Evaluation of Sediment Connectivity Indices to Improve the Prediction of the Spatiotemporal Variability of Sediment Yield for a Large River Basin (Wei River, China)

ABSTRACT Sediment connectivity between source areas and the main streams or local sinks is a complex and dynamic process, especially in large basins due to multiple heterogeneities and interactions between connectivity components. Sediment connectivity indices are promising tools to investigate sediment transport, especially in data‐scarce or large areas. The InVEST‐SDR numerical approach couples RUSLE gross erosion estimates with the Index of Connectivity (IC) to derive sediment delivery. However, involving functional connectivity and validating the results remains challenging. In a first estimate, we used the coupled InVEST‐SDR approach to calculate the annual sediment yield in the entire Wei River Basin (134,800 km 2 ) and three of its sub‐catchments. Then, we replaced the IC with the Aggregated Index of Connectivity (AIC), which includes functional connectivity aspects. Computational results were compared with observation data from 26 hydrometric stations to verify the performance of the simulations. Both IC and AIC performed well in predicting sediment yield, with R 2 > 0.91. The areas with the highest connectivity (90th percentile—P90) also showed high values of erosion: 54% of the P90 values were found in the three catchments with the highest observed sediment yield. The rainfall erosivity and soil permeability factors were found to be the main explanatory components of the difference in spatial domination of structural (no temporary changes) and functional (temporally dynamic) connectivity. This study demonstrated the accuracy of AIC for sediment transport and yield evaluation in a large river basin. This method is potentially beneficial for land management in large basin areas with insufficient data.