Modelling snow dynamics and groundwater-surface water interactions in mountainous, foothill and plain regions in North Saskatchewan River Basin, Canada

Snowmelt and groundwater-surface water (GW-SW) interactions are dominant controllers of hydrological cycles and water availability in northern latitudes that are characterized with ‘cold region hydrology’. Snowmelt is the main source of groundwater recharge and surface runoff, as snow accumulation holds the largest share of water resources in such cold regions. However, a comprehensive of uncertainty inreliability of snow depth and snowmelt modelling and projections of such GW-SW processes response to snow dsynamics under climate change are subjected to debate. To fill this gap, this study has been undertaken with the main objectives of (1) revisiting and characterizing the performance and uncertainty of using two commonly-used approaches for snowmelt modelling , namely Energy-Balance Modules (EBMs) and Temperature-Index Modules (TIMs), as well as two common Snow Density formulations (SNDs) that map snow water equivalent (SWE) to snow depth; and (2) assessing the changes response of snowmelt and GW-SW interactions to snowmelt dynamics under future climate change scenarios and emission scenarios, and evaluating the dynamics of GW-SW interactions in relation to snowmelt. For snowmelt analysis, we coupled the Soil and Water Assessment Tool (SWAT) model with EBM and TIM modules along with two SND formulations, by modifying its source code, in order to examine model representation of snow depth simulation. , and tThe Analysis of Variance (ANOVA) was used to assessfor spatiotemporalspatio-temporal variation of uncertainty by decomposition of the total projected snow depth uncertainty to its generating sources due to the use of EBM, TIM, projections in combination with five Global Climate Models (GCMs), two emission scenarios (RCP2.6 and RCP8.5), and two downscaling methods (DS1 and DS2). The analyseis were implemented in mountainous, foothills and plains regions of the North Saskatchewan River Basin (NSRB) as a large, snow-dominated watershed with high variability of climate, vegetation, and topography. Results showed that modeling performance in mountainous regions is poor under using all regardless of snowmelt approaches, i.e., EBMs, TIMs, and SND selection, modelingmodelling performance in mountainous regions is poor due to low quality of input climate data. However and SND selection plays an important part in performance and in reducing uncertainty of snowmelt snow depth modulein foothills and plains regions, where more accurate and high resolution climate data are available to setup the initial models. The uncertainty decomposition results showed that model parameter uncertainty, due to the use of EBM or TIM, dominantly controlled snow depth projections, particularly in mountainous and foothills regions. However, in plains regions, the uncertainty contribution of model parameters becomes less dominant and more variable in different months of the year. The results also showed that, regardless of time and season, model parameter uncertainty dominates all other sources in mountainous regions, whereas it becomes less prominent moving from mountain to foothill and to the plain, and the contribution of climate change models and scenarios becomes more important. After comparison of performance and uncertainties associated with different snowmelt modules, the most reliable module was chosen for calibrating the SWAT model. Thus, to answer the second research questionmain obejective, we developed the surface water and groundwater modelling through SWAT-MODFLOW model in order to assess the inter-relation of regional snowmelt and GW-SW interactions and their changes under climate change. Results predicted that under future climate change, earlier snowmelt is expected in mountainous regions, which directly affects the GW-SW interactions changes mainly in mountainous and foothills regionscatchments. On the other hand, correlation analysis of regional snowmelt and GW-SW interactions showed a higher correlation (R^2=0.494) in mountainous regions, compared to very low correlation (R^2<0.01) that was observed in foothill and plain regions. In foothills and plains regions, Vegetation cover and the resulting evapotranspirationhigher levels of evapotranspiration (ET), and as well as the dominant effect of rainfall on groundwater recharge and surface water availability in foothill and plain regions, are introduced as possible effects of low correlations between snowmelt and GW-SW dynamics in foothill and plain areas. It was also discussed that due to underground aquifer connectivity, the mountainous regions may receive less influence from any adjacent aquifers, while those in foothills and plains areas are cumulatively influenced by incoming groundwater flows from neighboring aquifers. However, this requires more research when a better aquifer boundaries are delineated and available for the modelingmodelling study. Overall, it is shown that the hydro-climate variability, topography and land-use in different regions play an important role in (1) the reliability and performance of snowmelt modules in reproducing snow depth and streamflow; (2) the cascade of uncertainty of snow depth projections; and (3) the inter-relation of snowmelt and GW-SW dynamics and governing processes that control GW-SW interactions.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.