Assessing sediment connectivity and its spatial response on land use using two flow direction algorithms in the catchment on the Chinese Loess Plateau

Severe soil and water loss have led to widespread land degradation on the Loess Plateau in China. Exploring the relationship between land use and sediment connectivity can be beneficial to control soil erosion. In this study, three catchments in the Yanhe River Basin on the Loess Plateau were selected to analyse the relationship between land use and sediment connectivity using grey correlation method. Index of connectivity (IC) was employed to quantify sediment connectivity, including two flow direction algorithms (D8 and D-infinity) and two final targets of sediment transport (outlet and main channel of catchment). Then, 11 landscape metrics were used to evaluate the land use spatial patterns of catchments. By comparing the IC value ranges, histograms and classes, and their relationship with remote sensing images of the two flow direction algorithms, we find that the D8 algorithm is more suitable for this study area. The results showed that the three catchments are characterized by high sediment connectivity in the grassland and forest close to the channel. In addition, the roads and bare land close to the channel also have high or medium sediment connectivity. Grey correlation analysis showed that landscape division index (DIVISION), fractal dimension index (FRAC-MN), aggregation index (AI), total class area, patch cohesion index (COHESION), and largest patch index (LPI) indices were the main factors that affect sediment connectivity at the class scale. At the landscape scale, the landscape shape index (LSI), Shannon's diversity index (SHDI), and gully density have an essential effect on sediment connectivity. This condition provides a way to control the sediment connectivity in the watershed by transforming land use type or changing its spatial pattern, but specific adjustment measures have to be further explored.

Automated summary

Severe soil and water loss on the Loess Plateau in China has caused widespread land degradation. This study explores the relationship between land use and sediment connectivity, finding that grassland and forest close to the channel have high sediment connectivity. Furthermore, roads and bare land near the channel also exhibit high or medium sediment connectivity. Landscape metrics such as landscape division index, fractal dimension index, and aggregation index are identified as important factors affecting sediment connectivity at the class scale, while landscape shape index, Shannon's diversity index, and gully density have essential effects at the landscape scale.