Use of GIS technique for determination of runoff coefficients for
Texas watersheds
Introduction:
The runoff coefficient can be defined as either the ratio of total depth of runoff to total depth of
rainfall, or as the ratio of peak rate of runoff to rainfall intensity for the time of concentration
(Wanielista and Yousef, 1993). The assignment of the runoff coefficient (C) is somewhat
subjective. At the time the rainfall producing runoff occurs, the coefficient varies with
topography, land use, vegetal cover, soil type, and moisture content of the soil (TxDOT, 2002).
The rational runoff coefficient is strongly dependent on land use and to a lesser extent watershed
slope as suggested by both Schaake et al. (1967) and ASCE (1992).
Methodology:
In this study there are 90 watersheds in Texas for us to estimate standard (table) rational runoff
coefficient. In the watershed database, we have the information on watershed area, longitude
and latitude of the USGS streamflow gauge station that is the outlet or pour point of a watershed.
From a previous TxDOT project, we had a geospatial database containing watershed boundaries
for 90 Texas watersheds, which were delineated using 30-meter digital elevation model (DEM).
ArcGIS was used for the extraction of land cover area of different classes within a particular
watershed. National Land Cover Data (NLCD) for 2001 were obtained for the Texas from the
USGS website http://seamless.usgs.gov/. There are eight panels of 2001 NLCD raster GIS layers
covering 90 watershed study area. After downloading raster data layers of 2001 NLCD for
Texas and loading the NLCD raster layers into ArcGIS, the next task was to cut out the NLCD
layer using watershed boundary for a particular watershed and find out areas of different classes
of land cover within that watershed. For completing this task we found that three different
methods under ArcGIS are available to use.
The first method was the extraction by mask method. When a particular watershed
boundary was selected in ArcGIS first, and then 2001 NLCD (raster layer) was loaded into the
ArcGIS, one can select or access “Extract by Mask” under ArcToolbox by clicking “Spatial
Analyst Tools”, then “Extraction”. Under the parameter input window for Extract by Mask,
“Input raster” should be the particular 2001 NLCD panel containing selected watershed, “Input
raster or feature mask data” should be the selected watershed boundary layer, and “Output
raster” should be the name of a new raster GIS layer that stores the cutout of 2001 NLCD layer
for selected watershed. When the cutout layer of 2001 NLCD for a particular watershed is
automatically loaded into ArcGIS, one can open the attribute table for that extraction, it has the
information on the number of counts of different land cover classes within the selected
watershed. Because 2001 NLCD GIS layer is 30-meter raster file, one can multiply the count by
900 square meters to find the area of each land cover class within a watershed.
The second method was clipping polygon method. For implementing this method, it is
necessary to convert the raster NLCD layer into the polygon feature layer (vector type GIS layer)
first. The “Clip” function of the ArcToolbox can be used by selecting “Analysis Tools”, then
“Extract”. Under the parameter input window for Clip, “Input Features” should be the particular
2001 NLCD panel containing selected watershed, “Clip Features” should be the layer containing
selected watershed boundary, and “Output Feature Class” is to provide a shape file name for
storing clipped NLCD area. When the clipped feature was loaded and the attribute table was
opened, one can add an “Area” field and use the Calculate Geometry function to determine the
areas for different grid (land cover) codes. Using the Statistics and Summarize functions, the
total area as well as the individual area of each land cover class for the watershed was obtained.
Both of the above methods were implemented using original projection for watershed
boundary and 2001 NLCD layers, which were NAD_1983_Albers projected coordinate system.
For the third method we redefined watershed boundary layer and 2001 NLCD raster layers into
UTM projection, which is more appropriate for area computation for relative small study area.
One can access “Project” function under the ArcToolbox by clicking “Data Management Tools”,
“Projection and Transformation” , then “Raster” and “Project Raster” for projecting 2001 NLCD
layers, or “Feature” and “Project” for projecting watershed boundary feature layer. The output
coordinate system NAD1983 UTM Zone 14N was selected for central Texas area. After
watershed boundary layer and 2001 NLCD raster layers were converted to UTM projection, the
clipping polygon method was followed.
All the above methods were first tested for two selected watersheds, and the results
obtained are given in Table 1. Area (DEM) is the watershed area within watershed boundary
delineated using DEM. Table 1 shows that percents of error (deviation) between total area from
clipped or masked 2001 NLCD layer and total area from DEM delineation are very small and
range from 4.9x10-6% to 0.08%. Table 2 shows areas for several land cover classes for two
selected watersheds developed by above three methods, and percents of deviation between two
methods range 2.8x10-4 % to 1.7 %. From the analysis of testing results all the above methods
produced almost the same results in areas for different land cover classes within a watershed, the
second method was selected and used for all the ninety watersheds one by one to get the total
area as well as the individual area of land cover class for each watershed. The results obtained
were copied and converted into excel spreadsheet for further analysis.
Table 1. Total watershed areas (m2) of two selected watersheds developed by above three
methods from clipped or masked 2001 NLCD raster or feature layers.
USGS
Area
Station
(DEM)
08139000 8115299.5
08140000 18966599.0
1st method
8116200
18966600
2nd
%error
method
1.1E-02 8115303
4.9E-06 18966603
%error
4.3E-05
2.1E-05
3rd method
8108868.1
18951530.0
%error
7.9E-02
7.9E-02
Table 2. Areas (m2) of a particular land cover class (11) for two selected watersheds developed
by above three methods from clipped or masked 2001 NLCD raster or feature layers:
USGS
Station
08139000
08140000
Land
Use
11
11
1st
method
102600
143100
2nd
method
100858
141638.35
3rd
method
101563
141637.95
error(1-2)%
1.7
1.0
Error(13)%
1.0
1.0
error(2-3)%
0.7
2.8E-04
Each different watershed may have different land cover classes distributed inside its watershed
boundary. It was found that there were total 15 land cover classes involved for the 90
watersheds studied. Table 3a gives detailed description for these 15 NLCD land cover classes.
Our next task was to assign runoff coefficient for particular land cover class. After the study of
various sources on selecting or specifying standard or table rational runoff coefficients, runoff
coefficients were assigned for the above 15 land cover classes as shown in Table 3b. From all
sources studied, typically we do not find runoff coefficients for most of 15 NLCD land-cover
classes, but we identified similar land use types to match them as shown in Table 3b. Assuming
that all of the rainfall is converted into runoff for open water and wetlands, the value of C
assigned to these land-cover classes is 1. For the other land cover classes a range of C values are
available in the mentioned sources under similar land use types. We took the average values for
them after determining under which land use type the particular class falls or closely matches.
Table 3a Detailed description of 15 land cover classes involved 90 watersheds (from 2001
NLCD website http://www.epa.gov/mrlc/definitions.htm).
NLCD
Description from 2001 NLCD website
Classification
Grid Code
11
21
22
23
24
31
41
42
43
52
71
81
82
Open Water - All areas of open water, generally with less than 25% cover of
vegetation or soil
Developed, Open Space - Includes areas with a mixture of some constructed
materials, but mostly vegetation in the form of lawn grasses. Impervious surfaces
account for less than 20 percent of total cover. These areas most commonly include
large-lot single-family housing units, parks, golf courses, and vegetation planted in
developed settings for recreation, erosion control, or aesthetic purposes.
Developed, Low Intensity - Includes areas with a mixture of constructed materials
and vegetation. Impervious surfaces account for 20-49 percent of total cover. These
areas most commonly include single-family housing units.
Developed, Medium Intensity - Includes areas with a mixture of constructed
materials and vegetation. Impervious surfaces account for 50-79 percent of the total
cover. These areas most commonly include single-family housing units.
Developed, High Intensity - Includes highly developed areas where people reside or
work in high numbers. Examples include apartment complexes, row houses and
commercial/industrial. Impervious surfaces account for 80 to 100 percent of the total
cover.
31. Barren Land (Rock/Sand/Clay) - Barren areas of bedrock, desert pavement,
scarps, talus, slides, volcanic material, glacial debris, sand dunes, strip mines, gravel
pits and other accumulations of earthen material. Generally, vegetation accounts for
less than 15% of total cover.
41. Deciduous Forest - Areas dominated by trees generally greater than 5 meters
tall, and greater than 20% of total vegetation cover. More than 75 percent of the tree
species shed foliage simultaneously in response to seasonal change.
42. Evergreen Forest - Areas dominated by trees generally greater than 5 meters
tall, and greater than 20% of total vegetation cover. More than 75 percent of the tree
species maintain their leaves all year. Canopy is never without green foliage.
43. Mixed Forest - Areas dominated by trees generally greater than 5 meters tall,
and greater than 20% of total vegetation cover. Neither deciduous nor evergreen
species are greater than 75 percent of total tree cover.
52. Shrub/Scrub - Areas dominated by shrubs; less than 5 meters tall with shrub
canopy typically greater than 20% of total vegetation. This class includes true shrubs,
young trees in an early successional stage or trees stunted from environmental
conditions.
71. Grassland/Herbaceous - Areas dominated by grammanoid or herbaceous
vegetation, generally greater than 80% of total vegetation. These areas are not
subject to intensive management such as tilling, but can be utilized for grazing.
81. Pasture/Hay - Areas of grasses, legumes, or grass-legume mixtures planted for
livestock grazing or the production of seed or hay crops, typically on a perennial cycle.
Pasture/hay vegetation accounts for greater than 20 percent of total vegetation.
82. Cultivated Crops - Areas used for the production of annual crops, such as corn,
soybeans, vegetables, tobacco, and cotton, and also perennial woody crops such as
orchards and vineyards. Crop vegetation accounts for greater than 20 percent of total
vegetation. This class also includes all land being actively tilled.
90
90. Woody Wetlands - Areas where forest or shrub land vegetation accounts for
greater than 20 percent of vegetative cover and the soil or substrate is periodically
saturated with or covered with water.
95
95. Emergent Herbaceous Wetlands - Areas where forest or shrub land vegetation
accounts for greater than 20 percent of vegetative cover and the soil or substrate is
periodically saturated with or covered with water
Table 3b. Runoff coefficients for various land cover classes
NLCD
NLCD Classification Runoff
Land use or description in
Classification Description
coefficient the source
11
Open Water
1
21
Developed, Open
Space
0.4
Residential: single family
areas(0.3-0.5)
22
Developed , Low
Intensity
0.55
50 % of area impervious
(0.55)
23
Developed, Medium
Intensity
0.65
70% of area impervious
(0.65)
24
0.83
31
Developed , High
Intensity
Barren Land
41
Deciduous Forest
0.52
42
Evergreen Forest
0.48
43
Mixed Forest
0.48
52
Shrub/Scrub
0.3
71
Grassland/Herbaceous 0.22
Business: downtown areas
(0.7-0.95)
70%IMPV_COV,HSG A (
0.5)
70%IMPV_COV,HSG B
(0.6)
70%IMPV_COV,HSG C
(0.7)
70%IMPV_COV,HSG D
(0.8)
Deciduous forest
(Tennessee) (0.52)
Forest (UK) (0.28-0.68)
Forest (Germany) (0.330.59)
Forest (UK) (0.28-0.68)
Forest (Germany) (0.330.59)
Woodland, sandy & gravel
soils (0.1)
Woodland, loam soils (0.3)
Woodland, heavy clay
soils (0.4)
Woodland, shallow soil on
rock (0.4)
Pasture, grazing HSG A
(0.1)
Pasture, grazing HSG B
(0.2)
Pasture, grazing HSG C
(0.25)
Pasture, grazing HSG D
0.65
Source
ASCE (1992)
TxDOT
(2002)
Schwab and
Frevert
(1993)
Schwab and
Frevert
(1993)
ASCE (1992)
Schwab and
Frevert
(1993)
Mulholland
PJ (1990)
Law (1956),
Hydrology
(1976)
Law (1956),
Hydrology
(1976)
Dunne and
Leopold
(1978)
Schwab and
Frevert
(1993)
(0.3)
81
Pasture/Hay
0.35
82
Cultivated Crops
0.4
90
Woody Wetlands
1
95
Emergent Herbaceous
Wetland
1
Pasture, sandy & gravel
soils (0.15)
Pasture, loam soils (0.35)
Pasture, heavy clay soils
(0.45)
Pasture, shallow soil on
rock (0.45)
Cultivated, sandy & gravel
soils (0.2)
Cultivated, loam soils (0.4)
Cultivated, heavy clay
soils (0.5)
Cultivated, shallow soil on
rock (0.5)
Dunne and
Leopold
(1978)
Dunne and
Leopold
(1978)
Finally a weighted C value was calculated in excel spreadsheet for each watershed using the
following equation:
m
C
C=
n 1
m
n
An
A
n 1
n
where
C = weighted runoff coefficient
n = nth sub area with particular land cover type
m = total number of land cover classes in the watershed
Cn = runoff coefficient for nth land cover class
An = sub area for nth land cover classes in a watershed
Table 4 shows a comparison of weighted C values for 36 watersheds developed from
current method with “Table C” values and observed runoff coefficient values from Kirt Harle
(2003) from another TxDOT project supervised by Dr. David B. Thompson. The “Table C” or
predicted runoff coefficient was determined by subjectively allocating C as per the percentage of
watershed area represented by the corresponding type of land uses and land covers (Harle, 2003).
Runoff coefficients were averaged on an area-weighted basis to compute an estimate of the
composite watershed runoff coefficient (Harle, 2003). Standard runoff coefficient tables from
TxDOT Hydraulics Design Manual (TxDOT, 2002) were used to sign predicted runoff
coefficient for each land cover and land use type (Harle, 2003). Observed runoff coefficients
were determined from plots developed using a frequency matching procedure (Harle, 2003).
Comparison in Table 4 shows that values of composite runoff coefficient C (“Table C”)
obtained using NLCD land cover layer and GIS technique are similar to “Table C” and observed
C developed by Harler (2003), and they are within the satisfactory small difference as regards the
average, standard deviation, maximum and minimum. Finally the composite runoff coefficients
obtained for all 90 watersheds are presented in Table 5.
Table 4. Comparison of the composite C values obtained against Thompson Table and Observed
C Values from Harle (2003)
Source
Watershed ID
Harle (2003)
Table C
Auburn
Table C
08096800
0.36
0.37
08098300
0.38
0.39
08050200
0.41
0.31
08058000
0.36
0.34
08052700
0.41
0.33
08042700
0.44
0.31
08063200
0.36
0.32
08137000
0.36
0.30
08156800
0.68
0.55
08158700
0.40
0.37
08158840
0.46
0.43
08178640
0.46
0.48
08181400
0.48
0.43
08182400
0.34
0.37
08187000
0.38
0.33
08187900
0.39
0.35
08159150
0.37
0.34
08157000
0.65
0.55
08157500
0.68
0.56
08158050
0.55
0.54
08158600
0.51
0.54
08158930
0.44
0.50
08055600
0.65
0.55
08055700
0.65
0.52
08056500
0.65
0.54
08057160
0.63
0.63
08057420
0.50
0.53
08057425
0.48
0.53
08061950
0.57
0.55
08048520
0.51
0.50
08048820
0.39
0.48
08048850
0.37
0.49
08177600
0.42
0.46
08177700
0.42
0.53
08178300
0.65
0.55
08178690
0.67
0.62
Summary of Statistic Results on
Difference of C Values between Different
Methods (RMS is the root-mean-square
deviation)
Harle (2003)
Observed C
Table C
Difference
0.34
0.67
0.58
0.65
0.54
0.32
0.53
0.35
0.7
0.30
0.4
0.35
0.38
0.28
0.16
0.41
0.26
0.42
0.45
0.57
0.38
0.43
0.52
0.55
0.78
0.72
0.64
0.57
0.6
0.68
0.40
0.51
0.52
0.42
0.42
0.61
Average
RMS
Maximum
Minimum
0.01
0.01
0.10
0.02
0.08
0.13
0.04
0.06
0.13
0.03
0.03
0.02
0.05
0.03
0.05
0.04
0.03
0.10
0.12
0.01
0.03
0.06
0.10
0.13
0.11
0.00
0.03
0.05
0.02
0.01
0.09
0.12
0.04
0.11
0.10
0.05
0.06
0.04
0.13
0.00
Observed C – Table C
Auburn
Harle
0.03
0.28
0.27
0.31
0.21
0.01
0.21
0.05
0.15
0.07
0.03
0.13
0.05
0.09
0.17
0.06
0.08
0.13
0.11
0.03
0.16
0.07
0.03
0.03
0.24
0.09
0.11
0.04
0.05
0.18
0.08
0.02
0.06
0.11
0.13
0.01
0.11
0.08
0.31
0.01
0.02
0.29
0.17
0.29
0.13
0.12
0.17
0.01
0.02
0.1
0.06
0.11
0.1
0.06
0.22
0.02
0.11
0.23
0.23
0.02
0.13
0.01
0.13
0.10
0.13
0.09
0.14
0.09
0.03
0.17
0.01
0.14
0.10
0.00
0.23
0.06
0.11
0.08
0.29
0.00
Table 5. Composite runoff coefficients for all the 90 watersheds
2
Watershed_ID A (mile )
08178555
0.06
08177600
0.32
08178690
0.42
08178736
0.69
08050200
0.87
08057415
0.97
08048530
0.97
08048550
1.11
08058000
1.20
08181450
1.23
08057130
1.29
08048540
1.29
08187000
1.44
08055580
1.90
08052630
2.05
08057500
2.09
08157000
2.21
08094000
2.38
08178645
2.45
08178640
2.46
08048600
2.57
08057440
2.62
08155550
2.67
08156650
2.71
08139000
3.13
08178300
3.27
08158880
3.57
08178620
4.05
08137000
4.09
08157500
4.17
08159150
4.46
08057020
4.53
08057160
4.59
08096800
5.06
08158380
5.26
08181000
5.54
08048820
5.66
08055600
5.69
08158400
5.70
08057435
5.92
08158920
6.29
08156700
6.34
08056500
6.35
08042650
6.56
08057120
6.57
2
Composite C Watershed_ID A (mile )
0.46
08156750
6.84
0.46
08182400
7.15
08057320
7.17
0.62
0.54
08140000
7.32
0.31
08061620
7.68
0.54
08057418
8.06
0.55
08057140
8.63
0.59
08158840
8.76
0.34
08187900
8.78
0.44
08057445
8.92
0.62
08057050
9.47
0.60
08178600
9.60
0.33
08057425
10.32
0.53
08055700
11.03
0.29
08158500
12.12
0.35
08158810
12.29
0.55
08158050
12.62
0.34
08158100
12.73
0.48
08156800
12.74
0.48
08048850
12.84
0.56
08061920
12.87
0.37
08057420
14.37
0.53
08181400
14.89
0.60
08048520
17.62
0.32
08063200
18.16
0.55
08158930
18.72
0.49
08177700
20.82
0.52
08158825
21.00
0.30
08136900
21.72
0.56
08154700
22.76
0.34
08098300
22.96
0.54
08158860
23.20
0.63
08061950
23.29
0.37
08042700
23.97
0.63
08158820
24.48
0.46
08158200
26.41
0.48
08158970
27.36
0.55
08108200
46.34
0.63
08158600
53.54
0.41
08137500
69.18
0.46
08052700
73.04
0.58
08155200
89.57
0.54
08155300
116.53
0.33
08158700
123.61
0.59
08158800
167.16
Composite C
0.58
0.37
0.53
0.30
0.62
0.54
0.59
0.43
0.35
0.53
0.54
0.46
0.53
0.52
0.60
0.38
0.54
0.51
0.55
0.49
0.59
0.53
0.43
0.50
0.32
0.50
0.53
0.39
0.31
0.51
0.39
0.44
0.55
0.31
0.40
0.53
0.51
0.38
0.54
0.30
0.33
0.39
0.41
0.37
0.37
References:
American Society of Civil Engineers (ASCE) (1992). Design and Construction of Urban
Stormwater Management Systems. New York, NY: American Society of Civil Engineers.
ASCE (1996). Hydrology Handbook; ASCE Manual of Practice No. 28. New York, NY.
Dunne, T. and L. B. Leopold (1978). Water in Environmental Planning. San Francisco,
California: W. H. Freeman and Company.
Harle, K. (2003) Identify size limitations for hydrologic models. Master’s thesis, Department of
Civil and Environmental Engineering, Texas Tech University, Lubbock, Texas 79409.
Hydrology, I. (1976). Water balance of the headwaters oin catchemnts wye and severen. 19701975. Report No. 33, Institute of Hydrology, UK.
Law, F. (1956). The effect of afforestation upon the yield of water catchment areas. Technical
report, British Association for Advancement of Science, Sheffield, England
Mulholland PJ, Wilson GV, J. P. (1990). Hydro geochemical response of a forested watershed to
storms effects of preferential flow along shallow and deep pathways. Water Resources Research
26 (12), 3021{3036.
Schwab, G. O. and R. K. Frevert (1993). Elementary Soil and Water Engineering. Krieger
Publishing Company.
Texas Department of Transportation (TxDOT), 2002. Hydraulic Design Manual. TxDOT
Hydraulics Branch of the Bridge Division, Austin, TX 78701.
Thompson, D.B. (2006), The Rational Method, Regional Regression Equations, and SiteSpecific Flood Frequency Relations, Texas Department of Transportation, Research Project 04405,Research Report -044051
Wanielista, M. P. and Yousef, Y. A., 1993. Stormwater Management, John Wiley & Sons, Inc.,
New York.