Land management controls on snow accumulation in a cold-region agricultural watershed

Over the past three decades, the Northern Great Plains (NGP) has experienced a novel wet climatic continuum influencing snow processes. Amid this wet condition, land management practices and other watershed properties also exert strong control on snow accumulation. However, the influence of climate, vegetation (crop stubble) and topography on snow processes and streamflow generations are poorly understood due to lack of field-based snow accumulation data. Most studies are limited to remotely sensed snow cover estimates or numerical modeling with very limited snow data. To gain an improved understanding of the hydrological processes to climatic variability and land management practices in the NGP, a series of detailed snow surveys were conducted at distributed locations (ten locations) in the Mauvais Coulee Basin (a headwater basin draining to Devils Lake) during the winters of 2017-2023. We also utilized a field-tested cold region hydrologic model (CRHM). The wettest winter year was 2023, with an average snow water equivalent (SWE) of 99 mm. In contrast, 2021 was dry, with only traces of SWE. The influence of land management practices (via remotely sensed tillage index and CRHM modeling), topography, and climatic parameters on snow accumulation were investigated. The relationship between the tillage index and SWE is stronger during wet winters, and SWE also exhibits systematic variation with elevation along the north-south gradient. Together, these patterns indicate that both land management and physiographic setting shape the spatial distribution of SWE. Consistent with this, air temperature and snowfall emerge as the most influential climatic factors controlling SWE, while a machine learning (ML) model applied to CRHM-simulated data further shows that crop stubble height exerts a stronger control on snow accumulation than topography.