GIS Calculation of Runoff Curve Numbers:
GIS Master Project
Spring 2006
By
Rocνo Huesca-Dorantes
Advisor: Tom Brikowski, Ph.D
Introduction:
The
study of hydrology at the watershed level is a main topic for geographic
research in the
Numerous
research projects have been done since the 1990s, focusing at the watershed
level:
reservoir and watershed
sedimentation, erosion, impact on lakes and coastal environment as well as the
effects of land use/land cover change over runoff, sedimentation, pollution,
streams and water bodies habitat and water
supply.[1]
Watershed is a basic
concept in all hydrologic designs. The best way to define a watershed is in
terms of a point: the watershed outlet. With respect to the outlet, the
watershed consists of all land area that sheds water to the outlet during a
rainstorm. Using the concept that water runs downhill, a watershed is
defined by all points enclosed within an area from which rain falling at these
points will contribute water to the outlet [2]
Rainfall
and snowfall are considered as the main form of Precipitation.
Precipitation can be
rain, snow, sleet and hail.
Abstractions
are losses that tend to reduce the volume of runoff in a watershed. The main
forms of abstractions are:
Infiltration,
depression storage, detention storage, interception, evaporation and
transpiration.
Of
all these, infiltration, depression storage, interception and transpiration
depend on ground cover, soil type and topography. Changes in land use/land
cover negatively influence these forms of abstraction.
Land
covers are especially important in Hydrology. The percentage of imperviousness
is a frequently used indicator of the level of urban development. High density
residential areas characteristically have percentages of imperviousness from
40% to 70%. Commercial and industrial areas are characterized by impervious
covers often from 70% to 90%.
Land Cover/Land Use: Qualitative description of land cover and use are
often transformed into quantitative index of runoff potential [3]
Objective
To
calculate hydrological parameter for an urban watershed using two different
approaches: A specialized hydrological analysis software, Watershed Modeling
Software (WMS) and Geographic Information System (GIS) approach.
Rationale
The
changes in land use/land cover are directly related to increments in runoff,
which deliver severe pollutant discharges to the watershed system, affecting
ecology and urban planning.
The
findings of this study might be of interest when no specialized software is
available for the calculation of hydrological parameters for a watershed.
Question
Can
calculations of hydrological parameters be performed in GIS?
Delimitations
The
area of study is
Limitations
Data availability from Catemaco lagoon:
Remote sensing images
with cloud coverage.
No shapefiles.
Data from
Shapefiles of soil type
and land use came with several text files that contain the attributes, but none
of them have explanation for codes meanings.
Shapefiles without area
information.
WMS
TR20 Model: Design to perform very detailed, small size urban watershed
Analysis.
Literature
review
Urbanization is a
global trend, and nowadays, almost half
the worlds population lives in urban areas, and that percentage is
expected to increase to 60% by 2030 (McGee, 2001).[4]
Urban development can
have a major impact in the local hydrology and water environment. Higher levels
of impervious surfaces result in higher volume of runoff with higher peak
discharge, shorter travel time and more severe pollutant loadings (Lee, 2003)[5]
Direct connectivity to
the drainage system is an important attribute of urban imperviousness. The
British Lloyd-Davies Rational Method assumes that "the directly connected
impervious area (DCIA) contributes 100% runoff to the whole urban catchments
(Lloyd-Davies 1906)."[6]
The size and land use
of a watershed are the most important factors for runoff estimations: size
usually remains constant over time while land use can change in a matter of
weeks. Of these factors, the one that significantly contributes to the
increasing rates of runoff, peak discharges, and time of concentration is the
land use.
Factors subject to change with general variations in
land use include the following:
- Permeable and impermeable areas
- Vegetation
- Minor topographic features
- Drainage systems.
All of these factors usually affect the rate and
volume of runoff that may be expected from a watershed.[7]
The effects on suburban
development on runoff (comparing these with the predevelopment conditions) are
very well known:
- Decrease in ground
water recharge,
- Increase in surface
runoff in annual stream flow,
- Increase magnitude of
peak runoff,
- Decrease of time of
concentration,
- Increase in the rates
of hydrograph rise and recession,
- Decrease in the mean
residence time of the stream flow. [8]
Urban
land covers are especially important in Hydrology. Many hydrologic design
problems result from urban expansion. The percentage of imperviousness is a
commonly used index of the level of urban development. High density residential
areas characteristically have percentages of imperviousness from 40% to 70%.
Commercial and industrial areas are characterized by impervious covers often
from 70% to 90%. In addition,
impervious covers in urban areas are not confined to the watershed surface.
Channels are often lined with concrete to increase the flow capacity of the
channel cross section and to remove flood waters quickly. Channel lined is
often criticized because it can transfer flooding problems from an upstream
reach to a reach downstream.[9]
Data
Sources
Land Use Shapefile
(1970):
WebGIS:
http://www.webgis.com/terr_pages/TX/luclutm/denton.html (UTM Zone 14 N, NAD 83)
Digital Elevation Model
(DEM):
USGS Seamless
Distribution Data System, Earth Resources Observation and Science (EROS) 2006.
http://seamless.usgs.gov/website/seamless/viewer.php
Soils Shapefile:
Geospatial Data Gateway
http://datagateway.nrcs.usda.gov (UTM Zone 14 N, NAD 83)
Software:
ESRI ArcGIS 9.1
Waterwhed Modeling Software (WMS) 7.0 (demo)
Analysis
and Methodology
Analyses in WMS
DEM, soil type and land
use shapefiles were analyzed using WMS, TR-20 Model, which is suitable for
urban watersheds (Figure 1).
Preprocessing:
Soils group shapefiles:
Find out which text
files was the one that include the required information: soil group and
hydrologic condition (muaggatt.txt)
Import it into Excel
(Table 1 in Appendix).
Made editions in Excel.
Export it as dbf IV,
so it could be added to ArcGIS as a table.
Join these tables with
the shapefiles using MUKEY field.
Dissolve-Merge-Dissolve,
using Soil Group field.
Clip to Watershed.
Convert Features to
Raster for the GIS analysis (Figure 2).
Preprocessing:
Land Use Shapefiles:
Dissolve-Merge-Dissolve:
Using the field LUCODE
to reduce the number of polygons, merging both shapefiles together (Dallas and
Sherman), and dissolve again to fuse the edge polygons.
Edition: to fill the
gaps between Dallas and Sherman Shapefiles.
Edition of the
attribute table to include de description of the land use (Table 2).
Clip to Watershed.
Convert Feature to
Raster for the GIS analysis (Figure 3).
Analysis
in GIS (Raster)
To calculate the Curve
Number (CN) is necessary to have:
Runoff Curve Number for
each one of the combinations of land use-soil type. (Table 3 in Appendix).
Areas of each soil
type-land cover combination (when converting features to raster, setting the
cell size to 30 meters will let you know the area).
In order to obtain a
unique code to identify each one of the soil type-land use combination, a
reclassification of both raster files has to be done (Table 3).
|
|
|
Land Use
Codes and Modified Codes |
|
|
|
||||||||
|
SoilGroup |
|
11 |
12 |
13 |
14 |
15 |
17 |
21 |
24 |
41 |
53 |
75 |
76 |
|
|
1 |
4 |
5 |
7 |
9 |
10 |
11 |
13 |
16 |
17 |
19 |
23 |
|
|
B |
1 |
1 |
4 |
5 |
7 |
9 |
10 |
11 |
13 |
16 |
17 |
19 |
23 |
|
C |
2 |
2 |
8 |
10 |
14 |
18 |
20 |
22 |
26 |
32 |
34 |
38 |
46 |
|
D |
3 |
3 |
12 |
15 |
21 |
27 |
30 |
33 |
39 |
48 |
51 |
57 |
69 |
After this is done,
assign the value for each specific combination of land use-soil group and calculate CN value
multiplying for the percentage of the area covered for that specific soil type-land
use combination (Figure 4, Table 4)
Results:
Figure 1:
Figure 1
Figure
2:
Figure 2
Figure
3:
Figure 3
Table 2.
- Land Use Codes and Description
|
|
|
|
LandUse |
|
|
11 |
Residential |
|
12 |
Commercial
Services |
|
13 |
Industrial |
|
14 |
Transportation,
Communications |
|
15 |
Industrial
and Commercial |
|
17 |
Other
Urban or Built-up Land |
|
21 |
Cropland
or Pasture |
|
24 |
Other
Agricultural Land |
|
41 |
|
|
53 |
Reservoirs |
|
75 |
Strip
mines, Quarries, and Gravel Pits |
|
76 |
Transitional
Areas |
Figure
4:
Figure 4
After
the analysis in WMS, the average CN Value for White Rock Lake Watershed is:
87.4
After
the analysis in GIS, the average CN Value for White Rock Lake Watershed is: 85.7658
(Table 4)
Conclusions
The outcomes obtained
in WMS were more accurate, but the analyst needs to enter specific hydrological
data.
It is possible to
perform hydrologic analysis using ArcGIS (without using ArcHydro);
one needs the adequate shapefile and tabular data values.
The results were close
to those obtained in Watershed Modeling Software.
This is a useful
approach if you do not have specialized software at your convenience.
Glossary[10]
Watershed: A watershed is a region
of land where water drains downhill into a specified body of water, such as a
river, lake, sea, ocean or wetland. A watershed includes both the waterway and
the land that drains to it. Each watershed is separated topographically by a
ridge, hill or mountain.
Runoff: Rainfall not absorbed by soil. Surface runoff is water, from
rain, snowmelt, or other sources, that flows over the land surface, and is a
major component of the water cycle. Runoff that occurs on surfaces before
reaching a channel is also called overland flow.
Impervious surfaces: Impervious surfaces are artificial structures, such as pavem