Modeling connectivity and sediment reduction with PROSEB: a study of prairie strips in agricultural catchments

Understanding how vegetative barriers influence sediment transport at the catchment scale remains challenging due to the partial dynamic nature of hydrologic dysconnectivity. This study introduces PROSEB, a model designed to assess sediment retention performance of buffer strip type structures under varying configurations in small agricultural catchments. The model evaluation integrates landscape hydrologic connectivity theory with six years of field data from 12 experimental catchments in Iowa, USA, ranging from 0.5 to 3.5 ha and containing single or multiple prairie strips covering 10% to 20% of the land area. Across 263 rainfall-runoff events, experimental sediment trapping efficiency in treated catchments ranged from negative values to over 90%, with cumulative efficiencies reaching up to 99%. The year 2011 presented notably lower average trapping efficiency, 61.8%, and included 12 events of negative efficiency, attributed to factors like strip breaching or rill development. Sediment dysconnectivity, P(C), values ranged from 0 (complete trapping) to 1 (no retention), revealing substantial temporal and spatial variability. Catchments with wider prairie strips at the outlet showed a higher probability of buffer dysconnectivity, P(B), though multiple narrow strips enhanced system redundancy. This conclusion is based on an interpretation of the results of the model. A logarithmic relationship between buffer width and trapping efficiency (R² = 0.46) supported differential performance of strip configurations. The model applied probabilistic functions to assess sediment connectivity and buffer performance, calibrated using both average and cumulative sediment trapping efficiency. Buffer performance values varied annually, with cumulative values presenting greater consistency. Despite challenges in predicting sediment yield at the event scale, long-term P(C) correlated strongly with measured sediment delivery (R² > 0.9 when control and treated catchments were combined). PROSEB's probabilistic nature captures key features of sediment dynamics and offers flexibility for evaluating vegetative conservation practices. Its capability to accommodate uncertainty and partial disconnection makes it well suited for both empirical assessment and scenario modelling, and its integration with erosion models could further enhance its predictive capacity for soil-conservation planning.