University of Nebraska - Lincoln

DigitalCommons@University of Nebraska - Lincoln

Biological Systems Engineering: Papers and
Biological Systems Engineering
Publications

5-1991

Hydraulic Roughness Coefficients as Affected by

Random Roughness
John E. Gilley
University of Nebraska-Lincoln, [email protected]

S. C. Finkner
University of Nebraska-Lincoln

Follow this and additional works at: https://digitalcommons.unl.edu/biosysengfacpub

Part of the Biological Engineering Commons

Gilley, John E. and Finkner, S. C., "Hydraulic Roughness Coefficients as Affected by Random Roughness" (1991). Biological Systems
Engineering: Papers and Publications. 122.
https://digitalcommons.unl.edu/biosysengfacpub/122

This Article is brought to you for free and open access by the Biological Systems Engineering at DigitalCommons@University of Nebraska - Lincoln. It
has been accepted for inclusion in Biological Systems Engineering: Papers and Publications by an authorized administrator of
DigitalCommons@University of Nebraska - Lincoln.
 Gilley & Finkner in Transactions of the ASAE 34 (1991)

HYDRAULIC ROUGHNESS COEFFICIENTS AS AFFECTED

BY RANDOM ROUGHNESS
J. E. Gilley, S. C. Finkner
MEMBER Assoc. MEMBER
ASAE ASAE

ABSTRACT resulting from tillage-induced surface micro-relief were

Random roughness parameters are used to characterize identified.
surface microrelief. In this study , random roughness was Random roughness is used to characterize surface
determined following six selected tillage operations. micro-relief. Height measurements were employed in
Random roughness measurements agreed closely with a procedure developed by Allmaras et aL (1967) for
values reported in the literature. calculating random roughness. To reduce the variation
Surface runoff on upland areas is analyzed using among measurements , the effects of slope and oriented
hydraulic roughness coefficients. Darcy-Weisbach and tillage tool marks were mathematically removed. The
Manning hydraulic roughness coe仔icients were identified upper and lower 10% of the readings were also eliminated
in this investigation on each soil surface where random to minimize the effect of erratic height measurements on
roughness values were determined. Hydraulic roughness the final resul t.
coefficients were obtained from measurements of discharge Several other techniques have also been introduced to
rat巳and flow velocity. characterize surface roughness (Luttrell , 1963; Currence
The experimental data were used to derive regression and Lovely , 1970; Podmore and Huggins , 1980; Linden
relationships which related Darcy-Weisbach and Manning and Van Doren , 1986; and Potter et aI., 1990). Several of
hydraulic roughness coe旺icients to random roughness and these methods have considerable potential for use in
Reynolds number. Random roughness values available in describing surface micro-relief. However, to date , these
the literature can be substituted into the regression other techniques have not received wide acceptance.
equations to estimate hydraulic roughness coefficients for a A description of previous studies involving hydraulic
wide range of tillage implements. The accurate prediction roughness coefficients on agricultural areas was provided
of hydraulic roughness coefficients will improve our ability by Engman (1986). Hydraulic roughness coefficients were
to understand and properly model upland flow hydraulics. developed from runoff plot data originally collected for
KEYWORDS. Hydraulics , Flow, Roughness coefficients , erosion studies. Friction factors were presented in a tabular
Tillage. format with a description of various surfaces and land uses.
Liong et a I. (1 989) developed a simple method for
assigning Manning hydraulic roughness coefficients to
INTRODUCTION overland flow segments in kinematic wave models. The
Hydraulic叫ess ∞ccoe
叫e
倪 proposed method was found to work well on a gaged basin.
surface runoff on upland areas. Calculation of time This procedure may be useful in estimating hydrographs
of concentration , determination of flow velocity for ungaged watersheds.
and simulation of runoff hydrographs require the use of Overland flow resistance for selected types of surface
hydraulic roughness coefficients. The ability to understand roughness induced by tillage operations was investigated
and properly model upland flow hydraulics is also by Sadeghian and Mitchell (1990). Darcy-Weisbach and
necessary for process-based erosion models. Manning hydraulic roughness coe旺icients for both rill and
Frictional drag over the soil surface , standing vegetative interrill areas were identified. Gilley et aI. (1990) measured
material , residue cover and rocks lying on the surface , hydraulic characteristics of rills on several sites located
raindrop impact , and other factors may influence resistance throughout the eastern United States. Regression equations
to flow on upland areas. Roughness coefficients caused by were developed which related Darcy- Weisbach and
each of these factors contribute to total hydraulic Manning hydraulic roughness coefficients to Reynolds
resistance. In this study , hydraulic roughness coe证icients number.
Laboratory measurements of hydraulic roughness
coefficients of surfaces covered with sand or gravel were
Article was submitted for publication in November 1990; reviewed made by Woo and Brater (1961) , Emmett (1970), Phelps
and approved for publication by the Soil and Water Div. of ASAE in
March 199 1. (1975), and Savat (1980). Similar tests were conducted
Contribution from USDA-ARS , in cooperation with the Agricultural under field conditions on natural landscapes by Dunne and
Research Division, Univ巳rsity of Nebraska, Li ncoln. Published as Journal Dietrich (1980), Roels (1984), and Abr由ams et aI. (1986).
Series No. 9436. In most of these studies
The authors are John E. Gilley , Agricultural Engineer, USDA-ARS;
and Steven C. Finkner, former Research Engineer, Dept. of Biological
Systems Engineering, University ofNebraska, Li ncoln.

VOL. 34(3): MAY-JUNE 1991 897

 Gilley & Finkner in Transactions of the ASAE 34 (1991)

overland flow is less pronounced as the depth of overland assumed to be approximately equal to flow depth. For these
flow becomes greater. situations:
The objectiv巳of this investigation was to develop
Rn三旦 (6)
regression equations for estimating hydraulic roughness v
coefficients using random roughness and Reynolds number
as independent variables. Relationships were identified for where flow rate per unit width , q , is given as:
predicting both Darcy-Weisbach and Manning hydraulic
roughness coe征icients. The equations were derived using
random roughness values varying from 6 to 32 mm , and q=? (7)
Reynolds numbers ranging from 20 to 6000.

The continuity equation for flow is defined as:

HYDRAULIC EQUA:τ10NS
The Darcy-Weisbach and Manning equations have been Q=VA (8)
widely used to describe flow characteristics. Both relations
contain a hydraulic roughness coefficien t. Under uniform
flow conditions , the Darcy-Weisbach hydraulic roughness where Q is flow rate. For broad sheet flow, water depth is
coefficient , f , is given as (Chow, 1959): given as:

f= 8gRS
2
(1) y=-~
---
Vb
(9)
V
e--==--
whHugSVR In this study, water depth was determined indirectly using
'也略

cvovd.tsoc
d

aanh tvda
α 町

oke-en
nm-UB
u'au
晦 ed v

gh
.川

3 e
z

bUAUAU
, equation 9, and measurements of Q, V, and b.
巳ivρd
uv出

The Chezy equation may also be used to characterize

1叫

flow. The Chezy hydraulic roughness coefficient , c , is

"
山

d
白

0
3
gIven as:

--nu)

，，
R=A

E飞
(2)
P

where A is cross-sectional flow area and P is wetted The Chezy hydraulic roughness coefficient can be
penmeter. determined directly from the Darcy-Weisbach hydraulic
roughness coe旺icient using the relationship:
The Manning hydraulic roughness coefficient , n , is
given as (Chow, 1959):
c=[制
l l)

/，‘、
2/LI/2
R-'·S
n= 一-一 (3)
V Information on existing flow characteristics is needed to
relate the Manning hydraulic roughness coefficient to
Manning and Darcy- Weisbach hydraulic roughness either Chezy or Darcy-Weisbach hydraulic roughness
coe证icients can be related using the following equation: coefficients , since:
,,,?&
I /6
F 4 1 l t t t t」

R一
句4d
、 ·1

g1- R
l l I Bι叶E E E E E 」

n= n=-一一 (1 2)
句

(4)
E

C
tt1

In this study, information is presented to allow expanded

Fl ow characteristics can also be described using Reynolds use of either the Darcy-Weisbach, Marming or Chezy flow
number , Rn , which is given as (Chow, 1959): equations on upland areas.

Rn= 立且 (5)
v
EXPERIMENTAL PROCEDURES
Field tests were conducted at the University of Nebraska
where v is kinematic viscosity. Kinematic viscosity can be Rogers Memorial Farm located in Lancaster County,
determined directly from water temperature. approximately 18 km east of Li ncoln , NE. The Sharpsburg
silty clay loam at the site (fine , montmorillonitic , mesic
For broad sheet flow conditions where flow width , b , is Typic Ar giudolls) formed on loess under prairi巳vegetation.
much greater than flow depth , y, hydraulic radius can be Average slope at the location was 6.4%.

898 TRANSA Cf10NS OF τ11E ASAE

 Gilley & Finkner in Transactions of the ASAE 34 (1991)

The experimental design consisted of two randomized artificial tmf carpe t. Runoff was diverted into an HS flume
complete blocks , with the first block being located with a stage recorder for measurement of flow rate.
immediately upslope from the second. Each experimental Flow velocity was determined using dye tracing
block consisted of six tillage operations performed at techniques. Approximately 0.2 L of fluorescent dye was
random locations within the block. The tillage operations uniformly injected across the width of the plot , 0.76 m
included an anhydrous applicator, chisel plow , disk , field upslope from the lower boundary. A peristaltic pump was
cultivator, moldboard plow, and planter. These implements used to continuously withdraw flow at four points spaced
were chosen to provide a wide range of random roughness equally along the collection trough. Discharge was then
conditions. circulated through a flourometer and a scale on the
instrument provided a visual display of dye concentration.
SITE PREPARATION Average time of travel was calculated as the length of time
Existing wheat residue was first removed from the study required for the dye concentration peak to reach the lower
area by burning and hand raking. Selected tillage boundary.
operations were then performed along the contour at the Five measurements of travel time were obtained at each
study site. One-meter-square plots were established within inflow rate. The mean of the five readings was used to
each tillage treatment using galvanized sheet metal borders calculate flow velocity at a particular inflow rate.
for the top and both sides of the plots. A collection trough , Calculated flow velocities were used to determined flow
located at the bottom of the plots , was used to collect depths and corresponding hydraulic roughness coefficients.
runoff. When not in use , the plots were covered with Additional details concerning experimental procedures are
plywood which was placed several centimeters above the given by Finkner (1988).
soil surface. The plywood covering prevented weathering
of the soil surface.
Soil surface stabilization was required to prevent RESULTS AND DISCUSSION
destruction of soil form roughness during test procedures. Random roughness values were first identified for each
After measurements for random roughness were obtained , of six selected tillage operations. Darcy-Weisbach and
the plot surfaces were stabilized using a biodegradable, Manning hydraulic roughness coefficients were then
latex-base soil stabilizer. The stabilizer was sprayed over measured on each of the soil surfaces where random
the entire soil surface using a hand sprayer. The stabilizing roughness values were obtained. Independent variables
material penetrated the soil approximately 5 mm , influencing hydraulic roughness coefficients were
e旺巳ctively binding the soil particles together with a water determined. Finally, regression equations were developed
permeable layer. for estimating Darcy-Weisbach and Manning hydraulic
roughness coefficients from values of random roughness
RANDOM ROUGHNESS and Reynolds number.
Differences in soil surface height were recorded using a
mechanical profile meter. The surface profile meter , similar RANDOM ROUGHNESS
to the device described by Allmaras et al. (1967) , could be Random roughness was calculated using the procedure
easily rolled above the entire plot surface on a rectangular outlined by Allmaras et al. (1967). Table 1 presents random
support frame. The support frame was of variable height roughness measurements obtained in the present study, and
and was leveled in the horizontal plane. The rectangular values reported by Zobeck and Onstad (1987) in a review
frame was supported by four 250-mm steel stakes which of available literature. Random roughness values in the
were securely anchored into the soil to provide a horizontal present investigation ranged from 6 mm for the planter to
ref，巳renee. The upper left corner of each plot border as 32 mm for the moldboard plow treatmen t.
viewed from the bottom of the plots was used as a vertical
bench mark , creating a three-dimensional referencing TABLE 1. Random roughness values for selected
system. tillage operations
The profile meter consisted of a single row of equal Random
length, 3.2 mm diameter steel pins positioned at a spacing Random Roughness
of 6.4 mm. When lowered onto the soil surface , the top of Tillage Roughness* (Present Study)
the pins formed a nearly continuous line which was traced Operation mm 口1 m

onto a strip of paper located behind the pins. The profile Large offset disk 50
meter and frame were oriented so that surface elevations Moldboard plow 32 32
were measured parallel to the contour of the study area. Lister 25
Chisel plow 23 21
Transects were spaced every 50 mm along the slope and Disk 18 16
transect traces were later digitized at 25 mm spacings. A Field cultivator 15 14
total of 629 surface elevations were used for determination Row cultivator 15
of random roughness for each one meter square plo t. Rotary tillage 15
H征row 15
Anhydrous applicator 13 8
FLOW VELOCITY Rod weeder 10
Following surface stabilization with the latex-base soil Planter 10 6
stabilizer, flow was added to the top of each plot at twelve No-till 7
rates ranging from approximately 2 to 300 L/min. Flow Smooth surface 6
inlet energy was dissipated at the top of the plots using an * Zobeck, T. M. and C. A. Onstad (1987).

VOL. 34(3): MAY-JUNE 199t 899

 Gilley & Finkner in Transactions of the ASAE 34 (1991)

The random roughness values shown in Table 1 Random roughness , Reynolds number, and slope were
represent best estimates for a particular tillage operation. retained as possible predictors. In an analogy to pipe flow,
Differences in soil texture , water content at time of tillage , random roughness provided a measure of the physical
or tillage depth may affect surface conditions. In addition , roughness of the flow boundary and Reynolds number
variations in the physical characteristics of the tillage furnished a flow property. Bed slope was included because
implements may result in different random roughness it has previously been found to influence hydraulic
values. roughness coe旺icients (Issard , 1944; Emmett, 1970; and
The anhydrous applicator and planter caused little Yoon, 1970).
disturbance to the relatively smooth surface which existed A simple multiplicative relationship which included
at the study site. Random roughness values for these two random roughness , Reynolds number , and slope as
operations were less than those previously reported. For the independent variables was tested. The effect of adding each
other tillage operations , random roughness measurements of these three variables into the prediction equations was
obtained in the present study were in close agreement with evaluated using multiple linear regression analysis. Only
values reported by Zobeck and Onstad (1987). random roughness and Reynolds number were found to be
The addition of rainfall may serve to reduce random significant at the 0.10 probability leve l. Therefore , only
roughness. To quantify this reduction , a relative random these two variables were considered in subsequent
roughness term (RRR) was defined by Zobeck and Onstad analyses.
(1 987) as:
DARCy-WEISBACH HYDRAULIC ROUGHNESS COEFFICIENT百，
Darcy-Weisbach hydraulic roughness coefficients at
RRR= 旦旦 (13)
RR o varying Reynolds numbers for the moldboard plow and
planter treatments are presented in figure 1. The trends
presented for the moldboard plow and planter operations
where RR is the random roughness of a surface following are also characteristic of the other experimental treatments.
rainfall , and RR o is random roughness immediately after In gene时， hydraulic roughness coefficients can be seen to
tillage. From published data on relative random roughness , decrease with greater Reynolds number.
Zobeck and Onstad (1987) developed the following The moldboard plow and planter treatments produced
equa tIon: the largest and smallest random roughness values ,
respectively. The largest hydraulic roughness coefficients
-0.026 cumulative rainfall usually occurred on those plots with the greatest random
RRR=0.8ge (14) roughness. The planter treatments with relatively low
random roughness values produced the smallest hydraulic
where cumulative rainfall is given in centimeters. roughness coe旺icients.
Equations 13 and 14 can be used to estimate random Within the same tillage operation, substantial variations
roughness of a surface following rainfall from information in hydraulic roughness coefficients were found. Roughness
on cumulative rainfall since the last tillage operation. elements were sometimes larger than water depth. As
Reynolds number increased , variations in flow patterns
INDEPENDENT VARIABLES INFLUENCING HYDRAULIC sometimes occurred. In addition , transition from laminar to
ROUGHNESS COEFFICIENTS turbulent flow conditions may have resulted during a given
Three criteria were established for the model equations test series.
used to predict hydraulic roughness coefficients. The Information from the six tillage treatments was used to
equations should be simple and easily solved using the derive the following regression equation for estimating
fewest number of independent variables necessary to Darcy-Weisbach hydraulic roughness coefficients:
obtain reasonable results. The independent variables should
be generalized , and applicable to conditions beyond those
found in the present study. Finally , the independent 000
variables used in the relationships should be easily F
E -←一-- Moldboard Plow
identified at other locations. D
ι2
0-一一-0 Planter
Variables which could significantly affect hydraulic 』

Qj 100
D
roughness coefficients included random roughness , u
<J)
Reynolds number , slope , type of implement operation , and <J)
Q)
10
hydraulic radius. However , not all of these variable吕would C二
i二
口3

serve as useful , generalized predictors. No common basis 3

0
庄
existed for relating the six tillage implements to other 五二
三J
machinery. Therefore , the type of implement operation
would not serve as a desirable generalized independent
D'"
<J)
由

variable. 主 0. 1
>、
ιJ
Including hydraulic radius in the model equation would 」

m
Q
require an iterative procedure to solve the prediction
0.01
equations. An iterative solution would be more time 10 100 1.00e 10.000
consuming and difficult to solve , and could introduce Rey门口Ids Number
difficulties with convergence. Thus , hydraulic radius was Figure l-Darcy·Weisbach roughness coefficients vs. Reynolds
also eliminated from the analysis. number forselected tillage operations.

900 TRANSACTIONS OF'!HE ASAE

 Gilley & Finkner in Transactions of the ASAE 34 (1991)

mmmm
如 V
C。-

MANNING HYDRAULIC ROUGHNESS COEFFICIENTS

neof Perfect Agreement
u-

-一- Li Figure 3 presents Manning hydraulic roughness

z。。。 ωωωEζ白3。庄 υ二3而」U、

coefficients as a function of Reynolds number for the

moldboard plow and planter treatments. The shape of the
curves shown in figures 1 and 3 are similar. Hydraulic
FOughness coefficients can again be seen to generally
" decrease with greater Reynolds number.SUIfaces with the
刊

largest random roughness values usua1ly had the greatest

hydraulic roughness coe征icients.
AZ UφH

The following regression equation for predicting

却

Manning hydraulic roughness coefficients was obtaind

υ-

using data from the six tillage treatments:

UOL且

O
D 20 40 60 80 0.742
100
Measured Hydraulic Roughness Coefficient - f O.I72RR
n=-一一--一一旦 (16)
Figure LPredtcted vs.measured Darey-weisbach hydraulic Rn 0.282
roughness coefficients.

where RR o is given in mm. In deriving equation 16, RR o

75
f=6 .30 RR6· (15)
values ranged from 6 to 32 mm while Reynolds number
Rn 66 1 °. varied from 20 to 6000. When estimating Manning
hydraulic roughness coefficients , th巳random roughness
after rainfall term , RR , should be substituted for RR o in
for RRo given in mm. In deriving equation 15, RRO values equation 16 if rainfall has occurred since the last tillage
varied from 6 to 32 mm while Reynolds number ranged operation. It is important to remember that equations 15
from 20 to 6000. If rainfall has occurred since the last and 16 should only be used within the range of values for
tillage operation , RR after rainfall should be substituted for which they were derived.
RR o in equation 15 to obtain the new Darcy-Weisbach The reliability of equation 16 for use in estimating
roughness coe证icien t. hydraulic roughness coe旺icients was evaluated. Figure 4
The experimental data was used to test the reliability of shows predicted versus measured Manning hydraulic
equation 15 for use in estimating hydraulic roughness roughness coefficients. A coefficient of determination , r2,
coefficients. Predicted versus measured Darcy-Weisbach value of 0.727 resulted from linear regression analysis of
hydraulic roughness coe旺icients for roughness coe仔icient predicted versus measured hydraulic roughness
values less than 100 are shown in figure 2. Li ne a r coefficients.
regression analysis of predicted versus measured hydraulic
roughness coefficients yielded a coefficient of
determination , r2, value of 0.898. SUMMARY AND CONCLUSIONS
Many of the hydraulic roughness coe征icients shown in Analysis of surface runoff on upland areas requires
figure 2 are relatively small. It can be seen from figure 1 identification of hydraulic roughness coefficients. Total
and equation 15 that hydraulic roughness coefficients vary hydraulic roughness on a site is usually a composite of
inversely with Reynolds number. Thus , as flow rate roughness coefficients caused by several factors. In this
increases , hydraulic roughness coefficients rapidly investigation , hydraulic roughn臼s coefficients induced by
decrease , even for those tillage treatments with relatively surface microrelief were examined.
large random roughness values.
CFCO-u

'一一..... Moldboard Plow l.lfle at Perf 肘ct Agreement

但EO -

o一一一。 P l a n te r
86420
。o
0
u-

-o
汇 @O U mm@ E L03O E O F

Em的白 C Z D 3
口00

飞、
1
。

庄0U一 -
币」3
UK工 E D F U -比
K CE币 飞4

30Ln

001 nu
10 100 1.00e 10 ,000 口 2 04 ~6 0.8
Reynolds Number Measured Hydraulic Roughness Coefficient - n
Figure 3- Manning roughness coefficients vs. Reynolds number for Figure 4-Predicted vs. measured Manning hydraulic roughness
目Iected tillage operations. coefficients.

VOL. 34(3): MAY-JUNE 1991 901

 Gilley & Finkner in Transactions of the ASAE 34 (1991)

Random roughness is calculated from surface elevation Engman, E.T. 1986. Roughness coefficients for routing
measurements. Information exists in the literature for surface runoff. J. Irrig. and Drain. Engr. , ASCE
relating random roughness values to single and multiple 112(1):39-53.
tillage operations. If cumulative rainfall since the last Finkner, S.C. 1988. Hydraulic roughness coefficients as
tillage operation is known , the reduction in random affected by random roughness. Unpublished M.S. thesis,
roughness caused by precipitation can also be estimated. Univ. ofNebraska, Lincoln.
A field study was conducted to identify random Gilley, J.E., E.R. Kottwitz and J.R. Simanton. 1990.
roughness and corresponding hydraulic roughness Hydraulic characteristics of rills. Transactions ofthe
coefficients over a wide range of conditions. Random ASAE 33(6):1900-1906.
roughness measurements were made following six tillage Issard, c.F. 1944. The surface profile of overland flow.
operations performed on initially smooth soil surfaces. Trans. AGU 4:959-968.
Random roughness measurements were found to be similar Li nden, D.R. and D.M. Van Doren , Jr. 1986. Parameters for
to previously reported values. characterizing tillage-induced soil surface roughness.
Following measurement of random roughness , plot Soil Sci. Soc. Am. J. 50:1560-1565.
surfaces were stabilized using a biodegradable , latex-base Liong, S.Y., S. Selvalingam and D.K. Brady. 1989.
materia l. Steady uniform flow conditions were then Roughness values for overland flow in subcatchments. J,
established for a wide variety of discharge rates. From Irrig. αnd Drain. Engr. , ASCE 1]5(2):203-214.
measurements of discharge rate and flow velocity, Darcy- Luttrell , D.H. 1963. The effect oftillage operations on bulk
Weisbach and Manning hydraulic roughness coefficients density and other physical properties of the soil.
were calculated. Unpublished Ph.D. diss. , Iowa State University , Ames.
Multiple linear regression analysis was used to identify Phelps , H.O. 1975. Shallow laminar flows over rough
the independent variables influencing hydraulic roughness granular surfaces. J. Hydr. Div. , ASCE 10](3):367-384.
coefficients. Hydraulic roughness coefficients were found Podmore , T.H. and L.F. Huggins. ]980. Surface roughne岱
to be significantly affected by random roughness and effects on overland flow. Transactions ofthe ASAE
Reynolds number. The field data were used to obtain 23(6):]43 4-]439 , ]445.
regression relationships which related Darcy-Weisbach and Potter, K.N., T.M. Zobeck and L. J. Hagan. 1990. A
Manning hydraulic roughness coefficients to random microrelief index to estimate soil erodibility by wind.
roughness and Reynolds number. The regression Transactions ofthe ASAE. 33(]):151-155.
relationships can be used for random roughness values Roels , J.M. 1984. Flow resistance in concentrated overland
varying from 6 to 32 mm , and Reynolds numbers ranging flow on rough slope surfaces. Earth Surface Processes
from 20 to 60∞. and Landforms 9:54]-55 1.
Several factors may contribute to total hydraulic Sadeghian , M.R. and J.K. Mitchell. 1990. Hydraulics of
resistance on a given upland site. Information is needed on micro锢braided channels: Resistance to flow on tilled
hydraulic roughness coe旺icients provided by each of these soils. Transactions ofthe ASAE 33(2):458 -468.
factors , their contribution to total hydraulic roughness , and Savat, J. 1980. Resistance to flow in rough supercritical
the effect of flow rate on hydraulic roughness coefficients. sheet flow. Earth SUI内ce Processes and Landforms
The accurate prediction of hydraulic roughness coe旺icients 5:102-122.
will improve our ability to understand and properly model Woo, D.C. and E.F. Brater. 196 1. Laminar flow in rough
upland flow hydraulics. rectangular channels. Journal ofGeophysical Research
66(12):4207-4217.
Yoon, N. ]970. The effect of rainfall on the mechanics of
REFERENCES steady spatially varied sheet flow on a hydraulically
Abrahams , A. D., A .J. Parsons and S.H. Luk. 1986. sm∞th surface. Ph.D. thesis , University of lllinois ,
Resistance to overland flow on desert hillslopes. J矿 Urbana.
Hydrol. 50:343-363. Zobeck , T.M. and C.A. Onstad. 1987. Tillage and rainfall
Allmaras , R.R., R.E. Burwell and R卫Holt. 1967. Plow- e仔ects on random roughness: A review. Soil and Til/age
layer porosity and surface roughness from tillage as Res.9:1-20.
affected by initial porosity and soil moisture at tillage
time. Soil Sci. Soc. Am. Proc. 31:550-556.
Chow, V.T. 1959. Open-Channel Hydraulics. New York: SYMBOLS
McGraw-Hill Book Co., Inc. A = cross-sectional flow area
Currence , D.H. and w' G. Lovely. 1970. The analysis of soil b = flow width
surface roughness. Transactions ofthe ASAE 13(6):710- c = Chezy roughness coe证icient
713. f = Darcy-Weisbach roughness coe征icient
Dunne , T. and w' E. Dietrich. 1980. Experimental study of g = acceleration due to gravity
Horton overland flow on tropical hillslopes. 2. Hydraulic n = Manning roughness coefficient
characteristics and hillslope hydrographs. Z. Geomphol. P = wetted perimeter
Suppl. Band 35:60-80. q = flow rate / unit width
Emmett , W,W. 1970. The hydraulics of overland flow on Q = flow rate
hillslopes. Professional Paper 662-A , U.S. Geological R = hydraulic radius
Survey. Rn = Reynolds number

902 TRANSA CflONS OFηrnASAE

 Gilley & Finkner in Transactions of the ASAE 34 (1991)
nnk
vyv===

ω曲 -
wwa IEl
vdn
00'm

v
eeω

-1-
····n
RR = random roughness of a surface following

dm
rainfall

m
oco-1V4

户l
山
6EE-

飞
RRo = random roughness immediately after tillage

v
丁A
RRR = relative random roughness
S = average slope

VIX. 34(3): MAY-JUNE 1991 903