Assessment of snowmelt and groundwater-surface water dynamics in mountains, foothills, and plains regions in northern latitudes

Snowmelt and groundwater-surface water (GW-SW) dynamics are the most dominant controllers of hydrological processes and water availability in northern latitudes, particularly in warm seasons. The dynamics and changes of snowmelt and GW-SW interactions and their interrelationship at the regional scale are poorly understood in regional analyses. To fill this gap, this study implements a process-based hydrological model, using a coupled Soil and Water Assessment Tool (SWAT) and MODFLOW model, to map snowmelt and GW-SW dynamics at regional scales. The primary research questions are "how are GW-SW interactions altered by snowmelt dynamics?" and "what is the effect of climate change on snowmelt and GW-SW dynamics?" We simulated snowmelt and groundwater dynamics across a large watershed encompassing different ecohydrological regions, including mountains, foothills, and plains, analyzing the spatio-temporal relationships between GW-SW interactions and snowmelts in historical (1983-2007) and future (2040-2064) periods. We used an ensemble of five global climate models (GCMs) for future projections based on two emission scenarios (i.e., RCP2.6 and RCP8.5). We also used the Least Absolute Shrinkage and Selection Operator (LASSO) method to examine the effects of snowmelt, evapotranspiration (ET), and rainfall on GW-SW interactions. These analyses are implemented in mountains, foothills, and plains regions in a large snow-dominated watershed in western Canada. Results show that earlier snowmelt is predicted in mountains and foothills regions under climate change scenarios, while less snow accumulation and melt is anticipated in plains region. Future projections of GW-SW interactions show an increase in groundwater discharge to streams in the winter and spring seasons, mainly due to earlier snowmelt in winter and spring in mountains and foothills regions. On the other hand, more surface water loss to groundwater in summer and fall is predicted, which can be caused by the low level of groundwater under future conditions in foothills regions. Correlation analysis between regional snowmelt and GW-SW interactions for historical simulations show a higher correlation (R-2 = 0.494) in mountains region, compared to very low correlations (R-2 < 0.01) in foothills and plains regions. LASSO results showed that ET was the main control of dynamics of GW-SW interactions in all regions of the study. Snowmelt was the next most important predictor of GW-SW dynamics in the mountains region, while rainfall was more important in the foothills region. Nevertheless, none of the three predictors strongly influenced GW-SW dynamics in the plains region, mainly due to low groundwater levels and a lack of connectivity with the surface water. Overall, it is shown that the ecohydrogeological and climatic variability of different regions plays an important part in governing GW-SW dynamics.