PREDICTION OF ICELAND'S SOIL EROSION USING GIS

Iceland is battling a soil erosion problem throughout the country due to a lack of significant vegetation cover, steep terrain, high rainfall, and frequent volcanic activity. The goal of this study was to estimate Iceland’s soil erosion by calculating the revised universal soil loss equation (RUSLE) using a geographic information system (GIS). The results were compared to a 1998 field study of Iceland’s soil erosion by the Agricultural Research Institute of Iceland and the Icelandic Soil Conservation Service. Remote sensing data were used to derive estimates for spatially distributed soil loss equation factors. Of these factors, two—the conservation practice and soil erodibility—were held constant for the entire country and thus not used. The cover management factor was computed using a 1 kilometer resolution Greenland and Iceland land cover map from 2000 published by the European Commission Joint Research Centre. The rainfall erosivity factor was estimated using annual rainfall averages from 1971-1993 that were obtained from the Icelandic Meteorology Office. These points were encoded onto a map of Iceland and projected into contours. The slope steepness and slope length factors were estimated using a digital elevation model (DEM) from the ETOPO2 Global 2’ Elevation data set. The cover management, rainfall erosivity, and topography factors were converted to a raster projection so that maximum RUSLE factors could be calculated to display predicted areas of high erosion. The individual RUSLE factor predictions were overlain to predict Iceland’s areas of maximum potential soil erosion. When compared to published field data, the GIS- generated RUSLE underpredicted the observed soil erosion for Iceland, indicating that the maximum factor combination is not driving soil erosion in Iceland. It appears that the land cover of Iceland is driving the soil erosion because the maximum potential soil erosion map for land cover appears to correlate well with the observed soil erosion.