Presenter/Author Information

Reynaldo L. Lanuza
Eduardo P. Paningbatan Jr

Keywords

catchment runoff and erosion simulation technology (crest), soil erosion, gisassisted modeling, geographic information system, pcraster

Start Date

1-7-2010 12:00 AM

Abstract

Most erosion models have been developed based on a plot scale and have limited application to a watershed due to the differences in scale. In order to address this limitation, a GIS-assisted methodology for modeling soil erosion was developed using PCRaster to predict the rate of soil erosion at watershed level and identify the location of erosion prone areas. The GIS-assisted hydrology and erosion models were validated at Tanghaga Watershed using the observed values from previous experiment. The predicted peak rates, Qp, showed a highly significant relationship with the observed Qp, with an r2 value of 0.75. For soil loss prediction, a significant relationship was also noted with an r2 value of 0.74. Sensitivity analysis using four parameters was done. The model response was most sensitive to Manning’s roughness coefficient (n) for Qp. An increase in n value from 0.02 to 0.13 resulted in a decrease of 546% in the predicted Qp. On the other hand, the predicted soil loss was most sensitive to the vegetative cover. Increasing the value of vegetative cover form 0.20 to 0.95 resulted in decrease of about 1,567%. The location of erosion hotspots was predicted within and along the tributary channels as well as in areas with low vegetative cover and steeper slope gradient. The capability of GIS-assisted model in predicting the location of erosion hotspots is a significant finding and this approach is a valuable tool in the formulation of a good watershed rehabilitation program.

Share

COinS
 
Jul 1st, 12:00 AM

Validation and Sensitivity Analysis of Catchment Runoff and Erosion Simulation Technology (CREST): A GIS-assisted Soil Erosion Model at Watershed Level

Most erosion models have been developed based on a plot scale and have limited application to a watershed due to the differences in scale. In order to address this limitation, a GIS-assisted methodology for modeling soil erosion was developed using PCRaster to predict the rate of soil erosion at watershed level and identify the location of erosion prone areas. The GIS-assisted hydrology and erosion models were validated at Tanghaga Watershed using the observed values from previous experiment. The predicted peak rates, Qp, showed a highly significant relationship with the observed Qp, with an r2 value of 0.75. For soil loss prediction, a significant relationship was also noted with an r2 value of 0.74. Sensitivity analysis using four parameters was done. The model response was most sensitive to Manning’s roughness coefficient (n) for Qp. An increase in n value from 0.02 to 0.13 resulted in a decrease of 546% in the predicted Qp. On the other hand, the predicted soil loss was most sensitive to the vegetative cover. Increasing the value of vegetative cover form 0.20 to 0.95 resulted in decrease of about 1,567%. The location of erosion hotspots was predicted within and along the tributary channels as well as in areas with low vegetative cover and steeper slope gradient. The capability of GIS-assisted model in predicting the location of erosion hotspots is a significant finding and this approach is a valuable tool in the formulation of a good watershed rehabilitation program.