Canadians depend extensively on groundwater for residential, municipal, and industrial use, making effective long-term groundwater management essential. Reliable planning requires robust estimates of hydrological fluxes, particularly groundwater recharge. Yet existing recharge estimates are often coarse, outdated, spatially inconsistent, and insufficiently reflect climatic and hydrogeologic variability. In seasonally frozen regions where snowmelt and cryogenic processes strongly influence infiltration, commonly used analytical spatial recharge estimation methods are challenging to apply because they often lack the appropriate seasonal constraints needed to represent these dynamics accurately or to forecast effects of climate change. This study develops a province-wide, climate-integrated recharge assessment for Nova Scotia using a physically based SHAW modelling framework applied across hydrological response units that represent key variations in soil, land cover, topography, and bedrock. Gridded historical meteorological data and bias-corrected climate projections drive simulations for 1985-2010 and 2070-2100, allowing evaluation of how shifts in precipitation regimes, snowmelt timing, frozen ground, and evapotranspiration may alter future groundwater recharge. Model outputs are validated against long-term annual water balances derived from watershed-scale baseflow estimates. Recharge-Duration-Frequency curves are generated for all hydrological response units to quantify drought-related recharge deficits. Results show pronounced spatial variability and an overall increase in diffuse recharge under late-century climate conditions. The study delivers a new framework for developing climate-responsive recharge maps in seasonally frozen regions, along with supporting datasets that inform groundwater management and planning under evolving climate conditions.