Performance of the Rangeland Hydrology and Erosion Model for runoff and erosion assessment on a semiarid reclaimed construction site

Abstract

The ability to assess the impact of management actions on soil and water resources is crucial to sustainable land management. The Rangeland Hydrology and Erosion Model (RHEM) was developed as an assessment and decision support tool for resource management agencies and has been used to estimate soil erosion at national, regional, and local scales for both disturbed and undisturbed rangeland soil and vegetation conditions. In this paper, runoff, sediment, and microtopography data were collected during a series of rainfall simulation experiments aimed at evaluating the effectiveness of revegetation on a shrub-dominated post-construction hillslope where the soil has had time to reconsolidate. RHEM's input parameters are simple to collect and consist of soil texture, slope configuration, and canopy and ground cover. Experimental results showed that a mix of the shrub species rabbitbrush (Chrysothamnus nauseosus) and the invasive annual grass cheatgrass (Bromus tectorum) were more effective at reducing runoff and soil loss than rabbitbrush alone. Plots with more than 45% of residue cover produced as much as 2.1 times less runoff and 16 times less sediments than those with less than 15% of litter. Microtopographic changes information allowed partitioning the erosion process into diffuse and concentrated flow processes. Analysis of this 3D data highlighted the central role of concentrated flow erosion in sediment delivery on rangeland hillslopes. RHEM's performance as expressed by the coefficient of determination, R², and the Nash-Stucliffe Efficiency (NSE) was R² = 0.84 and NSE = 0.27 for runoff and R² = 0.81 and NSE = 0.26 for sediments. This study demonstrates that RHEM can be effectively used by land managers and project managers to estimate soil erosion as a function of precipitation event on construction sites that have been revegetated to rangeland plant communities.