Journal of Philippine Development II_

I1_ NumberTwenty-Six,VolumeXV, No. 1,1988

SOIL AND WATER CONSERVATION PLANNING:

POLICY ISSUES AND RECOMMENDATIONS*

Wilfredo P. David*"

I. Introduction

Soilandwaterarebasicresources.Theirexploitationordevelopment
is a matterof survivalfor many, a_ escape frompovertyfor most,and an
opportunitytopursueadditionalpower,wealthand selfishinterestforsome.
The overexploitationof these basic resourcesarises notfrom igno-
rancebutmostlyfroman instinctforself-preservation.Forestsarecutdown,
thintopsoilsand fragileaquaticresourcesaredepletedand marginallands
are overgrazedin orderto meetshort-termneedsforfood,energy,clothing
and shelter. However, viewed in the wider and long-termcontext, the
consequencesof suchactions are disastrous.
For the past fifteen years, forested areas have been decreasing
annuallybyabout180,000hectaressothatpresently,we havelessthanfive
millionhectaresofforestlands.The hydrologyand productivecapabilityof
about a thirdof our total landarea have been impairedby excessivesoil
erosion.
Excessivesoilerosion,resultingfromthe manipulationof our water-
sheds, makes upland farmers more reliant on chemical fertilizers for
sustainedyield. In the extreme,hillsidefarmersconstantlymovefromone
area to anotherintheirsearchfor morefertilesoils. Excessivesoilerosion
also pollutesstreamsand rivers. The sedimentdischargesof Philippine
riverswhosecatchmentsaresubjectto uncontrolledmanipulationsexceed
30 tonsperhectareyear(David,1986). Reservoirsandpondsusedforflood

"Paperpresentedat the Workshopon UplandResourcePolicyProgram

sponsoredbythePhilippineInstitute
forDevelopmentStudies(PIDS),Department
ofEnvironmentandNaturalResources (DENR),University
ofthePhilippines
atLos
BaRos(UPLB)andInternational Development ResearchCenter(IDRC),March14,
1988at PIDS,Makati.
"*Professor
andDean,CollegeofEngineering ;_ndAgro-lndustrial
Technol-
ogy,University
ofthePhilippinesat LosBaRos,College,Laguna.
48 JOURNAL OF PHILIPPINE DEVELOPMENT

control, water supplyand powergenerationare reducedin waterqualities

and storagecapacities(e.g., Binga andAmbuklaoreservoirs).Sedimenta-
tionthreatensthestabilityof aquaticecosystemssuchas lakes,mangroves
and marshes.
The primarycause -- andeffect-- ofthemore seriousenvironmental
problemsispoverty. Poverty,environmentaldegradation,skeweddistribu-
tionofwealthand powerand selfishpursuitof sectoralinterestsare simply
differentaspectsofthesamesetofproblemsassociatedwithenvironmental
degradationand resources'development (Ramphal, 1987).
The generalmeansforpreservingPhilippinelandscapesand protect-
ingtheirlandandwaterresourcesaresoundpoliciesand programstowards
theallocationand utilizationofsuchresourcesarejustenoughto implement
thesepolicies.Such policiesand programsshouldbe ofa naturewhereby:
(1) growthor developmentrespectsthe limitsto environmentalresources;
(2) thebasicneedsofthegrowingpopulationfor food,energy,clothingand
shelter are met; and (3) economic mechanismsreward the good and
penalizethe bad users. Relieffrompopulationpressurespartlyimpliesthe
developmentand applicationof suitabletechnologiesfor increasedand
sustainedagriculturalproductionin uplandsand hil/ lands.

II. Soil Erosion Under Various Conditions In

Selected Watersheds

Soil erosionis the productof manyinteractivesubprocesses.In the

Philippinesas elsewherein the humidtropics,wateristheprimary agent of
soilerosion.The interactiveprocessescausingsoilerosionbywaterinclude
detachmentand transportby fallingrain and flowingwater. The vadous
parametersinfluencingthese subprocessesinclude:cover, soil physical
and chemicalproperties,raindrop energy,flow velocitiesand hydraulic
properties,covermanagement, slope, slopelength andconservationprac-
tices.
Soilerosionis oftenclassifiedbased ondescriptiveparameterssuch
as magnitude,source, locationand shape (e.g. sheet, rill,gully, mass
movement,construction and miningerosion).Thissectiondeals mainlywith
sheetand rillerosionwhichwillbe jointlyreferredtoas sheeterosionin this
paper. Thus soilerosionas definedhere refersto the removalof the thin
layer of top soilby raindrops,overlandflow or by shallow,channelizedrill
flows.

1. Soil Erosion Rates in Selected Sites

There is a dearth of informationon soilerosionin the Philippines,and

research onsoilerosionconsistsmo3tlyof uncoordinated, isolated,empiri-
DAVID: SOIL AND WATER CONSERVATION 49

cal studies. In most of these studies, important parameters defining the

conditions under which the resultswere obtained were not monitored.
Furthermore,nostandardconditionshave been specifiedunderwhichthe
variouspieces of experimentalevidencemaybe compared.
When synthesizingthe availableinformationon soilerosionrates in
the Philippines,itbecomesimperative-- ifsuch informationis to be useful
to make quantitativeor at leastqualitativecomparisonsagainstcertain
standards.In an attemptto dothis,we havetakenthe libertyof evaluating
availableinformationin the lightofthe mechanicsortheoryof the erosion
processandofmakingassumptionsregardingconditionsunderwhichsome
ofthe erosionstudieswere carriedout. Furthermore,the resultsof studies
conducted elsewhere but applicable to Philippine conditions are used to
augment the meagre information available locally.

a. Various Cover Conditions

Undisturbed forests lands are the best soil conservers. In Mollucan

Sau, Katoan Bangkaland Dipterocarpforest cover, Serrano et al. (undated)
reported very low sediment yields of 0.10, 0.20 and 0.34 t/ha per year,
respectively. Kellman (1968) reported an annual soil loss of 0.09 t/ha from
primaryforest at 20pel'centslope on Mt. Apo in Mindanao. The comparative
soil loss rate in 250 million hectares of contiguousforest in the United States
was estimated at 0.4 t/ha per year. The above figures indicate that the
average natural or geologic erosion in undisturbed catchments should be
less than 1.0 t/ha/yr.
As the natural forest cover is disturbed by natural causes or human
activities, erosion rates increase dramatically. Table 1 presents the ratios of
soil loss rates for various cover to those for grasslands from small erosion
plots on Mt. Apo in Mindanao andon Mr. Makilingin Laguna. Thedifferences
in the relative soil-loss ratios in the two studies reflect the differences in
slope, soil, plot size and climatological conditions, as well as some known
instrumentation errors in the hydroecology group study.
These studiespoint to severalinterestingobservations. These include
the following: (1) soil loss increases exponentially as vegetative cover
decreases; (2)an undisturbed natural grass cover offers good soil protec-
tion; (3) soil erosion in some badly disturbed areas can be significantly
reducedifthese areasare left undisturbedfor native grass species to quickly
establishthemselves; (4)the practiceof large-scale clearing of grasses and
shrubsto establishorchards and tree I_lantationsmay not be a sound idea
as far as soil protection is concerned; and (5) cover manipulation offers a
cheap and effective means of minimizing soil erosion.
Table 1 also shows the wide range of variability in soil erosion rates
associated with cover and cover management practices. The Kellman
50 JOURNAL
OFPHILIPPINE
DEVELOPMENT

study, for example, reported a 300 fold increase in soil erosion from primary
forest to a 12-year old kaingin. Thus; only a small proportion of a watershed
area needs to be mismanaged before a watershed's sediment yields
drastically increase.

Table 1
SOIL ER()SION RATES ASSOCIATED WITH COVER AND COVER
MANAGEMENT PRACTICES

Cover Condition Ratio of Soil Loss Rate to

that of Grassland
(Kel/man, 1968)1,t UPLB Hydroecology Group, 1978) _

PrimarYforest 0.50
Softwood fallow 0.72

Imperata grassland 1.00 1.0

Plantation forest 2.9

Secondary forest 10.9

New rice kaingin 2.11 27.3
New kaingin (mixed crops)
12-year old rice kaingin 150.0
Old mixed crops kaingin 77.7

"For 20% slope on Mount Apo, Mindanao.

_Average of 36, 50 and 70% slope on Mt. Makiling, Laguna.

A combination of high-intensity rainfall, steep slope, erodable soil and

poor cover can lead to extremely high erosion rates. A study by Veracion
and Lopez (1979) of old kaingin areas in Benguet, Mountain Province
showed soil-loss rates of 308, 318,360,396 and 414 tons pe r hectare per
year even with strip crops of pineapple, coffee, castor bean, tiger grass and
banana, respectively. These excessive soil losses from old kaingin areas
were reduced to an average of 251 t/ha/yr by leaving them untilled or
Undisturbed for a year.
In a plot study on the soil erosion rates in cashew plantations in
Palawan, Madarcos (1985) reported that the selection of the appropriate
type o! cover and cover management is equally important in reducing soil
loss during plantation establishment. The results of his study as summa-
rized on Table 2 showed that (1) the practice of ring weeding (no intercrop)
results in high soil-loss rates during the establishment stage of orchards; (2)
 DAVID:
SOIL
AND
WATER
CONSERVATION 51

Table2
SOILLOSSESON CASHEW.BASED CROPPINGSYSTEMSFORONE
FIELDCROPSEASONFROMSEPTEMBERTO DECEMBER,1982
(Madar¢os,1985)

Cover:Cashew/Intercrop Soil_
Loss in T/Ha/Yr
Slope=21.7% Slope=39.2%

Three-yearoldcashew/No
intercrop 27.01 54.72
/ Five-yearoldcashew/Nointercrop 19.29 40,96
Five-yearoldcashew/corn 13.46 24.89
Five-yearoldcashew/soybean 11.70 22.69
Five-yearoldcashewGuineagrass 9.96 19.10
Five-yearoldcashew/sweet
potato 8.13 15.48

intercropping with cash crops such as corn and soybean may also resultin
high erosion rates; and (3) for a given slope, the soil-loss rate varies
significantly with the type of cover.
A study on the effects of cover intercrops in citrus orchards on 35
percent slope in Taiwan by Chang and Cheng (1974) also showed that
soybean intercropping resulted in high soil loss rates of 20 to 30 t/ha/yr.
Their results also showed that Guinea grass, Bahia grass and rice straw
mulch effectively controlled soil loss.
In a study of sheet erosion rates in the 412,000 ha Magat watershed
using a modified universal soil loss equation (USLE), David and C011ado
(1987) estimated an average soil loss rateof about 50 t/ha/yr. The soil loss
rates associated with the various slope ranges and types of land use are
summarizedin Table 3. For anygiven slope, the soil loss rateswere highest
in the open grasslands, river washings or deltas, built-up areas and
cultivated uplands.
A similar study of the 83,000 has Pantabangan watershed by David
(1987) showed higher erosion rates for the various land uses as a result of
steeper slopes and more erosive rainfalls. As shown on Table 4, the
average erosionrate for the entirewatershed is 108 t/ha/yr.
A comprehensive review of published and unpublished data in the
Philippinesand elsewhere would indicate soil loss rates of varying orders of
magnitude for different types of cover. Expressed in terms of cover
coefficients (or multiplierscompatible with those used in computing soil loss
52 JOURNAL
OFPHILIPPINE
DEVELOPMENT

Table 3
ESTIMATED PRESENT SHEET AND RILL EROSION LOSSES IN TONS
PER HECTARE PER YEAR FOR THE VARIOUS LAND USES AND
SLOPE RANGES OF THEMAGAT WATERSHED
(David and Collado, 1987)

Slooe _an_e, % Area Weighted

Land use 0-3 3-8 8-15 15-25 25°40 >40 Average
for all Slope

1. Primary forest 0.048 0.16 0.32 0.63 1.28 1.03

2. Secondary forest ' 0.064 0.34 0.96 2.05 3,73 8,79 5,73

3. Open grasslands 2,44 9.79 32.73 55.57 126.24 240.6 122.01

4. Irrigatedpaddy 0.21 0.67 1.52 3.74 10,43 10,92 1.03

5. Rainfed paddy 0.21 2.69 4.04 7.07 14,03 2.75

6. Terraced rice,
irrigated 0.094 0.41 1.29 2.34 4.44 9,42 4.63

7. Terraced rice,
rainfed - 0.40 1.66 3.57 - 2.04

8. Diversified crops 4.13 9.00 - 68.97 144.46 10.95

9. Orchard 1.95 6,77 - 74.44 45.98

10. Built-uparea 5.11 12.0531.48 90.92 160,83 282,57 12.88

11. River washings 10.86 61.64 165.25 410.92 1125.22 93.81

AVERAGE FOR ALL

LAND USES 2.07 4.99 25.58 31.24 57.21 72.23 49.99
DAVID:
SOILANDWATER
CONSERVATION 53

Table 4
ESTIMATED PRESENT SHEET AND RILL EROSION LOSSES IN TONS PER
HECTARE PER YEAR FOR THE VARIOUS LAND USES AND SLOPE
RANGES OF THE PANTABANGAN WATERSHED
(David, 1987)

Area
Weighted
Average
Land Use 0,3 3-8 8-15 15.25 25-40 >40 For aft
Slope

Primary forest .86 2.25 2.22

Secondary forest .41 7.03 6.95

Open grasslands 10.01 25.72 141.70 139.95 264.03 210.72

Irrigated paddy .18 8.82 .45

Rainfed paddy .22 .81 5.56 25.26 ,64

Savannah 120.85 238.99 194.83

Kaingin 280.85 374.88 586.51 507.88

Diversified crops 177.77 177.77

Riverwash 10.17 985.97 418.39

Residential 3.68 169.30 161.17 333.48 103.26

AVERAGE FOR ALL

LAND USE .48 2.60 25.72 141.33 140.45 113.40 108.20

rates through the USLE), the soil loss rates for the more common cover
conditions in the Philippines are estimated as shown in Table 5.

b. Slope

The bulk of the empirical evidence on the influence of slope on soil loss
point to a power function, according to which the soil-loss factoror coefficient
varies directly with the percent slope raised to a power greater than one.
That is m

S = a+bS
*
54 JOURNAL
OFPHILIPPINE
DEVELOPblENT

Table 5
ESTIMATEDCROP COVER COEFFICIENT OR C VALUES FOR THE
COMMON .COVERCONDITIONS OF PHILIPPINE WATERSHEDS

Cover C Value RatioOver That

for Primary
Forest

Bare soil , 1.0 1000

Primaryforest (withdense undergrowth) 0.001 1
Secondgrowthforestwith good undergrowth 0.003 3
and highmulchcover
Secondgrowthforestwith patchesof shrubs 0.006 6
and plantationcropsof 5 yearsor more
IndustrialTree Plantation(ITP)

a) BenguetPine with high mulchcover 0.007 7

b) Mahogany,Narra, 3-8 yearswith 0.05-0.10 50-100
goodcover crop
c) Mahogany,Narra, 8 yearsor more 0.01-0.05 10-50
withgood undergrowth
d) Yemane, 8 years or more 0.08 80
e) Mixedstandof ITP plantspecies, 0.07 70
8 yearsor more

Agro-forestrytree species

a) Cashew, mangoand jackfruit,less 0.25 250

than 3 years, withoutintercrop
and withring weeding
b) Cashew, mangoand jackfruit3 to 5 0.15 150
years withoutintercrop,without
ringweeding
c) Cashew, mangoand jackfruitwith 0.08 80
intercropor native grassundercover
d) Mixed standof agroforestryspecies, 0.08 80
5 years or morewith goodcover
e) Coconutwithtree intercrops 0.05-0.1 50-100
f) Coconuts,with annualcropsas 0.1-0.30 100-300
intercrop
g) Ipil-ipil,good stand,firstyear 0.2 200
with nativegrass intercrop
h) IpiHpil,good stand,2 yearsor 0.1 100
morewith high mulchcover
i) Ipil-ipil,newlycut for leaf meal 0.3 300
or charcoal
DAVID:
5OILANDWATER
C,E_SERVATION 55

TABLE 5 (Con't.)

Grasslands

a) Imperata or themeda grasslands,well 0.007 7

establishedand undisturbed,with
shrub
b) Imperataor themeda grasslands, 0.15 150
slightlygrazed, withpatchesof shrub
c) Shrubswith patchesof open, 0.15 150
disturbedgrasslands
d) Well-managed rangeland,slightly 0.3-0.8 300-800
grazed coverof slow development,
firstyear
e) Well-managedrangelandcover offast 0.05-0.1 50-100
development,firstyear, ungrazed
f) Well-managedrangelands,slightly 0.01-0.1 10-100
grazed cover of slow development,
2 yearsor more
g) Well-managedrangeland,cover of 0.01-0.05 10-50
fast development,ungrazed,2
yearsor more
h) Grassland,moderatelygrazed, 0.2-0.4 200-400
burnedoccasionally
i) Overgrazedgrasslands,burned 0.4-0.9 400-900
regularly

Annualcashcrops

a) corn, sorghum 0.3-0.6 300-600

b) rice 0.1-0.2 100-200
c) peanut, mungbean,soybean 0.3-0.5 300-500
d) cotton,tobacco 0.4-0.6 400-600
e) pineapple 0.2-0.5 200-500
f) bananas 0.1-0.3 100-300
g) diversifiedcrops 0.2-0.4 200-400
h) new kainginareas, diversifiedcrops 0.30 300
i) old kainginareas, diversifiedcrops 0.60 800

Others

a) built-upruralareas, withhome 0.20 200

gardens
b) riverwash 0.50 500
56 JOURNALOF PHILIPPINE
DEVELOPMENT

where S is the slopefactor or coefficient,a and b are positiveconstants,S,

is the land slope in percent and m is greater than unity. The findings of
Madarcos (1985) suggest an equation similar to that of Smith and Whitt
(1947, 1948)where m is approximately 1.3. Table 6 presents the estimated
values of the slope factor S for various slopes as estimated from the Smith
and Whiff equation.

Table6
ESTIMATEDSLOPEFACTORSFORVARIOUSSLOPES*

Slopein SlopeFactor, Slopein SlopeFactor,

Percent S Percent S

3 1.0 36 25.1
6 2.3 48 36.7
12 5.9 6O 49.4
18 10.0 72 63.0
24 14.6 102 100.2
30 19.7 150 167.5

*BasedontheequationS = a + b SL4/3wherea andb areapproximately

0.1
and0.21,respectively
and SL istheslopeinpercent.

C. Soil Erodibility

The soil properties influencing soil erodibility consist of those that

affect the infiltration rate and permeability and those that resistdispersion,
splashing, abrasion and transporting forces of rainfall and runoff. A compre-
hensive study on soil erodibility by Wischmeier and Mannedng (1969)
showed that soil erodibility is a complex interaction of many physicaland
chemical properties. Among others,these include particle size distribution,
organic matter content, structure, pH, bulk density, pore space filled by air,
slopeshapeand steepness,aggregation, and chemistry of parent materials.
Manyof these parameters are not, however, taken into account in standard
soil sampling analyses.
Overall, particle size distribution and organic matter content rank first
and second as indicators of soil erodibility. Soils that are high silt, low in clay
and low in organic matter ar_.the most erodible1.

1An analysis of all known erodible soils in England, Canada, U.S. and India by
Evans (1980) showed that over 95 per-cent of these soils have clay contents of less
than 35 percent. There were no erodible soils in the sand class.
DAVID:
SOILANDWATER
CONSERVATION 57

Usingthe simplified equation of Wischmeier and Mannering (1969) for

estimating the soil erodibility index or factor on the basis of particle size
distribution, organic matter content and pH, the soil erodibility indices of
representative soil samples fromthe Magat and Pantabangan watersheds
were estimated and the results shown in Table 7. The range of these values
may be safely assumed as representative of that for most Philippine soils.

Table 7
REPRESENTATIVE VALUES OF SOIL ERODIBILITY
FOR VARIOUS PHILIPPINE SOILS

Soil _-exture pH O,M. Sand Silt Clay K value1

% % % , %

Loamy fine sand 5.5 2.4 75 6 19 0.07

Sandy loam 5.6 4.8 65 26 9 0.23
Sandy loam 5.7 2.4 55 31 14 0.30
Loam 6.3 5,2 34 29 37 0.19
Loam 5.6 4.6 50 34 16 0.27
Loam 5.8 2.1 44 42 14 0,38
Loam , 7,4 0.9 42 47 11 0.63
Silt loam . 5.8 6.0 30 50 20 0.30
Silt loam 5.7 3.0 25 55 20 0.36
Silt loam 5.8 1.1 24 59 17 0.60
Clay loam 5.3 4.7 33 35 32 0.22
Clay loam 5.4 3.0 38 32 30 0.24
Clay loam 5.6 1.8 35 35 30 0.30
Siltyclay loam 5.8 4.0 8 60 32 0.28
Siltyclay loam 5.4 1.9 2 61 37 0.35
Siltyclay 5.5 5.3 6 49 45 0.19
Siltyclay 5.6 2.1 6 46 48 0,27
Sandy clay 6.3 0.9 54 14 32 0.20
Sandy clay 6.1 3.5 57 10 33 0.09
Clay 5.4 4,9 17 27 56 0.13
Clay 5.2 3.0 15 29 56 0.16
Clay 5.6 1.2 16 30 54 0.26

1Kvalues were computedusing the equation

K = [(0,043) (pH) + 0.62/OM + 0.0082S - 0.0062C ] Si
where OM = organicmattercontentin percent,
S = percentsand
C = clay ratio = % clay/(% sand + % silt)
Si =, % silt/lO0
S8 JOURNAL
OFPHILIPPINE
DEVELOPMENT

d. Rainfall

There is very little information on the erosive power of rainfall patterns

in the Philippines. The generally accepted equation for estimating rainfall
erosivity indices developed for the USLE requires more detailed rainfall
information (recording rainlgage data) than is currently available. Following
the suggestion of Mihara (1951) and Hudson (1971), David and Collado
(1987) adopted the equation

n
Rj = A T_, Pim
i

where Rj is the erosion index for any _,ear j and Pi is the precipitation total
for day i when this exCeeds the threshold of 25 mm,
Using A and m values of 0.002 and 2.0, respectively, .the average
annual R values for selected stations in Northern and Central Luzon were
estimated and the results are shown in Table 8. (The use of an A value of
0.002 renders the R estimates compatible with those of the USLE.)

Table 8
SAMPLE RANGE OF VALUES OF RAINFALL ERODIBILITY (R)°

Location AnnualRainfall, R value

mm

1. Bontoc,Mt. Province 2280 174

2. Lagawe, Ifugao 2645 158
3. Hapid, Lamut,Ifugao 1838 107
4. Baretbet, Bagabag,Nueva Vizcaya 1770 112
5. Diadi, Nueva Vizcaya 1937 168
6. Ambuklao,Benguet 2391 165
7. Carranglan,Nueva Ecija 3122 130
8. Carrili,Nueva Ecija, Pantabangan 2665 329
9. Marikit,Pantabangan,N.E. 2247 197
10. Tanauan, Nueva Ecija 2136 138

*Basedon the equation

n
R = A • T. Pim
1

where R = dailyrainfall> 25.0 mm and i = counterfor thedays of the year, A and m

are 0.002 and 2; respectively.
DAVID:SOILANDWATERCONSERVATION 59

e. Influence of Conservation and Management Practices on

Soil Erosion

Table9 summarizesthe valuesofthecommonlyacceptedconserva-

tionpracticeor managementfactors. As isthe casewith the otherfactors
or indices, these are used as multipliersor correctionfactors for the
estimatedsoillossrates underconditionsof no conservationpractices.
Soilerosionratesmay be considerablyreducedwith the adoptionof
one or more conservationpractices. For example, contourfarming and
mulchtillagein areaswith21to 25percentslopewillgiveanoverailmultiplier
of 0.23 (0.90 x 0.26), a fourfoldreductionin the soil-lossrate.

2. Estimates of Soil Loss Rates under Various Combinations

of Factors

In the absenceof any applicablemethodfor estimatingsoilerosion

ratesin tropicalAsia, David(1976), Davidand Collado(1977) modifiedthe
universalsoillossequation(USLE) to suitlocallyavailableinformationand
prevailingenvironmentalconditions.This modifiedUSLE stipulatesthat

E = R • K • LS • C • P

where E = soilloss ratein tons/ha/yr

R = rainfallerosivityindexvalue (see Table 8)
LS = length_slopefactorwhichmay be approximatedonthe basis
of percentslope(seeTable 6)
C = c0verfactorvalue as shownin Table 5
K = soilerodibilityvalue as shownin Table 7
P = is the productof the conservationor managementfactors
being practiced(Table 9)

Hence,itispossibletoobtain"guesstimates" ofthesoillossratesassociated
withanysetofconditionsby estimatingthevaluesofthevariousparameters
in the modifieduniversalsoillossequation.
The above considerationshave some far-reachingimplicationson
erosioncontrol, land-use planning and allocation,and developmentof
farmingschemesfor sustainedproduction. Considerthe following:
(1) There exist effectiveand technicallysoundmethodsto control
soilerosion.These may includeproperselectionofcover, cover
management,conservationpractices,tillagepractices,soilame-
liorationandcontrolstructuressuchasterraces,diversionchan-
nelsand checkdams.
(2) The erosionrates of croppedareaswith steepslopes(greater
60 JOURNAL
OFPHILIPPINE
DEVELOPMENT

Table 9
APPROXIMATE CONSERVATION PRACTICE OR MANAGEMENT FACTORS

a) Tillage,terracing,contouringand stripcropping

Land Slope Terracing Contouring Contour Strip

% Bench Broad-_based Cropping

1-2 0,10 0.12 0.60 0.30

3-8 0.10 0.10 0.50 0.15
9-12 0.10 0.12 0.60 0.30
13-16 0.10 0.14 0.70 0.35
17-20 0.12 0.16 0.80 0.40
21-25 0.12 0.18 0.90 O.45
>25 0.14 0.20 0.95 0.50

b) Mulchingand cover management

Surface Cover Pe_rcent Percentage Cover of Mulch or

Cover Vegetation at Ground Surface
0 20 40 60 80 100

a) None 0 1.0 0.53 0,33 0.20 0.10 0.02

b) Tall weeds or 25 1,0 0.56 0.36 0.23 0.11 0.03

short bush,0.5m 75 1.0 0.71 0.53 0.40 0.22 0.06
effectiveheight

c) Bushor brushes 25 1.0 0.55 0.35 0.22 0.11 0.03

2 m effective 75 1.0 0.61 0.43 0.27 0.14 0.04
height

d) Trees, 4m effect- 25 1.0 0.55 0.33 0.21 0,10 0.03

ive height 75 1.0 0.56 0.36 0.23 0.11 0.03

c) Tillage and residue management

Tillage Practice P Value

a) Conventionaltillage 1.0
b) Zoned tillage 0,25
¢) Mulch tillage 0.26
d) Minimum tillage 0.52
DAVID: SOILAND WATER CONSERVATION 61

than 18 percent) can be reduced to acceptable levelsof_say, 8

tons/ha/yrwith sound erosioncontrolpractices. Hence, the
government'spolicyof classifyinglandswithslopesof 18percent
or more as forest lands or non-arablelands does not have a
soundtechnicalbasis. Thisobservationhasfar-reachingimpli-
cations,speciallyin theimplementationof the Comprehensive
AgrarianReform Program.
(3) Natural cover suchas forest and nativegrasses are good soil
protectorseven in areas with erodiblesoils,steep slopes and
erosiverainfall. If undisturbed,grasslandswillnaturallyevolve
intosecondgrowthforestsand eventuallyintoprimaryforests.
Thus there are verycheap methodsfor arrestingsoilerosionin
areasthat are unfitfor farming.
(4) A soil-lossthresholdlevel for sustainedyield must be inferred
from, among others,soil-inherentfertilityand fertilityregenera-
tivecapability,existingor intendedlanduses, slope-rainfall-soil
erodibilityconditions,economiccostof conservationmeasures
and off-siteeffectsof erodingsoil particles.
In supportof someoftheaboveobservations,inTable10are shownsample
calculationsof soilerosionratesunder variousconditions.It is quite clear
thatwithpropercoverselectionand management,areason slopesgreater
than 18 percent could be used for ecologicallysustainableagricultural
development.AsshowninTable 10,a combinationof broad-basedterraces,
mulching,zoned tillageand contouringcouldkeep the erosionrate below
5.5 t/ha/yrevenin areaswithmoderatelyerodibleSoils,erosiverainfalland
18-48 percentslope,

Table10
SAMPLECALCULATIONS OF THESOILLOSSRATE_'UNDERVARIOUS
CLIMATIC,SLOPE,LANDUSEANDCOVERMANAGEMENTCONOITIONS.

RainfallErosivityValue(R)
120 250

I. Soil erodibility,K (siltyclay ) 0.20 0.20

II. Slope(a)p_e__ent 18 30 48 18 30 48
(b) LSvalue 10.0 19.7 36.7 10.0 19.7 36.7

III. (R) (K) (LS)value 240.0 472.8 880.8 500.0 987.0 1835.0

IV. Erosionrate,t/hatyror (R) (K)(LS) (C)

withoutanyconservationpracticewhatsoever
62 JOURNAL
OFPHILIPPINE
DEVELOPMENT

Table 10, continued

Rainfall Erosivity Value (R)

120 250

a) Primaryforest
(C=0.001) 0.24 0.47 0,88 0.50 0.99 1.84
b) Wellestablished,
undisturbedgrassland
(C=0,007) 1.68 3.31 6.10 3.50 6.91 12.84
c) Cashew orchard,
5 yrs or more
(C=0,06) 19.2 37.82 70.46 40,00 78.96 146,8
d) Corn crop(C=0.4) 96.0 189.12 352.32 200.00 394.8 734,00
e) OIdkaingin(C=0.8)192,0 378.2 704.64 400,00 789.6 1468,0

V. Erosionrate, t/ha/yr, withthe followingconservationpractice(s)

a) Cashew orchard

(1) Establishgrass
intercrop,such as
centrosema,80%
_s_urface
cover
(P=0.11) 2.11 4.16 7.75 4.40 8.69 16.15

(2) Grass intercrop,

60% surface cover
(P=0.23) 4.42 8.-70 16.21 9.2 18.;16 33,76

(3) Broad-based
terraces(P-=0.20)
with cover inter-
cropat 80% cover
(P=0.23) 0.88 1.74 3.24 1.84 3,62 6.75

b) Corn

(1) Contourstrip
cropping
(P=0.40-0.50) 38.4 94.56 176.16 80,00 197.40 367.00

(2) Zoned tillage

(P=0.25) with
contouring
(P=0.90-0.95) 21.6 44.91 83.67 45,00 93.77 174.32
DAVID:
SOILANDWATER
CONSERVATION 63

Table 10, continued

Rainfall Erosivity Value (1:7)

120 250

(3) Zoned tillage,

contourfarmingand
mulchingat 40%
cover (P=0.40) 8.64 17.96 33.47 18.00 37.51 69.73

(4) Broad-basedterraces
(P=0.18-0.20), con-
touringand mulching
at 40% cover) 6.22 14.37 26.77 12.96 30.00 55.78

(5) Broad-basedterraces,
withmulchtillage
(P,=0.26)contour
farming 4.04 9.34 17.40 8.42 19.50 36.26

(6) Broad-basedterraces,
mulchingat 80% cover
(P=0.15), zoned
tillage (P=0.25) and
contouring 0.58 1.35 2.51 1.21 2.81 5.23

c) Old kaingin

(1) Contourstripcrop-
ping, mulchingat
60% cover (P=0.30),
zoned tillagecontour
farming 5.83 12.12 22.59 12.15 25.31 47.07

(2) Left undisturbedfor

naturbJgrassesto
establishthem-
selves, 2 to 4 yrs
after (C=0.01) 2.4 4,72 8.81 5.0 9.87 18.35

(3) Left undisturbed

for 10 yearsfor
secondgrowth
forestestablish-
ment (C=0.003) 0.72 1.41 2.64 1.50 2.96 5.50
L - _'lr -- .i
64 JOURNAL OF PHILIPPINE DEVELOPMENT

III. -Sediment Yields

Not all eroded soil particles will be delivered to the waterways of a

watershed. Some will be deposited for good or temporarily at various
locations within the watershed. Thus at a watershed the sediment yield or
outflow at any point across a stream and for a given periodof time may differ
significantly from the total soil loss due to erosion. Sediment yield as defined
here refers to the total annual sediment discharge at a reference point
across a stream system. This reference point defines the watershed
boundary and the basin area.
As mentioned before, the sediment yields from watersheds with
primary forests are low, averaging less than 0.5 t/ha/yr. This is because the
forest cover and the cover litter cushion the soil against raindrops energy,
intercept a certainamount of rainfall, improve soilstructure, aggregation and
infiltration, and increase the soil-surface resistance to overland flow chan-
nelization. Obviously, erosion rates and sediment yields from forested
watersheds or watersheds with good cover do not fluctuate too much and
are influenced primarily by channelized or stream flows.
As the watershed cover is disturbed and reduced, sediment yields
increase and fluctuate considerably. This is because the effects of other
parameters (e.g., climate, slope, soil erodibility, cover management and
conservation practices) and their interactions become more pronounced.
Table 11 presents the sediment yields at selected watersheds in the
island of Mindoro. The large variation in their sediment discharges may be
traced to variation in rainfall patterns, land use, productive capabilities, and
cover management. The watersheds in Mindoro Oriental experience
uniformly distributed and less intense rainfall, while those in Occidental
Mindoro undergo distinct wet and dry seasons.
Table 12,showsthe specificflows and land usesof a representative'
watershedin each of thetwo provinceson the island.Bucayaowatershed,
in OrientalMindorohas moreforestcoveranda smallerproportion.ofopen
lands.Ithas morecroppedarea owingto goodclimateandmore productive
soils.A higher levelof productivitynotonlyencouragessoilprotectionand
goodfarmingsystemsbutalso reducespopulationpressuresto overexploit
hinterlands.OrientalMindoroalso hasbetter roadsand portfacilitiesand,
hence, better access to the markets in the Metro Manila and Southern
Tagalog regions.
Over halfof thecatchment of the BugsuangaWatershedConsists0t
pastures or open areas. Used mostlyas range lands,these openareas are
overgrazed. The long dry spell aggravates the cover conditions in these
areas. It is also worth noting the fact that the specific dependable flows of
this watershed arejust about a tenth of those in Oriental Mindoro.
A Dreliminary study by David (1982) projected a sediment discharge
DAVID:
SOILANDWATER
CONSERVATION 65

Table 11
SEDIMENT YIELDS OF SELECTED RIVER BASINS,
IN THE ISLAND OF MINDORO
(David, 1984)

River Basin Catchment Area Mean Annual Sedi-

km2 , ment Discharges,
t/ha

A. MindoroOriental

1. Pula 161 4.6

2. Bongabon 369 6.7
3. Bucayao 300 4.0

B. Mindoro Occidental

1. Mamburao 144 348.0

2. Pagbahan 337 171.0
3. Bugsuanga 415 233.3

Table 12
SPECIFIC FLOWS AND LAND USES IN TWO MINDORO-
WATERSHEDS
(David, 1984)

WATERSHE_'
Bucayao Bugsuanga

Catchment area, km2 384 438

Specific dependable flows in
m3/sec_,/kmz, catchment
80% dependability level 0.052 0.0064
90% dependability level 0.037 0.0041
Major land use
Categories in %
Forest 45 16
Cultivated area 33 18
Grasslands (good stand) 4 45
Savannah (cogon and talahib
grasses with shrubs and
brushes) 18 18

The difference in catchment areas of the two watersheds in Tables 11 and 12

is due to different sets of reference or monitoring points on their stream system.
66 JOURNAL OF PHILIPPINE DEVELOPMENT

in the order of 30to 40t/ha/yt intothe Magat Dam reservoir.ActualreservoirL

sedimentation measurements during the period from 1983 through 1985
showed sedimentation rates of about 38.8 t/ha/yr. For the Pantabangan
Damwatershed, David and Collado (1987b) projected a sediment yield of
77 t/ha/yr during the period 1980 to 1982. Preliminary results of.actual
measurements by the NIA for the past several years ending in 1982 show
a sediment yield in the orderof 80 to 90t/ha/yr (personalcommu nicationwith
the NIA watershed management staff in December, 1987).
Both the Pantabangan and Magat reservoirs are foreign-assisted
(IBRD) multipurpose water-resources development projects. Each was
designed to irrigate over 100,000 has. The estimated sediment inflows into
these reservoirs are two to three times higher than those assumed in their
feasibility and design studies. Although various,governmental agencies
have been directing massive protection and reforestation efforts to these
watersheds for the past ten years, these efforts have very little to show for
themselves (David and Collado, 1987 and David, 1987). This is because:
(1) considerable sediments are already in transit and are near the reservoir
areas, (2) some of the current efforts are not properly focused, or are
designed to generate incomefor the agenciesconcerned rather than arres/t
soil erosion and (3) the agencies concerned lack rational policies as well as
management and regulating capabilities.
Evengiven an optimistic scenario of no further increase in erosion and
sediment discharges, the Magat and Pantabangan sediment storages are
likely tofill upin about30and 40 years, respectively (as against the designed
lives of 95 and 75 years, respectively). Unless measures are taken to
refocus, coordinate and expand the government's watershed protection
efforts, the sediment pools of these reservoirs will get filled up 20 years from
now,
Pollisco (I975)reported annualsuspendedsediment discharges of74
and 13t/ha for the Agno and Pampanga riverbasins, respectively. Various
NIA documentson river basin developmentprojects during the past 10years
estimate sediment yields rangingfrom 6.8 to 44.8 t/ha/year. Considering the
ongoing massive land-use transformations at these watersheds, their
sediment yields could be higher by several orders of magnitude at present.
A study by Mahbub (1978) of 28 major rivers in Indonesia reported
sediment yields in equivalent depthof soil eroded ranging from 0.03 to 23.0
ram/yr. The major rivers of Central Java had an average equivalent load of
4.0 ram/yr. At an assumed soil bulk density of 1.2 gm/cc, a sediment
discharge of 1.0 mrWyris equivalent to 12 t/ha/yr. Thus a range of 0.03 to
23.0/mm/yr is roughly equivalent to a sediment discharge of 0.4 to 276 t/ha/
yr.
Lin (1981) reported an average sediment production of 8.04 mm/yr .
from nineteen watersheds in Taiwanl. These watersheds range in catch-
DAVID: SOIL AND WATER CONSERVATION 67

ment size from 175 to 3,257 sq km, and in sediment yield from 1.5 to 20.1
ram/yr. His study also revealed that extreme flow events produce most of
the total annual sediment loads of rivers.

IV. Land Use and Conservation Planning

1. Land UsePolicies

The Philippine Constitution mandates the proper utilization and con-

servation of natural resources. The revised Forestry Code of the Philippines
(P.D. 705) stipulates government ownership, control, regulation and man-
agement of some 419 large watersheds whose combined area roughly
equals 70 percent of the total land area of the country. With the exception
of areas earmarkedfor geothermal explorations, the administration,control
and management of these areas are entrusted to the Department of
Environment and Natural Resources (DENR).
Lands not needed for for_estryor conservation practices are desig-
natedas alienableand disposable (Aand D)by the DENRand are thus open
Ior privateownership.To date, DENR retainscontrol of over50 percentof i
the country'stotal land area, a monumentaltask. There are indications,
however,that DENR's regulatory, and managementcapabilitiesare inade-
quate for thistask. The DENR, for example,controls about 80 percentof
the total land area of the 412,000 ha Magat watershed. In spite of the
reforestationefforts by the DENR and the BFD, the badlydisturbed open
areascontinueto increaseatthe expense offorested areason steep slopes
at a rateof about11,000 ha each year (Davidand Collado, 1987). In 1984,
the watershedgrasslandarea was estimatedat about155,000 ha, about
90,000 ha of whichhave slopesexceeding 25 percent.
With legislationon agrarian reform, more governmentlands will be
declaredA and D lands. Thiswillrequirethe developmentand implemen-
tationof a soundland-classificationand land-useallocationschemewhich
considersshortandlong-termlanduse objectives.The needto evolvesuch
a schemebecomesmore urgentconsideringits inherentrequirementsfor
massive baselineinformationand strengtheningof the linkages among
variousinstitutions.
Other major government land_useand watershed management poli-
cies include the sustained multiple-use forest management approach,
environmental quality management, and active participationof and partner-
ship with the people in managing natural resources. The selective logging
system is an example of the sustained multiple-use management policy.
The policies for sound environmental quality management are embodied in
the provisions of the Environmental Policy Acts of the Philippines. The
policies mandating active people's participation are implemented through
 6B JOURNAL OF PHILIPPINE DEVELOPMENT

the various upland livelihood programs such as the Integrated Social

Forestry(ISF) Programof the DENR, the DendrothermalProgramof the
NationalElectrificationAdministration,andthe People'sForestprogramof
the former Ministryof Human Settlements.
The generalland use policiesof thecountryare premisedon funda-
mentalconcern for conservation,development,controlby nationals,and!
socialjustice. The concernfor conservation,for example,determinesthe
publiclandsthatmaybeclassified asalienableanddisposable.Section13
of P.D. 705, for example,stipulatesthatonlypubliclandswith0-18 percent
slopemaybe alienatedanddisposed. Whilegeneralpolicyforconservation
is beyond debate, the specific regulationdesigned to implement it is
technicallyunsoundand, hence, highlycontroversial.The specificregula-
tions restrictingthe alienationof landswith0-18 percentslope(e.g. head
stream areas,-criticalwatersheds,coastal areas, river banks 20 m from
normalhighwaterline,waterbodiesof at least5 m inwidth,reforestedareas,
parks andwildlifesanctuaries;historicalsites,shrinesandforestreserves)
are generallylesscontroversialbut leaveroomfor finetuningand improve-
ment.
The policyof controlby nationalslimitsthe disposition,development,
and utilizationof publiclands to Filipinocitizensand corporationswith at
least60 percentownershipby Filipinos.The policyof developmentpursues
a goalof sustainabledevelopment,whilethepolicyof socialjusticeaimsto
insurethe dignity,welfare and securityof allcitizens.
There are countless laws, regulations,and programsconcerning
forestlanduse inthe Philippines.The moreimportantonesand thosethat
are relevanttotheobjectiveof sustainabledevelopmentincludeLOis 1258,
1260and 1262 (alldated 1982). LOI 1258 providesfor a rationalevaluation
and re-examinationof all governmentreservationsand parks in order to
providefora scientificreallocationschemeof forestresourcesforthebenefit
of the majorityof Filipinos. LOI 1260 providesa programto help upliftthe
plightof uplandfarmers by stabilizing.thetenuresof the landsthey till and
providinggovernmentalassistancein the developmentof these lands.
LOI 1262 aims to provide a rationale and scientificbasis for the
allocationof forest lands into more specificuses such as agro-forestry,
protectionforest, pastures, productionforests and parks or recreational
areas. It callsfor the implementationofa systemof landclassificationthat
willpromoteoptimum,andequitableutilizationof forestresourcesandother
publiclandsinorderto meettheincreasingneedsof ourgrowingpopulation.
Mostlandusepoliciesarefocusedonthemanagementof publiclands
andthe dispositionintobroadland usecategoriesof A and D lands. There
arefew specificpoliciesaimedat stipulatingthelanduseand management
of publiclands declared as alienable and disposable. The few existing
specificland use policiesare commodityorientedand aimed primarilyat
DAVID:SOILANDWATERCONSERVATION 69

promoting the production of food crops.

2. Issues on Land Allocation and Use

Case studies involvingthe Magat and Pantabangan watersheds

(David, 1987 and David and Collado, 1987) revealed many interrrelated
issuesand problemsonlanduseplanningandmanagementandwatershed
management.Beloware someof them.

a. Policies and Conditions Affecting Policy

Unrealistic Policies and Regulations. As shownin Table 10, Section

13 of P.D. 705 which,in effect,has restrictedthe utilizationof vasttractsof
landsfor agriculturalpurposesfor whichthey are suited,is an unrealistic
regulation.
The lackof specificpoliciesaimedat regulatingthe useand manage-
mentof publiclandsto be declaredalienableand disposableis causefor
alarm inthelightof governmentpolicyof distributingovera millionhectares
of public lands under the proposed ComprehensiveAgrarian Reform
Program. Specificland-usepoliciesaimed at sustainabledevelopmentof
such lands are needed, since most of these lands are only marginally
productive.Itis unlikely,however,thatsoundspecificlanduse policiescan
be formulatedinthe nearfutureunlessgreatereffortsaredirectedimmedi-
ately at generatinginformationwhich will allow a classificationof land
accordingto crop-suitability.As pointedout by Revilla (1984a), existing
public-forestpoliciesare too diffuse,subjectto variousshadesof interpre-
tation,sometimescontradictoryin nature, and in manycases unrealistic.
The causes of these may be traced to lack of data base; too many
implementingagencies whose functions are stipulatedby a maze of
decrees, directives,LOl's, and administrativeorders;and failure to fully
considertheinterdependenceand interactionsofthe socialand biQphysica!
aspectsof forestecosystems.The constantchangingof the guardat the
DENR makesitdifficultto reconcile,modifyandfine-tunethesepoliciesand
regulations.
Rural Poverty. There is no denyingthe fact that for any watershed
managementschemeto be effective,it mustbe premisedon satisfyingthe
basicneedsofpeopledwellingin and aroundthewatershed. At theMag-at
watershed,where theaverageuplandfarmfamilyincomeis abouta thirdof
the food-thresholdincome, there is bound to be continuingpopulation
pressuresto exploitmore hillylands. The governmentshouldchannela
cdticalminimumof natural,financialand technicalresourcestosatisfythese
basi_needsifit isto relievepopulationpressuresjustenoughto effectively
implementlongtermwatershedmanagementplansandprograms.Itshould
70 JOURNAL OF PHILIPPINE DEVELOPMENT

again be pointed out that the disturbed open areas in the Magat watershed
are increasing at a rate of about 10,612 hectares per year in spite of the
combined •reforestationefforts of the BFD and the NIA.
Ineffective Land Use Planning and Allocation. The country does
not have a comprehensive system of land-use planning and allocation. This
needis more acute in the case offorest lands. The primarycriteria at present
for classifying lands into either forest or alienable and disposable lands are
•slope and existing land use. Ideally, however, lands should be allocated on
the basis of capability and suitability for alternative uses and utilization
demands or objectives. Thus land_useallocation must be validated from
land physical properties such as climate, slope, elevation and soil fertility;
suitability for sustainable uses such as productivity, erosion rates and
biological stability under alternative land uses as well as socio-economic
goals, objectives and constraints. Comprehensive land classification and
allocation are constrained mainly by the lack of proper appreciation of the
interdependence and extent of interactions of the various components of
watershed ecosystemsby agenciesor partiesconcerned. This has resulted
in critical gaps in the manpower and skills needed for proper land classifi-
cation and allocation. Forexample, neitherthe DENR nor DA today has the
capability to quantify erosion rates and the effects of watershed modifica-
tions on soil loss rates and runoff. Yet protection from soil erosion and its
adverse environmental effectsissupposedly one of the main considerations
in declaring some forest lands unalienable.
Lack of Active Support for Forest Resources Conservation and Man-
agement by the People. This problem is related to many other problems
such as those mentioned above. Rural poverty, insensitivityof the govern-
ment to the needs and aspirations Of the upland poor, among others,
contributeto the uplanders'indifferenceto governmentforest resources
conservationand managementprograms.

b. Implementation Problems

Ineffective Planning and Implementation of Action Programs. As

Revilla(1985)pointsout,thevariousuplandlivelihoodprogramssufferfrom
seriousproblemswhichinclude: (1) toomuchfocuson marginallands;(2)
inadequateandfaultyplanning;(3)lowcapabil!tyto providesocio-economic
upliftmentof uplanders;and (4) inadequatemanpowerto effectivelyimple-
ment upland developmentprograms. The basic flaw of many of these
programsis the assumptionthat uplanddevelopmentprograms mustbe
financiallyviablein an out-of-pocketsense. In reality,they are not likelyto
be soand mustbe viewedas rehabilitationactivitieswhosebenefitsarenot
privatelyappropriableratherthanas usualbusinessactivities.On theother
hand,the off-siteexternalitiesare ignoredand, hence, the socialvalue of
DAVID: SOIL AND WATER CONSERVATION 71

encouraging such activitiesis underestimated. Demanding that such

programsfunctionas viablebusinessventuresistantamountto askingthe
poorestof the poorto shoulderthe Costof forestrehabilitation.
The casestudiesof theMagatand Pantabanganwatershedsby David
and Collado(1987) and David(1986) alsopointto ineffectiveplanningand
implementationofwatershedmanagementprograms.Projectedyieldsand
benefitswere unrealistic. In the case of the Pantabanganwatershed and
erosioncontrol project, the cost of reforestationwas too high and the
accomplishmentsso far have fallentoo short of the targets.
The proposedMagat watershedmanagementproject,even ifeffec-
tivelyimplemented,willnotsignificantlyreducethe sedimentinflowintothe
Magat reservoir. Itis premisedprimarilyonhavingtheprojectpayfor itself
throughincreasedproduction. It focuses only on a small portionof the
watershed. It projectsunrealisticallyhigh yields and a very short full
developmentpedod. Both projectsdo not providefor mechanismsfor
monitoringenvironmentalimpact.They donotfullyconsidertheerosionand
sedimenttransportprocessesinvolved.
Bonita (1981) cited the Forest EcosystemManagement Program
(PROFEM) as a classicexampleof Ineffectiveand unrealisticreforestation
program.Afterfiveyearsof implementationwhich startedin 1976, ithadyet
to translateitsplansand programsintoaction. The failureof the program
was tracedto lackof planning,lackof coordinationand failure to consider
the needs and aspirationsof forestlandoccupants.
Lack of Effective Coordination Among Agencies Concerned. Many of
the problemsin watershedmanagementare institutionalin nature.More
than 50 percent of all our land resources are still classifiedas forest
lands. The DENR hasbeenentrustedwiththe primaryresponsibilityforthe
managementofournon-urbanwatersheds.As alreadypointedoutin many
reports,the DENR as an institutionisratherweak and unableto performits
tasksofprotecting,regulatingandmanagingourforestlandresources.This
has resultedin the creationand/orinvolvementof manyotheragenciesin
the regulation,managementanddevelopmentof ourforestlandresources.
Amongothers,theseagenciesincludetheNationalIrrigationAdministration
.(NIA), the Department of Agriculture (DA), the National Electrification
Administration(NEA), NationalPowerCorporation(NPC), Departmentof
Trade and Industry(DTI), NEDA, NationalDendroDevelopmentCorpora-
tion (NDDC), PresidentialCommitteefor Wood industriesDevelopment
(PCWlD) and the ManilaSeedlingBank Foundation.
Coordinationbetweentwo ministriesalone is alreadydifficult, For
example, the coordinationbetween the DA and DENR in the upland
livelihoodprogramsis almostnon-existentin the Magatwatershedarea. It
is quite obviousthat the task of integratingand coordinatingthe Magat
watershedmanagementactivitiesisalmostinsurmountablegiventhemany
72 JOURNAL OF PHILIPPINE DEVELOPMENT

participating agencies with overlapping functions.

Poor Delivery of Social and Agricultural Support Services. Boththe
NIA and Bureau of Soils surveysin the Magalwatershed revealedvery
low average incomefor the uplanddwellers. The 1982 and 1984 surveys
conductedby NIA alsoshowedpoordeliveryof socialservicesand inade-
quateagriculturalsupportservices.Thereareverylittleagriculturalsupport
services in the formof marketing,extension,farm inputsdistributionand
farm credits in the uplands as these services are concentratedin the
lowlands. Althougha numberof government;initiatedrural organizations
exist,membershipintheseis,however,low. Thewatershedoccupantsare,
in general,skepticalof externallyinitiatedorganizations.
Highly Centralized System of Governance. Unlike many other
countries(e.g., Indonesia),the planningand implementationof watershed
managementprogramsare highlycentralized,The inherentdifficultieswith
sucha systeminclude:thelackof accountability forfailure,evenat fieldstaff
leveland the lackof participationof, and controlby localinstitutions(e.g.,
local governments)in reforestation.In this respect,perhaps, it is worth
consideringthe Indonesianexperience,wherethegreeningoragroforesta-
tionof certain areaswas relegatedto districtsand provincialgovernments.
The above issues and problemsare, in fact, restatementsof the
requirementsfor soundpoliciesand programswhichtakeintoconsideration
theconceptof sustainabledevelopment,liveUhoocl secudtyforthe poor, and
equitableand economicuse of basicresources,as well as the institutional
and technicalconstraintsin puttingdevelopmentalplans into action.The
literatureis repletewith suggestionsand recommendationsfor achieving•
sustainablegrowth.(See, e.g. Swaminathan,1987).2

v. Conservation Planning

Inthe contextofthegovernment'srecentdecentralizationpolicy,past
experiencein conservationwork,andthe realitiesofthe Philippinecountry-
side, conservationplanningshouldhave a clearerfocuson the competen-
cies assignedto its various levels.•Thismeans that coherent p-lansandI
staementof objectivesmustbeexpectedfromthenationallevel,whileplans
and designs of greater resolutionmust come from the regionalor local
levels.

• Swaminathan proposes a seven-point

actionplanfor achieving
sustainable
nutritionsecurityin developingcountries,namely:a) locallevelcodesfor the
sustainable and equitableuseof environmental
systems; b) sustainedJivelihood
security forthepoor;c) symphonic production
system;d)scienceandtechnology;
e) knowledgeand skillssharing;f) resourcemobilizationand utilization;
and g)
politicalcommitment andaccountability.
DAVID:SOILANDWATERCONSERVATION 73

Conservation planning on the nationallevelinitiallyinvolvesa general

assessmentof the hazardsof erosionand sedimentation,tothe extentthat
problemor priorityareas or regionsare identified. The levelof resolution
couldbe onpoliticalregionsorlargeriverbasins.Suchan assessmentcould
bedonebycompilingandanalyzingrainfall,soil,landslope,landusetrends,
erosionrates,suspendedsedimentloadsand cropand fisheryproduction
information.A macro*levelassessmentoftheon-siteand off-siteeffectson
soilerosion,bothin financialand environmentalterms,is thencarriedout.
Such informa.tion arethen usedas a data base forformulatingor reformu-
latinggeneralpolicies. Nationallevelconservationplanningalsoinvolves
the provisionof legislationnecessaryto implementsuchpolicies;develop-
mentof a nationalframeworkfor validatingand monitoringregionalplans
and programsand setting out of institutions,institutionallinkages and
• mechanismsincludingsupportservicesfor the effectiveimplementationof
regionalplansand programs.
•On a regional level, conservationplanningbasicallyincludesland
evaluationand classification,landuse planningand allocation,anddesign
of appropriateconservationmeasures and monitoringand evaluation
schemes. Appropriatemeasuresimplysoundnessfrom agronomic,engi-
neering, hydrologic,socio-economicand environmentalstandpoints.The
designofsuchmeasuresrequiresanunderstandingofthemechanicsofthe
interrelatedprocessesoferosionand sedimenttransport,methodologiesfor
quantifyingthe magnitudesof these processesunder varyingwatershed
conditionsand well-defineddesigncriteriasuchas thresholdor allowable
SOillosslevels.

1. Soil Loss Threshold Levels

Anallowablesoillosslevelintheoryimpliesoptimumand ecologically
sustainedlevelof productionfor a giventractof landforan indefiniteperiod
oftime. Athresholdlevel,therefore,is thesoillossrateat whichthe rateof
soilformationequalsthe rate of soilerosion. It is alsothe levelwhere the
benefitsfrom the utilizationof the land are optimum considering,among
others,theon-siteandoff-sitecostsofsoilerosion.In practice,however,this
theoreticalstateof equilibriumisimpossibleto achieve,butshouldratherbe
regardedas a movingtarget thatvaries in space and time,with land use,
naturalattributesof soil,marketforces,and off-site(downstream)effectsof
soilerosionas variables.
It is quite understandablethat mostof the recommendedallowable
soil-losslevelsareaimedat lessdifficulttargets. Usually,theoff-siteeffects
are notconsideredandthe levelisset primarilyin termsof soilfertility.That
is, the level refers to the maximumsoil lossrate at which SOilfertilityis
maintainedfor,say 15to 25 years. Suchdefinitionofa specificlevelusually
 74 ,JOURNAL
OF PHILIPPINE
DEVELOPMENT

allows for the use of organic and inorganicfertilizersand the projected

improvementsin agriculturaltechnology(e.g., developmentof shortdura-
tion, fertilizeruse efficientvarietiesand improvementin crop cover and
managementtechniques).
Evenwiththe lessrigorouslydefinedthresholdlevels,greatdifficulties
are encounteredin arrivingat generalconsensus. Soilfertilityis a highly
dynamicand complexprocess. Moreover,soilerosionrates alsoexhibit
great variationsacrossspace and time.
Soilfertilityisinfluencedbymanyfactors.Amongothers,theseinclude
soilstructure,waterretentionproperties,soilchemicaland physicalproper-
ties,andclimate.Thereisampleevidenceintheliteratureindicatingthatsoil
structureis the key to soil fertility(Baver, 1956). Continuedcultivation,
togetherwith its accompanyingerosionof top soil, and exposureof the
surfacelayerto raindropsresultin breakingup of soilaggregates,deterio-
rationof soilstructure,andcompactionof soilstrata. Other soilproperties
whichinfluencesoilproductivityare likewiseaffected,suchasporosity,soil
and air holdingcapacities,andmovementof water and airthroughsoils.
There issubstantialexperimentalevidencein Westerncountriestothe
fact that continuouscultivationreduces soil porosityby as much as 18
percent. Furthermore,a differenceinporosityofas littleas 10 percent(e.g.,
60 versus50 percent)issufficientto causeyield-differencesof as muchas
200 percent. In suchcases, totalporosity,aswell as non-capillaryporosity
. (foraeration),influencedthe yieldlevels. On theaverage,thedifferencein
the non-capillaryporositybetweenforestedand bare soilsrangesfrom10
to 15 percent.
Experimental evidence on the influence of cover disturbance and
managementpracticeson soilmoistureholdingcapacitiescan perhapsbe
best illustratedby first analyzingthe staticsof soil water. Considerthe
equation
d, = Pw x As x D
100
where d,, is the depthof water storedin a rootzonedepth D with moisture
contentbyweightof Pwpercentand a bulkdensityof AS. Assumingallother
factorsremainconstant,an increaseinbulkdensity(morecompactsoil)will
increasethewater holdingcapacityofa soil. Whenwe considertheforces
actingonthesoilwaterortheavailabilityofthestoredwaterforplantgrowth,
a differentpictureemerges. Considerfurtherthefollowinghypotheticalbut
realisticcases.
Case 1 2

AS,g/cc 1.3 1.1

D, cm 100 100
Total porosity,% 50 57
DAVID: SOIL AND WATER _SERVATION 7§

Moisturecontent,Pw, % at

a) Fieldcapacity 30 34

b) Wiltingpoint 15 13

Maximumwaterholdingcapacity,

dw,cm 65.0 62.7

Depthof storedwateravailable
for plantgrowth,mm 19.5 23.1

Soilcompaction,oran increasein bulkdensity,generallyincreasesthe soil

maximumwater holdingcapacity.However,it reducesthewater available
for plantgrowthand adverselyaffectsothersoilproperties.
Amoreimportanteffectof soilcompactionisthereductionininfiltration
capacityof the soil. Compactionresultingfromovergrazing,for example,
mayleadto a severalfolddecreasein infiltrationrate. Thustheopportunity
for rain water conservationin soil rootzoneis correspondinglyreduced.
Pocketpenetrometermeasurementsby Sims (1975) of soilcompactionin
ungrazedandgrazed imperatagrasslandsoftheUpperTalaverawatershed
in Central Luzonshowedaverage readingsof 1.3 and 4.0 tonsper square
foot, respectively.
Itisquiteobviousthat a thresholdlevelon soilfertilitymustnotonlybe
inferredfrom soil nutrientstatusbut alsofrom otherdeterminantsof soil
fertilitysuch as soil structure,water holdingcapacityand porosity. The
currentlyavailableinformationin thePhilippinesis,however,inadequatefor
this purpose.
The spatiallevelof resolutionofestablishingthresholdlevelsandother
designcdtedaforsuitableerosioncontrolmeasurecan be a ticklishissue.
The factorsinfluencingsoilerosionare mostlystochasticin nature. Some
structures(e.g. checkdams, terraces)as well as thresholdlevels,would
probablyhave to be definedonfrequencyof exceedanceorreturnperiods
basis. Others, such as erosionrates on farmers' fields, are realistically
definedon a seasonalor annualaveragebasis.
Erosion and sedimentationproblemsalso have important spatial
dimensions. For example, not all sheet erosion is delivered into the
watershedstreamchannel.Thus, a thresholdfor an individualfarm might
differfromthat fora largerarea or an entirecatchment.
A rate of 11 tonsper hectareis a widely accepted threshold levelfor
individual
farms
inthemiesternUnited
tates.However,
thesoils
inthese
7'6 JOURNALOF PHILIPPINE
DEVELOPMENT

areas are deep, veryfertileloesssoils. Forshallowerand lessfertilesoils,

levelsrangingfrom2 to 8 t/ha/yrhave beenrecommended.Suchthreshold
valuesare unrealistic,however,in areas where erosionrates are naturally
high(e.g., mountainousterrainswithhighrainfall). In such places,higher
levelsof asmuchas 25t/ha/yrhave beensuggested.Where thethreshold
isona basinbasis,suchas sedimentyieldsofrivers,a muchlowerthreshold
maybe setdependingontheestimatedratioofthesedimentdeliveryto the
grosserosionloss.
The above considerationspoint to several important issues and
implications.These include:
(1) levelsof spatialand temporalresolutionfor thresholds.
(2) establishmentof realisticthresholdsthat consideron-site and
off-siteeffects,landuse,land naturalattributes,and socio-eco-
nomic,agronomic,and hydrologicfactors.
(3) cost-sharingschemeswhere significantoff-siteeffectsare con-
sidered in establishingthresholdvalues. Usually, the poor
uplandfarmers areexpectedto shoulderallthe cost of conser-
vationmeasures.
(4) short,mediumversuslongtermgoalsas are neededfor setting
up thresholdvalues,data constraints,needs and practicality.

2. Off-site Effects of Erosion Control Measures

The downstreameffectsof erosioncontrolupstreamdependon the

sedimentdeliverycapabilitiesofthewatershedstreamsystem,theactual or
potentialutilizationofdownstreamresources,andthepotentialdownstream
hazardsto humanlives,structures,and properties.The sedimentdelivery
capacityofa watershedstreamsystemiscommonlyexpressedin termsof
the sediment delivery ratio, or the fractionof the gross erosionthat is
deliveredtothe pointof theriversystemordrainagearea underconsidera-
tion.
The sedimentdeliveryratioisdependentonthe sizeofthecatchment,
watershed characteristicssuch as relief, drainage network, length of
stream,sedimentproperties,theamountofsedimentintransit,thesediment
storagecapabilitiesof floodplainsand streams,the hydrauliccharacteris-
tics of the transport system and streamflowpattems. Although many
empiricalequationsfor sedimentdeliveryratio have been proposed,most
of these only considercatchment size and, hence, are of very limited
Iocationalapplicability.
In most watershedsin the Philippines,a considerablevolume of
sediments
isintransit
along
the
river
system,
The
segiment
gisGharges
o1
such riversystemsare mainlydependentonflow and sedimenrproperties.
In such riversas Magat and Pantabangan,95 percentof their sediment
DAVID:SOILANDWATER_SERVATION 77

discharges are carried out by 2 percentofthe extremeflows. In suchrivers,

controlof erosionupstreamwill have very little shortand mediumterm
effectsontheirsedimentdischarges.Thustheconceptofsedimentdelivery
systemasbasedonwatershedsize isof nopracticalvaluein predictingthe
impactof upstreamerosioncontrolon sedimentyields.
On the basis of the studiesof David and Collado(1987) and David
(1978),the sedimentdeliveryratiosoftheMagat andPantabanganriversat
theirdamsitesare intheorderof40 to50 percent. Theseareway abovethe
usualdeliveryratiosof lessthan1 percentforsuchlargewatersheds. The
reasonisobvious.The sources(sedimentintransit)are verynearthe point
ofreferenceand are alreadyin thewaterways.The onlyeffectivesediment
yield controlmeasures are streamflowregulationand river•trainingand
control.In otherwords,thefeasiblemeasuresare thosethatwillreducethe
streamsystemsedimentdeliverycapabilities.
The actual or potentialwater resourcesutilizationdownstreamare
manyand varied,givena certainwatershedand time frame. To illustrate
possible downstreameffects,considerthe case of a dam and reservoir
project. In the analysisof the effectsof erosionand siltationof a storage
reservoir,thefollowingstoragesas illustratedare usuallyconsidered.

S, (FLOODSTORAGE)

-...,

Sa(CONSERVATION
STORAGE) "-

_ SPILLWAYS

S 2 • -.-- ,... .,.,

whereS, = sedimentpoolornaturalorgeologic erosioninducedsedimentinflow

S2 = sedimentstoragepool as a consequence of acceleratedsoil eros_n

S1+ S= = totalsedimentpool
78 JOURNAL OF PHILIPPINE DEVELOPMENT

If the catchment is undisturbed and if there are no irrigatedareas or

otherconstraintsdownstream,then S3and S2 may bothbe usedfor active
or conservationstorage. But the utilizationof the watershed upstream
rendersS2 amountof storageunavailablefor irrigationand perhapspower
generationfor thelifespanof theproject.The effectsofacceleratederosion
inthiscase includethe lossofirrigationcommandarea and power genera-
tion correspondingto S2 volume of storage. In the case where only S3
amountof storageis required,the costis that of a largerdam and bigger
pondarea.
Wherethedam andreservoirareexisting,themoresignificantcostof
increasedsedimentyieldasa resultofacceleratedsoilerosioninclude:(1)
reductionin storagecapacitywithtimeand (2) increasedO & M cost dueto
siltationoncethe sedimentpool is filledup. Actuallya certain amountof
sedimentsin transitwithinthe reservoirarea getsdepositedintothe active
storagepool even beforethe sedimentpoolgetsfilledup.
The effectsdownstreamofthedam includelesssedimentdischarges
since from 94 to 97 percentof the sedimentsare initiallytrapped in the
reservoir. The financialvalue of this again depends on the actual and
potentialusesof the downstreamland and aquaticresources.

VI. Land Evaluation, Land Classification

and Land Use Planning

As mentionedbefore,thebasicprocedureforregionalorlocalconser-
vationplanningincludeslandevaluationand classification,landuse plan*
ningand allocation,and designand developmentof suitableconservation
structures.Sincethekeyto erosioncontrolisputtinglandresourcestotheir
best or proper uses, there can be no effective conservationplans and
programswithoutproperland evaluationand classificationand land use
planningand allocation.

Land Classification

Land ca"pabilityclassificationis the commonlyused scheme of land

evaluationfor conservationplanning.The rationalefor sucha classification
systemis that landshave identifiablebio-physicalcharacteristicsand that
the importanceof each characteristicvaries with locationand land use.
Thus some lands are more suited to certain uses than others and their
correctuseisthe bestmeansof erosioncontroland ecologicallysustained
production.
Many landcapabilityclassificationsystemsare used for land evalu-
ation in the Philippines. The Bureau of Soils and the Bureau of Forest
DAVID:SOILANDWATERCONSERVATION 79

Development use systems that are basicallyadaptationsofthatdeveloped

by the U.S. Departmentof AgricultureSoil ConservationService (SCS).
The NationalIrrigationAdministration(NIA) employsa modifiedversionof
the U.S. Bureauof Reclamationclassificationsystem,a systemthat was
developedprimarilyfor assessingthe suitabilityof landsfor irrigation.
The BureauofSoilsland-suitability classification
systemdivideslands
intoninecapabilityclasses. Each classis subdividedintosub-classeson
the basis of the extent and degree of use limitations,with emphasison
erosionrisk,slope,drainageconditionand soildepth. Underthisclassifica-
tion scheme,only four general land classesare considered suitablefor
arable farming (ClassesA, B, C and D). The primary criterion used for
suitabilityis slope. Class D landswith15-25 percentslopesare restricted
tOpasturesand industrialtree farms.
All landshavingmorethan25 percentslopeare classifiedas unfitfor
cultivation.Two landclasses (L and M) are classifiedas suitableonlyfor
pasturesand industrialtree plantations. These are the level swampy or
stonyareasand thosethat have from25-40 percentslope. The restof the
landclasses(N, X,Y) arerestrictedtoforestorwildlifelandsby virtueoftheir
having over 80 percent slopesor wet most of the time (swamps and
marshes).
In conservationPla0ning,the primarypurposeof landclassificationis
to providebase lineinformationregardinglimitationson landuseas wellas
guidelinesonproperuseandconservationandmanagementpractices.The
BureauofSoilslandclassificationsystemdoesnotgiveclearguidelinesbut
rathervague, general land use recommendationsand their management
requirements,whichdo not directlyrelate to the erosionprocesses and
subprocessesat workorto workingmechanicsof soilconservationmeas-
ures. The systemas suchdoes notpmviclea base fordecisionmakingfor
• differentialeffectivitiesofvariouserosioncontrolmeasures,modificatiollof
erosioncontrolmeasuresand adoptionof newtechnologies.Itsqualitative
and subjectiveassessmentof erosionriskscould restrictthe use of land
resourcesfor utilizationobjectivesforwhichthey are suited.

A Proposed Land Classification System

To be very useful and flexible, a land suitability classification for

conservation planning must provide a data base for quantifying erosion
rates and soil productivity levelsunder vawing conditions of slope, climate,
soil physical and chemical properties, land use and cover management
conditions. This implies collectionand/or mappingof information relating to
processor parameterSthat influencesoil erosion,sediment transportand
soilproductivitystatus.
The processesor parametersrelevantto soillossestimationinclude
@0 JOURNAL OF PHILIPPINE DEVELOPMENT

slope, rainfallerodibility,soil erosivity,land use and cover management

conditions.Thosefor soilproductivityare manyandvaried. Amongothers;
these includesoildepth,water holdingcapacity,drainageconditions,soil
structureand porosity,organicmattercontent, levels of major and trace
elements,rate of soil regenerationor formation,soilwater-air movement
parametersand soiltoxicitylevels. Many of these aredynamicproperties,
buttheir rangeoffluctuationcan be predictedfromtheoreticalor empirical
information.
The proposedland classificationscheme should guide alternative
location-specificdecisionson erosionrates and soilproductivitylevelsfor
varyingconditionsof land use,croppingsystems,managementpractices
and agriculturalproductiontechnologiesgivencertain physicalconstraints
(e.g., soiltexture,slopeand rainfallpatterns)that are expensiveor difficult
to modify. Thisagain impliesIocati0n-specificbasicinformation.
Inthe caseof soilerosionrates,the basicinformationincludesslope,
rainfallerodibilityvalue,paniclesizedistribution,soilorganicmattercontent,
pH andexistinglanduse(thoserelatingtofixlandusefixturessuchasbuilt-
up areas, cemeteries,roadsand of specialhistoricalinterests), Fromthis
informationthe soilerosion ratesmay be estimatedunder varyingland use
trends, cover conditions and conservationand managementpractices using
empirical equations such as the modified USLE.
The essentialfeatures of the proposed land classification schememay
be briefly enumerated as follows:
(1) Collectionof basic location-specificinformationon the proc-
esses and parameters known to influence soil erosionand soil
productivity. The mappable properties should be mapped pref-
erably on a 1:25,000 scale.
(2) Compilationof available informationor mapson soilerosionand
soilproductivity levels.
(3) Superimposition of base maps weighted on the basis of the
information in item (2) to produce alternative decisions or deci-
sion maps erosion hazards, erosion rates and land suitability for
various utilization objectives. These should include manage-
ment guidelines for the various land-use alternatives.
The data and data handling requirements of such a land classification
scheme seem to be insurmountable at first glance. Considerable ground-
work aimed at operationalizing such a scheme has, however, been carried
Outby the author and his colleagues at the College of Engineering and Agro-
Industrial Technology at UPLB. These include the following: (a)a data bank
on rainfall in the Philippines; (b)a computerized ResourceMapping System
(RMS) for resource inventory and location mapping on a gdct basis; (c)
computerized hydrometeorology packages for climatological and hydro-
logic data analyses and (d) a computerized land use decision mapping
DAVID: SOIL AND WATER CONSERVATION 81

system that produces decision maps by computerizedoverlaying,on a grid

togridbasis,basemapsweightedaccordingtotheirrelevancetothedesired
land use decision. The RMS package features various interpolations
schemes and producesbase maps of interpolatedvalueson a grid basis
from irregularly-spaceddata pointinputs.

A Proposed Land Use Planning and Allocation Scheme

Itissuggestedthattheproblemofefficientuseandutilizationoflimited
land resourcesto meet desiredobjectivesbe resolvedthroughprogram-
mingmodels.Figure1presentstheframeworkof landuseallocationmodels
compatibleandcomplementarytotheproposedlandclassificationscheme,

MATHEMATICAL PROGRAMMING MODELS COMPUTERIZED RESOURCE

a) Linear programming models- MAPPING SYSTEMS
straight or multiple objective These feature mapping
LP interpolation, and a real
averaging of irregularly
b) Dynamic programming models spaced data to generate
base maps.

DECISION MAPPING SYSTEM

Location specific geographical
delineation of resources This features super-
allocation_ imposition on a grid-to-
gnObasis of relevant
l RESOURCE ALLOCATION DECISIONS -- LOCATION
base maps toSPECIFIC
generate
decision maps such as
erosion potential and
land capability

&

I -1
_ITUATION MODELS

These leatura trend analysis, envi_ 4

ronmenta[ impact assessment, multiple
use possibilities, alternative goals
andpolicy analysis.

I MA.A EMENT
1-"!-
O,D
,MPL,NES
EMENT.T,ON
]
1

Figure1
CONCEPTUAL
FRAMEWORKOF THEPROPOSEDLANDCLASSIFICATION
AND LAND USEALLOCATIONSCHEME
I}2 JOURNAL OF PHILIPPINE DEVELOPMENT

The scheme generatesland useallocationand land use management

guidelines. Non-mappable and dynamic parameters such as socio-eco-
nomicfactors are best considered in the programming model. The program-
ming model may take the form either of linear or dynamic programming
,models. These allow for multi-objective or multipurpose allocation or
planning. The outputs of the programming models are gross areal alloca-
tions given certain objectives and constraints. This limits the number of
decision variables for feasibility of solutions. Location-specifica'ilocationis
carried out through the use of RMS and decision mapping systems. For
flexibility in upgrading management guidelines and for the purpose of
predicting the impacts of future changes in policies and management
technologies, a long term goal of using descriptive models is suggested,
DAVID:
SOILANDWATER
CONSERVATION 83

REFERENCES

Baver, L.D. Soil Physics. JohnWiley and Sons,Inc. New York, 1956.
Bonita=M.M. "Reshapingthe PhilippineForestry Sector.* Policy Paper No. 4,
ForestryDeveicpment Center, UPLB Collegeof Forestry,College, Laguna•
1981.
Chang,S. M. and C• S. Cheng• *Studyon the EffectsofMulchingand CoverCrops
On Slope Land Citrus Plantation.* Journal of Chinese Soil and Water Con-
servation (1974) 5(1); 82-95.
Cruz, W. "The EnvironmentalCost of Soil Erosion." Paper prepared for
presentationat the FARM Workshopsponsoredbythe World Bank and the
Canadian InternationalDevelopmentAgency,1987.
David, W .P. HydrometeorologySection, BenchmarkResource Inventoryof the
Magat Watershed,Magat WatershedPre-FeasibUityStudy. National Irriga-
tionAdministration,Quezon City,1982.
• "EnvironmentalEffectsof WatershedModifications."Paper pre-
sentedatthe PhilippineInstitutefor DevelopmentStudiesSeminar Workshop
on EconomicPoliciesfor Forest ResourcesManagement. Club Solviento,
Calamba, 1984
. "Erosionand Sediment Transport." Upland Resource Policy
Program. Technical Report No. 1. PhilippineInstitute for Development
Studies,1986.
. "HydrologicValidationof the PantabanganWatershedManage-
mentand ErosionControlProject." UplandResourcePolicy Program. Tech-
nical Report No. 3. PhilippineInstitutefor DevelopmentStudies, 1987.
David,W. P. and C. U. Co,ado, Jr. "AnAssessmentof Land and Water Resources,
Present Management Practices a_d Problems, and Future Management
Plans and Programsfor the Magat Watershed." Upland Resource Policy
Program. Technical Report No. 2. Philippine Institutefor Development
Studies,1987.
Evans, R. "Mechanicsof Water Erosionand their Spatial and Temporal Controls:
An EmpiricalViewpoint.* Chapter 4, Soil Erosion, M.J. Kirkbyand R•P.C.
Morgan, Editors•JohnWiley and Sons, Ltd. New York, 1980.
Hudson,N.W. SoilConservation. CornellUniversityPress, Ithaca,NewYork.1971.
Kellman,M. C. "SomeEnvironmentalComponentsof ShiftingCultivationinUpland
Mindanao.*(1969) Journal of Tropical Geography 28.
Lin, Yuan-Lin. "Statusof Forestry HydrologyResearch in Taiwan•_ Technical
Workshopon ForestInfluencein SoutheastAsia and the Pacific. East-West
EnvironmentandPolicyInstitute,East-WestCenter, Honolulu,Hawaii, 1981.
Madarcos, B. S. "Soil Erosionin Four Cashew-basedCrops Under Two Slope
Categories.* UnpublishedM.S. thesis, Universityof the Philippinesat Los
BaSosLibrary,College, Laguna, 1985.
Mahbub, B. ErosionIntensityof Several Major Riversin Indonesia. Directorateof
Researchand Water Problems,Bandung,Indonesia, 1978.
Mihara,H. "Raindropand SoilErosion." National Instituteof AgriculturalScience
BulletinSeries A. No. 1, 1951.
84 JOURNAL
OFPHILIPPINE
DEVELOPMENT

Pollisco,F. S.. "Impactof ForestResearchon NationalPublic." Paper presented at

the 1975 Forest Conference, UPLB, College of Forestry,College, Laguna.
1975.
Ramphal, S. S. "The Environmentand SustainableDevelopment." Journal, The
Royal Societyfor the Encouragementof Arts, Manufactureand Commerce.
(1987) Vol. 135(5376).
Revilla, A. V. "Policies and Strategies for a Viable Forestry Program." Paper
presented at the UPLB College of Forestry Homecoming Symposium. UPLB
College of Forestry, College, Laguna. 1984a.
"A Critical Look at the Upland Livelihood Programs in the Phil-
ippines." PolicyPaper No. 16. Forestry Development Center, UPLB College
of Forestry, College, Leguna. 1984b.
Serrano, R. C., T. R. Villanueva and B. D. Sinis. Surface Runoff and Sedimentation
under AlbiziaFalcataria (L.) Fosb., Anthocephalus Chinensis (lank) Rich ex.
Walp., Dipterocarp and Mixed Secondary Stands. Pterocarpus 2(1). (un-
dated).
Sims, B. D. "SoilCompactionand CattleGrazingin theUpper Talavera Watershed,
NuevaEcija." Unpublishedreport,Departmentof ForestResourcesManage-
ment, Collegeof Forestry,Universityof the Philippinesat Los BaSos.1975.
Smith,D. D. and D. M. Whitt. "EstimatingSoil Lossfrom Field Area of Clay Pan
Soils." Proc., Soil Science Soc. Am. 12, 485-490. 1947.
. "EvaluatingSoil Lossesfrom Field Areas." Agricultural Engineering
29, 349-396. 1948.
Swaminathan, M.S. "The EmergingGlobal AgriculturalScenario."Journal The
RoyalSocietyfor the Encouragementof Arts, Manufactureand Commerce.
Vol. 135(5376). 1987.
Veracion,V. R. and A. B. Loper "comparative Effectsof DifferentVegetative Cover
onSoilErosionRatesinKainginArea§ inBenguet."ForestRe$ources Institute
(FORI) AnnualReport. FORI, College, Laguna. 1979.
Wischmeier,W. H. and J.V. Mannering. "Relationof Soil Propertiesto itsErodibili-
ties." Proc, Soil Sci. Soc. Amer. 33(1). 1969.