Hydrological effects of the snow fraction and its ecohydrological explication within the Budyko framework

Abstract

Understanding the effects of climate and catchment characteristics on hydrological behavior is critical in hydrological sciences. Furthermore, improving the clarity associated with the physical meaning of the Budyko hypothesis on how it can be incorporated into ecohydrological processes remains a significant challenge, while the conciseness it affords in addressing hydrological responses has promoted its application. This study explores the hydrological effects of the snow fraction (fs) and assesses the ecohydrological meaning of the catchment parameter (n) within the Budyko framework, employing 1576 globally investigated catchments. Results showed that the Budyko hypothesis decreased systematic errors (NS = 0.76, RMSE = 181.01 mm a−1) when incorporating the fs compared with no snow fraction incorporated it (NS = 0.54, RMSE = 248.85 mm a−1), especially for catchments with higher fs. In these catchments, a higher fs was generally consistent with a higher normalized mean streamflow (Q), while the opposite was true for normalized evaporation (E/P). Under the Budyko framework, n and fs distribution generally exhibited opposite trends. Moreover, higher n was mainly distributed near water limited and the highest energy limited boundaries, while higher fs represented lower (P-Q)/P (i.e., the mean (P-Q)/P was 0.59, 0.57, and 0.38 for catchments with lower (fs < 0.15), medium (0.15 ≤ fs < 0.50), and higher (fs ≥ 0.50) fs). Statistical relationships among ecohydrological processes and catchment parameter suggested that: (i) higher storm depth (α) generally indicated higher normalized mean Q; (ii) a higher plant-available water holding capacity (κ) resulted in a reduction in Q/P, especially when κ was > 0.12; (iii) the effective rooting depth (Ze) exhibited a complex regulatory effect on P partitioning; and (iv) α, Ze, and κ controlled n by 65%, 22%, and 13%, respectively, in all catchments. The ecohydrological processes incorporated into catchment parameters will broaden the ability of the Budyko hypothesis to describe hydrological behavior in a changing world, especially in the ungauged catchment.