Report Number: 117
Year: 2007

Developing a GIS-Based Soil Erosion Potential Model of the Ugum Watershed

Soil erosion is defined as the physical degradation of the landscape over time. The process is initiated when soil particles are detached from its original configuration by erosive forces such as rainfall. The soil particles may then be transported by overland flow into nearby rivers and oceans. Prior research has demonstrated that large sediment loads damages the coral reefs (Rogers 1990).

Current developments in geographic information systems (GIS) make it possible to model complex spatial information. GIS is used in this project to determine how soil erosion potential varies throughout a watershed. Hydrological data is also analyzed to give some understanding of the watershed response to the primary erosive input: rainfall. The two goals of this research project were: 1) to develop a GIS-based soil erosion potential model of the Ugum Watershed, located near the southern village of Talofofo, Guam and 2) to develop a correlation between recorded rainfall, stream flow, turbidity and suspended sediment concentration.

A method was developed in this research which combines the Universal Soil Loss Equation (USLE) with the computer capabilities of a GIS, specifically the commercial software package ArcGIS®. The USLE calculates long-term average annual soil loss by multiplying six specific factors which describe the watershed characteristics such as rainfall, soil types, slope, and vegetation cover. The GIS is used to store the USLE factors as individual digital layers and multiplied together to create a soil erosion potential map. This combination provides a way to assess soil erosion potential of an area with existing data sources. A digital elevation model (DEM) is used to calculate the slope steepness and slope length factors. Existing soil survey maps created by the Natural Resource Conservation Service (NRCS) are used to define the soil erodibility factor. A prior research on determining the R factor values for Southern Guam area is used in this study (Dumaliang 1998). Recent satellite imagery is used to determine the extent of vegetation cover and conservation practices.

In addition to developing the GIS model, a preliminary hydrological analysis was conducted. Recorded data for rainfall, stream flow, turbidity levels, and suspended sediment concentration levels were compiled and graphically analyzed. General trends were examined by correlating one hydrological variable with another.

Shahram Khosrowpanah
Leroy F. Heitz
Yuming Wen
Michael Park