Soil erosion is one of the most serious environmental threats strongly influenced by the spatial pattern of land uses. This study was designed to evaluate the relevance of land use pattern and soil erosion using landscape metrics across Gorgan Watershed in northern Iran. Therefore, the revised universal soil loss equation was applied to evaluate and model soil loss and sedimentation in the region. Then, soil erosion relationship to land use pattern was analyzed using a variety of metrics including percentage of landscape, number of patches, largest patch index, and landscape shape index. The results revealed that potential of soil loss, sediment retention, and sediment yield for the whole watershed were 6.6, 2.4, and 1.5 t ha-1 year-1, respectively. The quantity of sediment retention was estimated at 4.3, 3.2, 1.0, and 1.2 t ha-1 year-1 in forest, rangelands, agriculture, and built-up areas, respectively. Similarly, sediment yield was 0.6, 1.6, 1.5, and 2.1 t ha-1 year-1, respectively. The results revealed that the soil loss increased with decreasing metrics of forest and rangelands while increasing metrics of built-up regions and agricultural lands accelerated the process. Moreover, we showed that land use type of patches was an important factor on soil erosion, and soil loss was also affected by area, number, shape, and density of landscape patches. Result of this study can facilitate monitoring of erosion-sensitive areas in the watershed which can help managers and decision makers to design more suitable measures for soil conservation.
Keywords: InVEST; Land use; Landscape metrics; RUSLE; Sediment; Soil erosion.