Assessing effects of landscape pattern on sediment yield using sediment delivery distributed model and a landscape indicator

The rationale behind this research concerns the need to better understand relationships between landscape pattern and soil loss processes. Landscape indicators are commonly used to delineate these relationships. However, most indicators were not developed on the basis of soil loss progresses, and therefore their specific relationships with soil loss are difficult to construct. We improved the Location-weighted landscape Contrast Index which was developed based on sediment source-sink theory. This indicator encompasses three factors of landscape pattern: contribution of land cover types to soil erosion: composition and configuration of land covers. To analyze correlations between the landscape indicator and soil loss processes, variables expressing soil loss status should be first quantified.Therefore we applied the sediment delivery distributed model which incorporates revised universal soil loss equation (RUSLE) and sediment delivery ratio. The methods were applied in the Heishui and Zhenjiangguan subwatershed in the Upper Min River. Modeling results showed that 90% of the study area experienced soil erosion larger than 10 t/(ha yr). While its sediment yield rate was smaller than most tributaries in the Upper Yangtze River. Results of Pearson correlation analysis indicated that the rainfall factor in RUSLE was the dominant control and explained 93% of variance in sediment yield rate. We suggested preserving and increasing percentage cover of forests to adapt to climate change for soil conservation. Besides precipitation, landscape pattern was a principle factor correlated with sediment yield rate. The landscape indicator was significantly correlated to sediment delivery ratio, and explained 98% variation in sediment yield rate not considering precipitation by dropping the rainfall factor. The landscape indicator indicated that the landscape pattern was generally favorable for soil conservation in the two subwatersheds. This advantage was mainly ascribed to superior sediment sink-source compositions. In the Heishui subwatershed, however, sediment source contributed more to soil erosion processes than sink. Spatial configuration of sediment source and sink related to flow length were the most unfavorable factor, mainly due to the crops located in river vallyes. We gave suggestion to further optimize the landscape pattern: reducing percentage cover of grasslands with high soil erosion rate; decreasing soil erosion rate of sediment source in the Heishui subwatershed; increasing lengths of flow path from crops to river channels. (C) 2011 Elsevier Ltd. All rights reserved.