Predicting Sediment and Phosphorus Delivery with a Geographic Information System and a Computer Model M.S. Richardson and A. Roa-Espinosa; Dane County.

Predicting Sediment and Phosphorus Delivery with aGeographic Information System and a Computer ModelDane County Land Conservation DepartmentM.S. Richardson and A. Roa-Espinosa

PurposeExcessive sediment and phosphorus amounts in surface water have negative impacts onits aquatic and wildlife habitat, recreational use, and aesthetic qualities. In order tomaintain or improve water quality, local, state, and federal agencies are identifyingsources of sediment and phosphorus and predicting their delivery to receiving waters. AGeographic Information System (GIS) database was developed for a 443-km2 (171 sq.mile) watershed in southern Wisconsin to provide inputs for WINHUSLE, a distributedparameter, water quality model based on the Universal Soil Loss Equation (USLE). Theresults of the model are used to identify locations for rural and urban conservationpractices that provide the most water quality benefit across the watershed.

Data Sources (digital and tabular)We used seven main digital data sources including: soils, farm tract and fields (1997),land use from the Dane County Regional Planning Commission (1990), NaturalResources Conservation Service (NRCS) wetlands, zoning from cities and villages(1997), orthophotograph (1995), and digital elevation model (10 m2 grid cell size). Weused three primary tabular data sources including: soils information from the WisconsinComputer Assisted Management and Planning System (CAMPS) and NRCS WisconsinTechnical Guides and farm tract and field data from the Dane County Land ConservationCooperator Tracking System (CTS) database.

Analytical TechniquesThe WINHUSLE water quality computer program was run in R:BASE (a relationaldatabase) using nine R:BASE tables. Three tables already existed from the CAMPS soilsdatabase. Six new tables were created from watershed specific data, most using GIS datalayers. All GIS processing was completed with ARC/INFO Version 7.1.2 fromEnvironmental Systems Research Institute, Inc. The three major GIS data layers usedwere: 1) land use/land cover (created from new and existing digital data), 2) soils (by soilmap unit), and 3) Digital Elevation Model (DEM). The land use and soils layers werecombined in the GIS to create USLE factors. The DEM was used to generate a digitaldrainage area data layer from a pour point layer with ARC/INFO GRID commands.Inputs for WINHUSLE from the USLE, drainage area, and overlay of USLE/drainagearea data layers were downloaded into ASCII files. Finally, the files were imported intoR:BASE, calibration R:BASE tables were adjusted, and two versions of WINHUSLE(1.4.6 and 1.5) were run. The poster was created using ArcView 3.0a and MicrosoftPowerPoint 97.

Application CategoryNatural Resources

0 2 4 6 8 Kilometers

260 - 275275 - 290290 - 305305 - 320320 - 335335 - 350350 - 365365 - 380

Major stream

Explanation

Elevation (in meters)

Watershed boundary

0 10 20 30 Kilometers

City, village, or major road

Major lake

Upper Sugar Watershed boundary

Explanation

N

Predicting Sediment and Phosphorus Deliverywith a Geographic Information System and a Computer Model

M.S. Richardson and A. Roa-Espinosa; Dane County Land Conservation Department; Dane County, Wisconsin, USA

Introduction:Excessive amounts of sediment and phosphorus in surface water, negatively impact its aquatic and wildlife habitat, recreational use, and aesthetic qualities. Sources of sediment and phosphorus include both agricultural and urban land uses. To maintain or improve water quality, local, state, and federal agencies are identifying sources of sediment and phosphorus and predicting their delivery to receiving waters.

Objectives:Our objectives were to predict sediment and phosphorus delivery for the Upper Sugar River Watershed (a 443-km2 watershed in southern Wisconsin) by:

1. developing a Geographic Information System (GIS) database for the study area;

2. using the GIS data as inputs to WINHUSLE, a distributed parameter, water quality computer model based on the Universal Soil Loss Equation (USLE).

Methods:The WINHUSLE (Wisconsin Nonpoint Hydrologic Universal Soil Loss Equation) computer program was run in R:BASE (a relational database) using nine R:BASE tables. Three tables already existed from a federal soils database. Six new tables were created from watershed specific data, most using GIS data layers.

The three major GIS data layers used were:

1. Land use/Land cover

2. Soils (by soil map unit)

3. Digital Elevation Model (DEM)

The land use and soils layers were combined to create USLE factors (see Figure 1). The DEM was used to generate a digital drainage area data layer. Inputs for WINHUSLE from the USLE, drainage area, and overlay of USLE/drainage area data layers were downloaded into ASCII files. Finally, the files were imported into R:BASE, calibration R:BASE tables were adjusted, and WINHUSLE was run.

Study Area:

State of Wisconsin, USA

Sugar River Watershed in Dane County

Wisconsin

Canada

Sugar

River

Watershed

Lake Belle View is a shallow lake created from a dam at the lowest point of the Sugar River Watersheds in Dane County. It has several water quality problems usually associated with impoundments including: sedimentation, turbidity, excessive rooted aquatic plants and attached algae, and free floating bluegreen algae to name a few. We hope the results of our study will be used to reduce the amount of sediment reaching the lake and thus address lake water quality problems without resorting to costly dredging.

Mexico

Dane County Land Conservation

Results and Discussion:The WINHUSLE model estimated that the outlet of the Sugar River Watersheds (located at Lake Belle View) receives ~6,700 MT of sediment and 20,000 kg of phosphorus per year. Sediment delivery was also calculated across the landscape and results were used to identify areas with relatively high delivery. These areas can be examined in the field, and management practices (including conservation tillage, buffer strips, terraces, or water and sediment control basins) can be implemented to reduce sediment delivery.

Combining GIS data with a computer model was an effective and efficient method to acquire sediment and phosphorus deliveries for a large area (443-km2). The most time consuming process was the creation of a digital land use layer; however, this layer can be used in other applications.

Although model results were not compared with monitored values in the Sugar River Watershed (due to lack of data), WINHUSLE sediment and phosphorus numbers have been shown to reflect monitored data in other watersheds in Dane County.

Figure 1. Flow chart of data used to run WINHUSLE.

Land use / Land cover

Soils (K factor)

Hydrologic Drainage Areas

Cumulative Sediment Yields

Sediment Delivery by Field

The water quality of the Upper Sugar River may be threatened by agricultural and increasing urban runoff in the watershed.

An example of a brown trout that could be found in the Sugar River Watershed.

Poster printed February 1999.

† Universal Soil Loss EquationA = R*K*LS*C*P

A - soil loss rate (MT/ha)

R - annual rainfall factor

K - soil erodibility factor

LS - slope length/slope gradient factor

C - land cover/cropping management factor

P - erosion control practice factor

To run WINHUSLE, the watershed must be sub-divided into hydrologic drainage areas -- at a minimum, at the confluence of every stream.

We used a DEM to divide our watershed into 504 drainage areas approximately 0.9-km2

in size.

K factor or soil erodibility represents the ability of the soil to resist erosion. It is a function of the surface soil texture and composition.

Note that the LS factor (slope length/slope gradient factor) is also based on soil information but is not displayed.

WINHUSLE generated both sediment and phosphorus yields (only sediment is shown) for each field or land use to the outlet of the hydrologic area(s) that it’s in.

WINHUSLE generated both sediment and phosphorus yields (only sediment is shown) for each of the 504 hydrologic drainage areas.

croplandgrasslandpasturewoodlandwetlandwaterroadfarmstead/sub-divisionlow residentialmedium residentialhigh residentialcommercialindustrialdeveloping

watershed boundaryExplanation

0 2 4 6 8 Kilometers

0.01 - 0.050.05 - 0.17

0.17 - 0.280.28 - 0.370.37 - 0.49

Watershed boundary

K factor (soil erodibility)

Explanation

Major stream

0 2 4 6 8 Kilometers

Watershed boundary

Explanation

Major stream

Drainage network

Internally drained area

0 2 4 6 8 Kilometers

Watershed boundary

Sediment yield (MT/year)

Explanation

Major stream

0 2 4 6 8 Kilometers

0 - 450450 - 12001200 - 21002100 - 34003400 - 6800

Internally drained area

Drainage network

Internally drained area

0 2 4 6 8 Kilometers

Major stream

Explanation

Sediment delivery (MT/ha/yr)

Watershed boundary

22 - 7011 - 226 - 110.2 - 60 - 0.2

Funding for this project was provided by the Wisconsin Department of Natural Resources and the U.S. Environmental Protection Agency.

Digital Elevation Model (DEM)

The Digital Elevation Model used had a vertical elevation to the nearest 0.3 meter for every 10 meters on the ground.

RoadsRoads

City/village zoning

City/village zoning

Wetlands

Wetlands

Census land useCensus land use

Farm fieldsFarm fields

Land use/

cover

Land use/

cover

SoilsSoils DEMDEM

Farm field database

USLE factors†USLE factors†

Hydrologic unit data

Hydrologic unit data

Run WINHUSLE

Sediment and phosphorous delivery

results

Sediment and phosphorous delivery

results

WINHUSLE inputs

WINHUSLE inputs