eh. C.4) the methodology according to 'Th.ylor is used for the overall spacing of a number of discontinuity sets in a rock mass and a weighted mean is used for the overall condition. For the development of the SSPC system (section D) various options are investigated to quantify the overall spacing and condition of a number of discontinuity sets in a rock mass. (48) In the 'initial point rating' system (eh. C.4) the parameter calculated following Taylor is multiplied with 25 to achieve a point rating for the spacings of a number of discontinuity sets in a rock mass. For the SSPC system a comparison is made between different approaches to calculate a quantitative parameter for the spacings of a number of discontinuity sets in a rock mass of which one is calculated analogous to Taylor (eh. 0.1.3.3). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 78 C 3 Pan:uneren in rock & UI"-.OCOI " " Hi !!. V to we:atl1erinj;! state. The values "'"'~'v"·""""" are studie-d guesses. The Method of excavation more Jlll~'""'""'-""'""'- 100 ru.s:coJntumltles if high stress levels occurred in the natural u.e,,e!O>pm:ent Table 8 shov.'S Poor conventional m to stress levels and therefore show fewer mechanical discontinuities than an excavated in the same rock mass. The excavation of a rock mass by hand and and thus Thble 8. Laubscher, 1990). for method of excavation also a .tu:asrm~ can cause severe damage of the rock mass et al., Rosenbaum et al., The shock wave from the delton;aticm to '-'H,;.•ALUJO\ rock masses 8) and Romana (1985, 1991) was thcmgjht have been and the effect5 page weath1~rlrll! {50) of criticism is about this chamcterization sllmdard is therefore discussed in system~ has on a rock mass are described in chs. A.2.4 and C.3.2.2. nr!'oi,P.?'l>! described. a COJ:lSiderable amount The om;sib1e noPlllcement of the. British Standard a newer V. (Sl\ In the research area no difference has been observed between natural exposures made exposures created hand. The due to excavation wit."! hammers is, however, considerable (52) discontinui.ties vu~;wun:tg of a river) and D.L4). ~v··~~·•a<~ in the research area, but no evidence of these effects could be found. The condition of r(mg,tme:ssl was found tD be of the type of excavation D.l.4). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C PARAME'l.'ERDEFlNlTlON AND lNl'l'lAL POINI' lUl'lNG SYSTEM 79 Blasting techniques have changed over the years; the number of boreholes blasted in one round and the size (diameter and length) of the boreholes have increased. Also, where in the past blasting was directed towards creating gasses (slow detonation explosives), nowadays blasting is more directed towards creating a shock wave (:fast detonation explosives). A rock mass will have most open discontinuities in the direction of the free face. A blast creating gasses will therefore work more in the direction of the free face than inwards into the rock mass. Shock waves work in all directions and therefore in more recent excavations, the rock mass is more damaged in the direction smay from the free face than in older excavations. In this research old :fashioned blasting (creating gasses rather than shock waves) has been incorporated in the class for pre-splitting and smooth wall blasting as the results are comparable. In the 'initial point rating' system (eh. C.4) the different classes and the values are partly based on the work of Laubscher but modified as described above. The values for the additional classes are studied guesses. In the SSPC system (section D) the values for all classes for the method of excavation have been determined by analysis of the data obtained in the research area. C.3 .6 External influences C.3.6.1 Surface run-off water Water run-of£<53> over a slope and through the near surface of a slope can lead to instability, but it is not proposed that surface run..off should be treated as a separate parameter in a classification system. All slopes used for referencing the classification system have been subject to rainfall and surface run-off water and thus the calculation method, parameters and weighting factors in the classification system include the influence of surface nm..;off water. For· example, surface nm"off water will have a lmger infiuence on a slope in a rock mass with a small block size than with a larger block size because smaller blocks are more easily flushed away by the water. Block size (discontinuity spacing) is a parameter in the classification system and because the classification system is referenced against existing slopes, and existing stability, the weighting factors for discontinuity spacing incorporate the influence of surface nm-off water. C.3.6.2 Snow and ice The influence of snow and ice in the weathering of a rock mass is discussed in eh. A.2.4. Snow and ice may also block seepage from the discontin.uities where discontinuities are outcropping at the slope face which may lead to water pressures in the discontinuities. Additionally snow and ice add weight to a slope. Snow and ice do not commonly occur in the research area, however, during the fieldwork in 1992 it snowed, followed by the failure of some small slopes. It can therefore be assumed that the slopes have been occasionally subject to limited amounts of snow and ice characteristic for the Mediterranean climate. Hence, the classification system and weathering parameters ·inoorpomte the influence on stope stability caused by these tifiiited"qtilmtities of snO\Vano ice, ·oecause existing slopes and existing stability are used for calculating the weighting factors in the system. A separate parameter is thus not necessary for snow and ice. C.3.6.3 Rock mass creep and stress relief Rock mass creep and stress relief can lead to new cracks in intact rock, develop integral discontinuities into mechanical discontinuities and open existing discontinuities. These efkcts are included in weathering (eh. A.2.4). Creep movements and stress relief can also cause displacements along discontinuities, resulting in non-fitting discontinuity planes (eh. C.3.3.2.6). This is included in the characterization of the shear strength along discontinuities. Large movements of the rock mass in a slope may cause an increase in the slope dip angle leading to slope instability. In a classification of slope stability this can be incorporated by taking the slope dip angle that will exist due to rock mass creep and stress relief at the end of the engineering lifetime of the slope. For these reasons a separate parameter for rock mass creep and stress relief is not necessary in the classification system. (S3) The presence and pressure of water in discontinuities in the slope and the influence this has on slope stability and how it can be implemented in a slope stability classification is already discussed in eh. C.3.3.7. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ C.3.6.4 External stresses External stresses working on the rock mass in which a slope is or will be excavated can make a slope unstable. External stresses do not originate in the rock mass of the slope, but are, for example, stresses due to a high hill or mountain behind the slope or tectonic stresses. Generally, it is impossible to determine external stresses without stress measurements and their influence on the stability of a slope can mostly be only quantified with detailed numerical or analytical calculations. Therefore external stress influence cannot be included in a classification system<54> and consequently the classification system developed in this research cannot be used for slopes in rock masses that are under infiuence of external stresses. C.3.6.5 ror The engineering lifetime, example 50 years, of a slope is more than sufficient to allow some types of trees to develop to full growth. Root wedging will dislodge blocks, allow water infiltration, etc.. The prevention of such growth falls within the province of slope maintenance, which is not dealt with in this research. C.3. 7 Summary - parameters in rock slope stability The review of parameters important in rock slope stability and to be included in a classification system for rock slope stability results in the following conclusions: Intact rock strength: Intact rock st:rengtb. in the classification system for slope stability can be established with a 'simple means' test in the field. A cut-off value should be used above which the influence of intact rock strength on the calculation of the stability of a slope is constant. Susceptibility to weathering: In the 'initial point rating' system (eh. C.4) susceptibility to weathering is incorporated by estimating the time it takes for a rock mass to go one degree down in weathering according to BS 5930 (1981). In the SSPC system (section D) the expected degree of weathering at the end of the engineering lifetime is estimated. Discontinuity shear strength: Roughness of discontinuity walls, alteration of discontinuity walls, type of infill material, and the occurrence of ka:rst are described in classes that can be established by visual Observation of outcropping discontinuities. Determining discontinuity properties and parameters: Discontinuities within an outcrop and within a geoteclmical unit should first visually be grouped into sets. Discontinuities with characteristic or mean properties (e.g. orientation, spacing, and properties describing the shear strength of each set) shookl,be~ whereafter these~~ean: be measured at an cmy accessible location. Single parameters describing the overall discontinuity spacing and condition of a number of discontinuity sets in a rock mass are described in respectively chs. C.3.4.3 and C.3.4.4. Exposure and slope specific parameters: The degree of weathering and the method of excavation of an exposure and a slope are established and are used respectively to correct for local and future weathering, and to correct for the damage due to the method of excavation with which an exposure or slope has been made or is to be made. External influences: No parameters are used for external influences such as surfilce run-off water on a slope face, snow and ice influences, rock mass creep and stress relief, external stresses, and vegetation. <54> Most slopes in the research area are in a rock mass that is unlikely to be under influence of external stresses and those few slopes in a rock: mass that might be under influence of external stresses have not been used for the development of the classification system. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 81 C PJ'lJJ.AMETERDBFlNITION AND INITL4L POINT RA:Illv'G Si"Sl'EJ1 4 1 The concept ' eK1JO~tUfle:' in an exposure are converted in.to "'"' ""11'1""'t"''"" par·ameteJrs for of of the of determines the 34 the form used to Results The all slopes. 'Ihe results are presented intervals of 10 points and the number of slopes obtaining of estimated stability as oeJ:ceJltru:!e the 'initial point unstable slopes from iS5l The estimated classes are described in Table 5 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ c. 4 lnmal point rating system 82 ., INITIAL POINT RATING SYSTEM .. REFERENCE ROCK MASS ASSESSMENT ~--- ... ~ . ------------------.. . .-----. '· ·' '· r---------------------------------1 ! 'EXPOSURE' ROOK MASS ! ::::::I ::::::J I ~~·~a; ' malhcd of 8XCIMIIIIon clegrM of Wlllllwrtng 'REFERENCE' ROOK-MASS (WIIhout~) ::::::1 : :::: : i ::: ::: iparamelilln: i { ::::::: 1- ~(SPA) ldfaconllnullles Olteldallon 1-+--.~::f*::.:....;· I cllaconlhlltiee ~(CD) . ... ! Ol1ldatlon = l I ~H--1...--.. ~.-i 'I ~ H---11~""-.:..._......;' l !Intact luck~ l i : : Intact luck atreriQih (IRS) ! ~to-~{SW) !tll~to~ ! 1-------·-------------------~ a $ a a a 8 S a U a 8 8 8 a 8 a B a a a a I ! a a a a • a a a a a S 'REFERENCE' ROCK-MASS I 1 I I J 1 !I' i i ~ J l'llllng: ~! ! I I I ,! RFR•(SPA+OO+IRS)"SW I I I r-+-1 !-+-1 L----·-----------------·----------1 D r-------------·---------·---------, ! ' l---------------------------------1 a a a a a a a a a a B U a a 8 a a a a 8 a B a a a SLOPE STABIUTY ASSESSMENT Fig. 34. Flow diagram of the concept of the 'initial point rating' system. visualy estimated stability (number of slopes in brackets) Ill stable • cla8s 1 (1 08) • ID • unstable wllh small problems In near future • cla8s 2 (48) Ul1ll'able wllh ~age problems In near future • cla8s a (18) unstable wllh small problems at preMnt- clan 4 (20) a . ~.wlth~..problemaapNMnt--.sM -"#. Percentages are from total number Of sbpes per visually estimated stability dus. 10 5 16 46 56 rating(-) 75 Fig. 35. Results of 'initial point rating' system ·with optimum weight factors based on 250 slopes (Definition of visually estimated stability classes - Thble 5, page 52). B Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ can be contributed to: measurements as in the estimated stability w.11ten1er these \\-ere '"""''"""''"' The 'initial point area. Therefore the abandoned. In <SOl system did not to a satisfying assessment of the stability of slopes in the research stability assessment has been a point rating classification for ""'""r"""' approach for a slope stabHity assessment is designed. The same effect is also pre,sent in some of ilie classification systems for as discussed in eh. B. LL Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ I DATE: I 119 1 TIME: : hrl exposure no: WEATHER CONDITIONS LOCATION map no: Sun: Jcloudy/fair/bright Map coordinates: northing: ~~----4-----------------------------~ Rain:Jdrytdrizzle/slight/heavy easting: LOGGED BY: METROD EXCAVATION (ME) Of DIMENSIONS/ACCESSIBILITY hand-made : R·~ Size total exposure: (tick) ~ ~ ~: =~~~~~ b X~f~tor ~ tiG -----;- ~ti~?'1s! lntfi resu t:di fractur!! act ~f: crUSI'I;t n act roe~: (m) L: Jh: jd: 11-lllllf-Df:l-led-:-on-t~h~is~f-,orm_:_ _ (m-)-+l-:----+j:h:--=-----1-Jd.,...:-----1 ~A-,c.;..;ces'--s-:-:ib-::i~l-:-it-y-:------+-----poo-'--r-/'""fa-,i:-r-/good __.L,--_ _ _~ V:fi6 0:97 i=u : • FORMATl ON NAME: DESCRIPTION (BS 5930: 1981) colour I grain size I ~~f~et~i~~~~~ I weathering I I NAME strength ---t------~-------------t------~----------~-----1 I J I I I I I I I ___· ·-· · ·-·"·_a_=b_•_dcl__tng ...;;_···-:·~·~~~-r .-·.-::-,+--·w-····-"-··+·-···-·_2_.-+_. •_ ..• _3_.-+-·-·_4--ir-··-····_s_·····-~w~~~~t 1 ~, 1-:: : -=-::-CON_,d-:-i;-:-~...,~:,....~-~~_s Dip (degrees) (tick~ 1-D..is:....c-on-t"""'i_nu_i_t_y_s_pac:_i-,ng-(-DS->-----(me...;;_tr-e-s-)+----+---+---t------11-----f~~t~~red ately Ialong strike (meters) g y persistence ~---:-___;;~,--------~----11-----t-----t----+-----+----lc~letely (meters) Jalong dip r ~~:~~ : • : • : • CONDITION OF DISCONTINUITIES Roughness large scale (Rl) Roughness small scale CRs) =rn, ·~~Ly wavy ~t1ahtly curved striight tr gn area 8.2 x~~2 8 m) Fig. 36. Initial point rating - exposure classification fonn. SUSCEPTIBILITY <SW)TO WEATHERING (ti!k)year year > ~8 ~5~ sample '1'10: Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE S'£4B1Ll11' PRORABlUTY CJ.A5'SIFIC4.1WN- SSPC' D SLOPE STABILITY PROBABILITY CLASSIFICATION. - SSPC 85 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOP13: SI:Wll1IY J!'ROIMBJL!'J''Y CUSSlFJCAT!ON ·· SSPC 87 1 The analyses fur the development of the SSPC CoJrrce]pt of the SSPC system D.l.l) Or1.entatm~n dependent stability D. 1 'Sliding criterion' D. L 2 .1) D.l on tl1e results obtained The analyses in. chs. D.l and D.2 result the SSPC which is presented of the &)'stem are presented and the results of the SSPC system are compared to other mass classification In the same also the merits of the rock mass strength parameters calculated with the SSPC are evaluated and to other methods to calculate rock mass Examples of the application of the SSPC system to four slopes in. the research area are given in eh.. D. 5. In two examples the the SSPC are compared to analytical and numerical calculations of the "'""''.n"" of The the SSPC system and the tollow in. eh. DA Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SSPC sysrem Ll rock mass classification det>en.dellt on to the assessment of 2 3 0.1.1.1 to ex:cava.tim.L 3 ch~ira1;tex:ize:d in an exposure pru:arnlel:ers nlerun.tn~ in the exposure such as Vl!el'ltllt~rmr!:! """'"'·"v'tL•u•u n:aeti1oo used to the exposure, are then cornpe:ns.;J;ted mass' to that of the theoretical mass that um.ue11ce zones cor.np~)llSlitlcr:n reflecting and excavated 37 shows exposures with various 38 shows a flow slope and the rock mass. 37. Sketch of exposures with vari.ous concept of the 'reference rock mass'. is done de~~re1::s that are of tmt"'<'lt~rl'!n''"" Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STA11lU1Y PROlMBILlT.t ClASS1FlC'4TION ~ SSPC 89 ;11 'Slope specmc parameters: !- Method of excavation to be used Expected degree of weathering at end of 1 [ engineering lff&-time of $lope I ! ·I of I i I , ~ . ~-·~ ·---··--·1\tOP~or--·--··-·--·~-···-·~·· ISlope rock rnus parame~ significant for slope ~bltlty: I I .. DISOOI'ltlnultles: Ol'ientat'ion and sets (spacing) er single Dlsoontinulty properties: roughness, infiJ, k.ru'st 1 i ~ Material properties: strmgth, wsooptlblllty to weathering Orientation I· i 38. Parameters in the slope assessment slope in. assessment is mass' of the prumn.eters the influence of future weathering within to of ~tmhifir.., D.L1.2 Determination prumneters & weighting resulting failure modes {plane '"~'u"'·"'' Slope failure mechanisms such as shear toppling bucld.ing) are related and are on the onentati<Jns Slope fuilure that are usually not related to the of the Examples of these latter causes are: intact rock \\later (5'1) The Rock Mass' and Rock Mass' are !he same if an is not considered. Then it also not necessary to use !he exposure and as these are tile same. is examined and fl.trure "'"'·orn,,,,.u"' of excavation and we:~m:3Tirlll' tmt~llic>d are often des,;gn<'d for a lifetime of about 50 years. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D.LL3 the 1) the relationships tlet~m the panu:neters and the rm;seJat in the data set, not be detected with a ...,...,,.._...."""' a function relating all possible rock mass and en~gm~~nm.g metho:d is that the of _..,...., ......,u If et at, (5&) Optimization is the art of obiammg the best result under circumstances (F.ao, If an amount of the data inhibits a c.onsequent error the neural network wm the factors until fit the data, fue individual characteristics and not the structure of the data set are fitted, weJi12.11!txmr £~ctors that are darn set tsee page l59l Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SJXBJLITYPROBABfU'ITCIASSlFlCA.TlON- SSPC l1g.....39~ D:if1re-reJat and transport of rock blocks failure in the middle. 91 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 92 D.l The development tf the SSPC system D. 1.2 'Orientation dependent stability' - sliding, toppling and buckling criteria Failures in a rock slope may depend on the orientation of the slope and the discontinuities. These are mostly related to shear displacement along a discontinuity. The main parameters governing this type of firilure are the orientation of the discontinuity in relation to the orientation of the slope and the shear strength of the discontinuity. The parameters described in the field that are likely to have a relation with the shear strength of a discontinuity are the parameters describing the roughness of the discontinuity (Rl and Rs), the alteration of the discontinuity wall (Al), the infi.ll material in the discontinuity (/m) and the presence of karst (Ka)<(l.)>. This chapter investigates whether these parameters tngether with the orientation of the discontinuity can be related to iillure modes of slopes due to shear displacement and whether this results in criteria that can be incorporated in the SSPC system. Three difrerent modes of slope firilure related to shear displacement along discontinuities are investigated: sliding, toppling and buckling. The relationship fuund fur iillures related to 'sliding' are such that a 1 sliding criterion 1 can be defined that relates the maximum dip of a discontinuity and parameters describing the condition of the discontinuity in the field (eh. 0.1. 2.1). This 'sliding criterion' has been verified with field and laboratory test values fur discontinuity friction and with friction values fur discontinuities fuund in the literature, which confums that the 1 sliding criterion 1 is properly defined. Analogous to the 'sliding criterion' a 'toppling criterion' is defined (eh. 0.1.2.2). A similar criterion could not be developed fur buckling. This is in agreement with field observations as buckling as a cause of slope firilure is seldom fuund in the fieldwork area. Almost none of the slopes are high enough to cause buckling in the rock masses of the slopes (eh. 0.1 .2. 3). The sliding and toppling criteria are incorporated in the SSPC system to predict the 'orientation dependent stability' of a slope (eh. 0.1.2.4). 'Sliding criterion 1 0.1.2.1 The 1 sliding criterion 1 <61 > relates the orientation of a discontinuity that allows kinematically sliding, to the parameters describing the condition of a discontinuity. The relation found in eh. 0.1.2.1.1 is refined in eh. 0.1.2.1.2 by examining ditrerent parameters and values used in the description of the condition of a discontinuity. The 'sliding criterion' with refinements of parameters is presented in eh. 0.1.2.1.5. 0.1.2.1.1 Initial 'sliding criterion' Failure in slopes related to sliding along a discontinuity means that the driving furce along the discontinuity is larger than the restraining shear strength of the discontinuity. In the 'initial point rating 1 system the shear strength is described in the field with the parameter rr:. rr:: is a multiplication of the parameters fur the roughness of the discontinuity (Rl and Rs), alteration of discontinuity wall (Al), infi.ll material in the discontinuity (/m), and the presence of karst along the discontinuity (Ka). The values used :for the parameters are those included in the exposure characterimtion :form of the 'initial point rating' system (Fig. 36, page 84). The driving forces in the direction of the slope.dip are .related to the (apparent)dip.of.the discontinuity in.thedirection of the slope dip~ The larger the driving force is, the more likely it is that a block of rock laying on the discontinuity will slide out of the slope. The discontinuity dip in the direction of the slope dip (/J) is defined as follows: if: I 6 I< goo then: p = arctan (cos 6 * tan p = apparent disctmtimlity dip 6 = dip directionsloJH - dip dip~) [14] m direction slope dip direction~ <60> Spacing of discontinuities was not expected to have an influence on the shear strength of a discontinuity, which was confirmed in this research as no influence of spacing on discontinuity shear strength was found. The influence of intact rock strength on the shear strength along a discontinuity (eh. C.3.2.1) is discussed in chs. D.l.2.1.1 and D.1.2.1.2, alteration ofdiscontinuity waU. <61> The 'sliding criterion' is published in Hack et al., 1995. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SFABlLITY PRO.BABlLITY CLASSIFICATION- SSPC 93 Fig. 41 shows the relation of the initial discontinuity condition parame*~ /~ ter (K) against {3 for 1 day-light- ~ ing' <62> discontinuities in slopes that .J" show no signs of present or future ~ # -at slope fuilures (stability class 1, .-" Th.ble 5, page 52)<63>. In Fig. 41 a lu *v* + vague relation is visible; fewer dis'*' ~ */"+ + continuities plot in the lower right ,..-":!:-"",."' + + ~ ~ corner of the graph if the brst para*""'/ meter is not included in the calcula+~ + tion of 'IC. For {3 between 30° and ... ,. •• 01• dot) 80° it is possible, by visual examin+ + ,..-"" + ation, to draw (by hand) a boundary *',.,.i',.,.--" + kardcwllhlantfactor condition line below which only five discontinuity condition values for *~ 0~----r--L--,----.-----r-----.----.-----.-----r-~ 0 10 20 80 40 50 80 70 stable discontinuities in non-karstic ~ (• appaNnt discontinuity dip In direction slope dip) {deg) rock masses are present. This boundRock types in whldllant phenomena are ckMIIopecl are plotted twice. TC Is conaldered Wllh and WllhOut kanlt factor. ary line is considered to be the 1 sliding criterion' . In Fig. 41 many Fig. 41. Discontinuity condition parameter (n::') vs p, for 'day-lighting' discontinuities other boundary lines would have been in stable slopes (stability class 1, Table 5, page 52). possible, but the linear relationship between 'lC and {3 as indicated in Fig. 41, is the most simple possible boundary. lu * ** ** ** * *' * * ** lOA * * * *!: * * ** * * * * i* * *~ ,.~// ,/ .-"' ** ~///* * //.::::~ .-"' I Discussion A significant number of discontinuities in brstic rock masses have a value for the 'lC parameter plotting below the boundary line in Fig. 41. It seems therefore that either including brst in the discontinuity condition parameter is not a proper approach or that the reduction of 'lC by the brst parameter is too strong. The discontinuities in brstic rock masses for which the 'lC parameter is calculated without the brst parameter, plot above the dashed line, except for one value<64>. Four of the five<65> values for non-brstic discontinuities plotting below the dashed line are cemented discontinuities in limestone (Tg21) (see below- cemented/cemented infill). Fig. 42 shows the initial discontinuity condition parameter (without considering the brst parameter in the calculation of the discontinuity condition parameter, 1q for di.f:rerent rock lithologies. The relation between 'lC and the apparent discontinuity dip in the direction of the slope dip does not show a dependency on the type of lithology. <62> 'Day-lighting' of a discontinuity means that the discontinuity has a dip less than, but in the same direction as, the slope dip, and is outcropping in the slope (see also glossary, page 241). <63> The accuracy of measuring dip and dip directions is such that the accuracy of dip and apparent dip values is not less than 5° (the accuracy of field measurements and derived data is discussed in more detail in eh. D.2.l), therefore only discontinuities are included for which applies that dip51opo > p + 5o. If the difference is less than 5o the dip.Jopo and p (apparent discontinuity dip) are assumed to be equal and the discontinuity plane forms the slope. The latter are obviously not a cause for slope instability due to sliding and cannot be used to determine a relation for sliding. Also are not included discontinuities whose apparent dip is almost vertical, e.g. discontinuities for which the apparent dip (/1) > 84 o. <64l The karstic discontinuity at fJ = 61 o is a near vertical discontinuity with a dip of 85° (in slope 9ln/9.1!2; discontinuity orientation 078° /85°). For (near) vertical discontinuities the accuracy in measuring the orientation of discontinuity and slope becomes very important. Small inaccuracies will lead to large differences in the apparent dip. Therefore it is not unlikely that a small error in the measurement of the orientation causes this discontinuity to plot below the dashed line. <65) At p =55, 60, 62 and 69° for slopes respectively: 9l/6/l/s3a, 91/6/l/s2 (2 x for two discontinuity sets) and 93/13/1. The fifth non-karstic discontinuity at fJ 75° has an apparent dip that is just over 5° less than the slope dip and is likely to be slope forming (slope 93/15/1; discontinuity orientation 128onso with slope orientation 142°/80° results in a difference between dip,~ope and fJ of 5.4 °). = Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SSPC system Fig. 42. TC without the karst parameter in the calculation of TC vs discontinuities in stable for different rock materials brackets are average estimated llLaCt fOCk ou''"S'"'J· f3 for in-between Refinement of initial values of the 'initial is calculated values de!;cntption of 'gouge scale roughness will for the shear the dis:coJn.tU:mity the calculation fur this class of infiH material should be changed to a small scale roughness parameter of 0.55 ',·"'-"".............,.,possible a pruranle'((;r criterion' for the lithologies represented in the criterion' , have value is -.u'u'.!'''""' to 1 («.>) For the four values of 1 then. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl'ABlUTY PROBAIJlUTY CLASSIFlCATION • SSPC 95 these discontinuities will plot above the 'slidiug criterion.'<61). This leads to the conclusion that the value for infill material for the class 1cemented/cemented infill' has to be 1.07. Whether this value determined for a calcitic type of cement, is valid for other types of cement or cemented infill (e.g. quartz, salts, etc.) could not be determined. Many types of cement (e.g. quartz, etc.) are, however, so strong that shearing through the cement or cemented infill will not occur under the level of stresses occurring in slopes up to 40 or 50 m high. Cement types that consist of an easily dissolvable and deformable material, like salts or gypsum, are unlikely to be permanent during the engineering li~e of a slope in most climates. Also their easily deformable character means that the discontinuity might move, not actaally by shear but by deformation of the cement. Therefore it is a safer approach to characterize these latter types of cement as non-cemented infill. Ko.rst (Ka) mretrospect it is clear that the brst pammeter (Ko.) should not be used to calculate 1C in the form as applied in the 'initial point rating' system. The parameter for b.rst is, in the 'initial point rating' system, dependent on the frequency of the occurrence of brst along the discontinuity planes. For a 'sliding criterion' this is obviously not relevant as ooly a single discontinuity is enough to make the slope unstable. Moreover it can be questioned whether the parameter for brst should be dependent on the size of the solution holes along the discontinuities. Although brstic solution along discontinuities reduces the contact area between the two sides of a discontinuity, the normal stress on the contact area increases linearly with the reduction. of the contact area and the shear strength resulting from friction remains the sameC68>. The contribution to the shear strength from the discontinuity cohesion reduces linearly with the reduction in contact area. Most discontinuities do not contain cement or cemented infill (causing real cohesion). This leaves discontinuities with an apparent cohesion that could have been inftuenced by karst. The discontinuities likely to show apparent cohesion. are those with a small scale roughness of 'irregular/stepped'. Discontinuities in limestones do, however, seldom have a stepped surface but rather a plane or undulating surface forwbich the apparent:eohesimristow or :nommsmt. ·Forwth~temmns 1r1Stnrety thaftlleiiifiuence orkarst is considerably less than initially expected and the values have to be increased accordingly. Discontinuities with karst features in stable slopes will not plot below the 'sliding criterion' if the value for the karst parameter is fixed at 0.92<69> (independent of frequency of occurrence and independent from the size of the solution holes). The 1C (condition of discontinuity) parameter should therefore be calculated including a parameter for karst along discontinuities that should have a fixed value of 0. 92. Persistence All discontinuities in unstable slopes that are prone to sliding according to the 'sliding criterion are persistent, but also other discontinuities in slopes in the research area are virtually always persistent. Non-persistent discontinuities or discontinuities that abut against other discontinuities are very seldom. Because of this the influence of non-persistence of discontinuities could not be investigated. It is suggested that non-persistent 1 <67J The influence of cement or cemented infiU on the friction along a discontinuity as calculated with the 'sliding criterion' can be compared to the Q-system (Barton et al., 1990b). In the Q·system the difference in friction angle between a discontinuity with tightly healed, hard, non-softening, impermeable filling (i.e. quartz and epidote) and a discontinuity with unaltered joint walls, with surface staining only, is between 4 o and 7°. The first value is for a rough undulating surface and the second is for a polished planar surface (roughness descriptions refer to small and intermediate scale roughness in the Q-system, Flg. A 97, footnote 147). In the 'sliding criterion' the difference between a discontinuity with cement or cemented infill and a discontinuity with no infill is between 4 o and 2.5 o, if a value of 1.07 is used for the class 'cemented/cemented infill'. The first value is for a straight (large scale roughness) rough undulating (small scale roughness) surface and the second for a straight polished planar surface. Thus, the value of 1.07 in the 'sliding criterion' results in a good correlation with the Q-system for the rough discontinuity surfaces but less for more smooth surfaces. (68) The shear strength could increase if, due to the larger stresses, the friction parameters change. This effect can occur, by example, for a weathered discontinuity wall where, due to the larger stresses, the penetration of asperities into the weathered zone reaches less weathered material resulting in higher friction angles. For pure limestones, however, no weathering of the discontinuity wall material has been observed. This is different for limestones that also conmin clay minerals because, as the limestone dissolves, the clay minerals may stay behind as a coating on the discontinuity wall. (69) values for the karst parameter to obtain equilibrium for the .karstic discontinuities which plot below the 'sliding criterion' in Fig. 41, are: 0.60, 0.74, 0.87, 0.88, 0.92, 0.89, 0.72, 0.90 and 0.90. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 96 D.l The developmenuf th£ SSPC system discontinuities should be treated as stepped discontinuity planes. The step has to shear before movement can occur, a mechanism comparable to the breaking of intact rock non-persistent discontinuities(7o). m Conclusions m The foregoing results the fOllowing conclusions on the refinement of parameters and values for the condition of discontinuity parameter (1t'): 1 If the infill material of a discontinuity is m class 'gouge > irregularities' , the small scale roughness parameter should be 0.55 (minimum possible value). 2 A parameter fur the alteration of the discontinuity wall is not necessary a fiillure criterion fur slope stability. 3 The value used fur cemented discontinuities with bonding between the discontinuity walls, or fur discontinuities containing cemented infill with bonding to both discontinuity walls should be 1. 07 fur discontinuities containing a calcitic type of cement. Values fur other types of cement could not be established, however, the fOllowing approach is llirely logical. If the type of cement is very strong the discontinuity should not be considered as a discontinuity mthe classification system. If the discontinuity contains cemented in1ill from which the cement easily dissolves, the loose material that may remain after dissolving of the cement, should be accounted fur as a non-cemented infill. Also if the cement or cemented in1ill easily deforms, the discontinuity should be regarded as a discontinuity containing a noncemented in1ill material. 4 The value fur the karst parameter should be 0.92, independent from the size and frequency of the karst phenomena. 5 It is suggested that non-persistent discontinuities should be treated as stepped discontinuity planes. m D.1.2.1.3 Correlation of the threshold friction values of the 'sliding criterion' to test and literature friction values The 'sliding criterion' is based on the assumption that the friction angle along the discontinuity plane, is equal or larger than p ( = apparent discontinuity dip the direction of the slope dip). This allows fur comparison of threshold friction values round with the 'sliding criterion' with test and literature values (Hack et al .• 1995, appendix ill). The correlation round between the threshold friction angles determined with the 'sliding criterion' and the friction angles obtained from testing or round m the literature confirm the correctness of the sliding criterion' and the discontinuity condition parameter (7C) describing the discontinuity shear strength. m I D.1.2.1.4 Reliability of friction angle values based on 'sliding' criterion The reliability of the 'sliding criterion' fur estimating friction values along discontinuities from field descriptions can be perceived from a visual examination of the data and graphs<7 1>. There are a total number of 155 characterizations of discontinuities that kinematically allow sliding from about 100 slope stability assessments. These have been carried out by difterent persons mdifterent years and it can be assumed that a consistent operator bias is absent mthe data set. The 'sliding criterion as defined above, is based on 98 % of the data plotting above the line. In Fig. 43 two other criteria fur sliding are indicated (at 95 %: upper dashed line and at 99 %: lower dashed line). The influence these changes have on the friction angle is marginal and gives a change of a few degrees only. The difterences can safely be neglected for an empirical field classification system and they lie also I <70> This approach is comparable to the treatment of non-persistent discontinuities in the Q-system (Barton et al., 1990b). In the Q-system non-persistent discontinuities are treated as continuous discontinuities, but the parameter for joint roughness is taken higher. The friction values found by Barton for non-persistent discontinuities are approximately 4 o (for a discontinuity with unaltered joint walls, surface staining only) to 8° (fur a discontinuity with softening or low friction clay coating) higher than the friction found for rough undulating but persistent discontinuities. In the 'sliding criterion' a straight (large scale roughness) non-persistent (which is thus classified as having a small scale roughness of 'rough stepped/irregular'). discontinuity without infill hasa friction angle about 10° higher than a straight rough undulating persistent discontinuity, while a straight non-persistent discontinuity filled with softening fine material has a friction angle about 6 o higher than a straight rough undulating persistent discontinuity filled with the same material. (7IJ It is also possible to perceive the reliability from the probability analysis in eh. D.2.2. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STABJUTY PROIJABJUTY CI..ASS1FIC4TION- SSPC within the measuring accuracy of, for example, a shear test. '1". .· 1 • • .• ••- ~ The reliability of the 'sliding criterion' as an estimate for shear friction parameters is, however, dependent on the .accuracy of the description of the discontinuity. During the research it was found that although different persons made the descriptions, these rarely differed more than one class. For example: instead of describing a surface as rough tmdulating it was described as smooth undulating. The dif&ence in the friction angle is then 3 o (rough undulating: 53 o, smooth undulating: 50°; large scale roughness straight and no infill and karst). Obviously if for all parameters the :;:::: ..... t I o.a OA fJ7 * 1 ** * • • •* • • * * * * * * I*• •* * * • .!.. o02 "" 20 eo 40 80 p (- apparent chcontlnulty dip In dlrecllon alope dip) (deg) Fig. 43. Discontinuity condition parameter (1t) vs fJ for 'day-lighting' discontinuities in stable and unstable slopes (visually estimated stability class 1, 4 & 5). •• nn class is consequently taken one lower, then the ~ in friction value for the discontinuity becomes larger. This has, however, not been observed to happen, rather the differences were randomly a class lower or higher for the different parameters, which resulted in approximately the same resulf!"15t Viltles~ m.~~~ rr: 0.1.2.1.5 •• Discussion and conclusion The correlation found between the friction angles determined with the 'sliding criterion' and the friction angles obtained from testing or found in the literature confirm the correctness of the 'sliding criterion' and the discontinuity condition parameter (l'L) 4escribiog the discontimJity shear strength. * * * * * * ** :. X * i** * * + *if:§. **+: + : : :*_..////+ ** i!:o-"'•+*..; * •* : + *!* ~~--// + * **++lE+ /4 S * + *X* +* *+ **+ I++ * *~"'X + .+, #'.,.,.." + I g,!l ji : + ' + * -.;. . .-" + § li~ +*+ + + +..., .... + +X ...... ~:!Y~~ •* _,/ ~.)C'~ + X. * .... + Fig. ·· 44(72) sh.OY.'s 1C ver8us fJ for OA * •.. 'day-lighting' discontinuities in both TC=0.0113: li (;indeg) stable (class 1) and unstable (class 4 0.2 ~~~~~--~~~ r--;*:;---. ... -.-:...-:(clus-;---::1::-)--! and 5) slopes. Some discontinuities .!. ~.4 .fl" X unlllable(clus4) from slopes with visually estimated ~ ~/ + unataiJie (class 5) stability 4 and 5, plot below the 0~----r---~r----.-----r----~--~r-----r-~-.~ ro 20 ~ 40 ~ ~ ro 80 0 dashed line and it is likely that slid~ (• apparent discontinuity dip in direction slope dip) (deg) ing is the cause of the instability of Fig. 44. Discontinuity condition parameter (1C) with refinements vs fJ for 'daythe slopes containing these disconti- lighting' discontinuities in stable and unstable slopes (visually estimated stability nwtles. The discontinuities in classes 1 , 4 & 5). unstable slopes resulting in points that plot above the dashed line can, however, not be the cause of sliding instability in the slope and other causes (like toppling, buckling, etc.) have to be investigated for these slopes. (12) :t~/~~ The two discontinuities in stable slopes which plot below the 'sliding criterion' are discussed in footnotes 64 and 65. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SSPC system < 90" > fJ + 50 D.1.2.2 del:>en:drru~ on the orientation the slope and the discontinuities is toppling. 'Ibprplit1g of is 45. Blocks on the surtace of the to the forces of the blocks behind. Int•'·'!"""r after with D.1.2.2.I Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STABILITY PROBMJILITY CLASSIFlOO'ION- SSPC 9!J programmes can incorporate the rotation and crushing in the toppling mechanism (UDBC, 1993, 3DEC, 1993, etc.), however, require detailed slope and rock mass parameters. Toppling can be formulated following eq. [16] (Fig. 45) if only the friction component of the shear strength along a discontinuity is considered (Goodman, 1989). This holds only for discontinuities dipping into the slope, and not for vertical discontinuities nor for discontinuities dipping in the same direction as the slope(73l. toppling m,,. if: -90" 'P > (90"- + dip.tl«a &wally + m,., . ~~y) + di,p.,. > , [16] tp =.friction along tliscontmuity plmte Only if eq. [16] is satisfied toppling can occur. Equation [16] forecasts toppling before it usually happens in reality because rotational deformation and crushing are not considered. If the dip direction of the toppling plane is not approximately opposite to the dip direction of the slope then the blocks at the side of the block prone to toppling will prevent toppling. Different empirically established boundary conditions are defined in the literature. The boundary condition formulated by Goodman (1989) is formulated as follows:(74) [17] 0.1.2.2.2 Discontinuity condition and toppling An apparent dip of the discontinuity plane in the direction opposite to the dip direction of the slope can be formulated: y = apparent discontinuity dip in directiM opposile to arctan [jcos (di,p directio11Biope - di,p direction~)! tM slope dip = [18] * tan (difJcliow ...,)] Fig. 46 shows the discontinuity condition parameter (n:') versus fJ determined with eq. [16] for discontinuities in stable and unstable slopes(7S>. The dipdiscontiauity in eq. [16] is replaced by y following eq. [18]. The discontinuity condition parameter (n:') has been calculated with the refinements as for the 'sliding criterion' (eh. 0.1.2.1.5). In Fig. 46 is indicated for all stable and unstable slopes whether the difl.erence in dip direction between the slope and the discontinuity fulfi.l the boundary condition formulated in eq. [17]. Analogous to sliding, a boundary line, the 'toppling criterion', can be drawn below which no values plot<'6l. For comparison also the 'sliding criterion' is shown. For a particular discontinuity surmce type with a discontinuity condition parameter (n:'), the fJ found via the 'toppling criterion' is higher than the value found for the same type of surface via the 'sliding criterion'. Rotational and crushing ~ts likely cause this di:fterence. Apart from one discontinuity<77), all values plotting oefowtlie 'toppling criterion' m1!lg."46 are within ilie,ooWiaarles set by eq. [17]:,""Tfieiet'Ore itis likely tb.ata boundary on the dip directions of slope and discontinuity is not necessary if for the discontinuity dip y ( = the cnJ The only form of toppling discussed is that caused by stresses originating in the rock mass in which the slope is excavated or will be excavated. Other forms of toppling, for example, toppling of vertical blocks, may occur if additional external stresses work on the rock mass, however, these are not considered in this research (eh. C.3.6.4). <74l In the literature also other lower and higher limits are reported, for example, 165° and 195°, or a differentiation in likelihood is used: for example, if the difference in directions is between 165° and 195° toppling is very likely whereas in the ranges between 150°- 165° and 195°-210° toppling may happen (both under the condition that eq. [16] is satisfied). In general, it is likely that the boundary is not absolute but that a gradual boundary should be applied. <7Sl Only included are discontinuities with y < 85° (discontinuities with y toppling according to the criterion formulated in eq. [16], see footnote 73). :<!: 85° are assumed vertical and cannot enable <76l Note that y is the apparent dip of the discontinuity in the direction opposite to the direction of the slope dip; the value is always positive. The 'toppling criterion' in this chapter is formulated as lP < -90" + y + dip~~ope. This in contrary to the 'toppling criterion' formulated in eh. D.3.3 which is, more generally, defined in terms of apparent dip of the discontinuity plane (AP): AP > oo for planes dipping in the same direction as the direction of the slope dip and AP < oo for planes dipping in the direction opposite to the direction of the slope dip. The 'toppling criterion' is then: lP < -90° - y + dip••. (77) Slope: 91/10/1002; (dip direction,.~ope- dip directio~) = 213°. This is just above the boundary condition of210°. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SSPG' system - i D.1.2 . 23 1111! llli "' 1111 Ill! lOO ' 1111! «>!1111 ll! !111 "' ~ lll!l .!'Jl illl 1111 "" $ $ ~~~~~ Ill! 1!1! * 1111 """*lllll I® Ill! Ill! ill Ill t !11 $ G Ill IQ I - Ill! Ill! $ 1111! fill l!ll ... "' illll$ lll! rill' ·' .!lfUi "w llll illll "' ·~ I Fig. 46. 1t: vs lP for discontinuities estimated class 1 and 4 & 5 r < 85" 1t: < 0.0087 * (-eo a + r+ > y "' apparent mscor.ltinauty . . v,.,..,.u•v<> parameter (TC) is """""''"'·"""""" in the slope """'"''.uu to t<mpiulg dip~·. intact and 2) the column axis is stnttglJit. flex. under the load of the rock Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STA.Bll.JTY PROJWJllJTY CLASSIFIOO'ION- SSPC ll' c * 1t2 * E * a2 UW * COS(dip-,.) * tan qp1 - 101 IIIT'(dtp•) = 2.26 * (UW * sin(dip-,.) - h.: = critit:tll ~ [20) ti = thicknas of layer JII'OM to bflclt:ling ~ UW .. rmtt Might rr.d matma1 ,.,. E .. intt.1t:t rr.d ~ coh .. coiJaiora ,..,. c;:) ~ pltme ,, .. ~ • • discontintdty pltme Equation [20] can be expand.ed to more complicated forms of buckling (three hinge or more beam models(78> for straight or curved slopes) but the assumptions necessary for the more complicated models are manifold and it becomes questionable whether the criti.cal slope height resulting of more complicated models represents reality. Following eq. [20] the slope height would have to be about 100 m to create tlexural buckling iirilure if the discontinuity spacing (din eq. [20]) is about 0.1 m(79). Heights in the order of 100 m are more than the heights of the slopes in the research area and thus tlemml buckling is unlikely. This agrees with field observations as slopes in the research area have not been D.Oted to fiill due to buckling. Buckling, hO\lVeVer, has been observed to occur in very localized zones in slopes (generally zones of less than 1 m2). In these zones cleavage planes in slates have become detached due to weathering, reducing the discontinuity spacing to about 1 mm, allowing localized buckling. A 'buckling criterion' has not been defined or incorporated in the SSPC system because the slopes in the research area are not iiriling due to buckling. Also in other areas it is likely that buckling causes only fuilure if the slopes are higher than those tor which the SSPC system has been developed . . . . . . . . . . . . D•.1.2h•.!i. . . . The 1 sliding criterion 1 and •toppling criterion' are valid for all discontinuities that fulfil the kinematic requirements (e.g. 'day-lighting' tor sliding and dipping opposite to the slope dip for toppling). Both criteria have been incorporated in the SSPC system for predicting 'oriemation dependent stability' of a slope. The values of the angle of friction determined from the 'sliding criterion 1 are comparable to the result of laboratory and field tests and confirmed by friction angle values reported in the literature. Therefore the values determined from the sliding criterion' can be used to estimate friction angles for discontinuity planes. 1 (78) The boundaries of the hinges or beams are formed by discontinuities with strike parallel to the slope strike but dip opposite to the slope dip. (79> This is in rock masses with rock types in the research area. Eintactrock = 45 GPa, UCSinlactrock = lOO MPa and 'Pi · = 45o, which are typical values for the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 102 D.l The development if the SSPC system D.l. 3 Orientation independent stability The slope failures that could be attributed to discontinuity shear displacement and are dependent on the orientation of the slope and the discontinuities have been analysed in the previous chapters. It has been shown that a number of the investigated slopes are unstable following the criteria set in the foregoing chapters for orientation dependent stability, but a large number of the slopes are not unstable following these criteria. This chapter examines whether the rock mass parameters of the slopes in the research area that are not unstable due to the criteria for orientation dependent stability, show a correlation with the visually estimated stability of the slopes (eh. D.1.3.1). Parameters that are analysed do not depend on the orientation of a discontinuity nor depend on the orientation of the slope and hence slope failures due to a combination of these parameters have been named 'orientation independent stability'. Moreover the rock mass parameter data from the slopes in the research area are examined to see whether a mathematical model can be formulated to predict the 'orientation independent stability'. Two mathematical models are analysed: a linear model and a shear plane model (eh. D.1.3.2). The rock mass parameters in these models that depend on the overall spacing and condition of discontinuities of multiple discontinuity sets in the rock mass, can be calculated in di:fterent ways. Three different options have been selected for the spacing as well as for the condition of the discontinuities (eh. D.1.3.3). The linear model is optim.ized with all difterent options for the spacing and condition of the discontinuities (eh. D.1.3.4) and the results are used in optimizing the shear plane model (eh. D.l. 3. 5). The good capability of the shear plane model to predict the 'orientation independent stability' of a slope and, however less significant, the possibility to interpret the shear plane model as a physical model that describes the mechanical behaviour of the rock mass of the slope at tirilure, are the justification to use the shear plane model for the SSPC system for determining the 'orientation independent stability' of a slope (eh. D.1.3.6). D.1.3.l An analysis of the rock mass parameters of the slopes that are not unstable following the orientation dependent stability criteria for sliding and toppling as discussed in eh. 0.1.2<80>, shows that there is a marked difference between stable and unstable slopes for the main parameters describing rock mass quality. Fig. 48 shows the frequency distributions of these parameters (e. g. intact rock strength - irs, spacing parameter(8!) - spamass<SZ), and condition of discontinuities parameter(81>- conmas•(82>) for stable and unstable slopes. All three distributions show a shift from higher to lower values from stable slopes via unstable slopes class 4 to unstable slopes class 5. It is therefore likely that unstable slopes that are not unstable following the toppling or sliding criteria, are unstable due to a combination of the parameters for intact rock strength, spacing of the discontinuities and the condition of the discontinuities. D.1.3.2 Models In the previous chapter is shown that the rock mass parameters describing intact rock strength, spacing of discontinuities and the condition of the discontinuities, correlate with the visually estimated slope stability for 'orientation independent stability'. A mathematical relation between these rock mass parameters and the visually estimated stability is likely to be also dependent on slope dip and slope height: (80) Only slopes have been used for the development of criteria for orientation independent stability with a probability to sliding or toppling instability following the sliding or toppling criteria ofless than 5 % (for probability analyses see eh. D.2). Slopes assessed to be unstable in the future (class 2 and 3) are not used because the results of the 'initial point rating' system (eh. C.4.3) showed that the assessment of future instability may be not reliable. This results in a total of 141 slopes that are used for the development of orientation independent stability criteria, from which are 94 visually estimated to be stable (class 1), 10 to be unstable with small problems (class 4) and 37 unstable with large problems (class 5). <81> In Fig. 48 the spacing parameter (spa,_,) is calcul.ated following eq. [13], and the condition of discontinuities parameter (conlllilU) following eq. [22]. To avoid confusion with spacing and condition of a single discontinuity set, the characteristic value for the spacing and condition of a rock mass with one or more discontinuity sets are denoted by the subscript 'mass', e.g. spa,_, and conmass. <82> Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 183 D SLOPE STABIUTY PROBA.BlllTY CUSSlFICATlON- SSPC vllually Mtlmalad . . . . (dlllss1) 11 !llllllble (olasa 4) !2l !llllllble (dlea 6) 8WII'IIIg8 (·): olasa 1: olasa 4: class 5: U6 ll*k'il! facliar (spa IIIMI) (-) cloflely epacad ,..___ 0.75 --+ 0.339 0.186 0.134 U6 0..1111 widely spaced --.<->: o1asa1: o.m olasa4: olasa5: 0.878. 0.818 condlllon ollilllocnllnuill facllar (oon 1111188) {-) Fig. 48. Frequency distribution of irs, spa1IIIJII and con1IIIJII. I (ii'B, spa_, . . . con_, dip-.. height.,.) = stability irs "' ~rock strength . , _ con_ = tlae spa:cm, ~ tl&e cONlitilm of tl&e discmttinrtities m tl&e rock .,.. ,!'!JI!!J.,.._:~~!Sl~t!ll!JM._!~ ··- #Jl.,. ':...~.J!L~ s}pl!£. [21] ma.sl! Obviously the number of possible relations that could fit is large. Two relations have been tested: 1) a linear model ~W-<i 2).~ s,b~~lAA~ mqg~l.(Fjg. ~Ol$~~bJlS y~~<tfur~lo~ D.l.3.3 !>mbill!l' .~~ulations.in.sQils. Options for spacing of discontinuities (spamas,) and condition of discontinuities (conmass) Most rock mass classification systems consider only the spacing and condition parameter of the most prominent discontinuity set or the discontinuity set with the most adverse influence on the stability of an underground excavation or slope (eh. B.3.4). This is too simple fur slopes, for Wlure is often not determined by one main discontinuity set but by more than one set. Multiple options exist to implement the spacing and the condition of discontinuities. Averaging or a form of weighting of the parameters for spacing and condition of discontinuities give a large number of possibilities so that a choice had to be made. Three options for the spacing parameter and three options for the condition of discontinuities parameter, leading to a total of nine different combinations, are analysed in the linear model to establish which :fitted the data best. The different options are analysed according to the following rules. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D.l The developmentcfthe SSPC system 104 Spacing of discontinuities value (spa_,): 1 Minimum spamoss equals the spacing value of the discontinuity set with the smallest spacing in metre. The value is taken as 10 m if no discontinuity set is present in the rock mass of the slope. 2 Average spaiiiiJU equals the average of the spacing values (in metre) of all discontinuity sets present in the slope. The value is taken as 10 m if no discontinuity set is present in the rock mass of the slope. 3 Thylor spanws. equals the spacing parameter calculated following eq. [13] (page 76) and Fig. 33 (Th.ylor, 1980). The value is taken as 1. 00 if no discontinuity set is present in the rock mass of the slope. Condition of discontinuities value (connwsJ: 1 Minimum conmass equals the condition parameter (1() of the discontinuity set with the lowest condition value. The value is taken as 1. 0 165(83> if no discontinuity set is present in the rock mass of the slope. 2 Average connws. equals the average of the condition parameters (1() of all discontinuity sets present in the slope. The value is taken as 1.0165<83> if no discontinuity set is present in the rock mass of the slope. 3 Weighted ___ lf.no discontinuity_setis..presentinthe. rock mass ofthe slepe the eonmD.. iS" taken as·l:O l6S'83f:-ifonly one discontinuity set is present in the slope coniiiiJU is taken as the condition parameter (1() of that set. If more than one discontinuity set exists in the slope, the condition parameter (con_,) is taken as the lower value of: - the condition parameter (1() of the discontinuity set with the lowest condition value, or - the lowest value of the weighted mean values of the condition parameter (1() of any two or three discontinuity sets present in the rock mass, weighted inversely against the spacing. Thus conmas• may equal a value based on only one or two discontinuity set(s) even if the rock mass contains more than one or two discontinuity set(s). For three discontinuity sets the weighted mean value equals: [Zl] spacing1 spacing2 spacinga The nine ditrerent combinations have only been analysed in the linear model because optimization times in the (non-linear) shear plane model would have resulted in an infeasible calculation time<84>. <83> 1.0165 is the maximum possible value of R:'. <84> The author does not think that this is a weakness in the analysis as the outcome of the analysis show that the most logical choices for spacing, e.g. Thylor, and condition, e.g. weighted, parameters are the best. Also the results of the whole SSPC system are so good that it is unlikely that these choices are erroneous. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ]) SLOPE SX4BlllTf PROR<tBlLJTY t7LASSLF10iTJON •· SSI.>C 105 D.L3.4 The linear {visr.w!ltv estimtwd elms smri:ah!le i1<•lS!JlaUv estimated class if if atld 5) are ;,.,,.. ,,~·'r"'l"'thr note; factors a3 through a5 are negative because the model (eq. [23]) becomes stable for smaller conflict with the values and unstable for larger values of a3 through by a Monte a5. This Thble 9. Factors for linear model with spa="·' 49 shows the percenTaylor and con'""'" calcula·· of slopes for the three tion see eh. D.2.3. ~:~u.cua.u~ot. "'"''"-·"·' and for The of discontinuities and the standard errors obtained dl:tierent options for the model for a spa'"= cal.cul.ated D.l 1 The lowest peJrceJt:!'i:aJ!l;es method is used fur the sp~tClllg ,,.,."·'-••no" pru:an:iete~r. the different are very small if the spamass 1s calculated following Taylor the lowest pet·centages for slopes, class 5, are found if an average or a weighted is whereas the lowest for unstable class 4, are if a minimum or weighted condition is used. A weighted mean a munber of discontinuity sets in a rock mass is thus the 'orientation m<ten;en,a.ein approach to the of discontinuities in the linear model to stability'. The methods of calculating spama,,,, following Taylor conm.oss with a mean are used in the op'!ImJtzatlon of the ilie following (SS) :fhe results are for without weight factors to compensate for the difference in Li.e numbe-rs of stable and unstable This is because: l) with factors showed small differences with those without 2) the difference in numbers of stable versus unstable stab1e versus 47 unstahie no differentiation is made between dass 4 and ciass 5 factor a.lso increases ihe influence of outli.ers on the result. (Sol This of GlHGUJ•am:;g spa,llliSs and con..,,. also avoids the with the and condition of discontinuities rock mass classification systems as discussed in chs. B.3.4.3 and B3.4.5. as included in some ofthe Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ OOI'Idition: spacing: ~a minimum Minimum, average and TaylOi' or w~ refer !tl11'16 method mcalou!atioo for SPI!cil"lg and oondltlol'l of dlaconlloot!les par~$ {il!i def!I'!OO on the fl':ll'gO!ng pagee). Contlni..IQIJ$ i!nes ar® mean valu!H\ and dashed iii'le$ are the standard mar ('~he lines are plotted for easy !.ll'l~ing and ruwa 00 l'l'leal'l!r.g in b~n t!w data poin1S). Fig. 49. incorrectly calculated stabilities with linear model calculation of mean values and standard error see eh. D.2.3.l). 0.1.3.5 D. 1.3.5.1 model The shear ....L,.......... creates more P'-'·~•n•vu.n""··"' for movement in movement between individual in. a soil. -'-''"....,'-'ll'" failure crit.erion' the ,. .....,..;,.,.... in the soil. This model is sunilar to the (S?J This is not contrary to the conclusions about water pressures and the criterion' in eh. D. 1.7. The water i.n the surfuce than from the rock n1as:s discontinuities is present and stems more from influx water from th.e behind th.e Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE mBIUTYPROBA.BIUTY CLASSIFIOO'ION- SSPC 107 model for inmct rock stl'eD8t:h approximated with the 'Mohr-Coulomb milure criterion' (eh. A.2.4). The failure mechanisms that cause orientation independent slope milw:e resemble for some slopes intact rock failure and for other slopes failures in a soil. This may allow the strength of a rock mass that fails through these orientation independent iillw:e mechanisms to be approximated by a 'Mohr-Coulomb milure criterion'. Fig. 50 shows a slope in a rock mass following this criterion. Mathematically this is formulated as illlows (Das, 1985): .. ll. .. 4 * .. * aln(dip,..) * COS( f)_.) colt_. 1 - COS(dlp....,.. - f>-) UJf 111111X H._ .. ~ [24] H._ = ~ ~ Might UJV • Unit W,;ght of rock lfttlS8 eo~~._. ~ ONl friction. cmgk of the rock lfUl.S'8 ,_ • The maximum possible height (H-J of the slope in relation to the dip of the slope (dtp~ is governed by the rock mass cohesion (con.,.){88> and friction ( ,_){88) if the slope dip is larger than the rock mass friction. The material above the slope plane fOllowing the 'MohrCoulomb failure criterion' (Fig. 50) will fail if the excavated slope height or dip is larger than permitted by this criterion. There is no maximum to the slope height if the rock mass friction is larger than the slope dip. ·. dlp.t.,. ~~-~~~~~": D.l.3.5.2 Shear. plane mo.~Lf5!!:~~? ,,...... Parameters in the shear plane model 'Pmou and coh,_ are assumed to be dependent on the rock mass parameters measured in the field, e.g. intact rock strength (irs), spacing of discontinuities (spa,_) and condition of discontinuities (con,_). In this research has been found that both tp,_ and coh,_ can be reasonably represented by a J.inear<89> combination of irs, spa,_ and con,_. Chapter C.3.2.1 discusses the lilrelihood that the influence of the intact rock strength on slope stability is bounded by a maximum, i.e. a cut-off value. Linear relationships for tp,_ and cohiiWSS with a cut-off value for the intact rock strength (irs) result in the following: coh_ = wO , _ = w3 * irs * irs + wl + w4 * spa* spa- + w2 + w5 * con_ * con_. with cut-off valu.e for irs: ?. ctt:!(;Jf"V~s~we .. trs .. inttillroct i6Sitfi (os ~ if irs > Cllt-of/ ~ ... in = cl&t-ojf valru rrrs mth.e]lelilf weight :{t:ll:tow. wO, wl, ••• w5 ~ 0 'Pmass• the friction of the rock mass, has a value within a range from 0 to 90° (0 to 1r.l2). fPIIWSS has to be normalized so that the value is never outside this range to be able to optimi.ze the shear plane model. The maximum value for is obtained for an intact rock strength (irs) equal to the cut-off value, the spa_ equal to its maximum value of 1.00, and the con- to its maximum value of 1.0165. Hence, the maximum for fPmass is expressed by: •-s 4'- (~) = w3 * Cllt-ojfwdue + w4 * 1.00 + wS * 1.0185 [26] tpiiiQ$$ in eq. [25] must thus be divided by 'Pmass (maximum) and multiplied by 1r./2. Large dif:lerences in the order of magnitude of parameter values may have an influence on the optimum values found in the non-linear <88> Th avoid confusion between friction and cohesion along discontinuities the friction and cohesion for the rock mass are denoted with respectively 91,..... and coh,_. <89> Notonly linear relations between,,_ and coh,_, and irs, spa,_ and.con_,.havebeen investigated. Also relations have been investigated of the following forms: -wl f..,..= irs *e.,._ • - - , _ = irs * (spa-)"'1 • (con.....,)"'2 The results are, however, not leading to better results than a linear combination. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ :relations for the values fur and and if>uu.,. an order of Therefore the ( "- .. I \ if - irs 100 :s: a6 - ir:s = intact rock strength if -~~- > 100 a6 - irs ,., a6 * iOO aO through a6 "" factors dipslopt. "' dip slope irs = intact rock ·""'..'1!!''" spa_ = parameter con_ "' c;mdition disccntinuities porameter = maximum possible UW '" Unit the rock mass and conmass is with a value as these are the of the linear model (eh. D.l the gave the best results in the is taken the same for all :rock masses in the research D.1.3.5.3 (\)(JJ Measured intact rock unit sub--unit, of the rocks in the research area are bet<;vf'.en 25.5 and 27.0 , The range is unit dete.rmh"'lations within one sub-unit. Rock of !he ~>"'"''"'"Y determinations have, for obvious reasons, not been done. However, fue of the discontinuities and the fuct that open and not filled disconi:inuities reason to assume of fue mck mass is lower than the intact rock unit Also fue karstic rock units an~ not to have a rock mass unit The actual value of the unit used in eq. is not imr,nric,;,nt Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SJ'ABJUTY PROBA.BJUTY CLASSIFICATION- SSPC 109 For each slope f. ... er =1 [28] ER= ,Eer1 J (vi.Rially esliwuJted stability: cliJS 1 is Sldble, cltm 4 is UMtiJble with sma1.l problems and cltm 5 is 11Mt4ble with large problems; between brocuts is ~ the stability calclllated with the shear plane model) ER would equal the total number of slopes used in the optimization if the shear plane model is the completely correct model for orientation independent stability, if the data set is ideal (no errors in any parameter of any slope) and if the filctors aO through a6 are at optimum values. The stability calculated with the shear plane model would .then be the . same as the visually estimated slope stability in the fi.~ld. for all sl~s. QbviQusly this is unlikely because the shear plane model is not a completely correct model and the data set is not likely to be ideal. There is thus always a certain percentage of the slopes for which the calculated slope stability following the shear plane model is not equal to the visually estimated stability in the field. Hence, the value of ER is always larger than the total number of slopes used in the optimization. The goal of the optimization is therefore to minimize ER. The values for aO through a6 in eq. [27] belonging to the minimum value for ER are then taken to be the values that best fit the data set. During the optimi:mtion process the ratios of H1~Hmax (fur slopes visually estimated to be stable) and Hma/Hs~cpe (for slopes visually estimated to be unstable) are limited to maximal 2. The ratio of 'Pmas/diPstope (fur visually estimated unstable slopes) is also maximal 2. These limitations are necessary to avoid a too strong influence of possible outliers. In particular Hmax becomes (enremely) large and inftuences the optimization very strongly for an outlier with 'Pmass smaller than, but almost equal to, the slope dip. The maximum possible height of the slope (Hmax) is infinite if the slope dips less than the rock mass friction ( 'PmasJ. As a consequence of this and of the use .of a cut-off value for the intact rock strength, the function in eq. [27]. is not continuous in the first derivative. Because of errors in the data (visually estimated stability, dip, height, intact rock strength, etc.) the function contains multiple minima. Optimi:mtion of a function that is not continuous in the first derivative and that also contains multiple minima, is difficult and it is often doubtful whether the absolute minimum can be found. The function is therefore examined graphically to find ranges for the factors in which the function is likely to minimize (decreasing ER). Then an optimi:mtion routine (Levenberg-Marquardt, Marquardt, 1963) is started with starting values for the :factors within the ranges graphically determined. The procedure has been repeated multiple times<91 >. Multiple optimizations without the outliers<92> result in minima which are <91 ) The order of magnitude of the factors is considerably diffurent. aO, al and a2 are about 10000 times larger than a3 through a6. This diffurence could have influenced the optimization results and therefore an optimization with scaled factors aO, al and a2 has been done (e.g. coh_ in eq. [28] is multiplied by 10 000 which results in aO, al and a2 to be divided by lO 000). The results are the same as with none scaled factors apart fur the divider of 10 000. This implies that the optimization is not sensitive for this order of magnitude differences in the factors. The Levenberg-Marquardt routine used fur the non-linear optimization which is part of the computer programme MathCad does not use scaling of the factors. r Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SSPCsysum D.l :model shear these reasons it was dectd!oo "'"'"''"'""'"' the rock mass '~-"'"'~""'"" with "H''"~ilv The ~;tlnlateoa "...~.nu~y likely almost at ""'i''u"J'"..'"'' vv............ "''"""""""'"' at"'""'''"''',.. plot near the in Fig. 51). This "'"'nr•ri« the correctness of model because optunJ:zatton no differentiation is made between slopes with stability classes also no 'a priori' is to steer the opun:1rnmcm 4 and 5, and D.L3.6 Ut:scutssJt011 and conclusions on (!12) Four give in all oplim.izaltimts a non-realistic result for me maximum or the friction ( 'Pmass) and are therefore considered to be outliers. The of !:hese why these should not be used. The four are: The slope is to and very near to a fault The slope is situated in an associated shear zone area; the 90/10/2.2 disJ;ontimuty orien!ations are The rock mass consists of 92!13il401 of slates but the rock mass characterization is done for sandstone. Doubt about fue some observers classified as unsta!Jie estimated class 5) others as small 92/18/lc "''""'~>~"·""'~ in !:he near future estimated class 2). The slope is estimated to he stable but aH calculations result in an aosomrely 93/ll.s/ils was then characterized two other persons who measured corwldiem.bly ::;paccmrgs, that resulted in a stable the shear modeL <93l Brown Procedures for calculation of mean value and st!.ndard errors are discussed in eh. 0.2.3.2. This allows for criterion' with the rock mass ·~'?'~"""'" calculated with Bier.iawski's '!UvfR s-ystem and the 'modified Hoekof !:he SSPC system Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SWJIUTY PRO&BIUTY Cl.ASSIFIC4.Tl0N- SSPC 111 to perceive for the linear model. The shear plane ~rs theretbre an appropriate method for the calculation of the 'orientation independent stability' and is used in the SSPC system. Using in eq. [27] the mean values for the meters, listed in 'Th.ble 10, and simplifying eq. [27] results in: col&._ (in h) "' in • - (in de~ = in if iJrtact * 94.27 + J!lfJd- * 2862.9 + con_ * 3593 * 0.2417 + SJX.l.- * 52.12 + con_ * 5.179 rock ~ < 132 MPa - in = intact rocJc at7Ugth (in MPa) •• in .. 1S2 If dip,. • [29] mm:imrw slope Might (H,_) is i1rjinite tile , . . , . slope Might is~ hy: $ .,_ 11,.. • 1.8 .. * * 10--t * col& - sln(dip!"f!) * CCII(fl-) 1 - C08(dip..,.. - .,_) spa,_ is calculated follO'IIVing Th.ylor (eq. [13], page 76, 1kylor, 1980) and coniii/ISS is calculated with the option for a weighted conmtW fuUO'IIVing eq. [22] (page 104). -· I I !• 0.01,_--.--.,---r--.,---r--.,---r--.,---r--~,~----.,---.---,.-~ 0 0.2 0.4 0.6 tp 0.8 mas/ dipslopB (-) 1 1.2 1.4 Fig. 51. H-/H,~opt vs rpnros/dips/op< (for the graph H,_IH,• has a maximum value of 100, and H,.,!H,1ope rpnros/dipslq>t ~ 1). = 1 for Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ uz SSPC system factors in the shear p!ane model factor mean value H ao 9427 a1 28629 a2 3593 2458 1083 a3 visually estimated mean value !%1 standard error[%] 8 2 28 8 note; factors and percentages for optimization without outiiers. 1'.tble .HL Factors for the shear that conflict with the mode! and percentages of with a calculated estimated calculation see eh. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE /:;.TA.Bl!JT"f PROBABILITY CV1SSIF!C'4'110N · · SSPC :U.3 1 D.1.4.1 for m the research area and METHOD OF EXCAVATION (ME) excavator unknown reseru:cn area. Expo!nm~s excavator were by a frnme. 'Hand-made' exposures .uu!u~~u"''" of the of excavation on the d1S1:::onlm1J.IIl1es has been blasting average ais,conttm'LUty spa1;wg per hthostJratiJsrajJhic "'""'·-'""' and per t';pe of discontinuity versus the method of "'"r'"''',.."""'" conventional with the following result.: fractured intaGt rock '[able H. 1'1itial classes for the met.'Aod of excava- ti.on. D.1.4.2 relation oe1we:en di~;conllnuity "ll"'"'"'l"; di!;contl:nuity sp~tcii:tg in exposures excavated with excavation to the d.Is~contlnutty sp2tcrntg in 'natural' exposures in the same sub-unit and with the same of rock mass weathering should give the required parameter (ME1) (eq. [30]). The 'natural' exposures are assumed to be representative the rock mass prior to thus· without influence of the method of excavation. ME = 1 j <95l _!3!:~~~~~L discontim~i.ty spacingMtiP'd = method of excavation The steel rod of the hammer which had been wiili a diameter of about (U5 m. (%) In eh. D.1.3.6 is esmbHshed that a shear model with a parameter for the overa!l of a number of dts•~ontunnty sets in a rock mass calculated to Taylor (eq. page 1980) is to be used in the SSPC system. Therefure the parameter tor the method of excavation is determined for spa"'"'"' caicu!ated to 52 shows, however, the average per type to show that the between to aH types of discontinuities with.out differences. spa,= calculated to rres;pect!ve of the type of dis,~on.timlity. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SS:PC syszem wall §2. EXllm~)les of average d!s:collil:l!'llnty weathered exposures. This procedure to the values is vv''""J""' can be followed in of dif'fer:ent not connected exposures same weam:enng, "'"'''""''J"'"' this is if no milerdlepem:ierlcies exist between discontinuity spacing and met.hctd. D.L4.2.1 on the of bedding or So, if this spiJtCn:lg ~.~"'~"'"'"'"' that the sub-unit in an exposure after exc:av:itlrm Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SJ:4JJILITYPROll4.1JlLlT'l CLA.SSlFlaT!ON • SSPC not illfiuence same !<m·>-Hinu to et1"0neons 115 t'll">m!!t<z""., average spa""'"' 53. The average spa,=, of a series of exposures in a sub-unit or unit excavnted with a method may more on the absolute spama!, rather than on t.l·w influence of the exca-vntion method. same order as the The for the :influence of the exc:av:1twrn :rnternoas \"""'""""'""' 3 is not 4 size will be excavated type of excavation may also have been based on other ex<;avauon ""-~'·"'~--''""'..'' etc .. For a unit with size and high intact knew nature of the the most widely ,.,..,~,.,•.., .... dependency excavation based on d1s:coJrrtu1mty S]Jac:mJ?;. '""'"'"''""''"""'"' in this way are in Same Order liS IOf all SUlJ-UllltS. (98l The probiem is musb:ated with the Assume two exposures in the same limestone unit; one is a natural The is 0.5 m in the first exposure and is 0.45 exposure and the other is excavated with good conventional m in the second exposure determined after excavation. The lithostratigraphic sub-unit is in both exposures medium limestone. A of the due to the excavation method would lead to the conclusion that conventional by a factor of 0.9 0.45!0.5). However, it may well be fuat the in the rock rnass in which the second and that fue reduction of due to conventional is exposure has been made was 0.7 m, thus thick 0.65 (= 0.45/0. (~9) "~'~'"""'"'''" to the classes for conventional etc.. These are never (lOO) a Sub-units of a unit with the most type of excavation method have a d;s•;ontunlltl/ b.'Utl·UJllttl influenced on an a res1Jltil:Jg in a type of u.:nJ'""l4"', e.g. open choice of excavation method. or fuund after excavation in exposures excavated with which is the (which 'was also due to !he method of excavation (there are no 'wider' Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SSPC system aeJ~elJta on tlle structure of the :rock rnass The between th.e Conclusion D.L4.2.2 The va:lues of the pru:an:1ettrr for the method of excavation one exposure made with with conventional blasting with as fue direct am:>rOl:lCfi are also not de~1em!ent on rock mass structure Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 117 D SLOPE STABIUTY Plf.OBA.BIUTY CUSSIFIOO'ION- SSPC excavation in any lithostratigraphlc sub-unit and with any degree of weathering. This increases the accuracy of the values for the method of excavation parameter considerably because more observations can be used. The above is implemented as ibllows. The spa,_, values of the exposures are averaged per lithostratigraphlc subunit (u), per degree of weathering (w) and per method of excavation (l), resulting in average (spa,_},., w,l' Then the ratios are determined between any two types of excavation per lithostratigraphlc sub-unit and per degree of weathering: [33) " = ~~ (BU-) unit w "' . , . , of Nd; l'llt.m' weotMring i. j • 1MI1Iotl of uc4Wition The ratios are independent of the lithost:ratigraphc sub-unit and of the degree of weathering (eh. 0.1.4.2.1), hence ratiou, ... i, j can be averaged: ~ratio~'.~* i:(~. *i:ratio~,..~,) [34] (.-b-) rmits per degree of weathering of df/emtt degrees of wtiOtbering With actual data the calculated average ratios are formulated in the ibllowing set of equations: = ~ of~~ U.,. W = 1f1111Jber x, %1+1 x,...2 * %1+1 * ratio 1.2. 1 * x, = average * x, = average ratio1+&. 1 * x, = average ratio1.,., 1 * x,...2 * %1+1 x,.,. X;.s * %1+2 * %1+1 * x, = ~ ratioi+S. I * x,.,. %1+3 * %1+2 * x,.1 x, = average ratio1.e, 1 * .x;.,. * x,.s Xt+2 %1+1 * x, = average ratio1.,.7, X,.s Xt+& "or'~ Tlllioi~i;l * * * * * mm t ... 0, t·;;•'ftw i i = 0, 1 .. 5 = 0, 1 .. 4 i = 0, 1 .. 3 i [35] = 0, 1 .. 2 i = 0, 1 i =0 x,... X;.e and the values ibr Xo, 1 .. 6 are ibund by optimization. The values (ME1) ibr the method of excavation are then: ME_, = 1.00 MAj + 1 1 =-- j .. 0, 1,.. , 6 [36} j = 0: pne1I1Nitic lwtm&er acamtor, j = 1 : pre-splitti:nglsmooth Wall blosting, c~ ~with~ j = 2: good, j = 3: opex ~. j = 4: di.ilodgetJ blocb, j = 5: fmctured intact rock, j = 6: crushed intact rock A weighting &.ctor is used in the optimization of Xo, 1 .. 6 because the numbers of exposures excavated with each particular method of excavation are not all the same: Mighting factor for average ratio~, i = 1114mber of exposures1 * number of ex.posures1 [37] The resulting values ibr the parameter ibr the method of excavation are shown in Fig. 54 and Table 12. The values and standard errors are calculated by using a Monte Carlo simulation on the above methodology (eh. 0.2.4.1). 0.1.4.3 Reliability of the parameter ibr the method of excavation The values ibr the parameter ibr the method of excavation are as reliable as the number of exposures and number of different units they are based on ('Dlble 12). The more exposures in dif.lerent lithostratigraphical (sub-) units the more reliable the values. The values for 'pneumatic hammer excavation', 'pre-splitting/smooth wall blasting', 'good conventional blasting' and 'conventional blasting with as result 'open discontinuities' are based on a considerable quantity of exposures in dif:terent lithostratigraphlcal (sub-) units. Values ibr conventional blasting Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D.l The de>'l!:lopment SSPC: system uu'"""'''"''" bas~.d on fewer exposures less scatter and u"'"""v""'""' than those the number of •...,,,,..,""'"" area. D. AA masses in other rock masses to however, the scatter in the data is too 54 as the rntio of a :rock m.us. uul.wnm.)l; with the method is not possible, to a Romana also found that the by 'good' conventional blasting. are compamble. Note that me;ch:arucal excavation is about to the """''""'o'-' eXj:)ected the method of excavation influences the drscolrlfumi!ty SIJacJtng. The values established for the method mass due to the mema1n parameter are used in SSPC damage of the of excavation. (WZl The SMR factorfor 'mecharucaJ excavation' is to pneumatic hammer excavation' in the SSPC system. f(Omana's SMR and Laubscher's MRMR have one dass for deficient or poor This class is at the of with result blocks' in the SSPC S)'Siem . Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STA.BIUTY PRO&IHUTY CI.ASSlFICATION- SSPC 119 0.8 ~ 0.8 calculllllill'd ME WIIUae for ISPC ayatem ~MW of CllciiiiiiiS ME values for SSPC aystam . . . . . for malhod oftl1aiVIIIon ~. 1980) ................ mel1oc:l of~ (Romana. 1991) NGill: Aomano cto. not 8piiJCify the type of mechanical a:ava~Dr. J OA 0.2 0~~---r-----.-----.------r-----.-----.-----~ pnewn.hammer natt.lr8l smoolh wal crushed dl8lodged fractured L------------~------------~ method of excavation Fig. 54. open ~ues for the parameter for the method of excavation compared to values from (199iy(tines in-between·data ~have no meaning; and serve only for identification). ··Laubseher·{!~tild~ METHOD Of EXCAVATION IMEII 1l ME method mean value natural/hand-made 1.00 ·pneumatic hammer excavation 0:76 1 pre-splitting/smooth wall 0.99 good ·o.n ~-----------,--------------1 1 ~-------------- I blasting conventional with the fol· lowing result: standard error 0.06 number of observations(3) lithostratigraphic sub-units (2)(3) units(2) 92 23 6 173 21 6 0.11 57 19 6 o:o1 · 131 28 7 ~_Cl.P!~~.!~~n_!~~i!I!S_ 0.75 0.08 54 20 7 ~--~~~~~~E~E~-- 0.72 0.08 14 8 4 1 fractured intact rock 0.67 0.11 18 7 4 crushed intact rock 0.62 0.15 5 4 3 544 30 7 ~-------------- 1 Total: notes: 1 2 3 Data used for calculation are the combined data gathered for the SSPC system and for the engineering geological mapping (see preface). Columns 'sub-units' and 'units' are respectively the number of lithostratigraphic sub-units and lithostratigraphic units used for the calculation of the ME values. Used for the calculation of ME values and included in the column 'lithostratigraphical sub-units' are only those in which at leasttwo different methods of excavations have been used in exposures with the same degree of weathering, so that excavation damage could be compared in the same lithostretigraphic sub-unit with the same degree of weathering. 'Th.ble 12. Parameter for the method of excavation (ME) (for calculation see eh. D.2.4.1). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 1 ,._,.,..,.,..,,..,,,.,. is r~ ....,,...,.,"'"''r~ in the literature are .......,,...,~ ........,. . . iu D.l D. L5. 7. i -'- exc:av<ltic~n it mass 'I.Veathering. A In some locations it is ~-'"'"""'"'L'"' to follow a unit through different of in one exposure and to """'~"'"''"" with rock mass was, however, not to establish ac<;ur.ately because the of exposures in which a could be weathering was very Alternatively the values for the for independent exposures. The from l.lU10Sitrllttlg:ra]phJlC sub-units may, however, not the of rock mass weawrenmg decrease of to weathering may be on the rock mass stmcmre or '","'"-'"'" (these problems are analogous to those discussed in eh. D.L4.2.1). The influence of both is, supposed to be 1 and can be neglected because: The 2 The influence of is one exposure with a clear decrease of the formation limestones and dolomites. These can be followed in Spi!C!Hl~ti fOf 3ll Of W<':~fh>rirw (lW) 'Soil or small either without defined mechanicP.l !aminated mechanical discontinuities, and have a low intact rock 'Soil discontinuities or wit!J units resemble more cemented soils 1:han a rock mass. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SJ)!lJlUTY PROBABlU'I'f CLASS1F1C.4'I'lON , SSPC 121 D.L5.2 i, j the and values .. 3 "'average i = 0, i .. 3 xi+1 *X; "'average i * x,+1 *X; ,~average * XJ,:i! * Xi+l *XI =average X;+2 X...:;, XI are found i = 0, 1 i The values fur j "' .. 3 weighting the = 0, 1 .. 2 "' are then: 0, 1,.. , 3 because the numbers of e,.'{posures for each [43] parrun1~1er a Monte D.l In..fiuence The .influence weathering has been investigated fur the fullowing parameters: :intact rock overall ,Y..,~m'"" calculated eq. page condition of weJighted condition :in a rock mass and the of a rock mass D.l. 3 .6). Appendix IV shov.'S examples of the influence <~>-H!;;.,.,,.,t g~~ott::Ci1Jmc•al parameters used in the SSPC lithology groups. grame:a and in the values groups are not and Table 13 units (!OS) Calcareous show a sirr..i.lar decrease of the values of parameters with mc1rea1;mg ,,,,.~"",,.,., indeptmdent of the contents of other than calcitic: minerals, e.g, minerals. It sl:iouid, however, be noted that pure limestones or dolomires do not cx.~cur in a and weathered form. The values for calcareous units for and of other than calcitic minerals. weathefi"A are therefore based on units that contain a certain Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SSPC system ~-!y of-thGrl!lg (SS mll:19fl1! moowi!lllely (BS -~1'\·1<~1\ comp! oi~llil'l~ Values ~or WE are shown for the dlfferel"'t !!l:llllk'li:!les and for the w~ering !n!'!uenoo Independent M' the !ithology which Is denoted with 'all'. 'Soil ""'-"'"'""""'' bElclwse the ~technical parameters of 'soli type' units s~ oot to be inf!ijE>ncad weathertrm Fig. 55. {)verview of the infiuence of scatter in the data and could be The in 55 and Table 13 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl'AJJILI.TY PROJWJILI.TY CLASSIFlCATION- SSPC U3 WEATHERING degree of rock mass weathering liBS 6930; 19811 intact rock strength overall spacing of discontinuities (11 (spatmtBS) condition of a single discontinuity (set) overall condition of discontinuitiss 111 Icon .......) rock mass I 11 number of observetions (31 subunits (2113) units (2) 12 7 5 0.95 168 20 6 0.91 0.90 27 12 6 0.89 0.64 0.69 6 3 3 0.80 0.38 0.31 2 1 1 215 24 7 WEimiJct WE sps:~ng WE l1ingJe WE con~T~BSS WEcohm.a WE,_ fresh 1.00 1.00 1.00 1.00 1.00 1.00 slightly 0.88 0.93 0.99 1.00 0.96 moderately 0.70 0.89 0.98 0.99 highly 0.36 0.63 0.89 completely(41 0.02 0.55 0.77 lbtal: notes: 1 2 3 4 lithostratigraphic Values have been calculated after corraction for damage due to the method of excavetion. Columns 'sub-units' and 'units' ara respectively the number of lithostratigraphlc sub-units and the number of lithostratigraphic units used for the calculation of WE velues. Used for the calculation of WE velues and included in the column 'lithostratigraphic sub-units' are only those in which at least two different degrees of weathering have been observed so that weathering effects could be compared in the same 'lithostratigraphic sub-unit'. · 'completely weathered' is assessed in granodiorite only. lablaU..YaluesJ.'ortbe parametet:-fur weathering•. - Influence of the weathering on condition of a single discontinuity (set) and on the overall condition qf discontinuities parameter in a rock mass No major difErences are evident between the influence of weathering on bedding or on cleavage and joint planes (Fig. A 102, appendix IV). The general decrease of the condition of a discontinuity (as well for a single discontinuity as for the condition of discontinuities parameter, con-3) with increasing degree of weathering is evident beginning with a slightly weathered rock mass, but is considerably less than the decrease of intact rock strength and spa_. - Influence of weathering on rock mass strength parameters The influence of weathering on the rock mass cohesion coh_ and friction fl'mtus is evident and is similar for both parameters. D.l.5.4 WE parameter in SSPC system In the SSPC system three rock mass parameters are of importance. For 'orientation dependent stability' the rock mass parameter influenced by weathering is the condition of a single discontinuity (set): WEstngie· For 'orientation independent stability' the rock mass parameters influenced by weathering are cohmtus and fPmtus• expressed in respectively WEcoh and WE,.-· Using several parameters for weathering in the SSPC system may be confusing and therefore in the SSPC classification system only one parameter for weathering is used: WEm ....s· WEmass is the average of WEcoh and WE,. mtus because the values for both are very similar. Th be able to determine the influence of weathering for a single discontinuity (set) with a WE,._ relation has been established between WEmtus and WEsingle: WE..,. = b .452 - 1.220 * e- ..,._ [44] (correlatitm coefficient • 0.999) Thble 14 and Fig. 56°06> show the mean values and standard errors for the parameter of weathering used in the SSPC system. 0 06> In the forms for the calculation of the SSPC system the parameter is denoted with WE without subscript as only one weathering parameter is used. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 124 D. I The development if the SSPC system WEATHERING(1l condition of a single discontinuity (set) degree of rock mass weathering (SS 5930; 1981) rock mass WE,_ WE coli,.. WE *tile WE-12) mean value standard error mean value standard error mean value standard error mean value standard error fresh 1.00 - 1.00 - 1.00 - 1.00 - sUghtly 0.99 0.04 0.96 0.06 0.95 0.06 0.95 o.os moderately 0.98 0.03 0.91 0.06 0.90 0.07 0.90 0.07 highly 0.89 0.05 0.64 0.11 0.59 0.12 0.62 0.12 completely(3) 0.77 0.09 0.38 0.11 0.31 0.10 0.35 0.11 notes: 1 2 3 Values have been calculated after correction for damage due to the method of excavation. WE- is the average of WE coh , _ and WE , __ 'completely weathered' is assessed in granodiorite only. Thble 14. Values for the degree of weathering for a single discontinuity (set) and for a rock mass as used in the SSPC system (for calculation see eh. D.2.4.2). 0.80 ' ' ..___ no soil ~! <50% soil 0.40 -M- WEcoh mua __.__ WE • rn&l8 WE maa • average of WE coh maa and WE • mass 0.20 --+- ~ (1990) rock mua adjuslment wedlenild from freeh afller112 year weathered from hah . . . > 4 years >50%soil all soil 0.00 - + - - - - - - - - . - - - - - - - - . - - - - - ' - - - - - - + - - - - - - - - l slightly moderately highly oompl fresh degree of rock mass weathering (BS 5930;1981) Fig. 56. Weathering parameters vs degree of rock mass weathering (refer for the rock mass adjustments following Laubsc her to Thble 7, page 60). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl'ABIUTY PROBABH.JTY CLASSIFICATION- SSPC D.1.5.5 125 Reliability The values established ror weathering are as reliable as the number of exposures (observations) and the number of difterent lithost:ra.ti.graphic (sub-) units on which they are based. The values ror 'slightly' and 'moderately' weathered are based on a large number of exposures and diffurent units, but the values ror 'highly' and 'completely' weathered are based on fewer exposures and units(107). Consequently, these values are expected to be less reliable. The standard error the weathering values (Thble 14) reflects the scatter and uncertainty in the data. These do not reflect whether the values are also applicable to other lithologies than those used ror the calculation of the values. However, considering the number of observations and the number of ~rent lithologies observed, the values are likely to be applicable also outside the research area. Some of the uncertainties in the data may stem from the practical difficulties of applying the weathering classification given in BS 5930; 1981 (see also appendix V). ror D.1.5.6 Comparison to literature values The rock mass adjustment :fitctors for susceptibility to weathering according to Laubscher' s rock mass classification system ('Thble 7' page 60) are adjustment mctors describing the future influence of weathering in a mining environment. At the time of excavation the rock mass is supposed to be fresh and to weather within a certain time span to another degree of weathering. The degradation of the rock mass due to weathering and thus the reduction of its mechanical characteristics is expressed, by Laubscher, as a mctor. The rock mass rating, calculated rollowing Laubscher, obtained the fresh rock mass after excavation is multiplied by this factor to obtain Laubscher' s rock mass rating in a weathered state. ror :A:lime;;;spmtor·me··we~tb:ermg~pmc~s could not ·t>eaermec.t·ror ·the· SSPC sysrem(clLD:l:6); however;· the influence of an increase in weathering on Laubscher's rock mass rating, e.g. Laubscher's factors ror susceptibility to weathering, can be compared to the influence of the degree of weathering on the mechanical parameters used in the SSPC system. Although the SSPC system uses different parameters than Laubscher' s rock mass rating a comparison is likely valid because both describe the mechanical characteristics of the rock mass. The adjustment factors of Laubscher are included in Fig. 56 and show that these factors ror a rock mass weathered from 'fresh' to 'slightly' or 'moderately' ror a time-span of more than 4 years, are the same as the parameter ror rock mass weathering (WE~ obtained in this research. For an increase in weathering to 'highly' and 'completely' weathered the factors according to Laubscher are larger, e.g. the influence of weathering on the rock mass parameters is less than according to the SSPC system. This is likely due to the di~rence in the influence of the condition of discontinuities on the final rock mass parameter. The condition of discontinuities, which is the parameter least influenced by weathering for 'highly' and 'completely' weathered rock masses, has an influence of about 34 % ollJ.he~MRMR rating,lYbiltti&Jhe.S~fC..system .tile. iuilwmc.e.of .lh~. ~oP.dition,ofdi3coll~.on the~ .mass friction and cohesion is in the order of 7 to 9 %. The correlation of the weathering parameter (WE) of the SSPC system with the adjustment factors of Laubscher supports tlie concept of 'the wealliermg parameter ~(WE) as defined in the SSPC system. :Laubscher' s viilues are based on research ror a di~t application (underground excavations) and on d~rent rock types than the SSPC system. The validity of the SSPC weathering mctor is thus likely not restricted to use for the mechanical behaviour of surface slopes in the rock types studied in the research area. D.1.5.7 Conclusions The preceding chapters have demonstrated how weathering influences the intact rock strength, the overall spacing of discontinuities and the overall condition of discontinuities. Weathering is of obvious importance in the estimation of the stability of existing slopes and in the rorecasting of the stability of new slopes for which the degree of weathering may increase in the future. For this reason the values for WEmoss ('Thble 14) and the relation between WE......... and WEs~ng~e in eq. [44], are incorporated in the SSPC system to correct geotechnical parameters ror past and future weathering. 001) Highly and completely weathered exposures have not been found for all formations, because of erosion, vegetation and agricultural use. Highly or completely weathered exposures of pure limestone or dolomite are nonexistent. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 126 D.l The development if the SSPC system The condition of discontinuities is considerably less influenced by weathering than the intact rock strength and the spacing of discontinuities. The influence of weathering for all rock mass parameters is low for an increase in the degree of weathering from 'fresh' to 'slightly' and 'moderately', but strongly increases for 'highly' and 'completely' weathered. This corresponds to the percentage (as indicated in Fig. 56) of the rock material which is decomposed or disintegrated into a soil following the definition of the degrees of rock mass weathering (BS 5930; 1981). 'Soil type' units seem not to be influenced by weathering. The scatter in the data is larger than a possible decrease of intact rock strength, spacing or condition of discontinuities. The correlation between the adjustment :fu.ctors of Laubscher and the weathering parameters of the SSPC system supports the correctness of the approach to determine the weathering parameters and it extends the validity of the weathering parameters also to rock types not occurriDg in the research area. 0.1.6 Susceptibility to weathering The susceptibility to weathering of a rock mass as a function of time is one of the parameters most difficult to determine. Not only is the parameter dependent on the lithology, texture and structure of the rock and rock mass material but also on the climate, quantities of water percolating through the rock mass, chemicals and salts dissolved in the water, the orientation of the etposure, etc.. The type and quantity of chemicals and salts dissolved may change in time due to change of landuse, change in fertilizer use, etc.. These influences cannot be incorporated in enough detail to give a parameter for susceptibility to weathering leading to a universally valid function of time. Slope :lirilure may, however, occur due to the rock mass weathering within the engineering ~e of the slope. For this reason the degree of weathering of the rock mass at the location of the slope that will be reached at the end of the engineering lifetime is estimated in the SSPC system. The rock mass parameters (cll, ru.S)aHhtlocationdthe slope are corrected forthe·~· ohock rill~Wweathering"expecred at the end of the engineering lifetime of the new slope, and the slope stability is calculated as if the slope is made in this more weathered rock mass. The determination of the degree of rock mass weathering for an existing etposure is, to a certain ettent, subjective. The accura.cy with which the degree of rock mass weathering can be determined at the end of the engineering lifetime is, however, not only partly subjective, but will depend heavily on the etperience of the observer. The accuracy of the estimation depends also on rock mass specific factors and local circumstances such as the regularity of weathering over the years of the rock mass considered, the quantity of etposures in the area, the dif:fimmces in time of existence of the exposures, the number of dif.lerent degrees of rock mass weathering present and the homogeneity of the rock mass. Susceptibility to weathering is a major factor in determining the slope stability at the end of the engineering lifetime of a slope excavated in a rock mass prone to weathering within the engineering lifetime of the slope (see also pf~Be 152). The SSPC system is,. not designed t() quantify. ~~eptibility to wqtpering as a function of time. howeVer, with the SSPC system the future stability of a slope can be determined if the future degree of rock mass weathering can be predicted. In most other classification systems for slope stability (eJteept Haines et al., 1991, eh. B.2.4. 7), the influence of future rock mass weathering is neither discussed nor quantified. D.l. 7 Water pressures in discontinuities Water pressures in a discontinuity counteract the normal stress across the discontinuity and therefore reduce the shear resistance along the discontinuity. Water pressures in discontinuities are therefore an important reason for slope instability in traditional limiting-equilibrium stability calculations (Hook et al., 1981, Giani, 1992, Fig. 7a and b, page 11). However, in eh. B.3.4.12 is shown that this influence may be considerably less than often assumed because of the stress distribution in a slope and the possible restriction of water flow and pressures to discontinuity channels. The reduction of the influence of water in more recent classification systems supports this (eh. B.3.3). Moreover in eh. C.3.3. 7 is shown that a classification system for slopes should contain a parameter for water pressures only if the system is used for the design of a new slope that will intersect a permanent water table. This led to the introduction of a parameter for permanent water pressure in the 'initial point rating' system (eh. C.4). Whether this parameter should be maintained in the SSPC system can be questioned. The friction angles determined with the 'sliding criterion' should have been considerably lower than laboratory and literature friction values if water pressures in the order of magnitude as normally assumed in traditional limiting-equilibrium Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ calculations had been present in the stable slopes that determine the 'sliding criterion'. The friction angles from the 'sliding criterion' are, however, very well compamble with laboratory and literature values (appendix ID) and there is no reason to assume any water pres&lm' inftuence. Obviously water pressures may have only been present in unstable slopes. This is, however, bigbly unlikely because the rock masses of stable and unstable slopes are not fundamentally difi:rent with respect to the possibilities fur water pressure build-up. In the research area it is thus unlibly that water pressures are important. This is also supported by the fact that virtually no evidence of water under pressure, such as water spurting out of discontinuities, has been observed in stable or unstable slopes, not even during or after heavy and prolonged. rainfall. The evidence of water in discontinuities has been some li.mileci and localized seepage out of some discontinuities. It is likely that more evidence of water under pressure in discont:in.W.ties had been observed if the instability of many slopes in the reseateh area had been caused by water pressures in discontinuities (see also chs. D.S.2, D.5.3, examples nand lli). Moreover, for the ~ority et the iilled slopes it is difficult to imagine how the discontinuities could ever have been filled, completely or tbr a large part, with water because the water can ftow out of the discontinuities sideways or via other connecting discontinuities. The pressure build-up in such rock masses is equivalently smaller and considerably less than those normally assumed in a traditional limiting-equilibrium calculation (see also chs. D.5.2, D.5.3, examples ll and ID). Notwithstanding the above it 'should be noted that most slope iillures occur during or directly after ra.infall, this also happened in the research area. This does not conflict with the observation that water pressures may be of less importance in slope &.ilures. Discontinuities will become saturated during rainfall. Lubrication and the reduction of the friction angle of infill m.a1erial that sofals under the inftuence of water, e.g. clay, cause the slope to fail. The observation that slopes often filii directly after ra.infal1 and not always during rainfall may be further evidence · ttmwJ~ottmm matefillttnbe reasOirtbf'&ilure: lftlitnwter pressures had beentliecause furtlie slope failure, iillures would occur during the rainfall because water levels drop after rainfall ceases. The saturation process of infill material is, however, time dependent because most softening infill material has a low permeability, and it is thus very well conceivable that the maximum saturation is reached after rainfall. A separate parameter fur water pressures in discontinuities fur the slopes in the research area is not incorporated in the SSPC system. The presence of water causing lubrication and softening of infi1l material is already incorporated. in the parameters describing the infill material in a discontinuity. Whether in other areas with more rainfall or dif'rerent rock types a parameter fur water pressures is needed cannot be conclusively answered. However, considering that the area has been subject to heavy and prolonged rainfall and the amount of different lithologies and rock mass types, it is not likely that a parameter fur water would be needed elsewhere. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ US D. 2 Probability anal)lres D.2 PROBABILITY ANALYSES The qwm.tity of data collected during the research allows for a statistical analysis of the relations found in the foregoing chapters. A probabilistic qwm.tification of the stability results in the slope stability probability classification (SSPC). Such probabilistic analyses require an analysis of the distributions of the input (field) data and parameters (eh. D.2.1). This is followed by probability analyses of the 'sliding' and 'toppling' criteria for orientation dependent stability (eh. D.2.2), of the linear and shear plane models for orientation independent stability (eh. D.2.3), and of the parameters for the method of excavation and the degree of rock mass weathering (eh. D.2.4). A discussion of the distributions and errors of field data used for the development of the SSPC system should consider what ~nt types of distributions and possible errors are present for each rock mass parameter measured in the field, for each parameter describing the geometry of the slope, and for the visually estimated stability. Rock mass parameters A rock mass parameter measured has a distribution that is the combined result of: the distribution of a parameter in a rock mass, and 1 2 the limitations of the distribution of a rock mass parameter imposed by the subdivision in geotechnical units, and 3 the error made in measuring a rock mass parameter in a geoteclmical unit. Parameters describing the geometry of the slope Tile· distributiml ~,a·parametel' 4eseribing,·thctgeometry ·of a 'Slope m. the combined result of: 1 the distribution of the geometrical parameter, and 2 .the ~t:ror ~e i!!m,~wing 11 geom~trical param~. Visually estimated slope stability The error made in visually estimating the stability of a slope. Rock mass pammeters The distribution of a parameter within the rock mass is not relevant for the SSPC system, which is applied per geotechnical unit (eh. C.2), and is not further discussed. The distribution of a rock mass parameter within a geotechnical unit depends on how the rock mass is subdivided into geotechnical units. A parameter within a geotechnical unit is never a single value but a certain range for a parameter is allowed. The allowed width of the range depends on the context in which the geotechnical unit is used (e.g. the risk of a slope fiillure), on the variation of a parameter in the rock mass, and on the experience of the observer (as discussed in eh. A.2.2). The error made in measuring a rock mass parameter within a geotechnical unit can be determined. Repeating a measurement multiple times at exactly the same location will result in a standard error of a parameter measurement. Clearly only one single location should be used as otherwise the distribution of a parameter in the geotechnical unit would be contributed to the standard error. A combination of the distribution of the parameter in the geotechnical unit and the error made in measuring the parameter is obtained if several measurements of a rock mass parameter are made all over the geotechnical unit. This is the distribution needed for a probabilistic assessment of slope stability. However, to obtain this distribution is often difficult, time consuming or impossible in many situations (as already discussed in relation with the discontinuity orientation in eh. C.3.4). Therefore, in Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl'ABIUTYPRO&BIUTYCLASSlFJCAT/ON- SSPC U9 the SSPC system the input field data should be the characteristic value for a rock mass parameter in a geotechnical unit. Ideally, the characteristic value will be the mean value of the combined distribution of the error and the distribution of a parameter in a geotecbnical unit. For the development of a probabilistic classification system the distributions of measured rock mass parameters are, however, necessary. Therefore, during the research, multiple measurements of the same parameter in the same geotechnical unit have been done by dif:terent students and staff members. The distributions resulting from these measurem.eDts are assumed to be typical for the error distributions<108> for the measurement of a characteristic value for a particular rock mass parameter within a geoteclmical unit. Most of the distributions are normal. Some, however, are discrete or show a non-normal behaviour near limit values of the ranges allowed for a parameter. In the probability analyses the non-normal distributions and the discrete distributions are replaced by a continuous normal distribution because the ciifterences between the obtained distributions and a normal distribution are generally small. The standard deviations of these normal distributions, either direct or ex.pressed as a percentage of the mean (characteristic) value, are taken as the standard error<'08> of the characteristic value of a rock mass parameter. Thble 15 gives these standard errors. The standard errors are not exact for all geotecbnical units because a geotechnical unit with a wider range of allowed values will likely also have a wider distribution of the characteristic value and thus a larger standard error. The error distributions of the characteristic values were, however, approximately identical in c:iifierent rock mass types in the research area and are assumed to be representative for the error made in measuring a characteristic parameter value in all geotechnical units. Po.rameters describiRg the geometry of the slope The slope height and orientation have been measured as described in eh. C.2.1. These have only rarely the same values everywhere along a slope. Also an error may be made while measuring the slope geometry. The combination of the two results in a distribution, called the 'error distribution' and the standard deviation of this distribution is the standard error<108> ('Iable 15). Derived parameters The distributions of parameters derived from a parameter or combination of parameters measured in the field are established by Monte Carlo simulations. The simulations are done by randomly selecting sample data points out of the distributions of the parameter or parameters that form the basis for the derived parameter. Enough samples are used to obtain a stable 'robust' distribution for the derived parameter. Most of the resulting error distributions<IOS) are normal<109>, but some distributions show a non-normal behaviour near limit values of the ranges allowed for a parameter, or are discrete. Such distributions are replaced by a continuous normal distribution because the differences between the obtained distributions and a normal distribution are generally small. ,.Wi...,, estimtzted . Y.Rft •• -n S*nJ.JI:fk, MM(_ _, The visually estimated stability is a discrete parameter (eh. C.2.2). It classifies the stability in stable or unstable with a further subdivision in future instability and present instability. The unstable classes are further subdivided in unstable with smaii problems and llilStaolewiffilaige problems. It m·oeenestablished in eh. C:4.3 that.the visual estimation of future instability is unreliable and therefore slopes estimated to be unstable in the future have not been used in the development of the SSPC system and are also not used for the probability analyses in this chapter. In the calculation of the relations and in the probability analyses only the difference between stable (visually estimated stability class 1) or unstable (visually estimated stability classes 4 and 5) slopes has been used. Thus, the visually estimated stability of a slope in the probability analyses can be only stable or unstable and this is assumed to be a certainty. <108> For reason of simplicity the distribution which is a combined distribution of an error and the variation of a parameter in a geotechnical unit, is denoted with 'error distribution'. Consequently the standard deviation of the 'error distribution • is denoted by 'standard error'. (I09J A normal distribution is expected because of the Central Limits Theorem from basic statistics (Davis, 1972). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ note O!STR!BUT!ONS OF FlELD PARA!v'IETERS (3) Tabie 15. Distributions of field and derived parameters (nu,mb,ers in brackets refer to the notes in the Notes on TaMe 15: Dip, and and cl>.aracteristic ctis'co.ntiJ'IUitv Analyses of field data have shown that the error distributions for dip and are normaL Standard errors for direction are from its mean The distributions for and height were also found to be The standard errors for as a percentage of the mean value. Intact rack In !he t1eid intact rock has been estimated by a classification scale. In eh. C.3.2.1.2 was concluded that tl:Je average of a series of estimated nearer 1:0 the charncreristlc ti.m:i the of a limited amount of UCS tests. Srudents and staff have estimated the intact rock in the m the same est:imstres of to range of the class 4 with a range from 12.5 to 50 :hfl»a the of the range is 31.25 1\.!Pa). The is taken ""'"'r~"'"'' value if the was estimated to be cm the 1>01.m1.1ary ,.,..,,Jt;.,,. distribution of the estimated with a funn which resembles a normal distribution. The standard deviations siandard errors) are in the range from 27 to 40 % of the mean value. The distributions are not nonnal at the extreme classes l and 7. In the pn)bod:>l!litv r;alculations an e1Tor distribution for aH classes of intact rock estimation is assumed with a standard error of32 %of the value or of then""'"""''"" vaiue of two classes, 2 (UU) of a ouJ-<!trecuon measun::ment increases with u"''~""'·~l.IQ; are, OH) The error distribution of the 'characterlstic' disicontil<mity apr1m;nmate.1y normaL The distribution of the dls,cor.ttil111Uty "P"''m'!>" if the of the rock masses in the research area was found to be dis:conttillU!ty set have not been Such Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D Sl-OPE SIABlL!Tf<E~'?.OJ:M1JlLl'IY CLASSIFlC4T!ON- SSPC 131 3 Condition A Monte Cado simulation to determine the. error in the condition parameter resulted in a sumda.rd error of0.068. 1400 l'!III.ftCiffii,Y OUt Of distributions of aH parameters £l>'.iP.rrnin1na TC parameters fur and small scale rmJJ~:lltneJSS parameter for each 200 distusbances are. introduced out of a uniform and discrete d.istril:mtion of one class below until one dass above the class, except for the classes at the extremes, fur ·•·,rhich a uniform and discrete distribution is used from one class abo-v1:1 the minimum one l:he estimated c.'!ass. The disturbances are t~e errors made in the of discontinuities D.l.2 .1 The result is fur most a d~e.rete which can be a nuimltl For so:me a distributicm with clear for distribution certain values is obtained but. the distribution of all 1."-an still be l1i"''""n''"· rrmted a normal distribution. Near !he extremes fur 1C tmJ:nrnll.unl: maximum: l the distributil:m is oot nofll'ml and is not of l:he mean 1C vaiue. The. diffe.rences are, very small, if is considered iliat the distributions are oot kno\'m in detailRnd are assumed to be uniform. For is assumed that the enor distribution is normal around l:he mean 1r: value and is from the mean 1r' value. The average of the standard errors of all resuits in a standard error of 0.068. This value is robust and rer1eatea simulations with randomized and disturbances resulted in maximum differences of 0.003. ~ 4 In the and criteria use is made of l:he apparent ti:iction along fP friction(~) is derived from the apparent of the piane in the direction of !:he slope for the and is derived frorn the in the direction for the eh. D.l.2, D.l sliding and The apparent each sample 200 disturbances are introduced r.ttui.c1miy m>Hl!re•cn<ms (Thbie note The result is a maximum standard error of about 5o for p fur distributed around the mean value and of l:he mean value. '""'l-'"'"·""" and disturbam:es differed less than 0.5 ° and fuus the standard errors are robust. The standard errors are on fP and are not distributed near fue end values of the range for p and however, the differences are less l:ha.n 2" and are neglected. 5 The error distribution of the average D. A of discontimrities as used in the orientation stability (eh. di.r>tril::a;ai®d with a m.aximmn standard error of 0.28 m. The !llrror distribution of the spa"'""' pariiimeli.u c.alcui1!ted page as used in the orientation stability a maximum standard error of0.003. The re.sults are obtained a Monte Carlo simulation. 50 distributed with discOJ!l.tir!t!iicv sets out of unifomr distributions for all "";""''""' "P''""·'lS" nu1d{1mly out of normal distributions with standard errors of 5 % of the value (Table 15, note 1). The results for of discontinuities and tor the spa"'"'"' parameter calculated to are distributed and mc!ep•en,dellt ti:om the average or from l:he value of the weighted spa,.ms parnmerer. simu!ations with nmdomized samples and disturbances result ln approximately the same values for the standard errors. 6 condition parameter and con,...,, parameter The error distribution of the average condition of discontinuities parameter and the conmc." pam;·neter as used in the orientation independent analyses D.l are normally distributed with a maximum standard error of 0. 050 res:oec:t1veiV 0.065. The results are obtained a Monte Carlo simulation. 50 are for three discm1tir1ui1:y set<; out of uniform distributions for all TC values between 0 and 1.0165. On each 50 disturbances are oe;roer.•:tef! randomly out of the error distributions. The error distribution for the condition of is normally distributed wil:h a note 3) and for !he of discontinuities is normally distributed with a standard error of 5 standard error of 0.065 %of the value fur the spacing note 1). simulations with randomized and disturbances values for the standard error with differences less than 0.005. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 132 D. 2 Probability anal)lses 0.2.2 Probability of orientation dependent stability The sliding and toppling criteria are based on a boundary line below which no discontinuities in stable slopes plot. Visual determination of these boundary lines as done in eh. D.1.2 is possible but does not quantify the reliability of the lines determined. Therefore an alternative procedure has been applied to the 'sliding' and 'toppling' criteria that is discussed in the following chapters. D.2.2.1 Probability of 'sliding criterion 1 Determining boundary line ~· 1b determine the boundary line for the 'sliding criterion 1 (eh. D.1.2.1.5, Fig. 44) 300 sets of data points (f', 1C) have been generated randomly out of the original data set for discontinuities in stable slopes, with on each original data point the standard error distribution in rp and in rr: (Thble 15). A number of data points (X) with lowest ratio of n;;., are determined from each set of data points. Data points with lowest 7t::/ rp are used because the boundary line should be the lower boundary of the data set (eh. D.1.2.1). Slope and intercept of a linear regression of these X data points are computed for each of the 300 sets of data points, resulting in 300 regression lines. The --mean-and standan:hmm of the ··'Slopes mcHmercepm--or these 300 lines are calculated. The number of data points (X) used for the regression, is varied from 2 through 30. Fig. 59 illustrates the procedure for X = 2. The mean and standard error of the intercept and the slope of the 300 lines are shown versus X in Fig. 60. If 6 points are used for regression, the values for the mean intercept and mean slope become robust, e.g. change only slightly if more points are used, and the standard errors become approximately constant. The value for the mean ~JQ3. ~g~iQ~§. ..~ .~te!l •. witll..th~ Yi&ually determined boundary with a slope of 0.0113 in Fig. 41 (eh. D.1.2.1). Having determined the numoer or pointsT6J neeessacy·ro compute-alower boundary, the next step is to compute the reliability of this boundary. I ~ t~ f2 • - 0 original dllta point 1st genendecl data Ht 2nd~ dlltalllll: ·-- .. am generated data aet -"~"i"""Jf-.......... lllannlluullydlpirullrecllort..,._(clef)·· ·· Fig. 59. Sketch showing the procedure ID calculate the boundary line for the 'sliding criterion' for X = 2 (e.g. boundary line based on 2 dam points). ...... 0.8 .,!.. ,-----------------------------.-0.016 slope - 0.0113 --- -_;:::.=.--~--=.::::...-~---- l0.6 f 0.4 I o.2 E 0 0.0121 o.008~ -~~-------------------------------------- 0.004~ ~~+----,----,----,----,--~~--~6 M 0 5 10 15 20 25 30 ~ I l I I I I I I \ I \ \ \ Determining lines of equal probability - -. + J.. 0.008 f4 0.008 ~ intercept '--,--~------------------------------------- 0.002~ For each of the 300 regression lines (which are 0+----,---,----,----~---,---~0 based on 6 data points with the lowest 7t::/ rp ratio) 0 5 10 15 20 25 30 the rr: value is computed for f' = 5o, 10°, 20°, x (number of points for regression line) (-) .. , .. , 80° and 85 o. The distributions of the rr: Fig. 60. Mean and standard error of intercept and slope of boundary values for rp = 5o, 10°, etc. are determined and lines vs X, for 'sliding criterion'. the cumulative probability is calculated for 5%, 30%, 50%, 70% and 95% (Fig. 61). The percentages indicate the probability that a discontinuity with a measured rp and rr:, will not cause a slope to be unstable due to sliding over this discontinuity. The probability lines, except 50%, are clearly curved due to the lower data densities for low and high values of rp. The probability lines are fitted to second degree polynomials with correlation coefficients over 0.999. The coefficients of the polynomials are listed in Table A 18 (Appendix I). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 133 D SLOPE STABlLlTY PROBABlUTY CLASSIFICATION· SSPC .. .. .. .. .. - 1.00 .................... ·········· .......... ; .......... ; .......... ~··········~··········~··· I ~ ~ :e • • discontinuity stable 0.80 ......... with respect to sliding : § ~ : ~ : ~ • : : ~ : . : ~ ......... ~ .......... ~ ........ : ~ ~ : • : ~ -~ : : : . . : : : t - . .,........... . .. . .......................... :e; 0.60 .......... ; .......... ; .......... ; .......... ; .......... : 0 . . . . . . .. .... ~ .......... ~.......... -~· ......... -~- ......... . 0.40 .......... ;, ......... ;, ......... ; . : : : ~ : : : : : ~ . . . : discontinuity unstable ~ with respect to sliding : : : : : : . : . : . : . : . : _ ~ F- : 0 0.20 ......... . Cl) . . . ...... ·: .......... : .......... : .......... : .......... : ..........:......... .. : . .. .. .. .. .. .. 000 ~--~~-'--~-----·~----~·----~·----~·------·~----~·----~ 10 20 30 40 50 60 70 80 90 0 AP {• apparent discontinuity dip In direction slope dip) (deg) . Fig. 61. Sliding probability for orientation dependent slope smbility. D.2.2.2 Probability of 'toppling' criterion Determining boundary line Determining the boundary line for toppling (eh. D.l.2.2.2, Fig. 46) is done in the same way as for the 'sliding criterion'. All .discontinuities which kinematically allow for toppling, irrespective of the orientation according to eq. [17] (page 99) are included. In Fig. 62 the mean and standard error of intercept and slope are plotted (analogous to Fig. 60 for sliding). For 6 data points, With the .--------------.,-0.012 z l i lE 0.2 1 :rJ;;~j::~~~t:.:~-:~ra:= 0 '1 .inte~ _______ ,.,.- /"' ~,."'"' /"' 52. ,."' __ , .... ______ _ ~ ~ o.008 ~ 0 'i 0.004 ..,!, ...... 0-+0--.,... - - . ,20 .- -25 -.-----30 -l- 0 5 - -1r0---, 15 ~o.A .-------------:c-=. \ .1-·· - \ ~·z~·~:: ::·:::: i more than 14 data points. Between 6 and 14 data points the value is, however, approximately constant. As the minimum number of data points is required, the increase of the mean intercept value above 14 data points is not important< 112l. // slope .. 0.0087 ~Jrk-a~~..~,~~m"~~,~··-···-·· and slopes become robust and the standard errors ----------------~-- 1i 02 o.oos ~ 0.004-t ~··~----------------------------- ~! u ..... . _ , . 0+---,---,--.---.--.---+0 0 5 10 15 20 25 30 x (runber or points for regression line) (·) Fig. 62. Mean and smndard error of intercept and slope of boundary lines vs X, for 'toppling criterion'. Determining lines of equal probability The cumulative probabilities fur toppling are computed analogous to the 'sliding criterion' and polynomials are fitted. The coefficients for the polynomials are listed in Thble A 18, Appendix I. The cumulative probabilities are the probability that a discontinuity in a slope is not the cause for toppling fuilure. The lines are plotted in Fig. 63. <112> The more data points are used in the regression the more the line moves into the data set. If all points of the data set are used, the line is the linear regression line of the whole data set. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 134 D. 2 Probability anal)lres -f ~ 1 00···········= ...........:.................... ; ................................................. . 11 - ~ c § =a : : : : : : . 0 80 .......... · J.... : : 95% . . discontinuity stabJe with respect to toppling : . 70% ··60% 30% . 5% 0.60 ··········~··········~······· .. ·~··········~·········· ......... · . 0 . . . .. .. .. discontinuity unstable ............ ........... ! .. with respect to toppling ..... .... ... .. . . . .. . . ... .. ... 0.40 .......... -~· ......... ~· ......... ~ . . • . ~ . .. : .......... ~ : ........... : ........ 4 ... : ........... . ~ 0.00~------~---------·-------·------·------------------------~ 0 10 20 30 40 50 60 - 90 - AP + dip a~ope (deg) 70 80 90 (AP • apparent discontinuity dip in direction opposite slope dip) Ego 63•.. Toppl.ing probability. for orientation dependent slope.-stability................................ 0.2.3 Probability of the orientation independent slope stability The probabilities of orientation independent slope stability are calculated for the shear plane model (eh. D.l. 3. 5). For the linear model (eh. D.1.3.4) only the mean values and standard errors of the :fu.ctors (aO through a5) are calculated. D.2.3.1 Probability of the linear model for orientation independent slope stability The linear model relates linearly the visually estimated stability class with the slope geometry parameters (dip•• , height.~opc) and the rock mass parameters (irs~ spaii!QU an(} con-J. A set of these data points is generated randomly out of the original data set with on each parameter of the original data points, an error distribution. The error distributions are normal distributions with mean values 0 and standard deviations as discussed in eh. D.2.1. The visually estimated stabiliti.es of the slopes belonging m the generated data points are the same as the visually estimated stabilities of the slopes belonging to the original data points. The fuctors (aO through a5) in the linear model (eq. [23], page 105) are calculated with this generated set of data points. The procedure is repeated with newly generated sets of data points, leading to new values for the fuctors. The mean values and standard errors of the fuctors belonging to all generated data sets are then calculated. New sets of data points are generated and the newly calculated fuctors are included in the calculation of the mean values and standard errors of the fuctors until the mean values and standard errors become constant. Fig. 49 and Thble 9 (eh. D.1.3.4) show the resulting mean values and standard errors. D.2.3.2 Probability of the shear plane model for orientation independent slope stability Determining mean values and errors for weight factors of the shear plane model A probability analysis analogous to the linear model is done for the shear plane modeL Sets of data points are generated randomly out of the original data set with on each parameter (dip,~ope, heightstope• irs, etc.) of the original data point an error distribution. The number of points in each of the newly generated sets is thus the same as the number of points in the original data set. The error distributions on the parameters are normal distributions with mean values 0 and standard deviations according to eh. D.2.1. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SI'ABIUTY PRO&BIUTY CLASS!FlCATION- SSPC 135 1~>~------------------------------~ i I ~r---,,~, ___________,__ J-----------------~---------- 1· I I I I Q I 0 20 40 IILII'II!Mr et optmiallolra (-) feaor.O(-) Fig. 64. Mean value and s1lmdard error ibr factor aO in shear plane model vs number of optimizations. Fig. 65. Distribution of aO after 72 optimiza- tions. The filctors aO through a6 in eqs [24] and [27] are (non-linear) optimized following eq. [28] (page 109). The procedure is repeated with newly generated sets of data points, leading to new values for the factors. The mean values and standard errors of the factors belonging to all generated data sets are calculated. New sets of data points are generated and the newly calculated factors are included in the calculation of the mean values and standard errors of the factors until the mean values and standard errors become constant. For factor aO this is shown in Fig. 64 and the distribution of aO is shown in Fig. 65. The mean values and standard errors of factors aO through a6-are~tisted in 'BlbleiO (eh. &.1-:3":5)~ Lines of equal probability The lines of equal probability for the orientation independent stability of a slope according to the shear plane model (Fig. 67) are obtained as folc 0 .. lows. 640,001 (j = 0 to 640,000) sample data points are randomly generated out of uniform distributions from all possible intact rock strength J. 95 values (0 through 150 MPa), spa-, values (0 :! through 1) and con,_ values (0 through 1.0165). dip.r~ope values are randomly generated out of the ::1! rang~J!QIIJ.J() 0 !<>. ~~~.11Jlc:i ~~jor lJ~~~~pcJl!~ randomly generated out of the range from 2 m through 25 m or 50 mC113>. For each of these sample data pomts·oo are calculated the (" - ;1·md (H,.J1 (fOllowing eq. [27], page 108) with the ~-~~---:r-----:T"C----.-----l0' 1o.o fil.ctors (aO through a6) equal to the mean values 1.0 ,_/dip-. (Th.ble 10, eh. D.l. 3. 5). The ratios of ( tp,_)1 over (dip,~op)1 and (H,.J1 over (H,~1 are calculated and Fig. 66. Example of distributions for the calculation of lines of equal probability for orientation independent stability for the shear plane result in the points: ( tp,_Jdips~ope. Hmo/H,~1 . model. fPmass and Hmox are also calculated with all the pairs aO through a6 found in the optimization of the shear plane model (see above). For the filctors aO through a6 pairs of aO through a6 are used, e.g. (a00 , al 0 , a20 , a30 , a40 , a5a. a6o), (a01 , al 1, a21, a31, a41, a5I> a61), etc., because the factors are likely not independent. There have been calculated 72 pairs of factors aO through a6 (i = 0 through 71) and thus for every pointj are calculated 72 points i, resulting in: {tp,_,dips/ope' Hmo/H,~1• 1• If (tp,_Jdip,"'p)1, 1 < 1 and (Hmo/Hs~J. 1 < 1 the point S <113> The calculations have been done for two ranges for the slope height: up to 25 m and up to 50 m. Most of the slopes in the research area have a height less than 25 m so that the probability lines for these slopes may be regarded as more certain than the probability lines for slopes with a height up to 50 m. Therefore the probability lines in Fig. 67 are continuous for slopes with a height up to 25 m and dashed for slopes with a height up to 50 m. For higher slopes no probabilities have been calculated as no field data are available in the research area. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 136 D. 2 Probability analyses represents an unstable slope whereas if ( tpmas/dip3Wpe}i. 1 ~ 1 or (HmaxfHswpe)i, 1 ~ 1 the point represents a stable slope. The points representing a stable slope are counted for every point j. The total is divided by 72 and multiplied by 100 %. Hence, fur every point (tpmas/dip8tope, Hnuz/HMope)1 is thus established the percentage of the points i representing stable slopes and thus what the probability is that a point ( fPmaa/dips~ope, Hnuz/Hswpe)i represents a stable slope. The procedure is illustrated in Fig. 66. Curves<ll 4) are :fitted with a least squares method through the points with an equal probability of 5, 10, 30, 50, 70, 90, 95 % and are shown in Fig. 67. 10~--~~~~-------------------------------~-~Dalhed probebllly ... ._~nc~~ca~e hit the number of llopea used for the ~of IM SSPC 8Y*m for 1heae MC11on1 of 1l'le graph 18 llmlled and the probability lines may not be 8S certain 8S the probability lines drawn with a continuous line. 95%...... probability to be stable > 95 % .·· :90% ·--.... ::::::·10% probability to be stable < 5 % 6% I i Example l, old road cut exposure .A ! probability to be stable • r:~ 75 % 0.1 0.0 0.2 0.8 0.4 0.6 SFRI I difJslope 1.0 Fig. 67. Probability of orienmtion independent slope smbility. \hlues indicate the probability of a slope to be slable. 0.2.3.3 Probability of the cohmau and 'Pnwss The distributions of rock mass parameters cohmass and 'Pmass of the shear plane slope stab1tity model ate Used rot the. qwmtilicanori onlie error in the parameter for weathering orily (eh. D.2.4.2). The error distributions .ofcohirioss aOO·f'us are detel'mined M· fellews. 4(H ·emple data points are generated randomly out of unifurm distributions from all possible intact rock strength values (irs, 0 through 150 MPa), from possible spacing of discontinuities values according to Thylor (spamau, 0 through 1), and from all possible weighted condition of discontinuities values (conmass, 0 through 1.0165). On each sample data point 401 disturbances are introduced out of the error distributions, which are normal distributions with mean values 0 and standard deviations confurm Table 15 (page 130), giving 400 data sets. For each data point of each set cohmass and 'Pmos. are calculated fullowing eq. [29] (page 111). The mean value and error distribution are determined for each sample data point. Fig. 68 gives an example of the data set of one sample data point. The error distributions are normal distributions except fur the distributions fur sample data points which are calculated from values at the extremes of the range, e.g. irs = 0 MPa or spamtJS3 :::; an .,..{deal Fig. 68. Example of the distribution of one dam set of .,.....,. = 1. 00, etc. For simplicity these are also <114> The formulae and fuctors used for the curves have no meaning other than giving a best line represenmtion through the points (the formulae and fuctors are given in appendix I, 'Thble A 19). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl'ABIUTY PROBt4BlUTY CLASSIFIOfl'lON- SSPC 137 assumed to be normal. The average of the standard errors of all data sets equals 7.5 % respectively 9. 5 % of the mean value of cohltUU, and fiman· 0.2.4 Probability of the values for the method of excavation and degree of weathering parameters The probability of the values for the parameters for the method of excavation and the degree of weathering (chs. D.1.4 and D.1.5) are found by Monte Carlo simulations. D.2.4.1 Probability of the values for the parameter of the method of excavation The method of~ ~uences only the spacing of the discontinuities (eh. D.1.4). The standard error distribution on the spa_ parameter calculated according to 'Th.ylor, equals 0.003 ('Dlble 15). The error distribution on the method of ex.cavati.on is as follows. The estimation of the method of excavation is assumed to be certain for the classes 'na.tw:al', 'pneumatic hammer excavation', and 'pre-splitting/smooth wall blasting' (these can normally be easily recognized in the field). The classes for the quality of conventional blasting are subjective and the assumption is made that there is a uniform and discrete distribution from one class above until one class below the estimated method of excavation, e.g. each class has a probability of 113. For the classes at the extremes, 'good conventional blasting' and 'conventional blasting with result crushed rock' the distribution is uniform and discrete from one class below respectively from one class above the estimated class through the estimated class, e.g. each class has aprobability of 1/2. A Monte Carlo simulation is run with randomly generated data sets out of the original data from the field with the error distributions as described above (Thble 15). Equations [33] through [37] (page 117) are calculated with those data, resulting in values for the parameter for the method of excavation. The procedure is repeated with newly generated sets of data points, leading to new values for the parameter for the method of excavation. The mean values and standard errors of the values for the parameter for the method of excavation belonging to all generated data sets are calculated. New sets of data points are generated and the values for the parameter for the method of excavation are included in the calculation of the mean values and standard errors of the values for the parameter for the method of excavation until the mean values and standard errors become constant (approximately 150 times). The results are listed in Table 12 (eh. D.1.4.2.2). D.2A.2 Probability of the values for the parameter ·Of the degree of weathering The de~ of \Ye&f!leri!lg influences all ro<:Jc lll8:~S parameters (as di:scussec! in eh. D.l. 5). The error distributions of the intact rock strength, spacing of discontinuities and discontinuity condition (1C) are calculated in eh. D.2.1 and the rock mass parameters cohman and fiiiiiiSS in eh. D.2.3.3. For all calculations of the error in the values quantifying the influence of weathering on the different rock mass parameters, the same procedure is used. This procedure is analogous to the procedure used for the parameter for the method of excavation (eh. D.2.4.1). The error distribution on the degree of rock mass weathering of the exposures is assumed to be uniform and discrete from one degree above until one degree below the estimated degree of weathering, e.g. each class has a probability of 113. For the degrees at the extremes, 'unweathered' and 'completely weathered', the distribution is uniform from one degree below respectively from one degree above the estimated degree through the estimated degree, e.g. each class has a probability of 112. A Monte Carlo simulation is run with randomly generated data sets out of the original data from the field with the error distributions as described above. Equations [39] through [43] (page 121) are calculated with those data, resulting in values for the parameter for the degree of weathering. The procedure is repeated with newly generated sets of data points, leading to new values for the parameter for the degree of weathering. The mean values and standard errors of the values for the parameter for the degree of weathering belonging to all generated data sets are calculated. New sets of data points are generated and the values for the parameter for the degree of weathering are included in the calculation of the mean values and standard errors of the values for the parameter for the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 138 D. 2 Probability anal)ses degree of weathering until the mean values and standard errors become constant (after approximately 100 recalculations). The results are listed in 'Thble 14 (eh. D.1.5.5). D.2.5 Conclusions The large number of field observations allowed for a probability approach of the SSPC system. The difrerent probabilities analyses calculated in this chapter have been incorporated into the Slope Stability Probability Classification (SSPC) system as described in eh. D. 3. Generally the error distributions of the rock mass field data are conservative. It should be noted that the same observations done by experienced users of rock mass classification systems would likely result in lower errors. In the opinion of the author this is no problem as the SSPC system is likely to be used by experienced and unexperienced users. Experienced users will note that the results based on the SSPC system may be conservative and will interpret the results accordingly. It is, however, highly unlikely that an unexperienced user would be able to recognize that the results are too optimistic and be able to correct for too optimistic results. A conservatism in the results is therefore rather advantageous. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SL'4EILI:IT f'RO.&lJllliT CLASSIFIC!J:l'ION- SSPG' 13' 3 for 'reference' rock mass correction for local in exposure of excavation used to make the exposure. The 'reference' rock damage due to mass state to characterizing the 'reference' rock mass can be coJnpare~d mass are of the that uwa.uae;~w due to the method of t:.'I:.Cavation to be used for excavation The probability to ch:ara,ctt:ru~atilon (H5) (116) should be calculated exa,mJJ•les are ~wesen:tea in eh. D.5. of an e'l{posure and calculation of the rock mass in consist for the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 140 D. 3 '17le complete SSPC system D.3 .1 Exposure characterization The complete exposure characterization form is presented in Fig. 71 (page 145). The entries in the form are discussed step by step in this chapter. ~ DATE: LOGGED BY: J!F Sun: Rain: -- !f/04/98 WEATHER CONDITIONS LOCATION I J Map coordinates: cloudy/fair~J!!:!ab! TIME;: lf)iJf) map no: northing: easting: !!!xfdrizzle/slightlheavy - ------ - --- -- NAME: colour ~ ./4 4Mit e11t - . ......... A AAI: 1!19.'Nf) 9115.6'10 IEIIU rv Size total exposure: /()() h: mapped on this form: (m) 1: H h: 9 9 d: d: 4 2 poor/fair/good ._,t~ tiJ8/ l exPosure no: 01 (m) 1: Accessibility: F0r!IVIAB 11.11'1 hr orain size I"- I DES IIBS 59 30: 1981 structure &. texture weetherino I -"'- WMI. _._~~MM :~ I NAME -~- I I The size of the exposure and the part of the exposure mapped on the form may be of help if at a later stage the significance of the description has to be determined. Accessibility and weather are recorded becanse experience teaches that if accessibility is poor or when the weather is poor the descriptions and measurements are less accurate. · · Expomre speeflic ptm:Uftlter: MetiWd of"acavmwifTMEJ The classes and values for the method of excavation have been determined in eh. D.1.4. METHOD OF EXCAVATION IMEl (tick) natural/hand-made pneumatic hammer excavation pre-splitting/smooth wall blasting convantional blasting with result: good o~n discontinuities dislodged blocks fractured intact rock crushed intact rock 1.00 0.76 ./0.99 0.77 0.75 0.72 0.67 0.62 Material property: Intact rock strength (IRS) I I Intact rock strength is estimated with 1 simple 1 field tests that are related to the strength classes of the British Standard (BS 5930, 1981) (eh. C.3.2.1). A standard geological hammer should be used (weight about 1 kg). A space is provided for sample numbers for intact rock strength laboratory testing. The values resulting from such testing should, however, be used with care, as discussed in eh. C.3.2.1.2. Exposure specific parameter: l*athering (WE) The degree of rock mass weathering is classified following British Standard (BS 5930, 1981, Table A 20, appendix V) (eh. D.1.5.7). WEATHERING IWEl (tick) unweathered slightly moderately highly comriletelv 1.00 .r0.95 0.90 0.62 0.35 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl'ABIUTY PRO&BIUTY CLASSIFICATION- SSPC I I DISCONTJNU111ES B=bedding C-Cieavage J-joint B1 (degreea) Dip direction Dip oe (m) I ----- 16 0.50 0.50 I I 5 . iM 16 0.4() I 4 93 92 _Ot/4_ ll~ (degreea) Specing (DSI 141 I -------------- I Discontinuity sets and the type of discontinuity, e.g. B(edding), C(leavage), J(ointing), etc. are established visually and indicated in the appropriate boxes. Characteristic orientations and spacings (DS) are measured and recorded for each discontiaity set. If necessary, scanline and statistical methods are used to establish mean values, although the comments made in cli. C.3 .4 concerning the accuracy of measuring methods should be considered. More forms should be u~ if more than five discontinuity sets are present. Single discontinuities (e.g. a single fault, etc.) are also recorded because the SSPC system can also be used fur a single discontinuity to determine the probability fur sliding or toppling fiillure. Spacing is obviously not applicable for a single discontinuity and an S (indicating single) is written in the appropriate space before the discontinuity set numberC11 7). Disconti.Jmity property: Persistence I I I B1 DISCONTINUITIES B-bedding C=Cieavage J=joint I persistence I along strike lmll 1 along dip lmll I 92 I 93 I I I I 4 I 5 I >24 I >2 I >2 L >eo I >2 J >2 I J I . . _____ !?!~~~~~~ty -~is~~J~~~ ~ lm4~~o,g dip1i.~~~~~-for each discon~~ty_(~t?t)_: !\_P~~-in~~a~g___ _ 1 laxger than' ('>')means that the discontinuity is continuous as fur as visible in the exposure. Discontinuity property: Large (Rl) and smtdl (Rs) scale roughness I I DISCONTINUITIES B•bedding C=Cieavage J=joint if1 I CONDITION OF DISCONTINUITIES wavy Roughness slight~ wavy curve large scale (RI) =~:tit_curvad Roughness small scale (Rsl -----~- -- rough stepped/irregular smooth stepped pelislled mpped rough unGula~ smooth undulating ron·anareaor·· ". f.ro~~ti'lll 20 x 20cm~ ---~-- smoo planar polished. Pllmar (}2 I 4 1}3 I 5 I I I I :1.00 :0.95 :0.85 :0.80 :0.75 :0.95 :0.90 0.15 0.10 0.10 0.10 0.10 :0.-Si- :0.80 :0.75 0.10 -~ .,__ "'''""'" :0.60 :0.55 ~~ ~~- I J I - The roughness of each discontinuity (set) is visually estimated according to Fig. 69 for laxge scale roughness (on an area > 20 x 20 cm2) and Fig. 70 for small scale roughness (on an area ~ 20 x 20 cm2). The tactile roughness classes, e.g. rough, smooth and polished, are established by touch. If the discontinuity roughness is anisotropic (e. g. ripple marks, striation, etc.) the roughness is estimated both perpendicular and parallel to the direction with the maximum roughness. The directions are noted on the form. If roughness profiles of both discontinuity sides are non·fitting (eh. C.3.3.2.6), this is noted on the form. The reduction of the friction along the discontinuity plane that is expected due to non-fitting may be estimated and samples for tilt or shearbox tests can be taken. Considering the difficulties and uncertainties related to shearbox tests, the estimation of the reduction of the friction angle, for example, with the Rengers envelope (Rengers, 1970, 1971) and tilt tests are almost always more appropriate than shearbox tests. The estimated or determined friction angle is converted into a value for the roughness parameter by multiplying this friction angle with 0.0113(118> (l17l Sometimes a single discontinuity may be better characterized and described as a separate geoleehnical unit. This may be necessary if the infill in the discontinuity is very thick. Often major fuults and iilult zones can be better classified as a separate geotechnical unit. The comments in eh. C.3.4.1 can be used as guidelines to decide whether to include a discontinuity in a discontinuity set or to classify a discontinuity as a separate geotechnical unit. (!18) Use is made of the 'sliding criterion' (eh. D.l.2.l). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 142 D. 3 The complete SSPC system if the estimated or determined friction angle is determined for the large scale roughness (Rl) only, or if the angle is determined for both the large and small roughness combined (RI and Rs combined). The friction angle is multiplied with 0.0151 if the friction angle is only applicable to the small scale roughness (Rs). A separate value for the parameter for in1ill material is not required if the estimated or determined friction angle is applicable to the discontinuity including the influence of in1ill material, for example, if a tilt test has been done with infi.ll material present. The same applies if karst is present. amplitude roughness ::::5-9cm wavy ::::5-Scm curved slightly curved : : 1.5 - 3.5 ~_]_ __ _ --,---- straight ::::1 m {i-angles and dimensions only approximate) Fig. 69. Large scale roughness profiles used for the slope stability probability classification (SSPC). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STABJUTY PROBUJIUTY CLASSIFICATION- SSPC amplitude roughness > 2- 3 mm stepped 143 1_ T amplitude roughness > 2 - 3 mm undulating l_ T planar = 0.20 m (dimensions only approximate) Fig. 70. Small scale roughness used for the slope stability probability classification (SSPC). Discontinuity property: lnftll material (lm) I DJSCONTINUmES B=beddln(l I c=tteavage J•Jolnt ir1 I 5 4 da d2 I I I I I I :1.07 cemented/cemented inflll ~;;~;~"niie-.i::fr~~~coa~e---~~~ material, lnflll ~free of 1 medaum :0.90 ~~~~~- --------r~~-----~~8~ 1.00 soft sheared material, :0.75 1 coaree :0.66 e.g. clay, talc. etc. 1 f!lldlum ----------------L~n~-----~~~ gouge < irregularities :0.42 :0.\~ ft~~~~arities :0.0 material llml I 1.00 I I 1.00 I Descriptions of the iDfiU material classes are given iD eh. C. 3. 3.4.3. If the iDfiU material is characterized as 'gouge > irregularities' the small scale roughness equals 0.55 (polished planar) (eh. D.1.2.1.2). Discontinuity property: Ksrst (l{lz) I I OISCONTINUITIES B=bedding C=Oeavage J=joint Karst ll<al 1none karst I t.'1 I I I :1.~1 :0.92 1.00 I (h I (}a I o.92 I o.92 I 4 I I I I I n The presence of karst features should be noted for each discontinuity (eh. D.1.2.1.2). 5 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 144 D.3 '1'he complele SSFC qstem I I I I SUSCEPTIBILITY TO WEATHEftiNG (SW) degree of weathering: ---------------- date excavation: remerks: -------------- ------------------------> 1/()~· ~----------- -------------- ~~~~~- I I H remarks: ~ Utt ~ 614 IIHI(. &u.. tll4tut """" 44IMfllt ( , _ , _ . . ) ~ - ......... Aft~~. MI!Mtfl#t~· --~ ~~ The assessment of the susceptibility to weathering (SW) in the SSPC system may be done by noting the degree of weathering in surrounding exposures that are in the same lithologic unit, together with the length of time these exposures have existed. If special circumstances have influenced the rate of weathering in the other exposures (tOr example: different orientation, permanent water ftow over the exposure, etc.) this should be noted in the space provided. Existing slope ? This intbrmation can be a reference tOr the reliability of the slope stability probability classification (SSPC) system. The stability classes are visually established. The description of the classes tOr the visual estimation of stability as used tOr this research can be used as guidelines ('Dlble 5, page 52). The classes that indicate a possible likelihood tOr firilure in the future, e.g. 'small problems in near future', class 2, and 'large problems in near future', class 3, may be difficult to distinguish. --- EXISTING SLOPE? dip-direction/dip --- -- --- -- 040/1'() r,()m heiaht: Stability (tick) stable ,/"1 small problems in near future 2 large ~:_roblems in near uture 3 ~ ~!1 £~'!:!! I .. : Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STABlU"l'Y PROJJtUJlU'l'Y CLA$SJFIOO'lON- SSPC Itick) natural/hand-made pneumatic hammer 811Cavation p,....spilttina/smooth wall blasting conventional blasting with result: good /()() h: 9 d: 4 24 h: 9 d: 8 (tick) unweathered slightly moderately highly completely 1.00 ./ 0.95 0.90 0.62 0.35 EXISTING SLOPE? ().75 o.ro o.ro o.ro o.ro (),Kf) !.()() !.()() 0.92 ().92 Roughness small scale (Rs) (on an area of 20 x 20cm2 1 lnfill material (lm) Karst !Kal !.()() ---------------1-·--·----------------Fig. 71. Exposure characterization (example I, old road cut exposure A, see eh. D.5.1). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ mass of The reference intact rock strength (RIRS) the intact rock strength of 132 MPa and that value it is ,.,.,,..,.,..,..,,.,... for the of mass weaute11mg at the location of the (!19) exposure sp~tC1llf2:S are used to calculate the <>v<>c~w•J<. mass..,.,,.., ......,."''"" form The Aw_,.,........15 For a rock ma.~ eq. and ·with one disccm:tJ:nuity set: ., 0.45 + 0.264 " with two discoi'!Jim4i:ty = 0.38 + se~: 0.259 * !og11l X~ "' 1 "'0.30 + 0.259 * ~. Hl + SPA "' (x " dis,;ominuiity A correction for the to the Jliiv'WJl'-"" "'0.20 + 0.298 * 0.333 * ~ three discontinuity sets with the and for the of ~ .<~rMrri11r.v.<~l (ll 9J in the reference is a~J1.nn:;u mass: RSPA =SPA I (1!9) Correction for is not nece.ssary for 'soH mechanical parameters of 'soil. the parameters tor the of in the exposure calculations of the reference rock mass and t'le etc. page 24 l) as the Therefore, for these type of units and at the location of the 1.00 in the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STA.BlUTY PROBABlLITY CLASSIFICATION- SSPC 147 Discommuity property: Re.fmmce condition of each discontinuity (set) (RJ'C) and Reference overall condition of discontinuities (RCD) Discontinuity: Roughness large scale ~~~~~~ssiJi~ii~~i!::::·::· ·'r:'~i.ll.~fl~'l.l.. Kerst CONDITION OF DISCONTINUITIES 92 93 4 8'1 IRII .. .... .fl,'ll...... (Fisi .tl,tfl.. . ............ llr:n..l 1.1)() urn:: & RCDI 6 . .f!,(f!.... ...... (J,tfl.... .. ... . (!,(!!... L f!..,((l. ...... .............. .. ···· ' ........li.K!. ... UX.l.. ...... ,............................. (Ka) ·· ·· ···"jjjjj······ · (), 92 (), 92 • RTC is the discontinuity condition of a single discontinuity (set) in the reference rock mess corrected for discontinuity weathering .. u.;To:::::ta=-l--:I"'"Ri'"""*Rs=-=.,:()I.5!J:!<--'---_;_---j • RTC .,. TC I sqrt(1.452 -1.220 " e"I·WEll 1m-::•-:-:Ka-.,...::TC:=;-jl-()!:!,6()=..-'--"'()~~S'-"'!tJ'-----;--..xu ().61 RTC Weighted by spacing: (),6() TC1 TC2 (),6() ----- + -- + - os 1 CD 052 (),6() TC3 0.59 0.59 - - + ---- + ------ DS3 = ------------1 1 1 DS1 DS2 DS3 ---- + -- + ---- ().1/() ().5() ().5() = ---------- - 1 (),1/() 1 1 ().5() ().5() =().59 + ---- + --- corrected for . RCD "' CD/ WE = 0.591 ().95 i The condition of discontinuity (1t) for each discontinuity (set) is a multiplication of the parameters for large (Rl) and small (Rs) scale roughness, infill material (/m) and karst (Ka): 1C RI * Rs * Im * Ka0 20>. The condition of discontinuity in the reference rock mass for each discontinuity (set) (IUC), is the condition of discontinuity (1t) corrected for the degree of weathering in the exposure. The correction parameter for the degree of weathering should be the correction parameter for the condition of a single discontinuity (set): RTC = 1C I sqrt(1.452 - 1.220 eWE) <119> (eh. D.1.5. 7). No distinction is made between continuous and abutting discontinuities. Non-persistent discontinuities (thus discontinuilies ending in intact rock) are characterized by changing the parameter for the discontinuity small scale roughness to 'rough stepped/irregular' ( = 0.95) (eh. D.1.2.1). CD is the weighted overall condition of a number of discontinuity sets in the exposure rock mass unit. RCD equals CD divided by the parameter for the degree of rock mass weathering (WE) cu 9>. = Anisotropic roughness The calculation of TC and RTC should be done for the minimum roughness and for the maximum roughness if the roughness is anisotropic. The condition of the discontinuities in the reference rock mass (RCD) and the reference rock mass friction (RFRI) and cohesion (RCOH), should be calculated with the average of minimum and maximum roughness. Reference rock mass friction and cohesion (RFRI & RCOH) REFERENCE UNIT FRICTION AND COHESION IRFRI & RCOHl Rock mass friction: RFRI = RIRS * 0.2417 + RSPA * 52.12 + RCD * 6.779 ...................................................................................................................................................................................~1. .. ::'. 1.f.*Q,g:~1'!.:f..f!.:!l.t.~Ji~,J~.t ..fl.,f#..•~ . .5.,.?..7.~t..... · ........ '1/a. . . . Rock mass cohesion: RCOH = RIRS * 94.27 + RSPA * 28629 + RCD * 3593 RCOH = 7fJ * 94.27 . + ().85 * 21,1.629 + 0.62 * 3593 = \ 19Tl5Pa Rock mass friction and rock mass cohesion are calculated according to the formulae on the form. D.3.2.1 Determination of number of geotechnical units in a reference rock mass In the exposures the rock mass is divided in geotechnical units and the parameters of each geotechnical unit are described. After correction for the degree of weathering and for the method of excavation, parameters are determined that characterize each geotechnical unit in the reference rock mass. However, if the differences between the geotechnical units in the exposure(s) are caused only by a different degree of weathering or a different method of excavation then after correction for the degree of weathering and the method of excavation, these can be combined in one geotechnical unit in the reference rock mass. A form of averaging is necessary because the (!20) The parameter determined with the testing of discontinuities including infill material or karst or both (eh. D.3.1) already includes the infiuence of these parameters. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ditrerent exposures or zones in the m one exposure and rock mass is not nc"Vc:ss~tn< dis· {),6(} 0.59 ().59 + -.... ~-- + ---~~ (),«} 0,5() o.so ........--·--··--··-----···· = 0.59 ~---~~"' CO "" ---·-···-······-·-·-··· 1 1 1 ---- + ,_____ + ---DS1 DS2 DS3 : ........ + ------ + -----(),4() ().50 ()_!}() 72. Reference rock mass calculation (example I, old mad cut exposure A, see eh. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SDUJIUTY PROIWJIUTY CLASSIFIC4TION- SSPC 149 D.3.3 Slope stability probability The stability of a new slope is assessed with the new slope made in the 'slope' rock mass. The parameters characterizing the geotecbnical units in the 'slope' rock mass are obtained by correcting the parameters characterizing the geotedmi.cal units in the·~· rock mass for the damage due to the method of excavation to be used for a new slope and are corrected for the decay of the rock mass due to future weathering< 121>. This latter is achieved by estimating or guessing the desree of weathering of the geotecbnical unit in the slope rock mass at the end of the ~ lifetime of the new slope. The probability of the slope to be stable is then calculated for a slope made m dlis 'slope' rock mass. For each geotecbnical unit m the slope rock mass the orieatati.on dependent and orientation ~t stability are determined. The orientation dependent stability assesses for eaeh of the ~ (sea) the probability for sliding and toppling along that discontinuity or discontinuity set, and the orioatatioo. ~ stability assesses the probability of a slope to be stable with respect to :fa.i.lure meclwri.sms that are not «irectly related to a discontinuity. The form to calculate the slope stability probability is prescmed in Fig. 74 (Me 153), with the data for the slope of example I that has also been used for the explanation of the exposure characterization and reference rock mass calculation. i map.. '.lo: LOCATION Map coordinates: .............. ! easting: . ··~" Slo~ FJII'etTy .... The SSPC system can only be used for a slope of which dip, dip-direction and height are broadly uniform. This means that if a slope is curved laterally, e. g. the dip direction of the slope is . .... .. . .. .... . . '!'!.$..-:«! ....................... m ......... _.D=E=TA=·1 l=S-=O.,_F..:S...,L~......,._IE- - , - - - - - - - - - - - i l ·· ······· ············· ········ ···· · ··············· ·························· ······ · ··· · ································· .............. ~19P.E! ..~.!P... ~irE!'"~iQI} .. (~I!9r.!J.E!li.I.:.Q~ . . $.!gp!! 4!P.J~1!9rl!l!§): . . .. f..Q varying, the stability of the slope has to be assessed in different .................. ... ............ . .. vertical sections where in each section the dip-direction is . . .. .J::!l'i91:!1.J~P~J.)Jml; ...... ...fl) broadly uniform. The same applies if a slope dip or height changes laterally along a slope. If the slope dip changes vertically the slope should be assessed in di.fterent horizontal sections for which the slope dip is broadly uniform. The height of the slope is the height from the bottom of the section assessed to the top of the slope. It may also be necessary to divide the slope in horizontal sections and to determine the slope stability per section if the slope is benched (eh. C:2.1).1fmore tban·one geotecbnical unit is present at the location of the slope (thus in the 'slope' rock mass) then the stability of the slope should be calculated per geotechnical unit. The height of the slope is . hilreii as tneneigtirrrom1neoommroTlliegeoteamrcarfunl assessed"rollierop ortfie ·STOpe: DETAilS OF SLOPE WEATHERING {SWE) METHOD OF EXCAVATION ISME) (tick) natural/hand-made pneumatic hammer excavation pre-splitting/smooth wall blasting conventional blasting with result: gp.~ discontinuities dislodged blocks fractured intact rock crushed intact rock ltickl 1.00 unweathered 0.76 slightly .t 0.99 moderately highly 1.00 .t 0.95 .. 0.90 0.62 ····· 0.35 0.77 completely 0.75 0.72 note: SWE = 1.00 for 'soil type' units, e.g. 0.67 cemented soil, etc. 0.62 The method of excavation which is going to be used for the new slope (SME) and the degree of weathering of the rock mass at the location of the slope (SWE) that is expected at the end of the engineering lifetime of the slope, according to the British Standard (BS 5930, 1981) (appendix V, Table A 20) for rock mass weathering, are noted. <121> If an existing slope is assessed the value fur the parameter for the method of excavation of the slope (SME) is equal to the value fur the parameter fur the method of excavation of the exposure (ME). The same applies fur the value for the parameter fur the degree of weathering (SWE = WE). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 150 D. 3 The comple#J SSPC system Slope rmit mune I SLOPE UNIT NAME:,,, «-,.,...,, ... ~ ±~ Material property: Slope inlsct rock strength (SIRS) INT.&.CT ROCK STRENGTH (SIRSI ..... ·············siRs ·:·itiris it;;;;;;·~~~··;;;~k·~~~;· ~· swe·i~~th~~;;:;::;~i~~~;·:···ifi·~·i£95·: The slope intact rock strength (SIRS) equals the refurence intact rock strength (RIRS) multiplied by the parameter fur rock mass vveathering at the location of the slope<122> (SWE) <119>< 123>. .......................................... ....... .. ...................................................................Q.!~C9.N:TINV!IY..~PAQNQ. .1$.$.P.AI .......................................................... SSPA = RSPA (from reference rock mass) • SWE (weathering slope) * SME (method of excavation slope) SSPA o; 0.35 * ().95 • ().99 = i (),39 The overall discontinuity spacing parameter fur the slope is determined by multiplying the reference overall discontinuity spacing (RSPA) by the parameter fur the method of excavation fur the new slope (SME) and by the parameter for rock mass weathering at the location of the slope<122> (SWE) <119>. Discontinuity property: Slope overall coNlition of discontimdties (SCD) .................................................................................................................. G9N.Pf!:!9N ..9f..PI$.G.Q!'ffi.NVm~~~$.~P.I ...... . SCD = RCD (from reference rock mass) * SWE (weathering slope) The slope overall condition of discontinuities (SCD) equals the reference overall condition of discontinuities (RCD) multiplied by the parameter fur rock mass weathering at the location of the slope< 122>(SWE) <119>. Rock mass friction tmd cohesion (SFRI & SCOH) . . .. Rock mass friction: SFRI = SIRS ...... ''' ' ~!,Qf.'~.VNJTffi!G.TIQN ANR G.QH.t;$.!9N * 0.2417 + SSPA * 52.12 + SCD * 5. 779 " ...... ... .. Rock mass cohesion: SCOH = SIRS " 94.27 1$.1"~! ~ $.C«;mt . ... , . . $.f.~!':':.l$~.Q,,~4l?.±!?J.'J.'~~~,J~±(2~9~~,]7~:::.! + SSPA " 28629 + SCD * 3593 : SCOH = 15 • 94.27 + OM • 28~29 + 0.59 * 3593 ·~ ! The rock mass friction and rock mass cohesion fu:r·the slope are calculated accordtn:g to the furmwa.e 99.~. IK6SKPa on ·the furm. (I:ZZJ The existing degree of rock mass weathering of the rock mass at the location of the slope should be used if the stability of an existing slope is assessed. The degree of rock mass weathering that is expected to exist at the end of the engineering lifetime of a new slope is to be used if the stability of a new slope is assessed. <123> A problem can arise if the stability of an existing slope is determined. The maximum of the intact rock strength for the reference rock mass of a value at 132 MPa causes that fur an existing slope the intact rock strength could become lower than the intact rock strength measured in the rock mass of an existing slope. To avoid this problem SIRS should be taken equal to the intact rock strength as measured and described on the 'exposure characterization form' with a maximum of 132 MPa. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SXiJJJLITY J='ROBA.l:tlUTY ClASSlFICAT'ION · SSPC Orientation dependent .wta.bilitv The , ....,.,,,,.,.,nt ilie dJlSCCI!ltiltlJll1-::f disc:ommuit~ Discontinuity property: condition discontinuity (STC) is at (set) should be used, for weathering of the entire :rock mass. not the chs(~ontimlity 151 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ r;-,.,.,;"'"'"" in same as is not occur. The resulting probabilities for the expected orientation dependent stability. This slope had 1). 10r---------------------------------------------~ probability to be stable > 95 % 95 o/!> : probability to be stable < 5 as f<Jnction of the to be stable: 3 mar "' maximum rougf!JI~ss am~ctum of minimum ro~;:gnJ'!eS'iS) A [46} Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SJMJJUTY PROB4.BJUTY CLASSIF1C4.TION- SSPC 153 The roughness parameters (ror lmge and small scale) calculated with eq. [46] should then be multiplied with the value for the parameter i>r infill material and with the value for the parameter i>r brst. The resulting value replaces RI'C in the calculation of orientation dependent slope stability. Unit weighti/UW The iilctors in the SSPC system have been optimized based on rock masses with an intact rock unit weight of 25.5 to 27 kN/m3 and rock mass unit weights of around 25 kN/m3 • A correction of the calculated maximum possible height (H-J should be applied if a rock mass unit weight is difrerent. This correction equals 25/Unit Weight"ma.s (in kN/m3). Rock masses with a high porosity and permeability (e.g. with a large storage capacity) may exhibit ~=~:"!.~=, ~-=~,;=:~=~~e DETAILS OF SLOPE WEATHERING ISWEl METHOD OF EXCAVATION ISMEl (tick) natural/hand-made pneumatic hammer excavation pre-splittlng/smooth wall blasting conventional blasting with result: !tick) 1 _00 1.00 unweethered ./ 0.95 0.76 slightly 0.90 ./0.99 mod-tely highly 0.62 0.77 completely 0 · 35 0.75 0.72 note: SWE = 1.00 for 'soil type' units, 0.67 e.g. cemented soil, etc. 0.62 ;:~ discontlnuities dislodged blocks fractured intact ro!Jk crushed intact rocK .- ' ~ NAME:.t41. . Slope dip direction (degrees): . .... . . . .................... Slope dip (degrees): Height !Hslopel (m): ---.~.~Mtt ORIEN'TA"'""ON INDEPEIIIDENT STABIUTY INTACT ROCK STRENGTH (SIRS) · ·· •SiRS::••Rifts.ii~~~:a:ffi~~~~--·~··swe:·7~~iii:fii~1riii i1i6iir ;;;•-•1fi··~·-i£95::;. ·' ·· 75·· ··············· ·· ············ .............................. ..... ....... ..................... • ............................................ .. .P!$.c.9.!IDNVIT.Y .!?.PAG!NGI$$.~L ...................................................... . SSPA = RSPA (from reference rock mass) • SWE (weathering slope) • SME (method of excavation slope) SSPA ·sc:o·~·-·Fico = 0.55 • 0.95 • 0.99 = ' 0.53 CONDITION OF DISCONTINUITIES (SCDl . itrom.reter&ilce·n:;c;k.··;n&&lii····sweiW&iitilEiiiilg.sloilei························································ SCD ·Rock. liiali& . tri<:ii·o;;:··s"F"Ri··;;;; ·sifis •o:2417···+-·ssf'A *52ji. +sco = 0.62 * ().95 = • ().59 SLOPE UNIT FRICTION AND COHESION (SFRI & SCOHl ····s57s·················· · · · · · · · · · · · · · · · · · · · SFRI = 75 * 0.2417 + 0.35 * S2.12 + 0.59 * 5. 779 = . ·Fioci< miisscotiiision: scoii··;;;;siils *94:2'7 +. ssi>A··· 2ss2s +. sco··· 3593······································ ·············· ························································· ····· ·················· ····· SCOH = 75 * 94.27 + 0.35 * 28629 + 0.59 * 3S93 = ..............................................................................................................Jf $ffii.<;J!!QP.f!l ~iP.; .MA>.<!MV.M .. ~bQPJ;:. HfJ!G,!:ITI.!:I.m!!.X.J ................................................................. . Maximum possible height: Hmax = 1.6 * 10"" * SCOH * sin!slope dip)* cos(SFAil/ (1-cos(slope dip- SFRI)) $'!. ~ . 1¥69¥ Pa ratios: ORIENTATION DEPENDENT STABILITY DISCONTINUITIES 4 .. .. • .. ~J... .. P.? .<.~agrees) J.f.t) ...L ......f/44 (degrees) L Ql... .f.6. ... oil> ciire.ciioil · · ..................................................... ........... .. ...... ..... .. . . . Dip wittl. Aliaiilt~i; ve~iC.~I~r.~~~~~ RTC !from reference form) sic ; ; ·i:ITC • 5Cirt:i1":452 ~ 1":226 * o 5 .:r:::::··:·::. . . .................... ~ ............. ~ .... •... (de!JrS.Iilt;l .... J~t ¥6 .................. L. JI.,~L.... ()_,60 ii;;.i~sweii ·· · ............ AP . --~- .f$. ...................... 0.60 . · iii9 ·· Probability stable: Determination orientation stability: calculation AP: Ill = discontinuitY diD. a - sloDe diD-direction -r - discontinuitv diD-direction: ll = a- 1:: AP = arctan loos ll • tan Bl stebllitv: sliding topplina stabilitY: slidina toDolina AP > 84° or AP < -84° (slope dip+ 5°) < AP < (slope dip-S 0 ) < ~P 'lsi.Ooa.dio + s • 0° < AP < 84° < (slope dip-S 0 ) with 100% 100% equal 100% 100% with use 9reph sidmg 100% AP < o• and f-90° - AP + slope dip) > o• against 100% Fig. 74. Slope stability probability calculation (example I, old road cut exposure A, see eh. D.5.1). use graph toppling Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ COMPARISON 1 SSPC DA.Ll D.4. 1.2 slope classification •• ~~·~~..,V et at, 1991, eh. B.2.4. factors of 1. 2 and ·"·"'·"/';"'" of the is in accordance has been used. This to be "A.L;'""'"'"'' 024l The data used in this assessed to be stable or unstable at present Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SIABlllTY PROB.4BILJ"l'Y CLASSJFlCAT!ON !:;'..'!PC 155 a: SSPC SSFC um~ -l!l!i mmmw pr~ (%} +----------- ·11l .lj H!llil'l!ll!!' ~alp· ~g lllf!ljl!) dip ---~---~---->- ul'llrtlllbi& $~ ~------ (~} _________ ____,... ~ll!G The vl$uafty esti~ :!rtl:lbilities of the research area compared to 1:1"16 """'......"'"'"""" dlffemnt classllcmlon sy$'1.ems. ~of~ pttr~lly~~!l:yciass. ?er0111nfllll11!® are ffOI'II kltal visually ~tta stabi~ty St!ilblll! (l'l!.ll'l'iberofllil:ljll$$: 109) IIIo a~gM otpms!lflt or I!!Wm alcpe falwrM S!mtll probltllml (111.1~ of~: tc) i'lw llo!J$ pr~ ~ lllgm of dvi* smmi ~RhM-~Ii:!!llili'Mtll'41~fallures !.~QQ ~{number of lliO!*t= 55) i'lw slop!i! prH!mtly ~signs ol ad!Ve ~~ ~ 75. of siope classification and c: Ro:mana's SMR and M8 !he f,lOtw'ltlai for i'llture large ml!urM different classification systems. a: SSPC system, b: um>m\re d1tt~re;1ce are Haim~s· ~,;uJullAucJ,cu ~-'"''""''....."""' for design cla.ssinc:au<ms systems is thus (!25) 1) 2) 3) 4) 5) The L:mhscher MRMR is calculated from the data collected for the SSPC system as follows: The intact rock estimation is used ior the I\1RMR. The MRMR fur the is calculated from ihe number of discontinuities per cubic metre = sum of number of discontrnuities per metre from all sels): RQD = 115 ·· 33 * Jv if Jv > 4.5 then = lOO % fur the spacing of discontinuities is calculated eq. [45] (page for a maximum of three dis:colilfulutlty sets. These are the sets wit.'l a maximum influence on the v-alue. The MR,\flt tbr ilie condition of discontinuities is calculated with the condition parameters from the di:Scmlli:r!uil:y set wiili the minimum condition value (thus with the :rrJI'Jmum IC value in the SSPC The vaiues used for the diffurent parameters :.u2d smaH scal.e a!terntion of waU and are those putms1:1ea Laubscher which a maximum water influence, e.g. the values pressure; > 125 d!s:co:ntllmilty set with me minimum condition value is not a set which is used of the of the discontinuities. we:am<~rn;g nor orientation are used but the adjustmeJi'l~ parameter and values for the method 8, page Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ DA.1.3 D.4.1.4 classification system c:onect!y clas- correctly clas- sified sti:lble sil'ied <mstable s!opes !visually estimated stabiilty slopes !visually estimated stability classification systems. as discussed therefore the better correllition the in the :re~H::an:n area. (120) 1) 2) 3) 4) RMR are calculated from the data coHected for the SSPC system as follows: estimation is used for the RiVl:R. for the is calculated from the number of discontinuities per cubic metre = sum of number of discontinuities per metre from all "" 115 - 33 Jv if Jv > 4.5 then = lOO % The RMR for the The RMR for the condition of discontinuit.ies is mken condition parameter 'lC D.l.2.l) from the SSPC system. The range of the values of the RMR for the condition of discontinuities is between 0 and 30 and the :range for the 1L:' parameter is from 0 to J) 165. ID obtain the for the condition of discontinuities for the RMR system from the 'lC value of the SSPC system the value has been with 30 I UH 65. The set with the minimum value for the condition of disconl:im.tities has been used to calculate the. Rt"fR the same set as the set with the minimum w,..~.~ ....,. Parameters related to orientation of the and discontinuiti.es F 2 and are calculated for the (.hSCOiltl!!Uii:Y which has a maximum adverse influence the The parameter for the excavation method The R\fR for 'vater is for a maximum water influence which leads ID a of 0 re: 5) 6) 7) Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE 5if)fBlU1Y PROlJAJ:JJ.LlTY CU'J;J,'IFWtTION SSPC: fu~ 157 ~ n::SI:2Jl!"G!L COJmn:re~;sn'e The ' str-e:ruzlth stress at a1 =2 * cohesion * tan (45° Romana •s SMR :mass of the eh. n 1 Both ~V'ill"f'TII1"l stress is expressed in the this is Jt\~A.U<\.U<A<VU + \ principal stress at o3 = minor »nnrtnnl stress a:t cohesion, friction "" cohesion and frict'.on this is o1 principal effective stress at o c ,. m~; and a are pwameters WMi J.VAUUUQIC""-' as: a~ "' minor principal effective stres..v at mtact rock strength the rock ma:ss tmd strl4Cture, e<mditil.m re'j:me!,en,r.<ttxve for the stress "''-'""'~";"'""'"'"'-'"" in the rock mass in the same as the toe of the slope but at some UA''""''"'"' the face. The major and minor principal stress values at failure are calculated only fur the purpose of the comparisons. \Vhether the calculated stress values at failure are true in reality is not evaluated and probably not very likely as the actual values will generally depend on more meters that are not considered in the "'""·"'""··'"'. . ''-'"""' as the of etc. (lZS) !ne minor stress for each is calculated with the Mohr-Coulomb failure criterion with the rock mass cohesion and friction calculated with the SSPC system and with the stress to the overburden pressure, e.g. to the of the slope with the unit This u 3 is then used in the calculation of the stress at failure to the Mohr-Coulomb fuilure criterion with the rock mass cohesion and friction calculated with the Rl'V1R system, and is used in the calculation of the stress at failure to the 'modified Hoek-Brown failure criterion'. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 158 D. 4 Resull3 and comparison .. . 10 10.-------------------------~ a. !. -= a. 1111 :I ! -; I + + + +i I I 4E "i 1 So 8 + 'C ; ...11:1. f Cl. 0.1 c I a: a: ::I ~ + 0.1 ++ + * + ID 0 0.01 0.01 ::r: 0.1 1 10 0.1 88PC major pmolpal lti'HS (WIPa) Fig. 76. Comparison of total major principal stress values at fiillure; left: RMR vs SSPC; right: 'modified Hoek-Brown failure criterion' vs SSPC. 0.4.2.1 SSPC system versus Bieniawski' s RMR system The values of the major principal total stress at ihllure (a 1) calculated according to the Mohr-Coulomb ihllure criterion (eq. [47]) with the rock mass cohesion and friction determined with the SSPC system, are compared to the major principal total stress values calculated with the Mohr-Coulomb ihllure criterion with the rock mass cohesion and friction determined with the RMR classification system (Fig. 76left). The total minor principal stress a 3 in each slope is calculated following footnote 128. The RMR rating is derived from the SSPC field data following footnote 126 and converted to fOCk mass cohesion and friction values. 0.4.2.2 SSPC system versus the 'modified Hoek-Brown fiillure criterion' The rock mass parameters in the 'modified Hoek-Brown ihllure criterion' (uc, mb and a, eq. [48]) are derived from Parame.ters m~s~ fc>r ~e SSPC syste~:129>. :fig. 76 rigl1t shows the major ptjnc!pal t«?tal stl;e§s at fiml!l'e calculated according to the 'modified Hoek-Brown ihllure criterion' versus the major principal total stress at fuil.ure according to the Mohr-Coulomb ihllure criterion calculated with the rock mass cohesion and friction derived from the SSPC system for each slope. The 'modified Hoek-Brown 1hllure criterion' is defined in terms of efrective stresses while the rock mass cohesion and friction from the SSPC system, used in the Mohr-Coulomb fuil.ure criterion, are defined in total stresses. A calculation done with efrective stresses for as well the SSPC system as for the 'modified Hoek-Brown ihllure criterion' showed virtually the same relation as shown in Fig. 76 right. <J29l The parameter Oc (inmct rock strength) is laken as the inmct rock strength field estimate from the SSPC system. Hoek et al. (1992, eh. B.2.3.5) also suggest a determination of the inlact rock strength by field estimation (although their classes and boundaries are slightly different from those used in the SSPC system). Hoek et al. derive the parameters mb and a from a matrix describing the rock mass 'structure' in four classes, and the 'surface condition' in five classes. For the analysis described in this chapter the 'structure' and 'surface condition' parameters have to be derived from the SSPC system parameters. The 'structure' parameter in the 'modified Hoek-Brown failure criterion' is related to the size of the blocks in the rock mass, and is therefore taken to be linear with the spaiii/JS3 in the SSPC system (e.g. spa,.,... > 0.75: 'structure' == class l, 0.50 < spa,11., < 0.75: 'structure' = class 2, etc.). The 'surface condition' parameter in the 'modified Hoek-Brown fuilure criterion' is related to the condition of the discontinuities and is therefore taken to be linear with conllftUS in the SSPC system (e.g. con,_, > 0.81: 'surface condition' = class 1, 0.61 < con-, < 0.81: 'surface condition' = class 2, etc.). The parameter mb is adjusted for the type of material of the inlact rock. Not all positions in the matrix are defined. In the comparison only those rock masses are compared for which the matrix gives values for mb and a. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SEIJJIUTY PROBABIUTY CLASSIFICATION- SSPC 0.4.2.3 159 Discussion The major principal total stress values at fuilure from the SSPC system correlate with the major principal total stress values at fuilw:e from the RMR system (Fig. 76 left) and the 'modified Hoek-Brown fuilure criterion'<130> (Fig. 76 right). The absence of a difference between calculations done with total stresses or calculations done with efktive stresses in the comparison of the 'modified Hoek-Brown failure criterion' with the SSPC system, may indicate that the SSPC system is defined in terms of efi:ctive stresses and that thus water pressures in the slopes in the research area have been small or absent. The overall reasonable correlation proves that the SSPC system methodology for non-oriented slope stability is justified. 0.4.3 Conclusions The calculation of the stability of a $lope with the SSPC system gives a more distinctive ditl.erentiation between stable and unstable than with the Haines and SMR systems and is .a clear advantage of the SSPC system over these classification systems. The correlation between the visually estimated slope stabilities and the predictions of stability of the SSPC system is better than the correlation with the other classification systems. This very likely proves that the SSPC system is more reliable in predicting the slope stabilities of the slopes in the research area. The 'strength' of a rock mass as determined with the SSPC system is good comparable to other methods. This proves that the calculation methodology used for the orientation independent slope stability incorporated in the SSPC system is justified. 030l The rock mass parameters in the Hoek-Brown failure criterion, 'structure' and 'surface condition', are calculated following procedures from the SSPC system. The good correlation suggests that the rock mass parameters in the Hoek-Brown failure criterion could standard be calculated following SSPC procedures. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D.5 EXAMPLES AND VALIDATION Four slopes in the research area are presented as examples of the application of the SSPC system. For two of the examples also an extensive analytical and numerical modelling, and a sampling program have been carried out which are included in this chapter to Wlldate the results obtained with the SSPC method. The worked out classification forms for the SSPC system of each example are included in appendix VI. The computer software programme used for numerical calculations is UDEC (Cundall et al., 1971, 1985, Hart et al., 1988, UDEC, 1993). This programme models the rock mass as individual blocks separated by discontinuities. The intact rock blocks are allowed to deform, rotate and translate. The movements of blocks along each other are governed by the shear and discontinuity stiffness criteria defined for the discontinuities. The programme calculates, therefore, a fairly realistic model of a rock mass. However in complicated situations that ~ reqUire a moaer~ contahiing~many indlvidUil blocks, ~c8.lcUlations~beeome extremely thne-consUDling. Th~ programme is two-dimensional which requires a transformation from a three-dimensional reality to a twodimensional computer model. In many situations this is virtually impossible therefore the programme has only been used for slopes where it was 'a priori' recognized that a simplification from three- to two-dimensions would not have a too large influence on the calculated slope stability. Slopes have been used that can be modelled in a vertical cross section perpendicular to the slope and in which the discontinuities determining the stability are approximately perpendicular and parallel to this cross section<131>. 0.5.1 Example I. Predicting the stability of a slope in Lower Muschelkalk: (Tg21) This example demonstrates how the SSPC system is applied to design a slope in a new road cut from old BiC'road eut··~·situated in:~Musehelblk·{"fg2ty"aticm 494·aloog the~roatt ·N.:.42(:) from Fatset to Reus. Fig. 78 and Fig. 79 show two exposures of Tg21limestone and dolomite along the old road. The first pho1:9 shows au. ~cJ!Vlited ro.lld. cut..made~ by small hok~hlasting (hlast.hok diameter ;; 2.5 cm, length.;;; 0. 75 m) probably blasted by gunpowder about 40 years ago while the second photo shows a similarly blasted exposure with a natural exposure above in the same unit along the same road. In the same unit a new road cut has been made in 1989 (Fig. 80). The new road cut was excavated by blasting (blast hole diameter 7.5 cm, length 8 m equal to the full slope height). The blasting was done with care. Fig. 77 shows a sketch of the locations of the exposures and of the new road cut. ~sure&. = = The rock mass characterization, reference rock mass calculation<132> and the calculation of the slope stability probability of exposure A are used for the description of the SSPC system (respectively Fig. 71, page 145, Fig. 72, page 148 and Fig. 74, page 153). The forms for exposure B are Fig. A 108, Fig. A 109 and Fig. A 110 in appendix VI. The calculations of the stability of both slopes result in stable slopes which (as the photos show) <131> Numerical distinct three-dimensional procedures and software programmes have been developed (3DEC, 1993, Cundall et al., 1985, 1988, Hart et al., 1988) but the cost of these programmes and of the requimd hardware (to obtain results in a reasonable amount of time) is so high that it was not possible to use these programmes in this research. It is also unlikely that such programmes will be used in day-to-day slope cutting practice in the near future. The values on the forms have been calculated with the computer programme SSPCCLAS in a higher precision than that shown on the forms. The rounding of the values may cause slight discrepancies in the calculations shown on the forms. <132l Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D.5,L1 cls.ssification Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hi2 "'"'""'~'"" L Natural exposure B old road. The natural exposure starts at about 2 m. from road level and blastholes in the lower part. overgrown. The lower part ofilie exposure is blasted. Note the small Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl:4BlLITYPlWBA.ElliT'l CLAS/SlFlGL.'rfON ·· SS.PC J!'ig. 80. """""''"'"I. New road cut C '"'"""""·'!! dip of the ro the about of road ro t!i.e left causes a s<::~;mu1g1y visible at the left. 163 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ "'""'ut!J'" IL Geometrical cross section of the of ilie and D.5.2.1 res:nectrve1v t~e exposure refurence The et'll:posure characterization a of 90" The the reference rock mass and in the to be flushed into the discontinuities from the tem~in surface and not to be pn~se11t was nmss. \v'hefuer the of the second set is taken as 15 or 5 m does not make a difference for Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE S'.DUJILlTY PROBA.BlLlTY CLASSIFICATION- SSPC D.5.2.2 165 Laboratory tests Shear tests samples of the discontinuity planes have been done with the Golder shearbox (Hencher et al., 1989). Samples have been obtained from the debris of the fiilled slope and have been sawn out of still standing parts of the road cut. The samples from the debris -were used for shear tests on non-fitting surfaces whereas the samples sawn out of the rock-mass were used to test fitting discontinuity sw::filces. Only samples could be tested which did not contain steps. No significant differences were round between tests on the bedding planes and on the other discontinuities. The shearbox friction angle from these tests is 45° (this is the average of six tests which are not corrected for dilatancy, standard deviation 1 °). The clay infill on the bedding suriwe as observed in the field has not been present on the surfilces of the samples for testing. For the debris samples this is obvious but also for the sawn samples the clay infi11 (which is very thin; 1 - 2 mm) was lost during the sawing and preparation of the sample. The laboratory shcarbox friction values for the bedding plane are representative for a rough planar surface (the sample with steps could not be tested) without infi11 and a large scale roughness equal to straight. This results in a friction angle of about 43 ° according to the 'sliding criterion •< 133>. The description of the bedding plane in the field is, however, straight, rough stepped with fine soft sheared infi11 and equivalent to about 35 o friction angle along the plane ('sliding criterion')<133>. The value from the laboratory shearbox test of 45° is thus in agreement with the sliding criterium for the sample tested, however, is not representative for the bedding plane in reality. That the di.f'mrence between the test result and reality is not larger is pure coincidence. The absence of steps on the surface of the samples is compensated by the absence of the infi11 material in the laboratory tests. This illustrates the limited usefulness of shearbox testing, even for discontinuities which have no large scale roughness. Slope stability by limiting-equilibrium back calculation D.5.2.3 A traditional limiting-equilibrium back analysis was made of the slope of example II (Fig. 83). The cohesion along the sliding plane is taken as zero. The length of the sliding block is defined by the second joint set (337/48) approximately perpendicular to the fiillure plane, the socalled 'internal joint'. In the calculations the spacing of this joint set and thus the length of the sliding block is varied between 3 and 15 m. Whether the fiillure occurred under the influence of water pressures in the discontinuities was also investigated. Three different levels of water in the 'internal joint' -were used in the calculation: hw = 100 %, 50% ;m4Z,5,i, (hw = the hei&btof~ ~ as percentage of!U.Jbeb&jght of the joint above the bedding plane). The friction angle along the sliding plane is calculated with: = fl fl arotan ( W W*sin Fig. 83. Example II. Limiting-equilibrium analysis. V) sint + Woost-U [49] =friction along sliding pl.aM W = wight of block t = dip of sliding pkme U • water force tit bottcmt of block V = water force tit rear of block Fig. 84 shows the relation between the length of the sliding block along the sliding plane and the friction angle for dif"ierent water heights (hw) in the 'internal joint'. The friction angle decreases if the length of the sliding block increases. This relation is less pronounced if the water level in the joints decreases. For a friction angle of 45 o (shear test result) along the sliding plane, sliding should not have occurred for a fu1115 m length block, even not if the 'internal joint' would have been completely filled with water (hw = 100 %). However, for a block length of 5 m sliding would just have been possible if the 'internal joint' was completely filled with water. For a friction <133l Condition of discontinuity for the laboratory samples: n:: = 0.75 (straight) * 0.65 (rough planar) * 1.00 (no infill) * 1.00 (no karst) 0.49. 'Sliding criterion': fl n:: I 0.0113; 9' = 43°. Condition of discontinuity for the bedding plane in the field: n:: = 0.75 (straight) * 0.95 (rough stepped) * 0.55 (fine soft sheared infill) * 1.00 (no karst) == 0.39. 'Sliding criterion': 9' = TC I 0.0113; 9' = 35°. = = Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ o 2 4 s a 10 12 14 1s ~~of the block [mJ 84: TI. The friction of the water in the second as function of block set and the elasto- discontinuities behave as an area contact The programme allows fluid flow and water pressure modelling to water pressures. Permeability parameters discontinuities are, 11o•we\rer. ......t .......,...,. 1' " in the guesses. The pa:racneters D. .S nn1Ultng-el::jm.uoln<tm and numerical become "'-'''""''"'""' than about 37"' to 38° without water pressures. ~m ff in the discontinuities had been assumed. Even if the oectau:tg li (l3 4l the main parameters for the are included here. Parameters not mentioned are at a default value as sug:geoited the manual of the programme and have no or minor influence on the result<>. For more detailed the reader is referred to the literature Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SfA?PE SI4.B!UTY.Pf:?.OJJAJ3.1UTY ClASSil7JC4..'tlON- SSPC 0 161 -.1> A ,.e ·1.0 ..~ ..l! 4.(1 I O.!l I Cl.!l U\ I U! b with back joint am:! wS!tar. dirl'!an:lllorts mmetres from the toe and xy-stresses while a classification Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ a4 in April I995 after the main fu.ilure of April 1992 and the renac.e on the left is the old Fig. 87: m. Geometrical cross section of the slope. Situation in fu.ilure occurred 1992 after the m.ain in ri"f'&....,..... t any direction. The included are those acc.:es:noJte locations at the bottom of the not for r.:inematic Qf i.ULn.o'>llU;e; be measured and are thus at Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE srAJJlUTY PROlWJlLITY CLASSIFIC4T!ON- SSPC 10 all discontinuities are continuous or have the same orientation throughout the rock mass of the slope. Fig. A 116b and c (appendix VI) show stereo projections and contour plots of the poles of cleavage and main joint discontinuity planes. The slope was excavated in 1989 during the construction of the new road alignment and the slope has been cut at about 70° in the direction 018° (comparable to the situation to the left in Fig. 86). In April 1992 the slope fiilled. The slope mce after fiillure has highly irregular surface. The overall dip of the slope became about 53 o with a slope dip of 41 o in the lovver part of the slope, 46 o in the middle part and 57° towards the top of the slope. The upper part of the slope became undercut (Fig. 87) and in 1995 also the undercut top part of the slope had partially fiilled (Ftg. 86) reducing the overall slope dip to about 45 o. Visually assessed the slope is now expected to be stable, although some minor blocks which are undercut and not fully supported, or which have already moved during the forgoing slides are expected to :fiill in the near future. The night beiOre the main fiillure occurred (April 1992) it had been raining and a very small amount of snow had :fiillen. The actual temperature had probably not been below zero at ground level. This slope can be analysed by a limiting-equilibrium method and numerically in two dimensions with some simplliications. Samples for UCS and shearbox testing have been sawn out of the rock and a detailed survey of the topography of the slope has been carried out. D.5.3.1 Slope stability probability by SSPC classification Because of the variation in dip of the cleavage plane (from 41° in the lower part of the slop towards 60° in the upper part of llie slope) the nuntber Of georecliiiical units in the rock :mass of the slope is infinite: The most unfavourable dip for the stability of the slope is, however, the 60° dip in the top part of the slope and, therefore it is sufficient to calculate the slope stability probability as if the dip of the cleavage is 60° throughout the slope. Fig. A 117 and Fig. A 118 (appendix VI) show the exposure characterization and reference rock mass calculation. Fig. A 119 shows the slope stability calculation before fiillure and Fig. A 120 after failure. These are based on average slope dip angles. The exposure characterization had been done before the slope fiilled in 1991. Already at that time it was obvious that fiillure was imminent and accordingly the slope was visually assessed as 'unstable with large problems'. The slope calculation shows that for an overall slope dip (road cut) of 70° (Fig. A 119) it could be expected that the slope would fiill because the calculated Ha- (3.2 m) is :fur below the real height (8.2 m) resulting in a probability to be stable of < 5 %. This instability is not caused by sliding along the cleavage discontinuity plane but results from the orientation independent slope stability probability. The friction along the cleavage planes is about 57° accordiDg to the 'sliding crlterion'<13S>. This is more 1:ban th~ ~t dip of the cleavage plane in the direction of the slope dip and sliding along this plane is not expected according to the SSPC system, even not along the steepest parts of the cleavage plane. The slope calculation (Fig. A 120) with an overall slope dip of 45 o which is the overall slope dip in 1995, resUlts in an SSPC slope stability calcUlation which is about UnitY (55 % stable). This corresponds with common sense as the stability of a slope is expected to be Unity after fiillure. A stability almost Unity also corresponds with the visually assessed stability in 1995. D.5.3.2 Slope stability by kinematic analysis A kinematic analysis of the mean orientations of the discontinuities (Fig. A 116, appendix VI) shows that for a slope with a slope dip of 70° sliding along cleavage planes is possible if the friction 5l~along the cleavage planes is ~~.2io for cleavage planes in the top of the slope (dip of cleavage plane 60°) and is less than 40° for the lower part of the slope where the dip of the cleavage planes is about 46 o. In the top of the slope, where the cleavage plane dips 60°, kinematically wedge fiillure is possible for the w!fges formed by the cleavage plane Cl with joint system-n:-r4 and J5 and for the wedges formed by J4 with J2, J3 and J5. The kinematically possible wedge fiillures formed by J4 are not relevant as J4 is nearly horizontal and the friction along the planes intersecting 0 3Sl Cleavage discontinuity plane: Ir: (condition of discontinuity) = 0.85 (large scale roughness 'curved')* 0.75 (small scale roughness 'smooth undulating')* 1.00 (no infill) * 1.00 (no karst) == 0.64. Friction along cleavage discontinuity plane: 0.64/ 0.0113 "' 57" ('sliding criterion'). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D.5.3.3 AC,COf'!lllJI.R; tO 45° a:nd 47"' resulting D.5.3.4 in of 57.4" under wet conditions and 43.9" under dry '-''.J'''"'"'"'""'· resistrulce dipp!rog ~ ~ faoll ! ' :~VIlli'! !!lope fJilm 0 20 a C~WIQ!<~!nimrl<ll~<lll) !~l 89: .I::IX<llmple ill. The factor of as fu.nction of the inclination of the inte:mal discontinuities Friction 4l ,..., 43 .5" tor base and side calculated Sarma' s meiliod. \ 136l The difference in friction (B7l Condition between """"''J"'"""'"'"J for the test and surfaces diminishes with less scale and no infiH or karst: TC = 0.75 * 0.75 surfuces. (it is too small tor a * !.00 * LOO = 0.:56. criterion': Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE SJXBll.JTY PRO&Bll.JTY CUSSIFIC4.TION- SSPC 171 The orientation of the internal discontinuities crossing the slid rock mass shows a variation due to folding. In order to study the inftuences of this variation of the orientation a sensitivity analysis has been carried out. A series of geometrical cross sections with difterent inclination angles (a) for the internal discontinuities has been modelled. Fig. 88 illu.stxates the way in which the orientation of is defined. The mctor of sarety of each geometrical cross section is calculated by Sarma' s method for non-vertical slices. The results of the calculations are illustrated in Fig. 89. As the inclination angle of internal discontinuities increases from a negative to a positive value (from dipping "with" to dipping "against" the slope fiwe) the actor of sarety decreases. The extent of the inftuence of the orientation and the friction angle of the internal discontinuities on the stability depends upon the inclination angle of the internal discontinuities. As the inclination angle of the internal discontinuities increases from negative through zero to positive (from dipping "with" to dipping "against" the slope :fuce) the influence of the shear strength becomes more pronounced. This is caused by the increase of the normal stresses on the internal discontinuities. The variation of the friction angle of the internal discontinuities has only a minor influence on the total sarety if the inclination of the internal discontinuities is in a range between + 10° and -10°. The analysis results in a safety iBctor of unity for a friction angle of about 42 o, virtually independent from the orientation of the internal discontinuities. a D.5.3.5 Slope stability by numerical analysis - UDEC simulation The UDEC programme (example II, eh. D.5.2.4) has been used for a numerical back calculation< 134> of a model with difimmt oriented internal discontinuities. The model in which the orientation of the internal discontinuities varies between +6.04 o and + 11.07° resulted in a tp of 43.7° (Fig. 90 a: unstable, b: stable). A sensitivity analysis comparable to the limiting-equilibrium back analysis was not possible due to the large calculation time necessary. As\Vifei'"maynavenaa m infiuenceon the fii1'tire0fthe slope a nuriiericiil back analysis~incliidirig water"lias been executed. The results were, however, totally unrealistic as during the calculations water pressures in discontinuities became larger than defined by the level of the water table. This is obviously not possible and the calculations have been abandoned<138>. D.5.3.6 Conclusions example III The classification, limiting-equilibrium and numerical calculations come to the same result: the original slope dip of approximately 70° was unstable. According to the SSPC system the slope was too high for a slope dip of 70° while none of the discontinuities was the cause for sliding or toppling instability. This is in contradiction to the limiting-equilibrium and numerical analyses which both show that sliding instability can occur. However, the sliding is only possible if the friction angle along the discontinuity planes is lower than the friction angles determined from testing and considerably lower than friction angles determined with the 'sliding criterion' , or if is assumed that high water pressures existed in the slope at the time of :fu.il.ure. The.'sliding criterion' gives a 1-easonably acctii:ate estimate Oftfie friction aloD.g the discontinuities (appendix HI). In this example this is confirmed by the tilt tests (45°) and laboratory shearbox tests (47°) for the friction angle without large scale roughness, for which the 'sliding criterion' results in 49°. In the limiting-equilibrium and numericiil analyses sliding is only possible if the friction angle along the discontinuities is around 43 o which is not only less than the test values, but also implies that large scale roughness would have been of no importance. This is unlikely. The existence of high water pressures in the slope at the time of :fu.il.ure is also unlikely for the same reasons as for example II (eh. D.5.2.5), e.g. water can flow out via connecting discontinuities and no evidence of water under pressure has been observed. Although water pressures are not the sole reason for the :fu.il.ure, the presence of water will have had a negative influence on the stability. Water will have caused a softening of the infi1l. material in joint discontinuity J4, will have lubricated all discontinuities, and will have created some, however, very limited, pressures in the discontinuities. Additionally the little bit of snow in April 1992 will have caused a (very) little additional weight on the slope. (t38) These erroneous results have been discussed with the manufacturers of UDEC however a reason could not be pinpointed and it is likely that these were caused by an error in the UDEC programme code (version 1.8). In later versions of UDEC this problem is reported to be solved, however these were not available fur this research. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 172 D.5 Examples and validation 10,000'~ 90,000 oydaa li.o ' &o ' ~ ' d.o ' u 11i.o -1111.0 -8.0 ~ 1.0 .!S. 1:.! ~ .11 .0 J~'"'-t· '0 ·1.0 . .. - • ---------l I I 1o,ooo oya~es I I -U .co · e.o · ~ · 1d.o lie ~ ~ ~ a.o a: friction angle 43.5 o. b: friction angle 43.9°. Velocities and xy-stress contours after 90,000 cycles. Horizontal and vertical distances from the toe of the slope. Sections are In the direction of the slope (018°). Fig. 90. Example 90,000 ~ d.o · ..., · I I 10.0 m. UDEC simulation. Displacement, velocity vectors and xy-stresses in the slope. The reason for fuilure of the slope is more likely that the rock mass as a whole has not been able to sustain the stresses in the rock mass caused by excavation the slope with a dip of 70°. After the excavation of the slope the new stress situation in the rock mass caused a progressive weakening by breaking pieces of intact rock, small movements along existing discontinuities, opening of cleavage planes-~ existisgjoints ~d possib,ly (ormil:J,g of ten8ion joints. this progressive weakening of the rock mass continued until in April 1992 the water and snow triggered the already weakened rock mass to full. The orientation independent stability probability of the SSPC classification system is empirically developed on the basis of existing slopes from which the rock mass of several must have been subject to progressive weakening of the rock mass. Therefore, the SSPC system can successfully assess the stability probability of the slope in this example. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl'ABlLITYPRO&BlLITYCI.,ASSlFlC4TION- SSPC 173 D.5.4 Example IV. Influence of weathering and method of excavation on the stability of a slope in Upper Muschelkalk: (Tg23) Example IV demonstrates how the SSPC system considers poor blasting and future weathering in a slope stability probability assessment. The slope is situated in Upper Muschelkalk (Tg23) limestone and dolomite in a road cut at km 492 along the road N-420 from Falset to Reus. A photo of the example slope is shown in Fig. 40 (page 91)039>. The slope is newly cut in Tg23 (Upper Muschelblk) in 1988. The slope has been excavated by blasting originally with a dip of about 80°. The present dip of the slope is between 60° and 70°. The length of the blasting holes cannot any more be determined, but it is likely that blasting has been done in one pass over the full height of the slope ( 14 m) with blasting holes with a diameter of about 7.5 cm. This procedure mr blasting has been sUmdard tor the road cuts in rock aloog this road (N-420) when the road was renewed in 1988, and it is likely that the same procedure has been followed mr this slope. = The Tg23 consists of interlayered thin bedded (visible in Fig. 40 just above the sitting person) and medium to thick bedded units. The same tb:in. bedded units are mund exposed in.nearby (less than 50 m away) old road cuts of more than 40 years old. Old road cuts made in the thin bedded units with dips of 60 to 70° and heights of about 5 m, are still (in 1995) stable and no or very little degradation of the rock mass is observed in these old road cuts. The rock mass in these old road cuts is still only slightly weathered. The method of excavation used for these old slopes is not known, but no renmants of blasting boreholes are visible at all, so that it is likely that these road cuts are ex.cavated by hand or by a small shovel. An exposure characterization of the thin bedded units is given in Fig. A 121 (appendix VI). The dip direction of the slopes in the old and new road cuts are approximately equal and the general position of !!le ol'!. ~~ cuts in ~~~~~ ~y ~~le to the new ~ . .cu!· . ~C?!h the old and.IlfiW .r<>ad cuts . ~ cut into a hill that flattens above the road cuts. Quantities of water flowing from above over the road cuts are, therefore, likely comparable, although this has not been tested. Also with respect to geology (mults, etc.) no major differences have been noted between the old and the new road cuts. The new road cut (Fig. 40, page 91) is clearly unstable, large parts show rill erosion and erosion of the thin bedded units causes undercutting of the thicker bedded parts, making these unstable. The general impression of the slope is extremely poor. On close examination those parts of the slope which appear to be 'soil' are in met the thin bedded units which are partly covered by top soil transported from higher parts of the slope. For another part the soil is derived from weathering of the thin bedded units. In some places these have been weathered to a moderate or high degree of rock mass weathering mr at least 0.5 to 1 m into the rock mass. The structure and coheren.ce of the rock mass, and in particular the structure and coherence of the thin bedded units, are disturbed by the method of excavation. Discontinuities have opened, blocks are displaced, and at many locations the intact ·~· is .f:radured and oeeasienaHy also crushed due"t& tlle blasting tor the excavati<m.l1le···slope is not unstable due to sliding or toppling along discontinuities. Although a back analysis of such a slope can never be very exact, the following reasonable assumptions can be made to explain the instability. The damage due to blasting has disturbed the structure of the rock mass so severely that water could flow through the near-sur:tace parts of the rock slope. This has caused the weathering of the thin bedded units<140>. The disturbed and moderately to highly weathered thin beds cannot sustain a slope with a dip of 60 to 70° at a height of about 14 m. (!39) The slope discussed here has not been used for the development of the SSPC system. Although located in the research area, the slope was considered to be too unstable to safely be analysed by students during the years used for collecting dam (1990- 1993). By 1995 most larger blocks had fallen of the slope (reducing the slope dip) and the slope was deemed safe enough to carefully assess the slope stability and rock mass. (l40) A rapid, within a few years, weathering of intact rock as well as the rock mass has been noted to occur in some of the units with a thin or smaller than thin bedding spacing of the Tg23 and Tg3 formations. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 174 D.5 Exmnples and validation 0.5.4.1 Slope smbiiity by kinematic analysis or calculation A kinematic analysis results in an assessment for the slope to be stable because the slope is not unstable due to discontinuity related sliding or toppling. A limiting~equilibrium or numerical analysis is extremely difficult for such a rock mass as it is almost impossible to obtain suitable samples for testing. It is also impossible to quantify the reduction in strength of the rock mass due to the loss of structure and coherence without large scale testing. 0.5.4.2 Slope stability by classification The SSPC system results in a probability to be stable of > 95 % for the old road cuts with a slope dip of 70° and a height of 5 m. The nevv road cut with a height of 13.8 m, with a 'slight' degree of rock mass weathering and 'dislodged blocks' due to blasting, results in a probability to be stable of less than 5 % for a slope dip of 80°. For a slope dip of 60° the probability to be stable increases to 85 %. If also the increased degree of rock mass weathering (highly) is taken into account, the probability to be stable decreases again to < 5 % for a slope dip of 60°. In the present condition the rock mass is clearly not able to support a slope with a dip of 60° (Fig. 40, page 91), and according to the SSPC system, stability will be achieved if the slope dip is decreased to 45° (probability to be stable 55 %). 0.5.4.3 Conclusions example IV This example shows that the SSPC classification of slope stability is also applicable in situations where the stability is governed by,damagtHiuo todl&method"ef~ and weathering,inftuenee;Ifthe slope' had been designed using the SSPC system the increased weathering would not have been anticipated as the old road cuts do not show this. However, the new road cut would never have been designed with the steep slope dip of 80° if sloppy executed blasting was going to be used. 0.5.5 General conclusions from the examples The kinematic analyses, and the limiting-equilibrium and numerical calculations executed for the examples give results for the stability which are, in general, comparable to the stability probability obtained by the SSPC classification system. However, assumptions have to be made in the kinematic, limiting-equilibrium and numerical analyses, e. g. water pressures, low friction angles along discontinuities, etc., which are not supported or which ~, ~ c,on~~ ~;y f!~l<i obse,~tiQJlS .9I: testmi.~. The .SSPC. ~las,si11~iJ..tipn, $Y$teDl.gives feasibkresults. without contradicting field observations or test results. Kinematic, limiting-equilibrium or numerical analyses would not have predicted the instability of the slope in example IV. The examples presented in this chapter are typical for the slopes in the research area. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE STA.BIUTYPROR4Blll11' CL.4SSlFlCAT.lON • SSPC 175 6 is the core of the SSPC system. The three and un~:l.ls1rurl[)eel mass, establishment of a conversion paramete1rs that ""'"''""'""""'""" 'slope' rock mass. The 'ex1posure unit the exc:av<mc•n to of the slope to be stable is then rl,.t,,..."",;.., ••ii SSPC has been designed in a a using the SSPC on slopes under conditions and in and rock masses, etc.. As for all empirical areas that are very different implies a risk. The SSPC system is, however, based on a large number of different slopes a wide rock materials rock masses, and may be slopes in mo:re rock mass for the design of The of the in a climate different from than those the climate where the has been developed may limited. The intensity and duration of the determine the water (and in and on the but whether it does not likely the maumuru P05iSIOl.e considered that v.rater pressures may be tonmrrtg in and on slope may, no,we>.rer. ............,..,""'"' prudent to check the system a mc!ep,en~crern: of the climate. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ strength estlmaltlon delten:nnlatJi.on as used in is more inhomogeneous rock masses than a li:nlited uwcuu'"'' is no reason and may oe<XHJmes rn:roo~isli:He due to small scale ml:i.OJJGOJ~er.telt:y, '-'"'"'"''".... ''u" can be this Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ D SLOPE Sl'ABIUTY PRO&BIUTY CUSSIFlOO'ION- SSPC 177 Strongly dlforming intact rock Rock types that are deformed very easily (gypsum, salts, etc.) are present in the research area and have been used for the design of the SSPC system. The smbility of the slopes in rock masses containing gypsum is, however, more governed by erosion and weatheriDg (in particular solution of gypsum) than by deformation of the rock material. The SSPC system cannot be used if the smbility of the slope is governed by deformation of the intact rock. Extemo.l stresses The system has not been designed tor slopes that are or will be excavated in a rock mass that is under influence of external stresses. External stresses do not originate in the rock mass in which the slope is or will be excavated, but are, tor example, tectonic stresses or stresses due to a high hill or mountain behind the slope. It may sometimes be possible to simulate the presence of a hill or mountain behind the slope by taking the slope height in the SSPC system as if the slope extends to the top of the hill or mountain. This has, however, not been tested and ofmn this heiaht will exceed the maximum height tor which the system has been designed (see above). System structure The system, in comparison with other rock mass classification systems, is more elaborate in structure and calculation. This is, however, not likely to be a drawback of the system in a time where computers are widely available both for office and field use. The system is suitable to be incorporated into a GIS environment. The parameters can be interpolated independ.ently and rock mass parameters and slope stability probabilities can be calculated at required locations. 1ime saving Fig. 91 shows time estimates for various methods to arrive at a stability assessment of a slope. Classification is an~attraetive option;minparticularNbecause m~ation may be. done whitesWlding"mfront of the stope. If curious results are obtained it is still possible to check the observations and data. •. Numerlcai'QJIQM&Ion baeci on clmlllcdorl c:W.a ... .. ... ... ' ' ' ' ' Fig. 91. Time estimates fur the stability calculation of a 15 m high slope. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 178 D. 6 Conclusions Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPEA'DlX I APPENDIX I TABLES SLOPE STABILITY PROBABILITY CLASSIFICATION (SSPC) 179 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX.! c~=~be:ti~ fica · ns lithostratigraphic time unit or formation sub-unit( 11 unit SSPC =~ce) conallsst 52 Jura llmest. 32 Tg3 dololrnite) 18 13 Tg3 shale 16 16 6 Keu~r Uj:!~r Musctlelkalk (Tg23) Ta23 v.thick Ta23 thick Ta23 medium lime.Ia23 thin stone Ta23 v.thin and Tg23 thick dolomite lam. Tg23 thin lam. 3 5 8 38 .22 26 2 14 Tg22 6 28 Middle Mu~0~~'f'lk Lower Muschelkalk ......... •••••m•••••••{Tg.211 Ta21 v.thick LimeTa21 thick stone Ta21 medium and dolomit6 w•'fa;tt1hirr"' Ta21 wthm Ta1 sst mass sandstone conglomerate slate fHslate) 19 Ji!~isst ium 1 17 Ta1 sst thin Ta1 sst v.thin 3 2 Tg1 congl. 7 Hslate thick 1 2 ~~ 4 10 21 2J 54 50 14 24 1 8 9 15 4 Hconal .Hut mass. Hsst v.thick Hsstthick Hsst medium Hsst thin (Hsst) Hsst v.thin Hsst thick lam. Honeiss 1 6 21 5 granodiorite 18 64 aolite 4 250 21 770 sandstone intrusive Total: LIMESTON A -frY· DOLOMI E. argillaceous to fine arenaceous, Red (occasionally greenish ~P$~ a~illaceous to fine arenaceous, g~:!ferous clayeys;a';t~ S ON ; large quantities of gypsum uo" occasionallY 0 . :r LIMESTONE AND DOLOMITE 6 1 1arn. Carboniferous (H) Off·white/l~re~~ellowish mm~-mNN,NNNNNNhWNNm. m To 1 sst thick Hslate thin Hslate v.thin Hsl,-:; thick m. Hslate thin Brownlvellowish CONGLOMERATEJitnd. SANDSTONE. g~i~te-.grey, arglllaceous to arenaceous, LIMESTONE AND 1 T • 20 - 100 cm thick off-white/l~e~ argillaceous to fine I arenaceous LIMESTONE AND DOLO IT . Red/green/greenish blue/brown/yellowloff-white, argillaceous to fine , : : a r s , calcareous sandy siity SHALES, with (small) I auan · avosum. 3 8 20 35 on-whiteti~Wr arenaceous, 27 48 1. .g........ 117 ... - (CALCARE 6 6 4 ~?Ju~~t Buntsandstone (Tg1) description(2l eel map. Tertiarv (Tg3) :Ui:l. Red/brown, coarse arenaceous (bottom) to fine arenaceous (top), SANDSTONE. Red/brown, rudaceous, CONGLOMERATE. Thick seqiJences I> 100 m) of d.Qrey, argillaceous, SLATE. Grev/brown MICRO CONGLOMERATES. 1 5 16 42 20 8 1 3 2 Grey/brown, SANDSTONES AND SILTSTONES. Black !white foliated). GNEISS. ~~~i~~T~· fine to coarse grained, GRANODIORITE (sometimes lti . D. arev. v. fine arained APLITIC DYKES Codes refer to the codes used on the geological map sheets of the area (Table 1, page 17) (sst =sandstone, congl = conglomerate). Notes: Lithostratigraphic sub-units are defined on bedding spacing for limestone, dolomite and sandstone. and on cleavage spacing 1 for slate. Cleav~ or beddioo soacing aooordin~~:to SS 5930 (1981): mass.... flO bedding visible, v.tflick "' > 2 m, thick = 0.6 - 2 m, medium 0.2 - 0.6 m, thin = 0.06 - 0.2 m, v.thin = 0.02 - 0.06 m, thick: lam. = 0.006 - 0.02, thin lam. = < 0.006 m. 2 Descriptions according to BS 5930 (1981) (1. =light; d. = dark; v. =very; mod. = moderately; extr. = extremely). = Table A 17. Formations, lithostratigraphic units and sub-units. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ,... ~ o- ;! ro:; .... I:. (0 --.. ill :!! i:o pmbabi!ity% ...00 w rJl (1;, :;) :-a. :> ... 5 :t> ;;!l :1 10 :ij' :;; .... ... :L ~- ~-- •••o--··- • ~ ------------ ····---·····------------ - ---·---·-- :;l ·~ - l.46e-02 -5.71 e-05 -1. 75e-01 1.49e-02 -3.15e·05 --1.13e-01 1.30e·02 -4.14e-05 -1.12e-01 1.33e-02 -1.75e-05 -6.99e-02 1.21e-02 -3.27e-05 -7.86e-02 1.25e-02 ·-1.06e-05 -2.11a-02 1.07e-02 -1.~l0e-05 1.18e-02 -4.70e-06 1.22e-02 9.73e-03 -!.Ul7e-06 '! .12e-02 5.06e-07 4.51e-02 8. -4.4'/e-07 60 70 0 a (I> V ... -1.08e-02 1.09e-02 ----~-----·- 80 --- --~- ------------~--- 8.80e-02 3.65e-06 --~------- ~---~·-·-~- ----------------~-~-.--------- ---------· -----------c -1.62e-01 ·2.76e··02 i'1) b -3.85e-05 --------------~------ 0 iii' a ! .58e-02 1.89e-02 1.00e-02 (") ~- b * lP + c " 1Jl 2 c ---------- [ a·:::l + -2.! l e-01 Ill n0 .... .... ----- toppling b :::. (!l .. ·-"•••-·---• 20 ~ 0 ··~•--•• -·~----··-----·------"·-·--- sliding a -ow TC "" a -----------~--·----·-~-~------~- 1.20e-05 ·--··-·- .. --------i.13e-01 ....... ---~-·--------~----~----- ··--· 7.43e·03 1.291.!-05 6.'73e-03 2.02e-05 4.89e-02 9.29e-03 1.86e-05 1.5~~e-01 5.60e-03 3.l7e-05 R l'i!'ln.I"\''J 8.26e-03 2.87e-05 1.98e-01 4.51e-03 4.32'!l-05 l 7.35e-03 3.74e-05 2.45e.. 01 3.19e-03 5.65e-05 -···--·>-··-------~---- 90 --·------·--· ---- ---- ·-----·--95 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX I Bmax B~~~spe ·r·JLr X= log.(x) 183 fl-.. diplllspe rtmges: probability % 10° S: 0.1 s: , _ < 1 dip. . dip,q,. S: goo and 2 m s B_,. s 50 n. fl-..• dip_,. mdegree&; Bmax• Bt~~Dpe mmetres p1 pO ..... R2 ····f ... without influence of uncertainty in weathering and method of excavation 5 0.8592 0.02732 0.25 10 0.9074 0.02341 0.46 20 0.9211 0.01205 0.11 30 0.9655 0.00444 0.25 40 0.9955 0.00219 0.17 50 1.0047 -0.00607 0.36 60 1.0260 -0.00941 0.78 70 1.0416 -0.01676 0.69 80 1.0665 -0.02341 0.83 90 1.1160 -0.04117 0.73 95 1.1978 -0.05644 0.49 ............. .. The formula and factors have no meaning other than representing a best fit for the points of equal probability within the indicated ranges. The scatter strongly increases for probabilities less than 50% which causes the low correlation coefficients (R2). Table A 19. Lines of equal probability for orientation independent slope stability (SSPC). ....... Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ A.PPENDIX 1! 185 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPElt'DlX ll Introduction The armrtn•·,.,.,,...,,. of intact rock C.3.2. 1), but "'t''"'"''"~""' in is limited. failures are failures are often related to shear failure are interlocked the the steps and may have an The influence c.rf for can take 187 never due to intact rock lf both sides of the steps on the steps have to be broken before A 92). This mechanism is related to the intact "T~'"'"'"'i·h im,.,.rtn~ltdn,n how intact rock "'"r"'"',~"" effect of steps on discontinuitias. Two are raised: must intact rock be measured to he certain that the 2 effect of steps on ",.,,_.,...,.t~ on can he estim;.;;md with accuracy ? ls there a certain v~lue !cut-off value} where above th!'!l intact mck is of no for the statolltHS!r!Q affect of steps, and can dimensions of steps, necessary to stabilise a The intention of the is not to an exact or factms for a classification system. How accuromly must intact rock be musun.wd to~ certain that the on dis,coJr1ti!)U!I'y on t;lops:s can be tJStimattUJ with enough accumcy ? The first question can be answered very simply. on cannot be observed and their location is unknown. As ail rock material and rock masses inhibit inhomogeneity in their intact rock it is obvious that the intact rock at the location of a step wm never be established A highly accurate method to is thus not necessary. establish the intact rock 1 2 Is tfmre a certain value (cut-off vlliueJ where above the intact rock strength is of no importance for the stabilising effect of steps, ami c~Nn dimensicms of steps, necessary to stabilise a slope, be established ? Consider a i1!i diSiCCJ•ntliriUitv nll'l11':!11'! riiftl!"'ir11l:'l the same direction as the dip and A 92). The surface or the is smooth planar for small scale roughness and straight for (for descriptions see large scaie step somewhere near the eh. C.4) except for a bottom end of the Most ~t,.nn~>~i pianes have numerous steps spread over but for step is used in this example. in this situation what are the minimum dimensions and what is the minimum intact rock of the step for which the rock of the step will not be sheared off or crushed? AI! formulae and calculations are for a cross section of 1 m the UW is the unit of the rock. The weight (iN) of the block above fig. A 92. Cross"·section of a with one step on a is: [50] This results in a normal stress on and a driving stress * height * the discontinuity plane ( * sin(jli) * oos(~) * height * The force the p!ane is: [52] Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 188 Assuming that there is no cohesion along the discontinuity, the restraining force from the shear friction (f!d) along the discontinuity plane !F.)' 141 , is: F11 .. 1..12 * lwight 2 * uw * (tanp - - 1-) * cos(p) * tan(;p4 ) tancx [53] The remaining force along the discontinuity plane (F,) is: F, = F.,. - F6 [54] The remaining force F,, if > 0, has to be counteracted by the step in order to ensure stability. The relation following the Mohr-Coulomb failure criterion between intact rock cohesion (cohesion), angle of internal friction (;p1) and the unconfined compressive strength (UCS) 11411 is: co#wion. = J ..1 * __U.;..C;;..;;'_ S _ 2 tan(45 + ~~) Assume that a shear plane through the step will be parallel to the discontinuity surface, then the force for shearing through a step with width sw is: Fb = (cohuitn&1 + SW • For equilibrium F, = Fb. The width "• [SS] (Fb) necessary * tan ;p1) * sw of the step width of the step necessary to prevent shearing of the step is then: F, SW "' - : - - - - - ' - - - - - colwsWn1 + o,. * tanq~1 [57] Therock..materialcanaisobe crushed by the stresses working on it. Thi&iselso a formofsheer feitttre;However, the shear plane in the step will be inclined with respect to the discontinuity plane. The height of the step controls this mechanism. The area of the upper side surface of the step is sh. ullh is the stress on this surface caused by the remaining force Ius~~ = F, Ish). un is the confining pressure on the step. This leads to a triaxial stress configuration. Using the Mohr-Coulomb failure criterion the equilibrium value for the UCS of the intact rock for which crushing will not take place is 1142,: [58] Interlocking by steps of discontinuity planes in slopes in the research area The height of the slopes in the research area ranges characteristically between 2 and 25 m with a maximum of about 45 m. For this example assume a blocl< of rock ori tne discontinuity with a height of 15 m, unit weight (UVV) 1143, = 25 kN/m 3 and the overall friction angle f()r the discontinuity plane (;pd) without the step is 25° (this is the lowest value measured in the research area). The intact rock cohesion (cohesion,) is 23 MPa and the angle of internal friction for intact rock is f!; = 40° (UCS 100 MPa). This f'; is not very critical because the normal stress on the step is small. Fig. A 93 shows the width of the step (sw) necessary for equilibrium , versus the discontinuity dip { {J) for various slope dips I a). The maximum step width of approximately 5 cm occurs for a slope dip (a) of 90° and a discontinuity dip ({J) of approximately 52 °. = fig. A 93. Width of step (sw) necessary for equilibrium, vs dip of discontinuity ({J) for various slope dips (a). <141 > The sbear strength along discontinuities and the strength of inlact rock in this chapter are calculated according to the MohrCoulomb criterion. This may be too simple and not accurate for most rock material, but it is accurate enough to illustrate the influence of steps on a discontinuity plane. <142> In this example the stress on the side of the step (um) is due to the whole block (there is only a single step) and is therefore very large compared to the normal stress on the step (u.). <143> Rock mass unit weights for the units in the research area are around 25 kN/m3 • Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX.ll Fig. A 94 shows the UCS value necessary for equilibrium (for which no crushing occurs), versus the height Ish) of the step. The curves are for each slope dip hd with a discontinuity dip for which the maximum F, is obtained (dF,J dfJ = 0, the maxima in Ftg. A 93). Provided that the step is wide enough to prevent shearingoff the step completely, then Fig. A 94 shows that for a UCS value of 1 00 MPa with a sAope dip («) of 90° and a discontinuity dip (/J) of approximately 52 ° (the maximum in Fig. A 93), a step height of approximately 13 mm is enough to prevent crushing of the intact rock material. The UCS has to be > 150 MPa for equilibrium if the step is less than ,. 3 mm high. Most rocks have an intact rock strength of less than 150 MPa so that the height of steps should be in the order of .. 3 mm or more to prevent crus.hing of the stepC1 441 • The conclusion is that a relatively small step (In width and height) is enough to stabilize a slope. 18!» 0~------~------~------~------~ 0 5 10 15 sh (height or step) (mm) fig. A 94. ucs~ vs height of step (sh). Dynamic effects At many locations in the research area steps on discontinuity surfaces in failed slopes have been sheared off. As this cannot at ways be explained by static force equilibrium {see above), alternatives as weakening of the intact rock material due to weathering, intact rock creep, progressive failure or dynamic effects should be considered as possible causes for the shearing off of steps. On many of the surfaces with sheared off steps no indication of weakening was observed visually or determined by Equotip measurements (eh. C.3.3.3). If the steps are not weakened then the shearing of the steps may have been caused by dynamic effects. mDynamie··effeet&"C80 bft~edr• axample; by earthquakesi1451;· blasting and vibrations caused by heavy road traffic, thunder storms, etc.. Blasting is likely the reason for steps to be sheared off during the excavation of the slope. Blasting, however, together with stress relief or rock mass creep can also have caused displacement& in the rock mass so that opposing step faces are not any more interlocked. The discontinuity shear strength is then determined by the friction along the discontinuity plane only. If opposing steps on a discontinuity plane are not in contact, it can be calculated that dynamic impact of steps creates stresses in the steps that cause shearing or crushing of the step. This is illustrated with the following example. Assume that equilibrium exists between the restraining force and the driving force of the weight of the block in Fig. A 92. Assume that the rock mass on top of the discontinuity in Fig. A 92 can move by 1 mm before the opposing step faces are in contact. For a slope dip of 90° and a discontinuity dip of 52° the energy of the rock mass acquired by moving over 1 mm is: Energy_, = mtJS.f * ~- * displacement I£.1'f{Y,_, .. Fr (~force altmg ~ IJlane) Eurn.., .. 1.1!1H . ~! 0.001m = 1100.N..m .. * 1mm • [59] This energy has to dissipate in the rock material of the step at the moment of contact between the opposing step faces. The energy will then be changed into elasto-plastic deformation of the rock pf the s:tep. The elasto-plastic energy of the deformed step is: [60] 0 44> This value is used as a guide for the indication of the scales for the roughness profiles (eh. D.3.1). <14Sl The research area is not known to have undergone any earthquakes in recent times. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ For example with an elastic modulus fE.-) of the limestone of the Lower Muschelkalk which equals 45 GPa11461 : 1100 N.m 0.013 m2 * 46 GPa * 0.05 m "' 276 MPa [61] The maximum stress during impact is then 276 MPa. This is three to four times the intact rock strength of Lower Muschelkalk and will lead to crushing or shearing of the step. Similar results are obtained for other units in the research area. Conclwions A highly accurate method to establish the intact rock strength is not necessary as far as the stabilising effect of steps on discontinuities is concerned because the location of steps is unknown and as every rock mass inhibits inhomogeneity in the intact rock strength it will never be possible to establish the strength of steps with a high accuracy. The above analyses are done for the situation that only one step on the discontinuity plane is present. This is hypothetical because (nearly) always multiple steps will be present along a discontinuity plane. The widths, heights and required intact rock strength of the steps necessary to stabilize a slope are then equivalently lower. In the field has also been observed thet steps are normally considerably wider than the minimum dimensions calculated above. This leads to the conclusion that the intact rock strength will usually be too high to allow shearing or crushing of steps. <146> length Laboratory test value fur Lower Muschelkalk limestone which is determined on a UCS sample: diameter =10 cm. = 4.5 cm and Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDJ}t: Ill APPENDIX Ill 191 CORRELATION OF THRESHOLD VALUES OF SLIDING CRITERION 10 TEST AND LITERATURE VALUES Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX Ill 193 ComJIIdiM of the tlnMold vlliws of '#lkllng Clitedon' to test and litenlture Motion values The 'sliding criterion' is based on the assumption that the friction angle along the discontinuity plane, is equal or larger than fJ ( = apparent discontinuity dip in the direction of the slope dip). In this appendix the threshold friction angles obtained from the 'sliding criterion' are compared to the friction angles resulting from laboratory and field tests done in the context of this research and to friction values found in the literature. The discontinuity condition parameter I TCI and the 'sliding criterion' in this chapter are calculated as defined in eh. 0.1.2.1.6 and include thus the refinements for the calculation of the parameter TC. In the following analyses the 'sliding criterion' is re-calculated for the different parameters in the 'sliding criterion'. For example, in the analysis of the small scale roughness (Rs) the 'sliding criterion' Is calculated for a situation that only small scale roughness is present, thus large scale roughness is straight, and that no infill or karst are present in or along the discontinuity. Sml!lll scale ~s (RBJ The threshold friction angles from the 'sliding criterion' are plotted versus the small scale roughness description in Fig. A 95 a and b for planes with large scale roughness straight, no infill and no karst. The threshold friction angles are then only dependent on the small scale roughness. Observed planes, measured and characterized in the fteld, with these specifications and that plot within a 20 % band of the 'sliding criterion' are plotted to verify that these planes actually exist in reality. 'Sliding criterion' compared to tilt and shearbox tests Fig. A 95a shows the results of fieid tilt tests (tilt angle) and Fig. A 95b shows the results of laboratory shearbox tests. The shearbox values are not corrected for dilatancy. Also plotted are the results of shearbox tests performed on (artificial) plaster samples (Grima, 1994). The linear regression lines between roughness description and friction angle found for tilt tests and shearbox tests are approximately the same. The tilt tests and shearbox tests show neither a dependency on rock material type nor on non-softening mineral coatings on the discontinuity surface (e.g. hematite coatings that were p.resent on the discontinuity surfaces of some of the slate samples). This is in accordance with the literature (eh. C.3.3.4.3). The graphs show a fairly large scatter which does not allow for a statistical evaluation; the linear regression lines in thegraphs are an indication of a trend rather than a correlation. A good fit between fJ, tilt angle and shearbox friction values cannot be expected. The tilt and shearbox tests are done on sample blocks extracted from the slope. The extraction process can easily break the cohesion and damage the discontinuity planes. In particular sharp asperities, that cause the highest i-angle, are easily broken. Secondly during extraction and preparation of tbe sample, the sample halves are nearly always taken apart. ~:~nd re-fitted Jor tbe ti.lt .Jlr shearbox tests. The cohesion that might have been present is broken and there-fitting will ~ftOt bee&.good· as the original in-situ fit of the sample halves. A not so good fit will result in a lower i-angle (Rengers, 1970, 1971, eh. C.3.3.2.6) and thus also in a lower tilt angle or shearbox friction value and as it is likely that the higher values are resulting from a high i-angle rather than a high fl value, the influence of the sample preparation is obvious. This is confirmed by the tests on the artificial plaster samples (Grima, 1994). The samples were made exactly according to the ISRM standard graphs (ISRM, 1978b, 1981 a) and testing started with perfectly fitting sample halves•. Each value is the average of 11 or 1 2 tests. The average values are considerably higher than the shearbox results on real rock samples but confirm the 'sliding criterion'. ~ 70r----------------------------------------------, a tilt le8ls ob8erved planes slate 0 slale & sandstone 6 sandstone • llrrieStc)M · o llmeSkine 1111 0..55 0.60 ().66 o:m 0.75 Q.SO.. uu0.86 0.70 0.75 0.80 small scale roughness parameter (·) 70 shearbox teats obseMci 0 slate c ~30 - 0.55 0.80 0.66 0.66 0.90 0.95 small scale roughness parameter(-) Fig. A 95. (a) fJ and tilt angle, (b) fJ and shearbox friction angle vs small scale roughness parameter (roughness parameter values see Fig. 71, artificial samples: Grima, 1994). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ criterion ' corrnJE!f'lfH:I to Jitero- ture ifPbostc values the literature for to Values values. between fPbas.~o for the dlfterent rock types in the !iterature are small and for many less than the range nulssured for ona type of rock. This critewas a!so found for the rion' which does not show any cant difference in friction values for 42). different rock types scale criterion' to small literature values lt is difficult to compare literature values for small scale with the 'sliding criterion' because the of the in the literature are not uniform, standards are oftGn not reported or a reference is to JRC number. The conversion of JRC numbers into the ISRM is and possible without ambiguity only for some (Barton, 1987, 1990b). However, an attempt to compare literature friction values with the threshold friction values obtained from the criterion' has been undertaken in Fig, A 97. The friction values for sman and intermedi· ate scale from the classification as by Barton et al. (1990b) are n""n"'!"rl•~nt on the alteration numparameter.. J. "" 1.on·m, sur· 0!1IY Si10Uid 01;! ,-,n:mn,,r~Hi with the 'sliding criterion' in The values are r~J><:ron:qnnJ in agreecriterion'. The ment with the Barton et al. are tilt tests that are reportad to be unreliable for faces {Barton et al.. 'l990b). stepped rion' is therefore ""'"""''"'""'~'~ literature values {Barton et 1990, 1971, Pereira, 1990) and criterion' vs in the i.e. quartz or Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX Ill ~~ Discussion influence of smaN scale roughness McMahon (1986) reported that small scale roughness is not important for shear friction along large (e.g. 30 to 250 m) discontinuity planes. This is based on comparison between peak and residual friction values from laboratory tests, intermediate and large scale field roughness mMSUrements and back-calculated friction values from failed slopes. Bandis (1983) found that the peak friction angle value {in laboratory tests) decreased with a larger test surface as did the difference between peak and re$lduat friction angles. During this research the friction values derived from the 'sliding criterion' are considerably higher than the values obtained by shear and tilt testing and they show an increase in friction angles with increasing smd scale roughness. The difference between the tendencies obtained during this research and those reported in these litefatw:e may be the following: 1) The literature values from real failed slopes are based on shearbox tests and roughness descriptions and measurements on discontinuities. After a sliding failure the discontinuity plane that failed wm have a different roughness profile and is unsuitable for back analysis (eh. C.3.3.2.6). For this reason it is good practice that both the roughness profhs are measured and the test samples are taken from other discontinuity planes in the same slope that are representatt'Ve for the failure plane. However, than the question arises: why did these planes not fail? Obviously a number of reasons are possible (differences in orientation, water pressures, etc.). lt is, howewr, afao possible that the friction along these planes is (slightly) higher than the plane that failed and that thus afao a iarger value for roughness friction is obtained. A friction value back calculated from the failed plane compared to the friction (roughness measurement and shearbox tests) from the non-failed discontinuities results in seemingiy less important roughness of the discontinuity plane. 2) The laboratory test results by Bandis (1983) are presently questioned because the results are based on averages while the scatter of results from individual tests is large. 1t is not unlikely that due to the equipment used lnon-comput erized shearboxl inaccuracies in the individual results masked the influence of small scale roughness. lt is doubted whether the conclusions would be the same if the tests are repeated in a modern computerized shearbox (discussion: Second international workshop on scale effects in rock masses, Lisbon, Portugal, 1993). 3) The scale effect between smaller and larger surfaces was also reported to be absent by Ohnishi et al. for artificiatnmples;-andth&retattonwasvague or absm'iffOra repetmorrof"the tests·orBandis on replicas of netural discontinuity surfaces COhnishi et al., 1993). 4) Another reason for the seemingly reduced influence of small scale roughness may be the handling of samples in laboratory and field tests. The larger the sample, the more difficult it is to perfectly fit two discontinuity halves together without damaging the asperities. The steepest asperities which are normally the smaller asperities, contribute most to the friction but especially the highest and sharpest asperities are most easily damaged and broken. Secondly the broken parts of these asperities may stay in the discontinuity and cause a (lower) rolling friction. Hence, the influence of small scale roughness seemingly reduces with larger sample size. I.Mge scale mughness (RIJ Threshold friction values obtained from the 'sliding criterion' fordiscominuities without infill and karst are shown in Fig.- A,SS versuakthe descdptions· for. large and small scale roughness. 'SHllifig crltenofi' compared to large scale field roughness measurements During the research a limited number of large scale roughness profiles have been measured. large scale i-angles (20 cm < base < 100 cm) measured on discontinuity planes in slate and limestone resulted in large scale roughness i-angles of between 6 ° and 1 0 ° for respectively slightiy wavy and wavy surfaces and 5° for slightly curved surfaces. These are lower than the threshold friction values for large scale roughness obtained from the 'sliding criterion' (fig. A 98). The large scale i-angles measurements done for this research have been done on exposed planes. The exposed planes are exposed because the material originally above has slid. This sliding may have reduced the large scale roughness f 'fc 70 CD ~-~. ----------- 80 --- . _. t-----,-----=-------...,--'--:.....,,--~~,_..-=-:-'--1 ~~qaaa~o~ba8clanlllldlnGidlllon o 1J ~so 40 for a clllcllntlrdywllll no 1111111 and no 11111111 __ ~ ::-:-:-:. _. _. IIWIIIy CIIMICI :-v ----- lllrllght 30 + + A ::SU:~IIhaldutaL(111110) -.trapllca balcflk:llonci'-.t smooth rough polished stepped Irough polished smooth undulating small scale roughness I rougl'l planar Fig. A 98. Friction vs large & srnali scale roughness and literature tHt test values of Chryssanthakis et al. (1990). i-angles. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ scale friction values '"""'"""''"'"' is «t~"'"'"'fi . The tilt test obtained from tl".a rRrl.!u:~lll~ that a cement and fine sand matrix wm be less are r"'"''nnn,.lr>iv not scaie tests in agreement with the and the the has been estimated underestim<~te lnfi!! matali<~l {lmj A 99 shows the threshold friction values for different infiil materials obtained from the criterion' as a contim.1ous line. The values are calculated for discontinuities with no r..arst, large scale ness and small scale ness '"'"''"'"'"'"' criterion ' - to litera- booll:mi!;o ~ iay<n values for infi!l material in natural discontinuities tun~ !n • mylmllle ~llfll~blll:lln. @kl.ml!lil 99 showsiitsrature va!us,s ...for !mmDM wl!ll ~tl!lng 1-$.! cm different infm materials {Hoek et al., .. Ui'llll!l!lll'!l <liaY!!'l!l!l!l "1 shear rp for filled natural ellscontinuities . As far as the infili thicknesses were '"""'nr1'<>n included in the are the residual friction ranges listed in the Q-system (Barton, 1988). The shear friction values based on the criterion' are correlate with the literature values. mm ~~&~join!$ 1981 ). The literature values are ~-$(liD -S!!ale fin® i gouge I imaguiarltles A 99. Friction angle vs inflll material iva!ues from Hoek et al., 1981, vertical lines from Barton, 1988). criterion' to literature values for infi!! material from artificia/sai'TinJes to compare threshold friction values obtained from the criterion' with tests on artificial discontinuities or on discontinuities with infiil materials as reported in the literature et al., 1990, Pereira, 1990). The materials and the circumstances under which these discontinuities were tested are, in general, very different from m:rtural materials and circumstances. th~ normal stress on the n"''"'""''"' often far than the stresses in the in this research. In are to the intm friction values resulting from the criterion'. The was not described iSRM but the friction (33") for a saw cut {planar surface) and !62 °) for the surface of the test without infi!l, were and from these values the the friction •u. ,-nr<~•n:n to the 'sliding criterion' could be back calculated and resulted in 5::mnnl"'~ used were not A 1 00) are onhn"'•"" amplitude criterion' (7 ° tests: 1 well be attributed to the > to a well in agreement with the criterion' 1). The high value {24°) for the thick infill '"'"'""'n"'r~,,..; differences between the circumstances Most failures of state thick fai!ure wm be failure it is likely that this leads to pore water pressures in the infm and thus low friction kaolin in the tests of et at was tested with a moisture content of 50 however, the of saturation is not se that these or less pore water .pressure. 2 3 have been tested in a not saturated state with no This reduces the for m."<cn:,un: .. The surface ln a is far tests. compared to the of water and thus reduction of pore water pressures in a The normal stress on the in the r.,-,,,·.,t··~rv tests is 50 MPa whereas the normal stresses normal stresses tor the tests in the ;esearch area are in the order of 0.01 MPa. The far in the Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX Ill 4 19'7 can lead to a collapse of the infill material structure and allowing easier discharge of (pore) water and reduction of pore water pressures. The shear velocity in real slopes is often far higher than in the laboratory tests (laboratory: 0.4- 1 mm/min), also reducing the possibilities for water discharge in slopes. lt is suggested that in slopes water pressures In the clay gouge cause an undrained shear behaviour whereas in the laboratory tests, with none or smaller water pressures, the shear behaviour is drained. The values found by Pereira ( 1990) (Fig. A 1 00) and Phien-wej (1990) (not in graph) for an open air dried, silty clay infill and oven dried bentonite infill (38° for 20 mm infill, roughness amplitude 10 mm) respectively, seem to support this suggestion. 70 eo 10 140 t. ' ' * 120 (thlclcnea 20 mm) * • 10 The values for non-cohesive soils of Pereira (1990) show that for the two larger grain sizes the friction angle is reduced rather than Increased. This effect is attributed by Pereire to rolling friction rather than shear friction (the silicious river sand wes rounded). 0 fig. A 100. Friction angle vs infm material compared to infill thickness laboratory tests. Papaliangas et al. (1990) tests with straight, rough undulating surfaces and Pereira (1990) with straight, polished planar sample surfaces. The friction angle values for discontinuities (Bieniawsld, 1989, Serafim et al., 1983) related to the descriptions in Bieniawski's (RMR) rock mass classification system (eh. 8.2.3.1) are difficult to compare with the threshold friction values found for the 'sliding criterion'. However, according to Serafim et al. the maximum friction along a discontinuity is 45 ° for a dry discontinuity and 37 ° for a wet discontinuity. The 'sliding criterion', laboratory and field tests, and the literature references cited in the foregoing chapters, allow for considerably higher maximum values and the merits of the values reported by Serafim et al. should be questioned. Apparent cohesion is not found for the 'sliding criterion'. This is expected for the more smooth discontinuity planes as the normal stresses in slopes are low compared to the intact rock strength, so that the asperities will mostly not -be sheared through, but.-.~ for the fougil4JF"stepped"aurfaces an a~,oohesioflwu expected but, however, not found. For larger test sample sizes the shear behaviour of a discontinuity is more ductile than brittle (Bandis et al., 1981, 1983, Muralha et al, 1990) and the apparent cohesion decreases. This may explain that apparent cohesion Is not present becat.lle the rock slopes studied have surfaces ranging between 3 mz and 300 m 2 which is, even for the smallest slope, considerably larger than the maximum discontinuity sample size ever tested. Cohesion resulting from infill material has also not been found. This may be due to the relative small number of discontinuities with a thick infill (gouge) which would have showed cohesion. The other parameters (roughness and karst) may mask the presence of cohesion. The generally good correlation found between the threshold friction angles determined with the 'sliding criterion' and the friction angles obtained from testing and found in the literature confirm the correctness of the 'sliding criterion' and the definition of the discontinuity condition parameter {TC). The 'sliding criterion' is therefore an appropriate method to determine the friction angle along a discontinuity, which is formulated as follows: n: ... 0.0113 * , n:' "' discontilwity cmrtlition ~r 'P "'friction agk along ~ (in degrees) [62] Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ AFPElv7JIX l'V ..t\PPENDIX IV INFLUENCE OF \\r:EATHERING ON GEafECHNICAL PARAMETERS l!W Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX IV '80 ~=-----*- r·.,.----..,.-------,,----.----===-..., 1180 I 1100 I •••••• Ieo A: 01)81'18 grained ------+-'l'al1vJtlft -+-'1111................. 140 • HllllltvJtlft Hllllltllolclom. --1"111-+Tll1_._ Tll1ll'ln I m~ 80 _., _ H-- -*- -e- r---.io 1 B: fine grained &. calcarious ~~+••m•m••••••••• I o+-----~----~----~===-~ 10+-----~----,-----~----~ hill ~ IIICiderlllly ~ clegNe of wetlltlelq (IS 5830;11111) ~ hill II'ICICienlllll ~ --of~ (BS 8830;11111) sllgliUJ ,40~--------~--~~~ fao ,., I 1 )10 t 0+--.- - r - - - . - - - - - , - - - - - ; hill Fig. A 101. Examples of average intact rock strength (field estimate) vs degree of rock mass weathering per lithostratigraphic (sub-) unit. 201 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ~~"~iintltW~fort!'il:'lin~lli'N~el~. ~ll {wnlilli!Oulll}, ·l.ll~t'l (~). J .. joint it:iashsd lino) e• t:X:amDil'~S !itlv><>t«~H•"<ra.nh,ir• A H:l3. {sub-) unit Of c ClVEJfag€; ,-l;.,•..,,.,..,.,; (SUO-) Unit i::!fld per type Of "'~'""n~• of average spam.,• vs of rock mass corrected for method of excavation). per Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX IV . : A:~ grained --------1 : :::~~~~---~---------~---., : -----.............. ,.. ~ .. o.e ta4 _..,.......,.. I u --+- (IIJ)IIIIilt- ·-+-- loi)H-0+-------~-------r------+-~~-rloi)~H---~--~ . ... ........................... B: fine grained : ............ : I ____ ,£ ---------~~~5~~~~~~:· r··------=:' . . . .. . f---.----=;r.:;::;=-===----.-_'--..,=..:..... ;:=-=r.:.___, -*- .., .. __ O+-~~~~~----·~-~..,~~~~--~ti*~~T-~----4 fNih hillll ...., IIIOdllllllrly .., --of~-~·) D: soli type fNah IIIIIUIIiiY II1CidlriMIIy lillllllv campi ._.af.,.,.._(ISiillllllll>1111) llflllhllv mccJenlllly lllghlr dag!aaf lllllhllllillg(IS5980;1881) ..... B - beddlnQ (conllnuoua), c .. cleavage (doftlcl), J .. joint (<IMhed line) Fig. A 104. Examples of the average condition of discontinuity parameter (TC) of a single discontinuity (set) vs degree of rock mass weathering per lithostratigraphic (sub-) unit and per type of discontinuity. lllgl!lly lllllCieraly lllghlr ~ cllglwoi'~(BS 11830;1111) 0011 11111111 cormctad for the irllluance ol' ll1e melhod ol' 8lrCaVIIIIon Fig. A 105. Examples of the average overall condition of discontinuities (conmassl vs degree of rock mass weathering per lithostratigraphic (sub-) unit. 203 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 1000 0+-------T-------r-----~r-----~ .._, llllllllnllly lilclliiY campi --d~ (1861110;'1181) CiifiiM&e IICiiiiiCilad tOr lhe lnbnOit d lhe m.tloci Of8ICCIIiVauon fllllh Fig. A 106. Examples of average (sub-) unit. cohm- vs degree of rock mass weathering per lithostratigraphic ~~------~.------~,r----,-,r-==~~ : A:. coarse grained : "1"";::----1111::,..._ : '---'r---=---i ' 10 0+-------~------~------~------~ frellll llfllhllr ~ higllly Cllqll degree d wealll8rlnll (8S 15130;1811) 0+-------~------~-------r------~ fllllh ~ l'l!llllllnllll highly degree« -'*In; (8S 11830;1881) 14~-------,----------------------~ 12 fiiiSh C: soil type lillgllllv IIICICIInlllly highly degree« 1llllldl8ring (8S 11830;1881) campi • - I s COII'8CI8d for ... inlluGince d ... melhod d 4IIXCilMidlon Fig. A 1 07. Examples of the average .,,_ vs degree of rock mass weathering per lithostratigraphic (sub-) unit. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX V Vv"'EATHERING CLASSIFICATION Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIXV W Term Description Fresh No visible sign of rock material weathering; perhaps slight discolouration on major discontinuity surfaces. I Slightly weathered Discolouration Indicates weathering of rock material and discontinuity surfaces. All rock material may be discoloured by weathering. 11 Moderately weathered Less than half of the rock material is decomposed or disintegrated to a soil. Fresh or discoloured rock is present either as a continuous framework or as core stones. m Highly weathered More than half of the rock material is decomposed or disintegrated to a soil. Fresh or discoloured rock is present either as a discontinuous framework or as core stones. IV Completely weathered All rock material is decomposed and/or disintegrated to soil. The original mass structure is still largely intact. V Residual soil All rock material is converted to soil. The mass structure and material fabric is destroyed. There is a large change in volume, but the soil has not been significantly degree VI transported. Table A 20. Degrees of rock mass weathering- BS 5930 (1981). m Introduction In the design of a slope the future degradation of the rock mass due to weathering is of major importance. In the SSPC classification system quantitative reduction values have been defined to accommodate for existing or future weathering; TheM· V8lues are related t& the degrees of rock mess weathering as described byf3S 5930 (19&1; Table A 20). This classification for rock mass weathering has been under criticism and different alternative classifications for weathering have been proposed since its publication in 1981 • The author has not noticed that any of these alternative classifications have been widely applied. Recently a new classification scheme for rock and rock mass weathering has been proposed by the Engineering Group of the (British) Geological Society (Anon, 1995). Whether the recommendations given by the Engineering Group will be widely accepted cannot be predicted, however, a comment on this scheme and possibilities to apply this scheme in the SSPC classification system is presented. The approach proposed by the Engineering Group (Anon, 1995) is composed of a general description of the weathering of the rock and rock mass (named: approach 1) and folio wad by different classification schemes (approaches 2 through 5) for different types of rock and rock masses (Table A 21). QUIIntificlltion of BS 5930 (1981) Although the scatter in the data is large it is shown in eh. 0.1.5 that it is possible to quantify the influence of weathering classified according to BS 5930 (1981 ). Some differences between the influence of weathering on the ,,gegt~al parameters ot.diffcnmt~tv.Pes of rocks and,,J~~~miWUitithave been notiG&d. hnw:ever,. thase differences .. are generally not large. Averaging over different lithologies and types of rock masses was possible and overall parameters for weathering influence could be calculated. Exceptions are the very weak 'soil type' units for which were found, in this research, that weathering does not influence the geotechnical parameters or has only a minor influence (the scatter in the data is larger than the influence of the weathering, eh. 0.1.5). approach 1 General description approach 2 Prescriptive classification for uniform materials approach 3 Prescriptive classification for heterogeneous masses approach 4 Prescriptive classification incorporating both material and mass features. approach 5 Classification of rocks that cannot be classified with approach 2 through 4, such as limestones developing karat. This classification may be based on associated characteristics (landform, etc.) of the rock mass but not on rock mass parameters directly. Table A 21. Classification approaches (Anon., 1995). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ AbMince of Wtlll'tllflling t:/eglwfJ8 Criticism on the BS 5930 classification of rock mass weathering focuses on the fact that it is not always possible to 'fit' the rock mass into one of the degrees of weathering or that degrees are not applicable to particular rock masses. This criticism particularly focuses on the percentages material decomposed or disintegrated into 'soil' which is one of the main criteria for the BS 5930. Some rocks do not produce 'soil'. As noted before (footnote 107), highly and further weathered rock messes following BS 5930, do not result from weathering of pure limestones or dolomites. The carbonates dissolve in surface and subsurface water. This may resuit in a karstic rock mass. Whether this should be classified .in a different weathering classification system is disputable. R~ liS 6930 ('IIJB'fJ by IIIII!IW ~~-following the~~~ Gmup of the Gedogiclll Society The newly designed scheme for weathering classification following the recommendations of the Engineering Group (1995) can be used for the SSPC classification system if the 'approaches' in the newly designed classification system are correlated to the old 8$ 5930 system in the following way (see also Table A 22): Appro11ch 2 - uniform matel'ial Grades I through V from approach 2 describe mainly the weathering of the intact rock in the rock mess. lt is proposed to correlate grades Ill through V of approach 2 of the new system to the degree 'moderately' in the old BS 5930.1n these rock types 'highly' and 'completely' weathered according to BS 5930 do not exist. Approach 3 - heterogeneous masses This approach can be correlated directly to BS 5930 if the term 'soil' in the description of BS 5930 is not teken too strict, but is teken equal to the material descriptions of grades IV - VI of approach 2 of the new system. Approach 4 - material and mass This approach can be correlated to the degrees of weathering in the old BS 5930, if class B of 'approach 4' includes both the degrees of 'slightly' and 'moderately' weathered in the BS 5930 classification. The value for WEapacmg (eh. 0.1.5) decreases considerably from 'moderately' (WE.rpd79 = 0.89) to 'highly' (W£~9 = 0.63) weathered. Therefore, the reduction of discontinuity spacing in the description for class C ('much closer !r.ii_'?~f:t.!PiiCil'lg'.} i! 111()~! 2~PI:If.~~-!o the ~~~ ()f.'.!:!ighly~ V!(titl:l~fe<t inlhe..BS..,5,930. classification . than . to the degree of 'moderately' weathered. Approach 5 - rock masses not fitting into approach 2 through 4 A classification based on, for example, landforms cannot readily be correlated to the rock mass weathering classification following BS 5930. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ :tCD ~ g..§2: :1> s·IONN CD PROPOSAL FOR THE COMPARISON OF THE WEATHERING SYSTEM BS 5930 (19811 WITH T~aNEW PROPOSAL FROM THE ENGINEERlNG GROUP OF THE GEOLOGICAL . SOCIETY 11995)(1 1 degree of rock m~is weathering - BS 5930 1981) :t ., ;:;.'a quantitative reduction values for weatherlrg ::r, .... !i i~~:t ~d' 0 ... (eh. 0.1.5 degree rock mass strength "0 n 0 0 .... 0 j ::r::::l approach 3 approach 4 . uniform materials (moderately strong or strong rock in fresh state) ~geneous masses of relatively strong and (mixt ; weak material) material and mass (moderately weak or weaker in fresh state) arade descriDtion zone description(21 class descriotion fresh 1.00 I fresh Unchanged from original state 1 100% grades I- Ill A unweatherad Original strength, colour, freeture spacing 11 slightly 0.95 siightly weathered Slight discolouration, slight weakening 2 ~~:tm oi"i" approach 2 WEmass en ... -- new proposal .J..orking group geological society (19951 CD 0 m::::1 .... IQ::r -· CD jO. CD~ ::1. .. c5 I G) a ac an ,Ill mol;ierately - Considerable weakened, penetrative discolouration - Larr:, pieces cannot be oken by hand IV highly -Large pieces can be broken by hand - Does not readily disintegrate (slake) when dry sample immersed in water V completely waathered - Considerably weakened - Slakes in water - Oriainal texture apparent ~thered• "0~ 0 3 -~~~ ... Ill ::rcn moderately 0.90 we,~thered CD :i: G')CD CD Ill 0 .... -::r 8~ --. ~· c5 cn2.. 0 -: n o -· :t ~­ $'cn 001 ::l CD ~ > 90 % grades Ill 10% grades IV- VI B 50 to 90% grades I -Ill 10 to 50%vrrades IV- 3 . highly 0.62 4 completely 0.35(3) 5 I l <c5 -w < I 30 to 50% grades I - Ill 50 to 70o/vrades IV - < 30% ~rades I - Ill 70 - 100 vrades IV - partialty weathered c distinctly weathered D de-struetu red (A) .... o CDCD_. 01CD _oo ..... Ill a .a 1: Ill ~ ~. ~ residual soil notes: (4) (11 121 (3) (4) .VI residual soi( Soil derived by in-situ weathering but having lost retaining original texture and fabric 6 I 100% grades IV- VI E residual or reworked ~!%.reduced strength, er fracture spacing, slig weathering penetrating 1n from fractures, brown oxidation Further weakened, much closer fracture spacing, grey reduction Greatly weakened, mottled, lithoreHcts in matrix becomi~ weakened and disordered, be dina disturbed Matrix with occasional altered random or apparent lithorelicts, bedding destroyed. Classed es reworked when foreign Inclusions are present as a result of transportation The correlation showed in this table is made by the author and is not a proposal by the Engineering Group of The Geological Society. Grades refer to approa(\11 2 (uniform materials). • Quantitative reduction values besed on granodiorite only. Not applied in this research. :lo. I ~ -.:: i Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ,.4PPEND1X V1 APPENl)IX VI EXAMPLES SSPC FORMS 211 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 212 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VI ~ HDATE: WEATHER CONDITIONS I Sun: cloudvlfair'lllilil!ll lr=Ra~in-:----;----SDU~~d~~~~.~~~~L-----i Ml::imN OF 213 LOCATION Map coordinates: DJ (ml 1: •CMEl total fllq)08UI'8: h: 6 d: 0.76 0.99 11-mepped~--:-on--:this:G.;-:-fo-rm-:--:-tm=l~t:--S==--~+-h-:---:6:----t-d-: "1.00 r --r-::---il 0.77 0.751~--~~----------~-----=77--+---------~--------~ poor/fair/~ 0. 72 Acceaaiblllty: 0.87 0.62 · NAME.: lilllll "' DESCRIPTJONJBS .5~ 3Q; 1~J.L colour !Ji.~ AA_.L. I~ ll-------~::-::-c=---'JN=:;p 1~2~·~~:! .L .... WEATHERING IWEI sample number(&): ~easily mhand :n:,.~~~tht~:.Csure (tick) unweathered slightly mOderately highly completely I 152:51~lo~ l1 Lumps broken bY ~ hemmer blow. LumP!! only chip by heavy hammer blows !Dull ringing 50· 100 MPa " 100- 200 MPa 1 sound! 1 Rock&' rlna on hammer blows. Soark 200MPa DISCON11NUITIES B=bedding C•Cieavage J•joint Dip direction !degrees> Dip > (degrees) Spacing IDSl 1.00 .,0.95 0.90 0.82 0.35 flv 4 _m 5 EXISTING SLOPE? ()6 dip-direction/dip 35K 1 45 heioht: 6.0m ~ lml K 6 ... . ;>j ... Stability ltickl lt---------,=~=~-=-=-=:==::::-::-::::==~-------'---=:.-'l'--'---=--=-....1---=;....;::...-.~.-___..___--ll stable .._______c_o_N_D.,..moN _ _o_F_D_IS_CONll _ _NU_I_n_E_s_____,,...,::-r----.,------,-----,----,-----!l small problems in ..1 00 near future 2 wavy : • large problems in Roughness sligclightly wavy :0 •95 near future 3 ufved :0.85 small problems 4 large scale (RI) 1 s~i!J~. curved :O.SO large problems 5 1 smuan~ :0.75 rough stepped/irregular :0.95 notes: Roughness :0.90 1) For infili '9ouge > smooth stepped pclished stepped :0.85 irregularities and small scale (Ha) rough undulatinp :0.90 'flowing material' small smooth undulating :0.75 0.95 scale roughness = 0.55. pcUshed undulating :0.70 21 If roughness is aniso(on an area p,t rough planar :0.65 tropic (e.g. ripple marks, 20 x 20 cm l Sft!\)~h})l,nar. :0.80 striation, etc.) roughness 11--------+o~ons~naaill:!.lo~JanarWI!.--:-:-=::-------':~0::,:.5::5:+----II----t---~l---l----ll should be assessed per:1.07 pendicular and parallel to cemented/cemented infill no infiU • surface staining : 1.00 the roughness and directions noted on this form. ncnecftili1g.,.&sii8ii7e(ifcciarse--:0:-95 31 Non-fitting of disconmaterlei; a.g. free of :0.90 tinuities should be 1 medium In fill cj~._~t.t~------_Lfl.Jl.! ____ :_O.:.I!li .......~.-.... ...... A 4u, 1.'marked in roughness soft ~'11111t81'fat; eoerse :0.76 " ,._,., · ·"V...,., ·"e'OJUmns. e.g. clay, talc, etc. medium :0.85 material (lm) 1 ______________J..f.!!l.! ____ ;.9...!lfi along strike ilfolig 'dli)"' . lml 0.4() (m) >fiB >( >.S o.ro /.()() /.()() > 0.6() > o.ro 0.4() .,, o.ro T gouge < irregularities GOIIfle > irreQularitiea Karst IKal t!owlna matenal none karst :0.42 :Q·lZ :u.uo :1.00 :0.92 /.'""' ...,., 0.92 0.92 lf-:------:---::---;------;:S:.:;U;::SC;::EPTI=:..!.,BI:U:TY~..:cro:...:.WE/1.==="-""""HEFR"'IING=~ IS\i'll:.::.~..----------------1! remarks: ,4/t ~ degree of weathering: date excavation: remarks: fig. A 108. Example I. Natural exposure B. Exposure characterization. -••fllf44. t. ..... "'•""-'• cu ~ ~ ~ - . H~or, ~ tAUO!Iftif•, de. -1 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 114 example l . . . " "" . . .. : : .. : ,f. :...~ : : : :: 1 discontinuity set .. :::# : ,:::::::: 08+-------~------~~~--~~~~~~N~.~.~~~ ,, OJ ... Cl 1 100 ~ 1000 discontinuity spacing (cm) = CD = ----------------· 1 1 1 DS1 DS2 DS3 --+---+- 0.64 - - + --- + --0.40 0.60 0.40 Rock mass friction: Fig. A 109. Example I. Natural exposure B. Reference rock mass calculation. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VI (tick) naturalfhEtnd-made pneumatic hammer GKC8Wtion pra·splitting/smooth wall blasting conventional blasting with result: good di"'"-nn•tin•~iti••" .r 1.00 o. 76 unweathered slightly 0.99 moderately highly completely Slope dip direction (degrees): 1.00 "0.951·········································································· 0.90 Slope dip ldegraesl: 0.62 0.35 Height (Hslope) (m): fig. A 110. Example I. Natural exposure B. Slope stability probability calculation. liS Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ example£ Probability stable: if SFm > slops dip L New road cut C, 85". calculation. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VI 217 SSPC. slope no; ~'!""' 1U4IIt Uti ff~U~' r;,o LOGGED BY: 11 LOCATION ..... ............................,. .. ......................... . Map coordinates: :map no: . •.. , :.~~.'!".~~.~ (J. . .... .. .. . . .'1!15::#.4.. · · ·· i easting: ····,a.t~rs DE"'AILS .OF SLOPE OF EXCAVATION ISMEl WEATHERING ISWEI (tick) 1.00 0. 76 0.99 natural/hand-made pneumatic hammer excavation pre-splitting1smooth wall blasting convantional blasting with result: g= .dl~ontinuitias diSIOdge(l blocks ./ ="n!:'ctct~k SLOPE UNIT NAME: 14121. 0. 77 0.75 0. 72 8::~ (tickI unwaathered slightly mOderetely highly completely 9$() Slope dip direction (degrees): 1.00 ./0.95 . 0.90 Slope dip (degrees): 0.62 0.35 Height IHslopel lml: note: SWE = 1.00 for 'soil type' units, e.g. cemented soH, etc. ?5 r.o ~. ORIENTATION INDEPENDENT STABILITY .........................................~.~.$."!':5~ro:H.!~'-"" . ·-··· -·-····················· -···· ............. ··············· SIRS .. l'!.l!'tS (from reference rock masst * SWE I weathering slope) • ?9 • 0/15 == 75 .................................. -.....................................................................................~S~V.ID..~ . t~t.MI ...................................... __ ,....... . SSPA = RSPA (from reference rock mass) • SWE (weathering slope) * SME (method of excavation slope) 035 * 0.95 * ().61 = SSPA = Qf..P.!~GQNI!N!J!TI!;$... 1~!;P.L .............. SCD = RCD (from reference rock mass) * SWE (weathering slope) ....... . . .. ... . . . ... ..... .... .. . .. .. ... .........GQNP!TIQN H • ••••••• •• • ... .. •••••• SCD ·Racit.lii'iliiS irictioil:··si'Ri ;;;; siRs • ·o:2417~Y~~2~~M~P.5~~~.~!QNJ$.ffi!..... ~®.HL · •••••••••••••11-<::--·,..,..,-w-•w,••••••m••.••••••••M•.o•-•• Rock mass cohesion: SCOH = SIRS * 94.27 · ••••••• ().22 •••••••••••••••• = ().65 * (). 95 • 0.62 - ·· ···· ········ ·· ••-•mm•--• SFRI ... 15 O 0~.2~.Lt.J?.Il.~. .§k.'IA~-~..§,.Z1.!L~ ..;, __ m•~•~••"'""'""ll·•. ••••• + SSPA * 28629 + SCD * 3593 . SCOH = 75 * 94.27 + (J# * 28629 + ().62 * 3593 = • !5596Pa .....................................................................................................................lt.~ffii :SJ!!!?.P..!!.!f.!P.; MA~.!Mv.M ...!?.!P.~J:!~!~HT..01~1!:!............................... ....................................................... Maximum possible height: Hmax = 1. 6 * 10... * SCOH • sin(slope dip) * cos(SFRII I I 1-cos(slope dip - SFRIII Hmax = 1.6 * 10... * retios: 15596* sinl75°1 * cosi.!B 0 JI(1·cosl?5• -$5°))= i T.6m .. ?$.o .. cc; .... , ...... (2//!l.. ....... . ...........................•.....•.•....•...•..............•...•........•................ J?J:!'!l./!1.19.P!!QIP.. :c'...~~J i :·. Hmax I Hslope • Probability stable: If SFRI >slope dip probability = 100% else use figure for orientation independent stability: ¥.6 m I T.O m = ~ ~ ORIENTATION DEPENDENT STABIUTY 5 Probability stable: Determination orientstlon stability: calculation AP:- if = discontinuitY -diP. = sloPe di~ -direction stabilitv: slidina tooolina > 84° or AP < -84° (slope dip+5°) < AP < vartical with 100% 100% (slol?!, dip-~~! < ~p < 1slooe d11> +5o, equal 100% 100% with use greph sliding 100 "' "' AP 840 0° < AP < (slope dip-5°) 100% 100% = discontinuitv diD-direction: 6 = a- 'f: AP == arctsn (cos 6 stsbilitv: sliding AP < 0° and 1·90° - AP + slope against 100 % diPI < o• AP < 0° and 1·90° • AP + slope against 100% dio) > 0° * tan Ill Fig. A 112. Example I. New road cut C, design slope dip 70°. Slope stability probability calculation. tooPiing 100 % use greph toppling Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 218 emmpleH (ml 1: 15 tml 1: h: 15 d: /() h: 15 d: I() EXISTING SlOPE? 0.75 large scale (RI) ().15 0.95 Roughness small scale (Rs) 0.95 0.95 ().'/() ().55 /,()() 0.92 f.()() Ion an area of 20 x 20 cm 2 ) In fill material (lm) Karst IKal ----------------~~---·- fig. A 113. Example 11. Exposure characterization. small problems in near future large problems in near Mure smaU problems large problems 2 3 4 ol'5 notes: 1) For infill 'pouge > irregularities ana 'flowing material' small scale rout,ness = 0.55. 2) I roughness is anisotropic (e.g. ripple marks, striation, etc.) roughness should be assessed perpendicular and parallel to the roughness and directions no~r this form. 31 Non· of diaconti· nuities shou d be marked in roughness columns. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VI ().19 OJ 1 10 • diecontlnulty spacing (cm) CD = - - - - - - - - 1 1 1 ----+--+DS1 DS2 DS3 1 1 1 --- + ------ + - ().$() 5.()() 5.()() Fig. A 114. Exampte 11. Reference rock mass catcutation. = ().46 219 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ no example 11 . . . . .. . . . . . . . . . . . ; map no: H~~rtti~.~~L 162 Slope dip (degrees): Height (HIIIope) (m): fig. A 115. Example 11. Slope stability probability calculation. 5.() Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VI + sl~ • el..,.._ c Joint 0 0 , 0 t + ••+. . +..,.•+..·> • ~ ... ···...t. .: oq 0 a: poles :< ( 6 :< ( 18 :< < 24 :< ( 30 :< ( 36 :< :< w 42 -... HEIIU:PHIDIE rarr"JJ\fi ( 0 :< < 3 :< ( 6 :< ( 9 :< ( 12 X < < 15 :< :< :< 18 2.1. -... UIR, IIEIUSP90 90 1100''11\fi c: joint systems Fig. A 116. Example Ill. Stereo projection. a: poles; band c: contours of poles and great circles of planes. Indicated orientations are dip-vectors. 221 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ m example Ill h: 1.00 15 d: ~0.76~--~--~~--~+-----~----r---~~---r----~--~1 0.99 2() h: 15 d: 2() sample numberls): < 1 MPa 1.25 • 5 MPa 5. 12.5 MPa 12.5 ·50 MPa ~ 50- 100 MPa 100- 200 MPa (tick) unweathered &tightly moderately highly completely Crumbles in hand Thin slabs break easily in hand Thin slabs broken by heavy hand pressure lumps broken by light hammer blows lumps broken by heavy hammer blows lumps only chip by heavy hammer blows (Dull ringing sound) Rocks ring on hammer blows. Sparks fly 1.00 ~0.95 0.90 0.62 0.35 EXISTING SLOPE? 2 Roughness fJ.F5 large scale IRil l()() ().75 3 4 /.()() ~5 Roughness ()//() ().65 ().55 f)}() l.tXI 0.17 . /.()() /.()() (on an area of 20 x 20 cm2 ) In fill material (lm) Karst (Ka) /.()() remarks: fig. A 117. Example m. Exposure characterization. /.()() Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX Yl 22:3 dis- DS2 DS3 CO = ·"···········~---····-1 1 1 ·····+--·--+····· DS1 OS2 OS3 (}.()2 ----h + ().2() ~---- 0.20 ---···············-············ 1 ------ + ----· + ...•... (U)2 020 0.20 = 0.65 111 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Ill flg. A H9. ilL Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VI !tick) 1.00 unweatherad rl 0. 76 slightly 0.99 mOderately highly completely fig. A 120. Example Ill. Slope stability probability calculation after failure. m Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ lV 4 UJO 0.75 0.75 0.75 0.60 0.95 0.55 1.00 Roughness small scale (As) characterization. 5 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX Hl 221 APPENDIX VII BLANK SSPC CLASSIFICATION FORMS Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VII tile 229 Blank forms that can be used for tile'ssfic system aretprovlded on following pages. The values for the reference rock mass and the probability of slope stability include expressions for spacing and discontinuity condition. These are calculated based on the discontinuity or combination of discontinuities that result in the lowest possible values for reference rock mass friction and in the lowest probability for the slope stability. This requires that calculations are done for each discontinuity set and for all possible combinations of discontinuity sets. This calculation is tedious and it is normally done by computer. However, a rock mass does not always contain more than one discontinuity set, or it is obvious which discontinuity set(s) will result in the lowest possible values, or a computer is not available. Therefore forms are provided which can be used for the calculations. One form should be used for each geotechnical unit. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 23t Blank SSPC classifiCOIWnjimns ..-.~ aeot.OGV ITCITU LOGOED BY: UDATE: WEATHER CONDf'llOHS Sun: cloudylfeir/bright I Rain: dry/drizzle/ I METHOD Of EXCAVATION !ME) (tick) naturaiJhand-made pneumatic hammer excavation pm-splitting/smooth wall blasting conventional blasting with rasult: good open discontinuities dislodged blocks frectUf&d intact rock crushed intact rock FORMATION NAME: colour grain size SSPC - SYSTEM TIME: hr 8Xj)OSUre no: LOCATION map no: Map coordinates: northing: easting: DIMENSIONS/ACCESSIBIUTY (m) 1: Size total aXI)osure: h: 1.00 0.76 0.99 mapped on this form: (m) 1: h: 0.77 0.76 0.72 Accessibility: 0.67 0.62 I NAME I sample number(s): INTACT ROCK STI'IENGTH (IRSI (tick) 1 Crumblss in hand 1 Thin slabs break easily in hand 1.25-5 MPa Thin slabs broken by heavy hand pressure 5- 12.5 MPa j lumps broken by light hammer blows 12.5- 50 MPa 1 lumps broken by heavy hammer blows 50- 100 MPa jlumps only chip by heavy hammer blows (Dull ringing 100- 200 MPa 1 sound) 1 Roc;ks ring on hammer blows. Sparks fly. > 200 MPa .. 1 DISCONTINUITIES B=bedding C=Cieavage J=joint .. 2 (degrees) Dip direction (degrees) Dip WEATHERING IWE) (tick) unweathered 1.00 slightly 0.95 moderately 0.90 highly 0.62 completely 0.35 < 1.25 MPa I < --- _l)pa_c::imllP_sL ____ ---~---· ..... ,------~ --------- --·· I along strike persistence along dip (ml I (m) (m) .. 3 .. 4 dip-direction/dip -- ----- -- --- large scale (RI) Roughness small scale (Rsl (on an area of 20x 20cm2 l material, e.g. free of 1 medium :0.90 material llml soft sheared material, e.g. clay, talc, etc. 1 coarse 1 medium :0.75 :0.65 Karst (Kal gouge < irregularities gouge > irregularities flowing material none karst _____________ .J..: _________ 1 fine :0.55 ------------------------------------------------------------- :0.42 :0.17 :0.05 :1.00 :0.92 SUSCEPTIBILITY TO WEATHERING ISWl remarks: '!_a_!e_~ca~_!i~'E _____ ------------------------- -------------- ----------------------------------- -------------------------------------- ----------------------------------- ---------------------- Fig. A 122. Exposure characterization. -----m 1 2 3 4 5 ~Wto CI.i_¥~~.:.~----~_!_.:_ __ ~~ '!_e_g~_!_E!_~a.!~E!!i!:.__-- ---, the roughness and directions noted on this form. 31 Non-fitting of discontif--itis-nct!l&be marked in roughness columns. :1.00 riOii sott&rling-&siiilaredTcoaiSe--:o-:-s5 Jnfill ------- notes: 11 For infill 'gouge > irregularities' and 'flowing material' small scale roughness = 0.55 21 If roughness is anisotropic (e.g. ripple marks, striation, ate.) roughness should be assessed per- ~wu iMil. no infill - surface staining ------- height: Stability (tick) stable small problems in near future large problems in near futurs small problems large problems :1.00 :0.95 :0.85 :0.80 :0.75 :0.95 :0.90 :0.85 :0.80 :0.75 :0.70 :0.65 :0.60 :0.55 wavy slightly wavy curved slightly curved straight rough stepped/irregular smooth stepped polished stepped rough undulating smooth undulating polished undulating rough planar smootli. planar polished planar .. 5 EXISTING SLOPE? CONDITION OF DISCONTINUITIES Roughness d: poor/fair/good DESCRIPTION (BS 5930: 19811 structure &. texture weathering I I I d: remarks: Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VII CALCUlATED BY: 231 'I exposure no: DDATE: I REFERENCE UNIT NAME: IN'W:T ROCK STRENGTH (RIM} I· ·•········ ····· · ······· ·············· ··ifiR8 >··132·MP& ilian.RiRs··;;;;··,32._.,.ms ;·ms·iifi. MP&ITWE icori:ec:tioil 10.: :we&t:h8rilllll;;;.·· ~:~·~ _., ... :.:: ;;;;,··· DISCONTINUITY SPACING IRSPAl 5 i SPA (see figure below) = .. 4 .. 3 •· ........ ! factor1 * factor2 " factor3 • .. 2 .. 1 DISCONTINUITIES * :: :).(:::: /: rQ[·~:~r::/ " ". . /. :1" , .. ' : : ::::::: : ::::::: ' ' . .. . . ,,, .. ' 3 dl8conl:lnulty sets ;;L;;,~1;~:,-~--: o.a ' /: :.t. ::: 'I' : /: :,,: :" m18l'm8dlate spaang : :://: :~-:--~ '~ /:::::::- muimUm spacing : .. ~.:t..,,,, ....... ''"'"" ........ / D.2 /: :.-:"1.:::/ ;,':: :::::: : :::::::: : ::::::: /'. / ,,;,1, .... '• : ,..;/,: :,..:"::: / : a1t-~~-'~/.~:~;~~~:~::r:/_~:~:~:~:~::~::T:_~:~:~:~:~:~::~::__~~~~ a1 1 10 100 diacontlnultv sPaCiM (cm) CONDITION OF DISCONTINUITIES (RTC & RCD) DISCONTINUITIES Rouoliliess iar9e scaie Roughness smaU scale · ·· iRif · ·· .. 2 .. 1 .. 3 : .. 4 .. 5 : IRsl +· ···················· ···,··················· ········· ............... ...................................................+·······················!···································'········ lnfitl material Karst (lml (Kal+ . : ; 1 , · · · · ;'· · · · · · · · · · · · · · ·:;·· · · · · . · · : RTC is the'dlscontinuity condition of a single discontinuity (set) in the reference rock mass Total IRI*Rs*lm*Ka = TCI : corrected for discontinuity weathering. ""=RT=c--------1----..._ ,----'--,----'.----'--------1 RTC = TC I sqrtl1.452-. 1.220 * e"I·WEll ll-=:,--.,.---.,..,.,--.,.,--,.,.,----=.:::-1----i----------'----.:.---~ Weighted by spacing: TC1 TC2 TC3 -·-- + - - + ---DS1 DS2 DS3 ---- + ---· + ··-- CD = ----··-··-·---·------ = --··----·-·---1 1 1 ---- + - - + -----DS1 DS2 1 1 1 = ....... ---- + ----- + ---- DS3 corrected for weathering: RCD (with a maximum of 1.0165) = CD I WE = .... I .... = . REFERENCE UNIT FRICTION AND COHESION IRFRI & RCDH) Rock mass friction: RFRI = RIRS * 0.2417 + RSPA * 52.12 + RCD " 5. 779 Rock mass cohesion: RFRI = ..... * 0.2417 + ..... * 52.12 + ..... * 5.779 =: Rcoii. ;;;; fiifis ;; 94:27 +fisiPA: * 28629 +'ficD * 3593 ........ . . . ...... . . . . . .. . . . . . . . . .. . ...... Pa RCOH = ..... * 94.27 + ..... * 28629 + ..... • 3593 = : notas: 1) For IRS (intact rock strength) take average of lower and higher boundary of class. 2) Roughness parameters should be reduced or shear strength has to be tasted If discontinuity roughness is non-fitting. 3) WE = 1.00 for 'soil type' units, e.g. cemented soil, etc.. 41 If more thanJhree discontinuity sets are present in the rock mass then the .. reference rock mass friction and cohesion should be calculated based on the combination of those three discontinuities that result in the lowest values for rock mass friction and cohesion. fig. A 123. Reference rock mass calculation. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ m Bltmlc SSPC classtfiClllionjorms I~GI\IEERING GEOLOGY ~~BY: HDATE: SSPC - SYSTEM Hslope no: D~ILS OF SLOPE WEATHERING iSWEl METHOD OF EXCAVATION ISME) (tick) (tick) 1.00 unweathered neturallhend-made pneumatic hammer excavation pre-splltting/smooth wall blasting convantional blasting with result: good open discontinuities dislodged blocks frectured intsct rock crushed intsct rock 1.00 0.76 slightly 0 · 95 0.99 moderatsly 0.90 highly 0.62 0.77 completely 0.35 0.75 0.72 note: SWE 1.00 for 'soil type' units, 0.67 e.g. cemantsd soil, etc. 0.62 'SiOP8CiiPI!i8i1re85i: = SLOPE UNIT NAME: ORIENTATION INDEPENDENT STABILITY INTACT ROCK STRENGTH (SIRS) ' ' ,,.,,,, sifis';·AiA5.itrolii.r&t&iiifio&'·n:;e:k·massl·· swe·iweil!tiiiiriilfi SiOP8i . ~,, :.:.:····:::::··'= ···· DISCONTINUITY SPACING !SSPAI ssi>A ;;;, Rsr>A itrom.r&flirence.rock m&&Si • swelweilt:ti&rinii.siOi)&i ,.,. sM'e.lmet:tioi.fof&Xi:iiVilt:iOil siCiP8l = .... * .. .. * .... SSPA ··sco ;;;··Fico iirolii r8t&r&i1e:&··;:o~:k·ma&iii =: CONDITION OF DISCONTINUITIES ISCDI ,.•..swEiweiltt.Eiiii\9 sioli&l., .... ,,,...............,. ,. ,....,...,......,.....,..... SCD= ..... * ..... =. SLOPE UNIT FRICTION AND COHESION !SFRI & SCOHI ,, SFRt,,. .... " &;24-H + .,';';''4 "'52.12: Flock mass C:oil&siaii:-scoii;;;,- ·sifls '*.94:27 +' ssfiA';;··:zas29'+ sco • 3593'''' ' If SFRI i'' .,;';'''* 5. 779 SCOH = .... * 94.27 + .... * 28629 + .... * 3593 < slope dip: MAXIMUM SLOPE HEIGHT (Hmexl =' o = : .... Pa 'Miiximlim.lioiisiili& il&ii1tii: iilnilx··; 1:a· *1ifi'. * scoii·• &iriisiope!iilii'* cosisi=Fiii'T.ii:e:osi'sioli&.liili :··si=Fiiii··· Hmax "' 1.6 * 10-< * .... *sin( .... 0 ) . cos( .... 0 ) 1 (1-cos( .... 0 • - .... 0 )) SFRI I slope dip = ..... o 1 ... .. o = ' "" ratios: Hmax I Hslope = ..... m I ..... m ProbabUity stsble: if SFRI .... m = : > = I ..... '*' slope dip probability = 100 % else use figure orientstion independent stsbility: ORIENTATION DEPENDENT STABILITY DISCONTINUITIES ·Dip direction ·· ····· · :~b;:A~;i~~t.:. V.!J.n_)~~~.-~r. ~9~~~ :: .. 2 .. 1 ······ ii:i8Qi885i · · · · · · ··· •·· · :Ja~~,~~f: , ............. -.. . . . H ··ei'c:···· .......... ····························· i!i!igi8'8ii)'····· .m: lffilm i:etereriee 10rm1 · ··· ··· · ·· · · ···················· : · ·Sic ;;;;. IUC..•.. slirt:i·1:452 -··1:22o· •· e;.:i~swe>>····· ....... ·····•········ Probability stsble: .... % .. 3 · • · ··· .... % .. • .. 4 • ............. . H. · , ...... ••• ...... H . ., . . . ' .. 5 . . . . . . ., . . , . . . . . . . H . . . . . . . . . . . . . . H . . . [ [ I .... % .... % .... % Detsnnination orientstion stsbllity: calculation AP: 11 = discontinuity dip, o = slope dip-direction, or = discontinuity dip-direction: a = o- 1:: AP = arctsn (cos stsbility: sliding toppling stsbility: sliding AP < 0° and (-90° - AP + 100% 100% against vertical 100% AP > 84° or AP < -84° slope dip) < o• AP < 0° and 1-90° - AP + with 100% 100% 100% against (slope dip+ 5°) < AP < 84° slope dip) > 0° (Slope dip-5°) < AP < equal 100% 100% (slope dip+ 5°) use graph with 100% 0° < AP < (slope dip-5°) sliding Fig. A 124. Slope stability probability calculation. a * tanJSI toppling 100% use graph toppling ...... Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ APPENDIX VII probability to be stable > 95 % probabil l33 95 %.: . . .· ...90% to be stable < 5 o/o 0.1 o~o 0.4 ·0~6 SFRI I slope dip fig. A 125. Probability of orientation independent slope stability. Values indicate the probability of a slope to be stable. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ with respect to sliding . ' ,, ~·~~' ·· ·····.,. ···· ··········.. ·· ·~~~-~~~•~•~-~•·~••••~" ooooOotl .. ~•••~•-••••~0> •••~~~•~•" ,, ~· •• ' • • • • • • . ., " , " " " •• _, ~ ~ ~ " Q'" • ,. 0 •• ' • • • • • '· ' ~ • ~ " 0 " 0 • ~ " • discontinuity unstable ~·-~ ···>···········:·~ ... ~~~-~-···.·········~··:·····~·~""'' .... ... ... . ' ' . ... . . ' ..' ... . .. 10 AP (=apparent discontinuity dip direction slope dip) (deg) stability. A 126. ~ •• " ~ " ~ •• ~ ~ .r. ~ ~ " • ~ " " . . ~ ~ 9 ••• 5 ~ " •• ~ " ~ " " - ~ ~ • • • • • • • • • ••••••• ~ ~ • • ~ " • •. " • • • • " •• ~ • Jj •· •• ~ ~ ••• ~ " ••• ~ • ~ ••• . ' ••• ~ ..... Q " " ••• " 95 discontinuity ..................... with respect to toppling ~ • ~ • " •••••• . ... ~ •• * • • • . • • • • •,• 0 .. .. .... " • " •• " . ... ... . t •••• ~ • < • . . ... • • •• • .. . • • 0 •• ~ " " •• 0 • .. . " •• 0 . ' ' •••••• ~ " ' ' ' 0 • ,_ " • ~ •• " . .. ... .. . ... .. . .. 40 50 • "'. - •••• ' ' ' 0 10 20 30 + dip fig. A 121. ~ • • " ~"' •• " •••• " •• 0 . .. . ......... 0 ••• ' , .. .. . .... " " " ••••• ~ • with respect to toppling ' ' 60 70 80 90 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ REFERENCEi 235 3DEC (1993). Three-dimensional distinct element code. ITASC.A group, Inc. ~~~····"~'"'""' Abelin H., T, Neretnieks I. & Moremo L {1990}. Results of a GEOVAL, on Validation of Flow and models. OECD Nuclear 14- 17. AG! (1976). . Anchor Books. 4 72 pp. Anon. (1970). The ""'~''"''""''.;,..,., report on: The of rock cores !?Arunnv_ 3. pp. 1 - 24. on: The Anon. t::nt1!nt3enrna and l:~eow•uv. (final D.G. & Wilson A.O. !19721. Case histories of three failures, California !9), - 299. Baardman B. !1993). Detailed modelling of discontinuity roughness in UOEC. Memoirs Centre rnt?Jm~"'"'nn t'T!'!nlr:•nv No. 109. De!ft, The Netherlands. 113 pp. Bandis lumsden A.C. & Barton N. (1981 ). studies of scale effects on the shear behaviour of rock lnt. Journal Rock Mechanics, Mining Sciences & Geomechanical Abstr: 18. pp. 1 - 21 . Bandis S.C., Lumsden A.C. & Barton N. (1983), Fundamentals of rock deformation.lnt. Journal Rock Mechanics, Mining Sciences & Geomechanical Ahstt: 20. {6}. Bandis S.C. (1990). Mechanical properties of rock joints. Rock Joints. eds Barton & sr"''"~'"~'"m~ . pub!. Balkerna, Rotterdam. pp. 125 ·· 140. Barton N.R. {1973a). Review of a new shear criterion tor rock 7, pp 287 • 332. Barton N.R. {1973b}. Review of a new shear criterion for rock Geology 7, pp 509- 513. Barton N.R., Uen R. & Lunde J. !1974}. Engineering Classification of Rock Masses for the ot Tunnel Rock Mechanics. 6. . Springer 89 - 236. "'"''"""'"''"'~""' with the of tunnel support Pro. on ... ,.,.,m.r:omrnn for Rotterdam. pp. 107 - 117. Barton N.R. {1976bl. Rock mechanics review. The shear of rock and roek lnt. .)oumaf Rock l'v1echanics, Mining Sciences & Geomechanical Abstr. 13. pp. 255 ·· 279. Barton N.R. & V. 1,1977}. Shear ofrock in Mining Sciences & Geomechanical Abstr. 10, pp. 1 · 54. Barton N . R., llliset F., Lien R. & Lunde J. i1980). of dimensions and support for underground installations. fnt. on Subsurface Rockstore '80. Stockholm. 2. ed. Bergman M. pub!. Pergamon, Oxford, 1981. pp. 553 - 561. Barton N.R., Banrlis S. & Bakhtar K. {1985). deformation and !nt. Joumal Rock Mechanics, Mining Sciences & Geomechanica! Abstr. 22. 13). pp. 121 Barton N.R. (1988). Rock Mass Classification and Tunnel Reinforcement Selection the Proc. Rock Classification Systems for Engineering ASTM Technical Publication 984. ed. Louis Kirkaidie. publ. American Society for and Materials, Philadelphia. pp. 59- 88. Barton N.A. & Stephansson 0. 11990al. Rock Joints. Pmc. !nt. on Rock Joints. . Ba!kema, Rotterdam. 814 A:S4>0C,,... f;;J~J:ltil'liJSimfJ. U6•0JG,If;fi:Slts, pp. Rock Barton N.R. & Bandis S. \i 990b). Review of predictive capabilities of JRC-JCS model in Joints. eds. Barton & Ba!kema, Rotterdam. pp. 603- 610. Bear J., Chin-Fu & Marsily G. de. !ads) 0 993). Flow and Contaminant Tra,ns,ooJ"t in Fractured l?ock. Academic Press, !ne., San 560 pp. Bekendam R. & ?rice D"G. 11993!. The evaluation of the stability of abandoned calcarenite mines in South Netherlands. Proc. Symp. !SRM £UROCK'93. Ussabon. . Baikema, Rotterdam . pp. 771 · 778. Berkhout T.J.G.M. i1985). Model tests to assess the deformation characteristics oi rock foundations. Memoirs 32. DelH, The Netherlands. 83 pp. Centre classification of rock masses. Tr;ms. South Afric<m Institution of Civil Bienlawski Z.T. {1973). Fnrllrl!!'!!rlruJ Hi, pp. 335 - 344. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Bieniawski Z.T. {1976). Rock mass classifications in rock engineering. Proc. Symp. on Exploration for Rock Engineering. Johannesburg. ed. Bieniawski. publ. Balkema, Rotterdam. pp. 97- 106. Bieniawski Z. T. (1989). Engineering Rock Mass Classifications. publ. Wiley, New York. 251 pp. Brekke T.L & Howard T.R. (1972). Stability problems caused by seams and faults. Proc. North American Rapid Excavation and Tunnelling Cont. Chicago. AIME, New York. Vol. 1 . pp. 25 - 41 . BS 5930 (1981). Code of Practice for Site Investigations. British Standards Institution (BSI). London. 147 pp. Burnett A.D. (1975). Engineering geology and site investigation- part 2: field studies. Ground engineering. July. pp. 29- 32. Carr J.R. (1989). Stochastic versus deterministic fractals: the controversy over applications in the earth sciences. In Engineering Geology and Geotechnical Engineering. ed. Watters. publ. Balkema, Rotterdam. 297 pp. Cervantes J.F.C.O. (1995). Behaviour of seismic P-waves in discontinuous rock masses. MSc. thesis. Engineering Geology. ITC, Delft, The Netherlands. 84 pp. Cindarto (1992). Rock slope stability. Msc. thesis. Engineering Geology. ITC, Delft, The Netherlands. 97 pp. Cindarto & Hack H.R.G.K. (in preparation). An example of analytical and numerical calculated rock slope stability. ITC, lnt. Inst. for Aerospace Survey and Earth Sciences, Delft, The Netherlands. Cording E.J. & Deere D.U. (1972). Rock tunnel supports and field measurements. Proc. Rapid Excavation Tunnelling Cont. Chicago. AIME, New York. pp. 601 - 622. Chryssanthakis P. & Barton N. (1990). Joint roughness (JRC") characterization .of a rock joint and joint replica at 1 m scale. Rock Joints. eds Barton & Stephansson. publ. Balkema, Rotterdam. pp. 27- 33. Cunha A. Pinto da (1990) (ed.) Scale effects in rock masses. publ. Balkema, Rotterdam. 339 pp. Cunha A. Pinto da (1993) (ed.) Scale effects in rock masses 93. publ. Balkema, Rotterdam. 353 pp. Cundall P.A. (1971 ). A computer model for simulating progressive large scale movements in blocky rock systems. Proc. Symp. on Rock Fracture. ISRM. Nancy, France. publ. Rubrecht, Nancy. Cundall P.A. & Hart R.D. (1985). Development of generalized 2-D and 3-D distinct element programs for modelling jointed rocks. Mise. Paper SL-85-1. US Army Corps of Engineers. ltasca Consulting Group, Minneapolis, Minnesota, USA. ___<;_•·!~Q~I-~~~~.JJ.JJ._~.~l~.FQI'!!!!,t!at!Q.'L.Cit~_t!JI~!i!.9!~nsiona! . . distincl.elementmodel. •.. lnt. . JoumaLRaclc Mechanics,.Mining Sciences & Geomechanical Abstr. 25, No.3, pp. 107 - 116. Das B.M. (1985). Principles of geotechnical engineering. publ. PWS publishers, Boston. 571 pp. Davis J.C. 11986). Statistics and data analyses in geology. publ. Wiley, New York. 646 pp. Deere D.U. (1964). Technical description of rock cores. Rock Mechanics Engineering Geology 1. pp. 16 - 22. Deere D.U., Hendron A.J., Patton F.D. & Cording E.J. (1967). Design of surface and near surface constructions in rock. Proc. 8th US. Symp. Rock Mechanics. ed. Fairhurst. publ. AIME, New York. pp. 237- 302. Deere D.U.& Deere D.W. (1988). The ROD index in practice. Proc. Symp. Rock Class. Engineering Purposes, ASTM Special Technical Publications 984, Philadelphia. pp. 91 - 101. Deere D.U. (1989). Rock quality designation (ROD) after twenty years. US. Army Corps of Engineers Contract Report GL-89-1. Waterways Experiment Station, Vicksburg, MS, 67. Den Outer A., Kaashoek J.F. & Hack H.R.G.K. (1995). Difficulties with using continuous fractal theory for discontinuity surfaces. lnt. Journal Rock Mechanics, Mining Sciences & Geomechanical Abstr. 32, No.1, pp. 3- 9. Eissa E.A. & l;ien Z. (1991). Fracture simulation and multi-directional rock qualit\( designation. Bull. Assoc. Engineering Geologists. 28 (2). pp. 193 - 201. Equotip (1977). Operations Instructions. 5th edition. Proceq S.A., Zurich, Switzerland (1977). Fec;;~er E• .& fumQer..s N. U 971). Measumment of large. s.cal.e.r.oughnesses ohock..pl.imes by..meaAS,Qf p,:efilegF~ and geological compass. Proc. lnt. Symp. on Rock Fracture. ISRM. Nancy, France. 1.18. publ. Rubrecht, Nancy. Fishman Yu.A. (1990). Failure mechanism and shear strength of joint wall asperities. Rock Joints. eds Barton & Stephansson. publ. Balkema, Rotterdam. pp. 627- 631. Fookes P.G., Gourley C.S. & Ohikere C. (1988). Rock weathering in engineering time. Quarterly Journal of Engineering Geology. 21 . London. pp. 33 - 57. Franklin J.A. (1970). Observations and tests for engineering description and mapping of rocks. Proc. 2nd lnt. Cong. on Rock Mechanics. ISRM. Belgrade. 1. Franklin J.A., Broch E. & Walton G. {1971 ). logging the mechanical character of rock. Trans. lnstn Mining Metall 80. Section A - Mining Industry, A 1-9. Franklin J.A., Louis C. & Masure P. (1974). Rock mass classification. Proc. 2nd lnt. Cong. Engineering Geology, IAEG, Sao Paulo. publ. Associacao Brasileira de Geologia de Engenharia, Sao Paulo. pp. 325 - 341 . Franklin J.A. (1975a). Safety and economy in tunnelling. Proc. 10th Canadian Rock Mechanics Symp. Queens University, Kingston, Canada. pp. 27- 53. Franklin J.A. (1975b). Rock Mechanics. in Civil Engineer's Reference Handbook. ed. Blake. publ. NewnessButterworths. Franklin J.A. (1986). Size-strength system for rock characterization. Application of Rock Characterization Techniques in Mine Design. New Orleans, Louisiana. ed. M. Karmis. SME-AIME, New York. publ. Society of Mining Engineers, Uttleton. pp. 11 - 16. Gabrielsen R.H. ( 1990). Characteristics of joints and faults. Rock Joints. eds Barton & Stephansson. publ. Balkema, Rotterdam. pp. 11 - 17. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ REFERENCIE 237 Gama C. Dinis da (1989). Analysis of marble fractures induced by stress concentrations at quarries. Proc. lnt. Cong. on Geoengineering, Torino. 2. pp. 805- 810. Gama C. Dinis da (1994). Variability and uncertainty evaluations for rock slope design. Proc. 1st North American Rock Mechanics Symp, Austin, Texas. pub!. Balkema, Rotterdam. pp. 547- 655. Gaziev E. & Erlikhman S. (1971 ). Stresses and strains in anisotropic foundations. Proc. Symp. on Rock Fracture. !SRM. Nancy, France. Paper 11-1 . publ. Rubrecht, Nancy. Genske D. D. ( 1988). Ansatz fiir ein probabilistisches Sicherheitskonzept ungesicherter Felsbiischungen im Rheinischen Schiefergebirge. Dr.lng. Dissertation. Bergische Universitat, GesamthochschuJe Wuppertal, Fachbereich Bautechnik.. (8). 210 pp. Genske D.D. & Maravic H. von (1995). Contaminant transport through fractured rocks: The state of play. Proc. 8th Cong. on Rock Mechanics. ISRM. Tokyo, Japan. publ. Balkema, Rotterdam. pp. 799- 801. Giani G.P. {1992). Rock slope analyses. publ. Balkema, Rotterdam. 361 pp. Goodman R.E. (1970). The deformability of joints. Determination of the in-situ modulus of deformation of rock. American Society for Testing and Materials. Special Technical Publication. 477. pp. 174- 196. Goodman R.E. & Bray J.W. (1976). Toppling of rock slopes. Proc. Conf. on Rock EnginHring for Foundations and Slopes. 9th speciality cont. Boulder, Colorado. ASCE, 2. Goodman R.E. & Shi G.H. (1985). Block theory and its spplication to rock engineering. pub!. Prentice-Hall, Englewood Cliffs, New Jersey, USA. 338 pp. Goodman R.E. (1989). Introduction to Rock Mechanics. pub!. Wiley, New York. 662 pp. Grima M.A. (1994). Scale effect on shetlr strength behaviour of ISRM roughness profiles. Msc. thesis Engineering Geology. ITC, Delft, The Netherlands. 100 pp. Hack H.R.G.K. (1982). Seismic methods in engineering geology. Memoirs Centre Engineering Geology. No. 9. Delft, The Netherlands. 170 pp. Hack H.R.G.K. & Price D.G. (1990). A refraction seismic study to determine discontinuity properties in rock masses. 6th Congr. lnt. Ass. Engineering Geology. Amsterdam. pp. 935- 941. Hack H.R.G.K., Hingera E. & Verwaal W. (1993a). Determination of discontinuity wall strength by equotip and ball mmu . ~EI~d. tests. IIJ.~u~tiJimal Rt:JCI<. Mf!.~..IJBnie&M[f!i!1fl$Ciences.~ Geomech~!Jif;.al A~tr..3QJ2}, pp. 15.1 - 155. Hack H.R.G.K. & Price D. G. (1993b). A rock mass classification system for the design and safety analyses of slopes. Proc. Symp. ISRM EUROCK'93. Lisbon, Portugal. pp. 803- 810. Hack H.R.G.K. (1993c). Slopes in rock. Proc. An overview of engineering geology in the Netherlands. ed. DIG. Technical University Delft, The Netherlands. publ. Balkema, Rotterdam. pp. 111 - 119. Hack H.R.G.K. & Price D.G. (1995). Determination of discontinuity friction by rock mass classification. Proc. 8th Cong. on Rock Mechanics. ISRM. Tokyo, Japan. publ. Balkema, Rotterdam. pp. 23 - 27. Haines A. & Terbrugge P.J. (1991 ). Preliminary estimation of rock slope stability using rock mass classification systems. Proc. 7th Cong. on Rock Mechanics. ISRM. Aachen, Germany. 2, ed. Wittke W. publ. Balkema, Rotterdam. pp. 887 - 892. Hakami E. (1995). Aperture distribution of rock fractures. Doctoral Thesis. Division of Engineering Geology, Dept. of Civil and Environmental Engineering, Royal Inst. of Technology. Stockholm, Sweden. 106 pp. Hammersley J.M. & Hanscombe D.C. (1964). Monte Carlo methods. Methuen. London. publ. Wiley, New York. 178 pp. Hart R., Cundall P. & L.emos J. (1988). Formulation of a three-dimensional distinct element. lnt. Journal Rock · Mechanics, Mining Sciences· & Geomechanical Abstr. · 25; pp. 111- 126. Hencher S.R. & Richards LR. (1989). laboratory direct shear testing of rock discontinuities. Ground engineering. March. pl), 24 . ~ . 31. Hoek E. & Brown E.T. (1980). Underground Excavations in Rock. lnstn of Mining and Metallurgy, London. 527 pp. Hoek E. & Bray J.W. (1981). Rock slope engineering. 3rd edition. lnstn of Mining and Metallurgy, London. 358 pp. Hoek E., Wood D. & Shab S. (1992). A modified Hoek-Brown criterion for jointed rock masses. Proc. EUROCK'92. ed. J.A. Hudson. publ. Thomas Telford. pp. 209- 214. · Holtz W.G. & Ems W. (1961 ). Triaxial shear characteristics of clayey gravel soils. Proc. 5th lnt. Conf. on Soil Mechanics and Foundation Engineering. Paris. Vol. 1 . pp. 143 - 149. Hsein C.J. (1990). A performance index for the unified rock classification system. Bull. Assoc. Engineering Geologists 27 (4). pp. 497- 503. Hsein C.J., lee, D.H. & Chang C.l. (1993). A new model of shear strength of simulated rock joints. Geotechnical Testing Joumal, GTJODJ. 116). pp. 70- 75. Hudson J.A. (1992). Rock Engineering Systems. publ. Ellis Horwood ltd., England. 186 pp. Hutchinson, J.N. (1992). landslide hazard assessment. Proc. 6th lnt. Symp. on Landslides. Christchurch, New Zealand. (3). ed. D.H. Bell. publ. Balkema, Rotterdam. pp. 1805-1841. ISRM (1978a). Suggested method for determination of the Schmidt Rebound hardness (part 3) and suggested method for determination of the Shore Scleroscope hardness (part 4)./nt. Joumal Rock Mechanics, Mining Sciences & Geomechanical Abstr. 15, pp. 95 - 97. ISRM (1978b). Suggested methods for the quantitative description of discontinuities in rock masses./nt. Joumal Rock Mechanics, Mining Sciences & Geomechanical Abstr. 15, pp. 319- 368. ISRM (1981 a). Rock Characterization, Testing and Monitoring, ISRM suggested methods. ed. E.T. Brown. publ. Pergamon Press, Oxford. 211 pp. u u •• Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ ISRM (1981 b). Basic geotechnical description of rock masses. lnt. Journal Rock Mechanics, Mining Sciences & Geomechanical Abstr. 18, pp. 85 - 110. Janbu N. (1973). Slope stability computations. Embankment dam engineering. eds. Hirschfeld & Poulos. publ. Wiley, New York. pp. 47-86. Japan (1992). Rock Mass Classification in Japan. Engineering Geology, Special Issue. eds K. Kitano et al.. Japan Society of Engineering Geology. 57 pp. Kirsten H.A.D. (1982). A classification system for excavation in natural materials. The Civil Engineer in South Africa. 24. pp. 293 - 306. KNGMG (1990). Geological Nomenclature. Royal Geological and Mining Society of the Netherlands. ed. W.A. Visser. publ. Martinus Nijhoff, The Hague. 540 pp. Kovari K. (1993). Gibt es eine NOT?. Geomechanik-Kolloquium, Salzburg. 42. pp. 17. (pre-print). Lajtai E.Z. (1969). Shear strength of weakness planes in rock. lnt. Journal Rock Mechanics, Mining Sciences & Geomechanical Abstr. (6), pp. 499- 615. lama R.D. (1978). Influence of clay fillings on shear behaviour of joints. Proc. 3rd lnt. Congr. IAEG. Msdrid. Vol. 2. pp. 27-34. Laubscher D.H. ( 1977). Geomechanics classification of jointed rock masses - mining applications. Trtms. lnstn of Mining & MetaHurgy. (Sect. A: Mineralindustry) 86, pp. A-1-A-7. Laubscher D.H. (1981). Selection of mass underground mining methods. Design and operation of caving and sub-level storing mines. ed. D.R. Stewart. AIME. New York. pp. 23 - 38. Laubscher D.H. {1984). Design aspects and effectiveness of support systems in different mining conditions. Trans. lnstn of Mining & Metallurgy. (Sect. A: Mineral industry) 93, pp. A-70-A-81. Laubscher D.H. (1990). A geomechanics classification system for rating of rock mass in mine design. Journal South African Inst. of Mining and Metallurgy. 90, No. 1o, pp. 257 - 273. lauffer H. (1958). Gebirgsldassifizierung tor den Stollenbau. Geology Bauwesen. 74. pp 46-51. lee C. & Sterling R. (1992). Identifying probable failure modes for underground openings using a neural network./nt. Journal Rock Mechanics, Mining Sciences & Geomechanical Abstr. 29 (1), pp. 49- 67. lee Y.H., Carr J.R., Barr D.J. & Haas C.J. (1990). The fractal dimension as a measure of the roughness of rock diaeontinuity profiles; /nt; Jourmtl Rock Mechanics, Mining Sciences & Gfiiimei::hariicalAbstr. 2 7, pp. 453 464. Louis C. (1974). Reconnaissance des massifs rocheux par sondages et classifications geotechniques des roches. Ann. Inst. Techn. Paris. no. 108. pp. 97 • 122. Marclia K.V. (1972). Statistics of directional data. publ. Academic Press Ltd., London. 357 pp. Maurenbrecher P.M., James J. & De lange G. (1990). Major road cut design in rock, Muscat Capital Area, Oman. Mechanics of Jointed and Faulted Rock. eel. Rossmanith. publ. Balkema, Rotterdam. pp. 929- 936. Maurenbrecher P.M. (1995). Stereographic projection wedge stability analyses of rock slopes using joint data. Mechanics of Jointed and Faulted Rock. eel. Rossmanith. publ. Balkema, Rotterdam. pp. 623- 626. Marquardt D.W. (1963). An algorithm for least squares estimation of nonlinear parameters. Journal of the Soc. for Industrial and Appl. Math., 2, pp. 431 - 441 . Mazzoccola D.F. & Hudson J.A. (1996). A comprehensive method of rock mass characterization for indicating natural slope instability. Quarterly Journal of Engineering Geology. 29. pp. 37- 56. McMahon B.E. (1985). Some practical considerations for the estimation of shear strength of joints and other disc.ontinuities. Proc. Jnt. Symp. on f~ of rockjoiwtB. Sjork.fiden, Sweden. pp. 475- 485. Moye G.D. (1967). Diamond drilling for foundation exploration. Joumallnstn of Engineers Australia. CE9. pp. 96 100. Muller L (1978). Removing misconceptions on the New Austrian Tunnelling Method. Tunnels Tunnelling. 10. Feb. pp. 29- 32. Muralha J. & Pinto da Cunha A. (1990). About lNEC experience on scale effects in the mechanical behaviour of joints. Scale effects in rock masses. eel. Pinto da Cunha. publ. Balkema, Rotterdam. pp. 131 - 148. Muralha J. (1991 ). A probabllistic approach to the stability of rock slopes. Proc. 7th Gong. on Rock Mechanics. ISRM. Aachen, Germany. 2. eel. Wittke W. publ. Balkema, Rotterdam. pp. 921 • 927. Nathanail C.P., Earle D.A. & Hudson J.A. (1992). A stability hazard indicator system for slope failures in heterogeneous strata. Proc. Symp. JSRM EUROCK'92. Chester, UK. ed. Hudson J.A. pub!. British Geotechnical Society, London. Neretnieks 1., Eriksen T. & Tahtinen P. (1982). Tracer movement in a single fracture in granitic rock: some experimental results and their interpretation. Water Resources Research. 18. pp. 849 • 868. Neretnieks 1., Abetin H., Birgersson l., Moreno l., Rasmussen A. & Skagius K. {1985). Chemical transport in fractured rock. Proc. Advances in Transport Phenomena in Porous Media. Nato Advanced Study Institute, Newmark, Delaware. pp. 4 7 4 - 549. Ohnishi V., Herda H. & Yoshinaka R. (1993). Shear strength scale effect and the geometry of single and repeated rock joints. Scale effects in rock masses 93. eel. A. Pinto cla Cunha. pub!. Balkema, Rotterdam. pp. 167-173. Pacher F., Rabcewicz l. & Golser J. (1974). Zum der seitigen Stand cler Gebirgsldassifizierung in Stollen- unci Tunnelbau. Proc. XXII Geomechanical Colloq. Salzburg. pp. 51 -58. Palmstrem A. (1975). Characterization of degree of jointing and rock mass quality. Internal Report. lng.A.B. Berclal A/S, Oslo, pp. 1 - 26. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ REFP:RENCE.~ I I 239 T., Lumsden A.C., Hencher S.R & S. (1990), Shear of modeH~:1d filled rock Rock Joints. eds Barton & Balkema, Rotterdam. pp. 275 • 282. Patton EO. {i 966). modes of shear failure in rode Pro(,~ 1st on Rock Mechanics. !SRM. 1. ed. Rocha M. pp. 509 • 513. Pereira J.P. {1990), Shear of filled discontlnuities. R(JCk Joints. ads Barton & . Balkema .. Rotterdam. pp. 283 ·· 297. F.C. ('1973) The use of in structural 8rd edition. . Amo!d, London. 90 pp. N., Shrestha U.B. & Rantucci G. {1990). Effect of infi!i thickness on shear behaviour of rock Rock Joints. eds Barton & Rotterdam. pp. 289 - 294. Pooi M.A. (1981), Rebound methods of in fieid and Memoirs Centre Fn<"Yin:eer.ina <:ieotD'OV. 5. The Netherlands. Price D.G., De C. & Pooi M.A. {19781. Field instruments for !AEG. 1\.J!sdrid. AEGA!, Madrid. Price D.G. {1984}. The detarmination of material and ITHISS 'w"'"'"'''~"'"' of rock. Genera! report, Session 13. Proc. 27th lnt. (Moscow}. VNU Science Press. 17. pp. 241 - 260. Price D.G. {1 992). Qmmtificstion of rock block form in BS 1981. Oral communication. Technical The Netherlands. versus mechanical discantinuities. Oral communication. ,..,.,....,.,.,,..h,Technica! Price D. G. (1993). DeHt, The Netherlands. Price D.G. i1995). A tac:As:·t~oi method for the classification of rock mass '"""'"'"'''"'y''"'" Journal of Em'1inA~er.fna t.e.r:uouv. 26. pp. 69 - 76. RAr~r!l'!'r"" N., Hack Brouwer T. & Kouokam E. lln Price map of Falset, Spain. !TC and TU Deift, The Neth~Nfands. Rabc:ewicz L 11964). The New Austrian Method. Water Power. Nov. pp. 453-457. Rabcewicz L & Golsar T. !1972). of the NATM to the works at Tarbeia. Water Power. Mar. pp. SB • 93. Rao, S.S. i1979}. and applications. Eastern Ltd., New Delhi. 711 pp. Rasmussen T.C. & Evans D. D. {1987). Meso-scaie estimates of unsaturated fractured rock fluid flow parameters. Proc. 28th US on Rock lvlechanics. Tuscon. eds Farmer Daemen J.J.K. & Desai C.S. Rotterdam. pp. 525 - 532. R,..,,.,.,..,.., N. (1970}. Influence of surface Proc. 2nd fnt on Rock Mechanics. lSRM. P..-.llnr,ui.-. ""'""'"..,.'N. (1971 ). Unebenheit und und Universitifit Kar!smhe. des Institutes fUr Bodenmechanik und Felsmechanik der Universitat Fridericiana in Karsruhe. (47). 129 pp. Robertson A.M. (1988). Estimating weak rock strength. A/ME- SME Annual meeting. Phoenix, AZ. Rode N., Homand-Etienne F., Hadadou R. & Soukatchoft V. 0990). Mechanical behaviour of of cliff and open Rock Joints, eds Barton & Rotterdam. pp. 693- 699. r~nm~m"" M. {1985). New adjustment ofthe Bieniawski classification to Proc. lnt. pp Rock Mechanics Works. Romana M. (1991). SMR classification. Proc. 7th 2. ed. Wittke W. . Ba!kema, Rotterdam. pp. 955- 960. Rosenbeum M.S., Rcee E.P.F. & Wi!kinson·Buchanan F.W. 0994). The influence of excavation on the '"'"".,"""'of unlined tunnel walls in Gibraltar. Prac. 7th fnt. IAEG, Ussabon. eds Oliveira R. et aL Baikema. pp. 4137 - 4144. nt""'""~JYtl' J.C. & Preston R.L (1978). with classifications of rock. Proc. lnt. Tokyo. pp. A3: 1 - 7. Z. & Eissa E.A. !1991). Volumetric rock designation. Journal Geotech Engineering. 117 (9). pp. 13311346. $en Z. !1992). Rock quality charts based on cumulative intact Bull. Assoc. 29 (2). pp. 175 - 185. Sarma S.K. 0 979}. Stability of embankments and slopes. ASCE Journal of the Geotechnicel ;..n,..,".,'"'"'""'n Division. 105(GT12), pp. 1511- 1524. Scavia C., 8ar!a, G. & Bemaudo V. 119901. Probabi!istic stability analysis of block failure in rock lnt. Journal Ffock Mechanics, Mining Sciences & Geomechanical Abstr. 27 i6), pp. 465 ·· 478. Schneider B. (1967). nouveaux de reconaissance des massifs rocheux. to Armales deL 'Inst. Tech. de Batiment et des Travaux Publics. 20, no. 235-236. pp. 1055 - ! 093. Selby M.J. (1980). A rock mass strength classification for purposes: with tests from Antarctica and New Zealand. Zeitschrift fi1r 23. pp. 31 - 51 . M.J. (1982). materials and processes" pubi. Oxford Press, OxJord. 264 pp. Classification of Bienlawskl. Proc. !nt. Serafim J.L & ?ereira J.P. (1983). Considerations of the Jm:.lt:m'lrnnnri Constr. . Ba!kema, Rotterdam. pp. 33 - 43. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ elements and of the natural aspects. Proc. 4th South American iNSMi with somf; c;n on Ffot.'k Mechanics. de ChiieL pp. 255 ·· 266. and of works. Prm::. 4th South Americ&n on Rock Mechanics. <>n!·,,.,•.,., de Chile. pp. 267,. 278. Shuk, T. {1 994cl. Natural Colombia. (oral commun!cationi. Shuk, T. {1994d}. Natura! (basic Sfluk, Colombia. {In nrl'>nl'l<r~t,lon SNmr\-.:n,n B. (19651. Index tests for rock. of Science and Je.::tmo!i:1nv London. on the role at rock Swindel!s C.E (1985). The datectkm of blast induced mechanics. Zacatecas, Mexico. pp. 81·88. mlnes, 1 in the Shabanie and H. W. 0 980!. A classification Zimbabwe. M. Phi!. Thesis. Univ.of Rhodesia. '"'"''"'''" K. ('! 946). Rock defects and loads on tunMl support. Rock n.u:•nP-.nma &. ·r. White. Commercia! OH. pp 15 ·· 99. •~··?,.,.,;..," A. D. ( 1965). Sources of error in (Hi). The Institution of Civil London. pp. 287 • 304. li.ilinov R. & Mo!okov L {1971 ). Role of on Rock Fracture. !SRM. France. pubL UDEC !1 993). Universal Distinct Element Code. group, !ne. VoU: User's manual and Voi.2: Verification and Minneapo!is, Minnesota, USA. of hillsides or scarps. Ground Movements and Vecchia 0. (19781. A simple terrain index for the Structures. ed. J.D. Geddes. Pentech Press ltd. pp. 449 • 461. Verwaa! W. & Mulder A. !1993). rock with the hardness tester. lnt. Journal Rock Mechanics, Sciences & Geomechanical Abstr. 30, pp. 659- 662. Weaver J.M. (1975). factors significant in the assessment ot The Civi!En;rJin;fier in South Af'rica. 17. pp. 3 1 3 - 3 HI. Welsh S.P. (1994). The of on the shear llltrl'>ntlth of rock discontinuities. Phd.-thesis, JJt:>n::n''rrr.,P.n of Earth of Leeds, UK. 257 pp. Sciences, Wickham G.E., Tiedemann H.R. & Skinner E.H. !1972). determination based on aei::>w,mc '"~.r~;,.1;,,..,.,, Proc. Rapid Excavation Tunnelling Cont., A!ME. New York. pp. 43 - 64. model - RSR concept, Pmc. Wickham G.E., Tiedemann H.R. & Skinner E.H. {1974). Ground support A!ME. New York. pp. 691 - 707. Excavation purposes. 1rans. Res. Rec. 783. pp, Wil!iamson D.A. {1980). Uniform rock classification for qe,ote;cnmc:a 9- 14. 21 {3). pp. 345 Wi!liamson D.A. 0 984). Unified rock mass classification system. Bull. Assoc. Em'lin<"!er.fno (ieOio'aJs "' 354. Yufu z. ('J 995). Principal conversion methods for rock mass classification systems used at home and abroad. Bulf. lnt. Geologists. 51 . pp. 81 - 88. Assoc. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ GLOSSARY 241 are not used in the literature. Therefore definitions of Definitions tor rock, rock mass and their used are listed below to avoid ln section .A the rnain terms and relations are described ln more detaiL is as ·far as necessary for the of the research and for the SSPC svstem and is based on the Terms of the .American emoPJ1ca1 Institute 1976) and the and of the Netherlands (KNGMG, 1980). *".,,,...,;.,..,.llr•n" which is J-angle See 'bi-linear shear criterion'. "imsk in this res!l<m::h "basic denotes the friction of a which friction does not ctnJse nn'"'"''nn of the Confusion has arisen in the literature about fPbasic· Some authors use 'Pba&c also for the "m of rock materia!, for 'Pro$1duat (which is the " obtained after or use the term for artificial be the same as the 'Pbasic of a but this is not surfaces (saw cuts). The " of these surfaces mH~es:sainiv so. See further 'bi-!inear shear criterion'. ((), See 'bHinear shear criterion'. Abutting discontinuities See Anisotropy The d""''"'"""'~"''"'" Si-linear shear criterion of shear is easiest with the 'bi·linear shear criterion' {Patton, 1966). For more for shear discontinuities literature. The shear is for a with a regular set of '""'""""·•h••"' formulated by Patton in the 'bi-iinear shear criterion' 128). The angle of friction is a material n<>n<>,nrli>nn on the structure, texture, type of material, !r'ITI~ril)!":i<lnr:l Of the is described by the = arctan !Jv/IJh). !n G 1 28 the roughness are the triangular asperities. Depending on the steepness of the asperities and the norma! stress across the the break rather than are overridden. The is then described by the rock materia! parameters cohesion and friction tj:'m" If there is no or agent {for example, cement) between the waiis the cohesion is described as apparent cohesion. The cohesion, or a part of it, may be rea! cohesion if a cohesion or G 'BHlnear shear criterion' for a with a set of (modified after Patton, agent is present. The parameters not the same as the cohesion 1966i. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 242 and friction i ~) ori~r~tetion orientation is the rn.ean of the orientations of the discontlnuities in a Crnuactaristic !::lh'l:co:ntilr>t The of discontinuities within one set of discontinuities ls defined as the G 129). The characteristic mean of the Characterization of a unit. A characterization is not Characterization is the "I'r"~"""""'" a classification. inclined to the rocks, is pressure, and re<:ry:stat!liz:ation of the rocks. Un this used for Carboniferous rocks Classificatio11 Classification is the characterization standard parameters which are related to an .A of the parameters acc:ormna to standard rules will lead to a recommendation for Cohesion {apparent/ For the strength 'bi-linear shear criterion'. mass and soil see 'Mohr-Coulomb failure criterion'; for dlscontinuities see 011 atlons. is denoted under pressure is zero as = 0) pressure. The G 130b). Creep in rock mechanics is a are different term. Various forms of or time deformation processes ;,vhich or chemical processes are a!! described as creep (eh. A.2.4). denotes that a has a less than, but in the sarne the difference between in the ,..,,_...,, '"''" '"" and dip. should be steeper than the Deformation Deformation of intact rock or a mck mass is the in vo!ume direction as, the rih,-ri;,.,• ..,H,.,,, should and be less Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ if sheared the """"'""" or chemica! characteristics of rock materiaL dsnotes a isolated fault,. isolated crack or etc.) that is not part of a set or, if part of a discontinui-· ty set, then the between the different discontlnuities is so that for purposes the may be considered as a feature.. set A c.;eow~.:;w~al denotes a series of discontinuities of and mechanical characteristics as well as their orientation are are: sets of sets, etc.). discontinulties are discontinuitias discontinuities t~>r which there is no "r'""'"" to the rock materiaL Intact rock may discontinuitles . Mechanical discontinuities are of weakness. fractures, faults, etc. are mechanical discontinuitias if the • Mechanical dlscontinuities to the discontinuity or the shear strength lower than in the rock material. the n~;,,,..,,,.,;,.,"'' Mechanical discontinuities will in be the boundaries for 'banks' of intact rock. The term bank is, however, not used as the definition of a bank is based on In this enltoUJg;ca• characteristics. 'discontlnuities' is used for mechanical discontlnuitles except where otherwise stated. lifetime Engineering lifetime denotes the lifetime of about 50 years. ex.pe~::re,o existence of an engineering structure. for a Failure mechanisms and modes Processes to siope failure are divided into different mechanisms that are sub-divided into different modes . For example, slope failure mechanisms are shear displacement, deterioration of rock material, intact rock creep, etc.; the failure modes of the shear mechanism are plane wedge failure, and, to some extent, walls are are sheared off, deformed wa!ls have been formation The primary unit of formal or . Most formations possess certain distinctive or combinations of distinctive mrnoiiOa!c<:m features. Boundaries are not based on time criteria. friction For the ,.,.r,<>n<wh n""'''"rirrtinn of mck, rock mass and soil see 'Mohr·Cou!omb failure criterion'; for discontinuities see 'bi··llnear shear criterion'. Geotechnical unit See unit • ueotleC!1!11 rock, as a result of wear as a WBI!S; thUS the initial Shear the rock ci"r"''"'''Th do not make Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Identification Identification describes the effect that a relation is defined that includes more parameters then necessary to relate the data. The parameters are not determined by the relation. For example: y =(a +b) *x y = dola a. b = parameters Both a and b can never be determined from this relation whatever the number of (x, y) data pairs. (Obviously for determination of (a + b) only one data pair (x, y) is sufficient.) In optimization of complex relation(s) identification problems might not be recognized leading to ambiguous results. x. Inhomogeneity Inhomogeneity is the spatial variation of properties of intact rock or of a rock mass. Intact rock Intact rock biocks are blocks of rock for which: 1) The physical and mechanical properties are roughly uniform. 2) The particles (mineral grains, rock grains, etc.) are bounded by a cementing agent which causes a block of intact rock to have a tensile strength. 3) An intact rock block does not contain mechanical discontinuities. Isotropy Isotropy designates that properties of intact rock or of a rock mass are not direction dependant. lithology - lithological The science of the rocks; in this study lithology denotes the typa of minerals, their origin or sedimentation environment. lithostratigraphic (sub-) unit See unit - lithostratigraphlc lubrication Lubrication by water may reduce the shear strength of discontinuities. The effect may be caused by the water itself which changes the mechanical characteristics of some materials. Another more general effect is that the presence of water will cause a reduction in shear strength because the surface stresses of water will cause a reduction of the normal stresses on the discontinuity walls. The quantity of water is not necessarily so large that an overall water pressure is established. lustre The appearance of a stone's surface (or of a mineral in general) in reflected light. Refraction index and perfection of polish possessed by the stone are the main factors affecting lustre, while hardness is also of some importance. Mapping unit See unit - mapping. Mollr::eGWomb tiiilure criterion The 'Mohr-Coulomb failure criterion' consists of a linear envelope (eq. [63)) touching all Mohr's circles representing critical combinations of principal stresses in the rock or rock mass, or soil (Fig. G 131). cohesion and tp ,;...,_ = cohesion + o...,_ * tan ( tp) ~ and agk of intemal friction of I'M material (63] are I'M Expressed in the 'Mohr-Coulomb failure criterion' the unconfinecl compressive strength (UCS) equals: ucs = 2 * cohaiM • tan(45° + -i) [64] The relation between minor (u3) and major (u1) principal stress at failure is: Non-fitting discontinuity See fitting discontinuity. Non-persistent discontinuities See persistence. a- fig. G 131. Mohr-Coulomb failure criterion. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Orientation See characteristic discontinuity orientation. Overfit Overfit describes the effect that a relation is defined that includes more parameters then necessary to relate the data. In optimization scatter on the data will cause that multiple, equally good, solutions are found. Each solution is a solution on different (clustered) subsets of the deta set. None of these solutions need to be the solution for the full data set. Outller An outiier is a data point which is clearly detached, or out from the main set of data points. Persistence (Fig. G 132) - Persisttmt discontinultitiS Persistent discontinuities are formed by a continuous discontinuity plane. Shear displacement takes place if the shear stress along the discontinuity plane exceeds the shear strength of the discontinuity plane. If unfavourable orientated it is often a sliding plane in slopes. - Abutting discontinuities Abutting discontinuities are discontinuities which stop at the intersection with another discontinuity plane. Abutting discontinuities might continue at the other side of the intersecting discontinuity, however, with a displacement to give so-called 'stepped planes' 1148l. Shear displacement along the discontinuity can take place if 1 ) the shear strength along the discontinuity plane is exceeded and 2) the blocks of rock against which the discontinuity abuts can move. mm - Non-persistent discontinulties Non pereistemdisoontinuities are discontinuitiee ending in intact rock; Before movement of the blocks on both sides of a non-persistent discontinuity is possible, the discontinuity has to extend and break through intact rock material. As intact rock material has virtually always a far higher shear strength than the discontinuity, a nonpersistent discontinuity will have a larger shear strength than a persistent discontinuity. Fig. G 132. Persistent, non-persistent and abutting discontinuities . Porphyritic, porphyrite A textural term for those igneous rocks in which larger crystals are set in a finer groundmass. Rock mass A rock. mass is a mass of rock blocks with or without discontinuities. A rock. mass may be homogeneous or inhomogeneous • Based on rock. mass parameters the rock mass is divided in homogeneous geotechnical units. Rock (mass) failure A rock mass is supposed to have failed if the rock. mass deforms more than allowed for a safe engineering application. Shear strength The shear strength is the shear stress at failure of a sample under a shear stress. See for shear strength along a discontinuity 'bl-linear shear criterion'. Slaty cleavage See cleavage. Slickensided Usually striated surface of rock. produced by friction. Soil typa units 'Soil type' units describe units which consist of loosely cemented grains or small particles, generally either without clearly defined mechanical discontinuities or having highly irregular and thinly laminated mechanical discontinuities, and having a low intact rock strength. 'Soil type' units resemble cemented soils rather than a rock mass. (!48) Stepped discontinuity planes should not be confused with discontinuity planes with steps. A discontinuity plane with a step is described in appendix ll. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Spacing See characteristic discontinuity spacing. S'lriated Surface of rock characterized by fine, narrow, curved or straight parallel grooves. Stylolite A term applied to parts of certain limestones which have a columnlike development; the columns being generally at right angles or highly inclined to the bedding planes, having grooved, sutured or striated sides, and irregular cross sections. Stylolites result from solution under pressure of limestone. The clay particles which were origina!Jy in the limestone, remained on the solution surface. Susceptibility to weathering See weathering. Tactie roughness Roughness that can be felt by using fingers. Tensile strength The tensile strength is the tensile stress at failure of a sample under a tensile stress. liiaxW compressive strength See compressive strength. Unconfined Compressive Strength (UCS) See compressive strength. Unit The following definitions are used in this study: - Uthostmtigraphic unit A layer or a body of layers characterized by consisting dominantly of a certain lithologic type (sand, clay, sandstone, shale, granodiorite, etc.). - Uthostratigraphic sub-unit A lithostratigraphic unit which characteristic bedding or cleavage spacing is within the ranges for discontinuity spacing as given by BS 5930 (1981) (Table A 17, page 181}. - Geotechnical unit A geotechnical unit is a part of the rock mass in which the mechanical characteristics of the intact rock material are uniform in each block of intact rock and the mechanical properties (including orientation) of the discontinuities within each set of discontinuities are uniform. Anisotropy of properties, if present, is uniform (eh. A.2.2). - Mapping unit The divisions made on an engineering geological map. (note: in this study 'lithostratigraphic sub-units' are a subdivision defined on bedding or cleavage spacing, of the 'lithostratigraphic units' found in the research area.) Weathering Weathering is the chemical and physical change in time of intact rock and rock mass material under influence of atmosphere and hydrosphere (temperature, rain, circulating ground water, etc.) (eh. A.2.4). A distinction is made between 1 ) the degree (state) of weathering (at a certain moment) and 2) the susceptibility to weathering (in a certain time-span). Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Sl'MBOlS 141 SSPC indicates that the"'""'''""''"''"'"" is used in the ciassificatkm system.'"'"'·".-·""" and codes used in the forms for the 'initial system {eh. C.4) are not included. used in to indicate that the parameter has no dimension in 67l friction angle CD e Hmax Im iRS or irs Ka X )( ME RCD RCOH RFRI R!RS RI Rs RSPA RTC SCD SCOH SF RI SIRS SME SPA spamafi SSPA STC st.dev. SW of internal friction of a rock mass apparent of of a in the direction of the direction of the s!ope dip {AP < O"l {SSPCl sets in an exposure rock parameter for the overall condition of a number of """'"""'"'"' mass unit !SSPC) cohasion oi' a rocil; mass parameter for the overall condition of a number of sets ln a rock mass characteristic (in metres) between the discontinuities ln one set in an exposure rock mass unit ISSPC) natural base of logarithms ie == 2.7182818 ... ) maximum height of a slope if SFR! is lower than the !SSPC) of a !SSPCl angle of for discontir.uities parameter for infi!i materia! in an exposure rock mass unit !SSPC) intact rock in the SSPC system used for the intact rock of an exposure rock mass unit parameter for l<.arst a rli<:.0nntim in an exposure rock mass unit (SSPC) natural of x (base e) lnru•riith1m of X {base 1 0} of excavation used for an exposure {SSPC} parameter for the ovaraH condition of a number of rli"'"""'ti sets in a reference rock mass unit iSSPC) cohesion of a reference rock mass unit !SSPCl of lntemal friction of a reference rock mass unit {SSPC) intact rock of a reference rock mass unit ISSPC) parameter for the scale of a in an exposure rock mass unit {SSPC) parameter for the sma!i scale of a discontinuity in an exposure rock mass unit {SSPC) parameter for the overall spacing of a number of discontinuity sets in a reference rock mass unit (SSPC) parameter for the condition of a \set) in a reference rock mass unit iSSPCl sets in a rock mas:s parameter for the weighted overaH condition of a number of ur!it !SSPC) rock mass unit (SSPCl cohesion of a of internal friction of a rock mass unit !SSPC} intact rock of a slope rock mass unit \SSPC! parameter for the method of excavation used for a new ISSPCl sets in an exposure rock mass unit parameter for the overai! of a number of n•<'"'"'"t' ISSPCl sets in a rock mass parameter for the overall "'""'"""'"" of a number of sets in a rock mass unit {SSPCl of a number of parameter for the overall parameter for the condition of a r!i<>rnntono rock mass unit {SSPC) standard deviation usc:ep1tibillity to Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ SVVE {~ TC parameter for the condition of a WE!.-~ parameter for the ,.,,.,.,t,h<>ei"'"' parameter '~<"'"'th,.,,rin,.., parameter ~AIE\~-tha·wil'"l parameter '"'&"MM•<"in'" parameter of !set) in an exposure rock mass unit !SSPC} of an exposure rock mass unit {SSPCl for the condition of a {set) in a mck mass unit for the cohasioo of a rock mass unit for the of intema! friction of a rock mass unit for a rook mass Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ INDEX :mEc se, 1eo, 235 AbeHn 41 ~ 235, 238 AGi 235, 241 235 A!pine 17, Hl anisotropci 4, 6, 7, 9, 13, 30, 36, 39, 58, 68, 139, 141, 145, 147, 152, 213, 218, 222, 225, 230, 237, 241, 246 A? 99, 151-Hl3, 215-217, 22:0, 224, 225, 232, 247 aplite 181 Amolct 8, 235, 239 asperities 14, 55, 63, 64, 63, 64, 63, 64, 60.58, 95, 93, 195, 197, 236, 241, 243 ASTM 235, 236 Baardman 64, 65, 235 S!lkht!lr 235 cohesion as, 42, 70, 92, 93, 95, 96, 99, 100, 130, 131, 151, 161, 169, 193, 197, 209, 24'1, 242, 247 apparent discontinuity 95 rock mass 107, 123, 147, 148, 150, 153, 154, 157, 158, 214, 215-217, 219, 220, 223-225, 231, 232, 24'7, 248 concept 5, 24, 26, 44, 54, 74, 81,83, 87, 88, 125, 240 conglomerate 16, 17, 181 construction materials 23 Carding 32, 236 creep 1 "I, 13, 14, 79, 80, 189, 242, 243 Criterium 165 Cuntlal! H30, 236, 237 Cunha 63, 72, 238, 238 oaemen 2.39 ball rebound 68, 237 Das 107, 236 Bandis 62, 63, 68, 195, 197, 235 Bar!a 239 day-righting 93, 94, 97, '!01, 187, 242 Bllrr 238 Sarton 25, 26, 28, 31-35, 38,-40, 44, 62, 63, 65, 68-70, 95, 96, 194, 196, 235, 236, 239 Bear 41, 235 Bekendam 58, 235 Bargman 235 Bmkhout 12, 235 Bemaudo 239 bi-linear shear criterion 63, 241-243, 245 bias 37, 44, 49, 52, 90, 96, 156, 175 Blenisv.tSki 21, 25, 26, 28, 29, 31-35, 37-40, 44, 69, 110, 154, 156, 157, 1l'i8, 197, 235, 236, 23!l Birgersson 235, 238 Sl.lliO 235 Slake 236 blasting 25, 31, 38 . 40, 42, 53, 56, 78, 79, 84, 91, 113--lH:l, 137, 140, 145, 149, 153, 160, 173, 174, Hl9, 213, 215-218, 220, 222, 224-226, 230, 232 Bray 237 Brekke 69, 70, 236 Broch 236 Brouwer 239 Brown xiv, 17, 27, 36, 38, 110, 154, 157-159, Hl1, 209, 237 Buchanan 239 buckling 87, 89, 92, 97, 100, 101, 106, 176, :243 Bumett 23, 27, 55, 23£ Carr 65, 236, 238 Cervantes 42, 236 Chang 237 channelling 235 Chin 235 Choubey 235 Chryssanthakis 195, 196, 236 Cindarto 164, 236 clay 7, 10, 12, 16, '17, 28, 32, 70, 71, 83, 84, 95, 96, 121, 127, 143, 145, Hl4, 1135, 194, 196, 197, 213, 218, 222, 226, 230, 238, 243, 246 cleavage 6, 1618, 55, 58, 74, 101 . 114, 115, 120, 123, 141, 143, 145, 68" 170, 172, Hl1, 213, 218, 222·226, 230, 242, 243r 245r 246 climata 16, 18, 27' n, 79, 26, 175 C!ipper 50, 253 coating 71 , 95, 95 Da\lis 75, 12.9, 236 dBasem 50 De Goeje 239 De Lange 238 Deere 22, 24, 35-37, 76, 236 deformation 7, 1 -~, 12, 14, 23, 24, 27, 34, 40, 44, 58, 59, 63, 64, 63-66, 69, 95,. 98, 99, 177, 189, 235, 237, :l42 Den Outer 65, 236 Desai 239 discontinuity abutting 39, 62, 147, 241.- 245 alten>tion 92, 94, 155 condition of a single discontinuity 26, 34, 39, 45, 77, 93-WO, 105, 110,130, 131, 137,147, 148, ·151, 152, Hili., Hi6, 1€.5, HlS, 170, Hl3, 197, 203, 214, 219, 223, 229, 231 condition of discontinui.tias 27, 39, 44, 54, 78, 84, 102-105, 107, 110, 113, 121, 123, 125, 126, '130, 131. 1:'!6, 141, 145, 147, 148, 150, 153, 155, 156, 203, 213, 214-220, 222-226, 230·232 dip ·11, 92, 93 .• 96, 99, 100, 131, Hi1·Hi3, 188, 189, 193, 215, 216, 217,220,224, 225, 232,242 inf!!l 6, 7, 1('..-12, 14, 24, 29, 32, 34, 35,37-39,41,45, 62, 63, 69-71, 75, 80, 84, 92-97, 106, 120, 127, 130, 131, 141, 142, 143, 145, 1•H, 148, 153, 155, 164, 165, 167,168-171,176, 193~197, 213,214,218, 2Hl, 222, 223, 226, 230, 231, 239, 240, 247 integral 6, 14, 78, 79, 239, 243 large scale roughness 65-67, 95-97, 141, 142, 165, 169, 110, 171, 187, 193, 195, 196, 247 mechanical ii, v, 3, 4, 6, 9-11, 14, 36, 38, 42, 49, 54, 55, 64, 78, 79, 102, 106, 118, 120, 125, 146, 235, 236, 238, 239, 243-246 non~fitting 67, 78, 79, 141, 145, 148, 165, 170, 213, 214, 218, 219, 222, 223, 226, 230, 231' 243, 244 orientation 27, 37, 74, 93, 128, 242, 245 persistence 27, 29, 34, 35, 38, 44, 62, 84, S5, 96, 141, 145, 147, 213, 218, 222, 226,230, 241, 244, 245 roughness 6, 25, 26, 34, 35, 39, 62-64, 63-68, 71, 74, 75, 78, so, 84, sz, 94·117, no, 131, 141-.143, 145, 147, 148, 152, 153, 155, 165, 169-17"1, 187, 189, 193-197, 213, 214, 218, 219, 222. 223, 226, 230, 231, zas, 236-239, 241, 246, 247 249 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ set 6, 7. 25, 26, 29, 38-40, 43, 71, 74-77, 102-104, 130, 141, 149, 152, 155, 156, 161, 164, 168, 229, 242, 243, 247 single 6, 74, 75, 95, 102, 122, 121, 123, 124, 141, 147, 148, 203, 214. 219, 223, 231, 243 small scale roughness 39, 66, 67, 94-96, 131, 141-143, 146, 147, 152, 155, 1EI9, 170, 187, 193-196,213,218, 222, 226, 230, 247 spacing 12, 22, 23, 31-33, 35-38, 54, 74-77, 79, 80, 84, 101, 113-116, 118, 120.122, 130, 146,148, 150,153, 202, 208, 214-217, 219, 220, 223-225, 231, 232, 242, 246 tactile roughness 88, 67, 141, 248 visible roughness 65, 87 wall 12, 14, 34, 39, 41, 63. 68-71, 92, 94-96, 155, 237 dolomite 10, 17, 125, 140, 145, 146, 148, 150, 153, 160, 164, 173, 181, 213-220, 226 Earle 238 Ebro river 18 Eissa 38, 236, 239 EIHs 237 eotian 17, 18 Equotlp 55, 68, 69, 189, 236, 237, 240 Eriksen 238 Erlikhman 237 Ettenne 239 Evans 239 excavator 84, 113, 118 Exposure Rock Mass (ERM) 88, 89, 147, 247, 248 failure machantems and modes 90, 91, 176, 243 Feirhurst 236 Feleet v, xiv, xv, 4, 15-17, 49, 160, 168, 173, 239 Fermer 239 · reckiir sa;·as:~236 · FJShman 68, 236 fitting 67, 78, 79, 141, 145, 148, 165, 170, 193, 208, 213, 214, 218, 219, 222, 223, 226, 230, 231, 241, 243, 244,253 Fookes 60, 236 formation 16, 18, 71, 84, 120, 140, 145, 181, 213, 218, 222, 226, 230, 243 fractal 65, 238, 238 Franklin 27, 31, 38, 38, 78, 236 friction discontinuity 92, 131, 237 rock mass 107, 109, 110, 125, 147, 148, 150, 151, 153, 214, 215-217, 219, 220, 223-225, 229, 231, 232 •basic 63, 70, 194, 241 Gabrielsen 76, 236 Gama 40, 90, 237 Gazlev 12, 237 Genske xv, 41, 130, 237 geometry 33, 34, 49, 51, 63, 64, 81, 83, 128, 129, 134, 139, 149, .1a4,. ..175, 238 geomorphology 18, 29 geotechnical unit 6, 7, 8-11, 14, 51, 52, 54, 55, 57, 59, 61, 62, 65, 74, 77, 60, 88, 128-130, 139, 141, 146, 147, 149, 161, 175, 176, 229, 243, 246 Giani 11, 98, 100, 126, 130, 194, 237 glaciation 18 gneiss 17, 181 Golser 238, 239 Goodrnan 11, 12, 62, 69, 72, 98, 99, 194,237 gouge 71, 84, 94, 96, 143, 145, 168, 196, 197,213,218, 222, 226, 230, 243 Gourley 236 granodiorite 4, 16, 17, 56, 122, 124, 181, 209, 246 Grima 63, 193, 237 gypsum 4, 16, 17, 95, 177, 181, 194 Haas 238 Hack 1, i, ii, xiv, xv, 42, 68, 70, 92, 96, 236, 237, 239, 253 Hadadou 239 Haines 28-30, 34-36, 40, 42, 44, 126, 154-157, 159, 237 Heksmi 41, 237 Hammersley 90, 237 hand-made 84, 113, 119, 140, 145,149, 153,213,215-218, 220, 222. 224-226, 230, 232 Hanscombe 237 Hencher 165, 237, 239 Hendron 236 Hercynian 16·18 Herda 238 Heyes 235 Hingera 237 Hirschfeld 238 Hoek xiv, 11, 12, 27, 36, 38, 62, 75, 78, 98, 110, 126, 154, 157, 158, 159, 196, 237, 253 Hoek·Brown faHure criterion xiv, 27, 36, 38, 154, 157-159 Holtt 69,237 Homand 239 Howard 236 Hsein 23, 63, 237 Hudson 27, 28, 31, 90, 237, 238 Hutehinson xiv, 237 i-angle 88-68, 193, 241, 247 IAEG 236, 238, 239 ice 34, 42, 45, 79, 80, 175 identificetion 119, 176, 244 ILWIS 253 inhomogeneity 7, 176, 187, 190, 244 isotropy 7, 244 ISFIM 6, 22, 23, 37, 60, 62, 63, 65-69, 76, 193, 194, 196, 235, 236, 237-240 ltasca 235, 236, 240, 253 ITC ii, xiv, xv, 4, 42, 49, 52, 145, 148, 149, 153, 213-220, 222, 223-226, 230.232, 236, 237, 239, 253, 254 James 238 Janbu 170, 238 JCS 235 JRC 65, 66, 194, 235 Kaashoek xv, 236 ······"·"~~t Si},:<'QJ~_!!!, 71,1:l(), JM,J!4::J~7, 1Q8.1l8. 130. 1;H. 1~1~1~1~1~1~1~1·1~1~ 193,194-197,207,208,213,214,218,219,222,223, 226, 230, 231, 247 kinematic stability 101, 188-170, 174 Kirksldie 235 Kiraten 31, 238 Kitano 238 KNGMG 238, 241 Kouoksm 239 Kovliri 27, 238 Lajtai 238 Lama 69,238 landuse 126 Laubscher 26, 29, 30, 34-36, 38-40, 42, 60, 82, 65-72, 76-79, 116, 118, 119, 124-126. 154. 155,238 Lauffer 24, 28, 35, 238 Lee 65, 90, 237, 238 Lamas 237 lien 235 limestone 4, 10, 17, 93, 95, 115, 125, 140, 145, 146, 148, 150, Hi3, 150, 164, 173, 1&1, 190, 195, 2lS..220, 226, 246 lithostratigraphic unit 18, 24, 36, 115, 246 loeas 16 leset 235 Louis 27, 235, 236, 238 lumsden 235, 239 lunde 235 lustre 65, 244 Manolopoulou 239 mapping xiv, 16, 55, 119, 181, 236, 239, 243, 244, 246, 254 mapping unit 244, 246 Maravic 237 Mardia 75, 238 Marquardt 109, 238 Marsity 235 Masure 236 Maurenbrecher 74, 238 Mazzoccola 31, 238 Mcmahon 195, 238 method of excavation v, 4, 28, 35, 42, 44, 78-81, 84, 87-89, 113, 114-122. 124, 128, 137, 139, 140, 145-150, 153, 154, 155-157, 173-175, 183, 202, 213-220, 222-226, 230-232, 247 Mohr 12, 106, 107, 157, 158, 166, 188, 242-244 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ INDEX Mohr-Coulomb failure criterion 12, 106, 107, 157, 158, 166, roughness 6, 25, 26, 34, 35, 39, 62-64, 63-69, 71, 74, 75, 78, 188, 242-244 Molokov 240 80, 84, 92, 94-97, 130, 131, 141-143, 145, 147, 148, 152, 153, 155, 165, 169-171, 187, 189, 193-197, 213, 214, 218, 219, 222,223, 226, 230, 231, 235-239, 241, 246, 247 large scale 13, 18, 65-67, 84, 95-97, 141, 142, 145, 147, 148, 165, 169-171, 174, 176, 187, 193, 195, 196, 213, 214, 218, 219, 222, 223, 226, 230, 231, 236, 247 small scale 39, se. 67, 84,94-96,131,141-143, 145,147, Monte Carlo simulation 90, 105, 117, 121, 129, 131, 137, 237 Moreno 235, 238 Moye 22, 238 MRMR 26, 29, 30, 35, 118, 125, 154, 155 Mulder xv, 240 Milller 27, 238 Milralha 90, 197, 238 Nathanail 31, 238 NATM 21, 27, 35, 239 Naretnieks 41, 235, 238 neural network 28, 90, 238 non-fitting 67, 78, 79, 141, 145, 148, 165, 170, 213, 214, 1-1~1~1a1•1~1~1~1D1K~~ 218,219,222,223,226,230,231,243,244 observer bias 52, 156 Ohikere 236 . Ohnishi 195, 238 Oliveira 239 optimization 30, 90, 102, 104, 105, 108-110, 112, 117, 121, 135, 239, 244, 245, 253, 254 orogeny 16-18 outlier 109, 245 overfit 90, 245 Pachar 21, 27, 238 Palmstrem 38, 155, 156, 238 Papaliangas 196, 197, 239 Patton 63, 236, 239, 241 Pareira 194, 196, 197, 239 periglacial 18 permeability 7, 14, 41, 58, 127, 153, 166 Phllm'~;197; 239 Phillips 74, 239 Pinto 236, 238 planning 240 pneumatic hammer 113, 117-119, 137, 140, 145, 149, 153, 213, 215-218, 220, 222, 224-226, 230, 232 Point Load Strength 27, 31, 36 Pool xv, 68, 239 porosity 7, 14, 58, 153 porphyritic. porphyrite 245 Poulos 238 pre·splitting 78, 79, 113, 114, 116, 117, 119, 137, 140, 145, 149, 153, 213, 215-218, 220, 222, 224-226, 230, 232 Preston 239 Price xiv, xv, 6, 13, 14, 16, 23, 68, 76, 235, 237, 239 Proctos 240 Profilograph 236 ~· 25, 28, 33, 35, 38. 40$ a5~. 96, 1S4, 196, 235 Rabcewicz 27, 238, 239 Rantucci 239 Rao 90, 239 Rasmussen 41, 238, 239 RCD 147, 148, 150, 153, 214-217, 219, 220, 223-225, 231, 232, 247 RCOH 147, 148, 161, 214, 219, 223, 231, 247 Reference Rock Mass (RRMI v, 4, 72, 81, 88, 89, 139, 146-150, 152, 153, 160, 161, 164, 169, 214-217, 219, 220, 223-225, 229, 231, 232, 247 Rangers xv, 63, 65, 67, 68, 141, 193, 194, 236, 239 RES 21, 27, 31, 240 RFRI 147, 148, 161, 214, 219, 223, 231, 247 Richards 237 RIRS 146-148, 150, 153, 214-217, 219, 220, 223-225, 231, 232, 247 RI 84, 92, 141, 142, 145, 147, 148, 195, 213, 214, 218, 219, 222. 223, 226, 230, 231, 247 RMR 25, 26, 28, 29, 32, 33, 35, 38-40, 54, 110, 154, 156, 157, 158, 159, 197 Robertson 28, 29, 32,34,35,39, 40, 54,239 Rocha 239 Rode 68,239 Romana 28, 29, 32, 35, 39, 40, 42, 78, 118, 119, 154, 156, 157, 239 Rose 239 Rosenbaum 78, 239 Rossmanith 238 · 214, 218, 219, 222, 223, 226, 230, 231, 247 tactile 66, 67, 73, 141, 246 visibie 65, 67 RQD 24, 25, 35-38, 44, 76, 155, 156, 175, 236, 239 Rs 84, 92, 94, 141, 142, 145, 147, 148, 193, 213, 214, 218, 219, 222, 223, 226, 230, 231, 247 RSA 25 RSPA 14&-148, 150, 153, 214-217, 219, 220, 223-225, 231, 232, 247 RSR 25, 35, 240 RTC 147, 148, 151-153, 161, 214-217, 219, 220, 223-225, 231, 232, 247 Rudledge 32, 239 safety factor 29, 35, 154-156, 170, 171, 236, 237 sand 13, 16-18, 194, 196, 197,246 sandstone 4, 17, 58, 110, 181,246 Sarma 170, 171, 239 scale effects 63, 69, 195, 235, 236, 238 scanline 38, 75, 76, 141 Scavia 90, 239 Schmklt hammer 29, 30, 36, 55, 68 Schneider 12, 239 SCOft··tso;151; 153,.215"217;220; 224',225, 232, 247 seepage 41, 42, 72, 79, 84, 127 Se~y 28, 29, 35, 36, 38,60, 62,239 San 24,38, 236,239 Serafim 157, 197, 239 SFRI 150, 151, 153, 215-217, 220, 224, 225, 232, 247 Shab 237 shearbox 63, 68, 141, 164-166, 169-171, 175, 176, 193, 195 Shi 237 Shrestha 239 Shuk 28, 30,40, 240 siltstone 16, 17, 181 simple means 36, 55, 56, 58, 59, 75, 80 Skagius 238 Skinner 240 slate 17, 58, 59, 118, 168, 181, 193, 195, 222-225 slickensided 27, 245 . alidine u. 13, 28, 32.. 3&, 40. 56. 63, 66, 75, 63, 87, 89, 91, 92, 93-102, 106, 110, 126-128, 13Q-133, 141, 149, 152,153,161,163-171,173, 174,176,182, 191,193, 194, 195-197, 215-217, 220, 224, 225, 232, 234, 243, 245 sliding criterion 87, 92-101, 106, 110, 126, 127, 132, 133, 141, 165, 166, 169-171, 176, 191, 193-197 slope dip 29, 30, 51, 52, 79, 92. 93, 96, 100, 99-102, 107, 109, 131,149,151-157,161,164,169,171,173, 174, 176, 187,188,189, 193,215-217,22~224,225, 232,242, 247 failure 54, 89, 92, 98, 100, 107, 126-128, 176, 196, 243 height 29, 35, 40, 41, 51, 52, 101, 102, 106, 107, 129, 130,135,151, 153,154, 160,161, 177,215-217,220, 224,225,232 orientation 28, 31, 35, 39, 40, 87, 90, 93, 100, 156 slope stability rating 83 Slope Rock Mass !SRMl 88, 89, 149, 151, 247, 248 SME 149, 150, 153, 215-217, 220, 224, 225, 232, 236, 239, 247 smooth wall blasting 78, 79, 84, 114, 116, 117, 137, 140, 145, 149, 153, 213, 215-218, 220, 222, 224-226, 230, 232 snow 34, 42. 45, 63, 79, 80, 169, 171, 172, 175 Soukatchoff 239 Spain v, xiv, xv, 4, 15, 239 SRM 88, 89, 156 SSPA 150, 153, 215-217, 220, 224, 225, 232, 247 SSPCCLAS 50, 160, 253 STC 151-153, 215-217, 220, 224, 225, 232, 247 251 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 252 Stephansson 235, 236, 239 stereo projection 75, 169, 221, 253 Sterling 238 Stewart 238 Stimpson 68, 240 strength intact rock strength 7, 11-14, 23, 26, 27, 29, 30, 32-36, 40, 42, 44, 54, 55, 54-59, 58, 59, 68, 69, 78, 80, 84, 87, 92, 94, 102, 107-109, 113, 115, 118, 120, 122, 121, 1~1a1a1a1~1a1~1•1~1~1~ 1-1~1~1·1~1~1~1~1·1~ 197, 201, 213-220, 222-226, 230-232, 245, 247 point load 27, 31, 36 shear 6, 7, 10-14, 25, 28, 32, 34, 39-42, 54, 60, 62, 63, 65, 66-74, 77-80, 92, 94-99, 148, 164, 167, 171, 176, 188, 189, 196, 214, 219, 223,231,235-241,243-245 tensile 3, 6, 7, 13, 39, 59, 166, 243, 244, 246 triaxial compressive strength 242, 246 true triaxial compressive strength 242 Unconfined Compressive Strength 22, 23, 36, 55-59, 58, 59, 68, 69, 130, 164, 169, 175, 176, 188-190, 242, 244, 246 stress 11-14, 24-26, 33-35, 40-42, 44, 45, 58, 59, 63, 64, 70, 76, 78-80, 95, 98, 126, 154, 157-159, 166, 172, 187, 188, 189, 190, 196, 237, 241, 242, 244-246 stress relief 13, 14, 34, 40, 41, 44, 45, 76, 79, 80, 189 striated 245, 246 stylolite 246 susceptibility to weathering 14, 34, 40, 44, 54, 60, 61, 80, 83, 87, 125, 126, 139, 144-146, 175, 213, 218,222,226, 230, 246, 247 SW 84, 144, 145, 188, 213, 218, 222, 226, 230, 247 SWE 146, 149"153. 215"217,220~224, 225,. 232.248 Swindells 78, 240 Tiihtinen 238 TC 92-98, 100, 99, 100, 104, 120, 121, 130-133, 137, 147, 148, 155, 156, 165, 169, 170, 182, 193, 197, 203, 214, 219, 223, 231, 248 Terbrugge 237 Terzaghi 24, 25, 35, 37, 75, 240 Tiedemann 240 tilt 141, 142, 170, 171, 193-196 topography 15, 16, 164, 169, 173 toppling 40, 87, 89, 91, 92, 97-102, 106, 128, 130-134, 141, 149, 152, 153,161,168,171,173, 174, 182,215-217, 220, 224, 225, 232, 234, 237, 239, 243 toppling criterion 87, 92, 98-101, 133 triaxial 36, 188, 237, 242, 246 Tsang 235 Tulinov 69, 70, 240 tunnel 7, 8, 21, 24, 25, 27, 28, 33, 34, 39, 41, 45, 62, 74, 235, 236, 239, 240 UDEC 99, 160, 164, 166, 167, 171, 172, 235, 240, 253 underground excavations xiv, 4, 21, 25, 26, 28, 31, 32, 34, 35, 41, 42, 45, 54, 60, 66, 71, 76-78, 90, 103, 118, 125, 235, 237-240, 254 unit weight 108, 153, 187, 188 URCS 22,23,36 Vecchia 28, 29, 34, 35, 240 vegetation 13, 15, 16, 80, 125 Verwaal xv, 68, 237, 240 Visser 238 visually estimated slope stability 51, 52, 102, 109, 128 Wlllton 236 water xiv, 10, 11, 13, 14, 24-26, 29, 34, 35, 40-45, 58, 63, 68, 69, 70-72, 71, 72, 79-81, 84, 89, 106, 126, 127, 144, 155, 156, 159, 165, 166, 165-167, 171-176, 195-197, 208,209,238,239,244,246 pressure 10, 11, 26, 34, 40-42, 45, 71, 126, 127, 166, 176, 196, 244 weathering rock mass 56, 60, 78, 87, 113, 120, 122, 124-126, 128, 137, 140, 146, 147, 149, 150, 152, 173-175, 201-204, 207-209, 239 Weaver 31, 42, 240 wedge 29, 89, 91, 106, 169, 170, 238, 243 Wej 69, 120, 197, 239 Welsh 70, 240 White 10, 17, 140, 145, 181,213,218,226,240 ~ckham 25,35,42,240 239 ~lliamson 22, 23, 240 ~lson 235 ~ttke 237-239 Wood 237 Yoshinaka 238 Yufu 32, 240 ~lkinson Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ INDEX 253 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 254 FIGURES Fig. fig. fig. Fig. Fig. 1. Intact rock vs rock mass. 7 2. Anisotropic rock mass. 3. The influence of discontinuities on the stability of a tunnel in the progress of construction (after Arnold et al., 1972). 4. Rock mass components. 5. Different geotechnical units in a single slope. Greenish and blueish grey layers consist of calcareous shale and brownish, pinkish off- 7 8 9 white layers consist of dolomite and limestone. Fig. 6. Section through the slope of Fig. 5. Fig. 7. Block on slope with and without watar pressure lW is the weight of the block, cohesion along discontinuities is zero). Fig. 8. Stress distribution (bulbs of pressure - lines of equal major principal stress) in a rock mass due to a vertically oriented plane load (after Gaziev et al., 1971). Fig. 9. Example of a cyclic plate-bearing test on fractured rock (after Schneider, 1967). Fig. 10. Dint• ,..,;D.,... vs discontinuity spacing for plate diameter 8 cm on a model rock mass (after Berkhout, 1985). Fig. 11. Rock mass under stress. Fig. 12. Research area. ~- ·Rg: ·13. Slock·sizinind·fofm ·aasci"iptionaccoraffigto BritlsflStanaarcnBS-5930; 19!!1fwith ratios for iiTock form .. !Pfice;-1-9921. Fig. 14. Terzaghi- rock load classification (after K. Terzaghi, 19461. Fig. 15. Design chart to determine slope dip and height using MRMR classification data (after Haines et al., 1991}. Fig. 16. Correlation between Bieniawski IRMRl and Barton (Q). Data from case histories with RMR and a-system (after Bieniawski, 1989). (Continuous lines indicate correlating classes of rock mass quality.) Fig. 17. Bias of RQD due to orientation of borehole. Fig. 18. Influence of discontinuity condition. lt is not clear which discontinuity set has the worst influence on the stability of the tunnel. Fig. 19. Standards for the geometry of a slope. Fig. 20. Estimated intact rock strength vs strength values determined by UCS tests. !The dashed lines in A and C indicate the relation if estimated strength equals UCS strength.) (Number of UCS tests: 941) Fig. 21. Average estimated intact rock strength vs average UCS for granodiorite units with various degrees of rock mass weathering. Fig. 22. Percentage of UCS test values falling in a range different from the estimated range value. 10 10 11 12 12 12 13 _15~ 23 24 30 33 37 39 51 57 58 58 Fig. 23. Ratio of average intact rock strength perpendicular over average intact rock strength parallel for UCS and field intact rock strength estimate per unit (values in brackets are the numbers of UCS tests respectively estimates). Fig. 24. Influence of roughness on displacement without shearing through asperities (left figure: unconfined; right figure: confined). Fig. 25. Displacement of block (shearing through asperities and deformation). HQ. 26. RougbMII 4aWm ~ for &iAole JMok ~ M4 -41eek ~etell.., V8l'ticel diseantlnuity trlghtt. Fig. 27. Parallel roughness profiles of one discontinuity plane. Spacing between profiles" 1.5 cm (after Baardman, 1993). Fig. 28. Interpretation of regular forms of roughness as function of scale and angle. Fig. 29. Equotip rebound valye$ on We,llthered d.i$COntinuity walls progr.e.ssively ground down to fresh rock (after Hack et al., 1993a). Fig. 30. UCS vs Equotip (after Verwaal et al., 1993). Fig. 31. New slopes in different conditions with water table. Fig. 32. Geological and structural geological analyses to obtain discontinuity properties. Fig. 33. Discontinuity spacing factors (after Taylor, 1980). Fig. 34. Flow diagram of the concept of the 'initial point rating' system. Fig. 35. Results of 'initial point rating' system with optimum weight factors based on 250 slopes (Definition of visually estimated stability classes - Table 5, page 52). Fig. 36. Initial point rating - exposure classification form. Fig. 37. Sketch of exposures with various degrees of weathering, diftarent types of excavation and showing the concept of the 'reference rock mass'. Fig. 38. Parameters in the slope stability probability classification (SSPC). Fig. 39. Different failure mechanisms making a single slope unstable. Over the whole slope relatively small sliding, toppling and transport of rock blocks during rain and a 'wedge' sliding failure in the middle. Fig. 40. Poor blasting, weathering and surface (rill) erosion making a single slope unstable. This slope is discussed in more detail as example IV in eh. D.5.4. Fig. 41. Discontinuity condition parameter (TC) vs p, for 'day-lighting' discontinuities in stable slopes (stability class 1, Table 5, page 52). Fig. 42. TC without the karst parameter in the calculation of TC vs p for 'day-lighting' discontinuities in stable slopes for different rock materials (values in-between brackets are average estimated intact rock strength). Fig. 43. Discontinuity condition parameter I TC) vs p for 'day-lighting' discontinuiti&s in stable and unstable slopes !visually 11stimated stability class 1, 4 & 5). Fig. 44. Discontinuity condition parameter (7C) with refinements vs p for 'day-lighting' discontinuities in stable and unstable slopes (visually estimated stabHity classes 1, 4 & 5). Fig. 45. Toppling. Blocks on the surface of the slope are pushed out due to the forces of the rotating blocks behind. lnterlayer slip and deformation or crushing of block corners govern the rotation of the blocks (free after Goodman, 1 989). 59 64 64 64 65 67 68 68 72 74 77 82 82 84 88 89 91 91 93 94 97 97 98 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ jigums, tables &: COTYfJUJer programmes Fig. 46. TC vs lP for dlscontinuities dipping opposite slope dip direction in visually estimated stability class 1 (stable) and 4 & 5 slopes (unstable); boundary condition refers to eq. !171. Fig. 47. Flexural buckling failure (laysrs flex under the load of the rock above) (free after Giani, 1992). Fig. 48. Frequency distribution of ir.s, $pll- and con_. Fig. 49. P8roenteges incorrectly calculated slope stebilities with linear model (for calculation of mean values and standard error see eh. D.2.3.11. F~g. 50. Shear plane model for dip- > ,_. Fig. 51. H,../H- vs fl_,ldip- (for the graph H,../H- has a maximum value of 100, and H,../H- .. 1 for tp,....ldip- :a: 11. Fig. 52. Exampiea of average discontinuity spacing per lithostratigraphic sub-unit and type of discontinuity vs the method of excavation in slightly weathered expoaures. Fig. 53. The average sp11- of a series of exposures in a sub-unit or unit excavated with a particular method may depend more on the absolute IIPII,_ rather than on the influence of the excavation method. Fig. 54. Values for the parameter for the method of excavation compared to values from Laubscher (19901 and Romana (19911 (lines inbetween data points have no meaning, and serve only for identification). Fig. 55. Overview of the influence of weathering on different geotechnical parameters. Fig. 56. Weathering parameters vs degree of rock maea weathering (refer for the rock mass adjustments following Laubscher to llible 7, page~. riiJ. . 57. Example of the distribution of irs estimates (made in exposure: 95/stu/2). Fig. 58. Example of the distribution of one sample of TC. Fig. 59. Sketch showing the procedure to calculate the boundary line for the 'sliding criterion' for X = 2 (e.g. boundary line based on 2 data points). Fig. 60. Mean and standard error of intercept and slope of boundary lines vs X, for 'sliding criterion'. Fig. 61. Sliding probability for orientation dependent slope stability. Fig. 62. Mean and standard error of intercept and slope of boundary lines vs X, for 'toppling criterion'. Fig. 63. Toppling probability for orientation dependent slope stability. F~g. 64. Mean value and standard error for factor 110 in shear plane model vs number of optimization&. Fig. 65. Distribution of 110 after 72 optimizaticns. Fig. 66. Example of distributions for the calculation of lines of equal probability for orientation independent stability for the shear plane model. Fig. 67. Probability of orientation independent slope stability. Values indicate the probability of a slope to be stable. Fig. 68. Exampla of the distribution of one data aet of,_. Fig. 69. Large scala roughness profiles used for the slope stability probability classification ISSPC). ""'119~ 1(5: $iil811 sclile roughness liSia fOi tfii Slope StabilitY prOI:iability classification !S!PCI. Fig. 71. Exposure characterization (example I, old road cut exposure A, see eh. 0.5.1). Fig. 72. Reference rock mass calculation (example I, old road cut exposure A, see eh. 0.5.11. Fig. 73. Example of future orientation independent slope stability as function of the future degree of weathering. Fig. 74. Slope stability probability calculation (example I, old road cut eXI)osure A, see eh. 0.5.1). Fig. 75. Comparison of slope stability asseasments by different classification systems. a: SSPC system, b: Haines' slope stability classification and c: Romana's SMR rating. Fig. 76. Comparison of total major principal stress values at failura; left: RMR vs SSPC; right: 'modified Hoek-Brown failure criterion' vs SSPC. Fig. 77. Example I. Location sketch. Fig. 78. Example I. More than 40 year old road cut A (blasted height about 8 metres from road level). Fig. 79. Example I. Natural exposure B along old road. The natural exposure starts at about 2 m from road level and is partiy overgrown. The lower part of the exposure is blastad. NOte the small gunpowder blastholes in the lower part. Fig. 80. Example 1. New road cut C (bedding dips about 6°; gradient of road to the left causes a seemingly larger dip of the bedding to the right). Blast holes are clearly visible at the left. Fig. 81. Example 11. The sliding piane is clearly visible on the right; left side still standing part of the road cut (scale: road lining about 0. 1 m wide). · ·f'itt•·D2¥Exampie tt. Geometrlcal-·~·-·1heslope lin the direotiofHJf.·1he•·'Ofthebeddingend~·~2"h· Fig. 83. Example 11. Limiting-equilibrium analysis. Fig. 84: Example 11. The friction angle as function of block langth and the height of the water in the second joint set (337/48). Fig. 85: Example 11. UDEC simu.lation •. EniiUII.ed part of the toe of the slope showing displacement. velocitv.and xy-stresaes along sliding plane. Fig. 86. Example The slope in April 1995 after the main failure of April 1992 and the partial failure of the top part of the slope (the terrace on the left is the old road). Fig. 87: Exampie Ill. Geometrical cross section of the slope. Situation in april 1992 after the main failure occurred (section in direction 018°). Fig. 88. Example Ill. Definition of inclination angles for the internal discontinuities. Fig. 89: Example Ill. The factor of safaty as function of the inclination angle of the internal discontinuities. Friction angle+ "' 43.5° for base and side friction; calculated by Sarma's method. Fig. 90. Example Ul. UDEC simulation. Displacement, velocity vectors and xy-stresses in the slope. Fig. 91. Time estimates for the stability calculation of a 15 m high slope. Fig. A 92. Cross-section of a slope with one step on a discontinuity plane. Fig. A 93. Width of step (sw) necessary for equilibrium, vs dip of discontinuity IPl for various slope dips lolrl. Fig. A 94. vs height of step lshl. Fig. A 95. (a), and tilt angle, (b), and shearbox friction angle vs small scale roughness parameter (roughness parameter values see Fig. 71, artificial samples: Grime, 1994). Fig. A 96. ,_values following 'sliding criterion', Barton 11973al and Giani 11992). Fig. A 97. Small scale roughness literature values (Barton et al., 1990, Goodman, 1970, Rangers, 1971, P&reira, 1990) and 'slidinQ criterion' vs surface description. Fig. A 98. Friction vs large & small scale roughness and literature tilt test values of Chryssanthakis et al. (1990). Fig. A 99. Frioticn angle vs infill material (vslues fmm.Jioek.etat, 1981, vertical linea from Barton, 1988). Fig. A 100. Friction angle vs infiil material compared to infill thickness laboratory tests. Papaliangas et al. ( 1990) tests with straight, rough undulating surfaces and Pereira (1990) with straight, polished planar sample surfaces. Fig. A 101. Examples of avsrage intact rock strength (field estimate) vs degree of rock mass weathering per lithostratigraphic (sub-) unit. Fig. A 102. Examples of average discontinuity spacing vs degree of rock mass weathering per lithostratigraphic (sub-) unit and per type of discontinuity. m. ucs...- 100 100 103 106 107 111 114 115 119 122 124 130 131 132 132 133 133 134 135 135 135 136 136 142 14!f 145 148 152 153 155 158 161 162 162 163 163 1iM. 165 166 167 168 168 170 170 172 177 187 188 189 193 194 194 195 196 197 201 202 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Fig. A 103. Examples of averaQellfJII,_ vs degree of rock mass ~ring per lithostratigrephic (sub-) unit (spa,.... corrected for method of excavation). Fig. A 104. Examples of the awraQe condition of discontinuity pan~meter In:! of a single discontinuity (aetl vs degree of rock mass weathering per llthostratigraphic (sub-) unit and per type of discontinuity. Fig. A 105. Examples of the averaQO OWI'IIII condition of discontinuitiea Icon_! vs dag111e of rock mass weathering per lithostratigraphic (sub-) unit. Fig. A 106. Examples of average coh,_ vs degree of rook maas weathering par lithostrstigrephic (sub-) unit. Fig. A 107. Examples of the averaQO .,_vs degree of rook mass weathering per lithostretigraphic (sub-) unit. Fig. A 108. Example I. Natural ~ure B. Exposure ~cterlzatlon. Fig. A 109. Example 1. Natural exposure Refererw:e rock mass clllcuiation. Fig. A 110. ExamPle I. Natural exposure B. Slope stebllitv probability calculation. Fig. A 111. Example I. lllaw rollld cut C. design slope dip 85°. Slope stability probability calculation. Fig. A 112. Example I. lllaw roiiJd cut C. dellgn llope dip 70°. Slope stebllity probability calculation. Fig. A 113. Example 11. Exposure oharacteliution. Fig. A 114. Example 11. Reference rock mass celculetion. Fig. A 115. Example 11. Slope steblllty J)I'Dbabllity calculation. Fig. A 116. Example Ill. Stereo projection. a: poles; b and c: contoure of poles and great circles of planes. Indicated orierrtations are dip- a vectore. 202 203 203 204 204 213 214 215 216 217 218 219 220 221 Fig. A 117. Example Ill. Exposure cheracteriation. Fig. A 118. Example lit. Reference rock mass clllculation. Fig. A 119. Example Ill. Slope stability probability calculation before failure. Fig. A 120. ExamPle Ill. Slope stability probability calculetion after failure. F19. A 121. Example IV. Exposure characterization. Fig. A 122. Exposure charecterization. Fig. A 123. Reference rock mess calculation. Fig. A 124. Slope stability probability calculetion. Fig. A 125. Probability of orlentetion independent slope stability. Values indicate the probability of a slope to be stable. Fig. A 126. Sliding probability for orientation dependent slope stability. Fig. A 127. lbppling probability for orlentetion dependent stope stability. Fig. G 128. 'Bl-lineer shear criterion' for a discontinuity with a regular set of triangular shaped asperities (modified after Patton, 19661. Fig. G 129. Discontinuity spacing. Fig. G 130. Compreellve strength. fig. ~~Mohr..coulomb . faih:mHiriterion.. " Fig. G 132. Persistent, non-persistent and abutting discontinulties. 222 223 224 225 226 230 231 232 233 234 234 241 242 242 m244 245 TABLES "'able 1. Geological table and description and main engineering charecteristics of the lithology of the Falset area. 'labia 2. Characterization of intact rock strength according to BS 5930 (19811, ISRM (1981b) and URCS (1980). Table 3. Rock mass parameters of interest for engineering structures in or on rock. Table 4. Parameters and their influence in existing classification systams. Table 5. Stendards for the visual estimation of slope stability and the number of slopes per stability class. lllble 6. Estimation of intect rock strength. lllble 7. Adjustment values for susceptibility to weathering for classification of stability of underground excavations in mining (after Laubscher, 1990). 'lilble xaJustman'tii'for metlloif of exeavatfon !'attar Laii.ilisC:l11ii.l9Mf:"" lllble 9. Factore for linear model with s,.,_ following 18ylor and weighted con_ (for calculation see eh. 0.2.3.1 ). lllble 10. Factore for the shear plane model and percentegSS of slopes with a calculated stability that conflict with the visually estimated · stebility lforeeleUistton·ses eh. D;2.3;2). Table 11. Initial classes for the method of excavation. Table 12. Parameter for the method of excavation (ME) (for calculation see eh. 0.2.4.1 ). lllble 13. Values for the parameter for weathering. lllble 14. Values for the deg111e of weathering for a single discontinuity (set) and for a rock mass as used in the SSPC system (for calculation see eh. 0.2.4.21. lllble 15. Distributions of field and derived pare meters (numbere in breckets refer to the notes in the text). lilble 16. Comparison of slope stebllity classification systems. lllble A 17. Formations, lithostrstigrephic units and sub-units. Table A 18. Coefficients for polynomials of equal probability for sliding and toppling criteria (correlation coefficients > 0.999) ISSPC). lilble A 19. Lines of equal probability for orientetion independent slope stability ISSPC). Table A 20. Degrees of rock mass weathering- BS 5930 (1981). Table A 21. Classification approaches (Anon., 1995). Table A 22. Proposal for correlation of the degree of rock mass weathering following BS 5930 ( 1981) and quantitative weathering with the proposal of the Engineering Group of the Geological Society (Anon., 19951. s: 17 23 34 35 52 56 §0 18 105 112 113 119 123 124 130 156 181 182 183 207 207 209 Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ Manuhc'iurer AppUcation Clipper Computer Allsoci;:;tes tntematiorml, !ne., One Computer Associates Plaza, lsiandia, New York 1 788-7000, USA. Obasa !11 ?Jus & IV Bor!and, 100 Sor!and Way, Scotts California 95056-3249, USA. \f<.~liey, Compiler land programming language) for caicu~ation and database prcgrommas for slope stability probability classification. Relational database::; for field and classifica- tion data. DiPS 2.0 E. Hoak & M. Diederichs, Rock Enginearing Group, Dtlpt. Civil Engineering, Univ&rsity of Toronto, Canada. iLWIS, version 1.4 iTC (lll't. Ins!, f9r Aer()l\lP!lol!: S!.!rvmr 1.H1d Earth Sciences}, P.O. Box 6, 7500 AA Enschede, The Netherlands. Mathciid for Wtndows, version 5.0 MathSoft !ne. '101 Main Strest, Cambridge, Massachusetts, 02142 USA. Optimization of linear and non-linear tunc· tions in orientation independent slope stability and probability anelyses. SiidaWrite Pius, version 2.0 Advanced Grn!)hlcs Soitware Inc. 5825 Avenkla Em::inas, Suite 105, Car!sbad, CA Curve fitting for orientation dependent and independent slope stability. m Stereo projection and contour piote for example m. 92008·4404, USA. SSPCCLAS UDEC, version 1 .8 H.R.G.K. Hack, Section Engineering Geology, !TC !lnt.!nst. for Aerospace Survey and Earth Sciences), Kanaalweg 3, 2628 E!'J De!ft, The Netherlands. Calculation programma for SSPC system \written in Clipper). Manugistics Inc., 2115 Er;st .Jefferson St., Rockvillll, Maryland, 20852, USA. Histograms and norma! distributions for probability <~~naiyses. UDEC, Universal Distinct Element Code. Distinct element analyses for examples !! and n:lillSCA Coosulting group. Minneapolis, m. Mirm!l1llotlil, USfo•. Wordl"arfect, version 5.1 + Coral Corporation. Coral Building, 1600 for DOS Text and layout. Car!ing Avem.le, Ottawa, Ontario, K1Z 8!17, Canada. WordPerfect Presentations, version DOS :to tor idem Figures. Hack, R., 1998. Slope stability probability classification; SSPC; 2nd version. Price, D.G., Rengers, N. (Advs). PhD thesis, University of Technology Delft; International Institute for Aerospace Survey and Earth Sciences; ITC, Delft, Enschede, The Netherlands. (43). ISBN: 90 6164 154 3. p. 275. http://repository.tudelft.nl/ 258 CURRICULUM VITAE The author is employed by ne in Delft in the section Engineering Geology since 1990. The author is a staff member of the research school 'Centre for Technical Geoscience' and is. president of the Dutch Association of Engineering Geologists (fugeokring). The author is a member of the Institution of Mining and Metallurgy and a Chartered Engineer. Besides the research done for the SSPC the author is also involved in research projects for engineering geological mapping, detailed seismic wave behaviour around discontinuities in a rock mass, and in research towards the optimizing of the use of three-dimensional geographical information systems (GIS) and knowledge base systems for engineering geology. The author woriaira.s engirieemig geriiogisfiilid geoteeliiiiciil engiiieer :Ui the Middle &St, Far &Si ind Africa. Several years he worked as senior rock mechanics engineer in the underground copper mines of Zambia. The author studied geology at the University of Leiden, The Netherlands, (BSc. 1979) and engineering geology with exploration geophysics at the Technical University Delft and the University of Utrecht, The Netherlands (MSc. 1983). He obtained the Doctor degree at the Technical University Delft in 1996.