Revising the SCS-CN model by using the effect of biocrusts to refine the results

Abstract

Water resource shortage typically creates serious challenges in dryland ecosystems. Accurate prediction of runoff during rainfall events is intrinsic to the scientific management of water resources and soil erosion control especially in dryland ecosystem where the vascular plant is sparse. Furthermore, biological soil crusts (biocrusts), the ubiquitous surface coverings in many dryland ecosystems, play a vital role in the water cycle and thus in runoff generation. However, the relationship between biocrust coverage and runoff yield and the effect of biocrust on SCS-CN model were not fully researched. Accordingly, we collected 364 sets of runoff data from 13 rainfall events across three watersheds of the Loess Plateau of China. The relationship between runoff yield and biocrust coverage under natural rainfall was examined, and the revised SCS-CN model, which accounted for the previously proposed biocrust coverage, was verified. We found that (1) runoff decreased as biocrust coverage increased at rainfall intensities below 43.2 mm·h⁻¹. Specifically, a 10 % increase in biocrust coverage resulted in a runoff reduction ranging from 6.9 % to 19.8 %. (2) The simulation accuracy of the revised SCS-CN model was significantly improved by considering the effect of biocrusts. Compared with that of the classic model, the Nash-Sutcliffe efficiency coefficient (NSE) of the revised model increased by 100.2 %, while the Percent bias (PBIAS) and the Root mean square error (RMSE) decreased by 97.0 % and 65.7 %, respectively. (3) The revised SCS-CN model showed better accuracy for runoff when the coverage of biocrusts ranged between 40 % and 60 %. Accordingly, we recommend using a curve number (CN) of 66 and an initial abstraction ratio (λ) of 0.05 for sites with biocrust coverage ranging between 40 % and 60 %. This study confirmed that biocrusts inhibit runoff under natural rainfall conditions. In addition, for the first time, the revised SCS-CN model that accounted for biocrust coverage was verified by natural rainfall in three watersheds. These results advance the understanding of the runoff effect of biocrusts and demonstrate that the revised SCS-CN model is effective for runoff prediction in dryland ecosystems.