The accelerated warming in the Arctic poses serious risks to freshwater ecosystems by altering streamflow and river thermal regimes. However, limited research on Arctic River water temperatures exists due to data scarcity and the absence of robust methodologies, which often focus on large, major river basins. To address this, we leveraged the newly released, extensive AKTEMP data set and advanced machine learning techniques to develop a Long Short-Term Memory (LSTM) model. By incorporating ERA5-Land reanalysis data and integrating physical understanding into data-driven processes, our model advanced river water temperature predictions in ungauged, snow- and permafrost-affected basins in Alaska. Our model outperformed existing approaches in high-latitude regions, achieving a median Nash-Sutcliffe Efficiency of 0.95 and root mean squared error of 1.0 degrees C. The LSTM model learned air temperature, soil temperature, solar radiation, and thermal radiation-factors associated with energy balance-were the most important drivers of river temperature dynamics. Soil moisture and snow water equivalent were highlighted as critical factors representing key processes such as thawing, melting, and groundwater contributions. Glaciers and permafrost were also identified as important covariates, particularly in seasonal river water temperature predictions. Our LSTM model successfully captured the complex relationships between hydrometeorological factors and river water temperatures across varying timescales and hydrological conditions. This scalable and transferable approach can be potentially applied across the Arctic, offering valuable insights for future conservation and management efforts.

Understanding the dynamics of heat transfer mechanisms is critical for forecasting the effects of climate change on arctic river temperatures. Climate influences on arctic river temperatures can be particularly important due to corresponding effects on nutrient dynamics and ecological responses. It was hypothesized that the same heat and mass fluxes affect arctic and temperate rivers, but that relative importance and variability over time and space differ. Through data collection and application of a river temperature model that accounts for the primary heat fluxes relevant in temperate climates, heat fluxes were estimated for a large arctic basin over wide ranges of hydrologic conditions. Heat flux influences similar to temperate systems included dominant shortwave radiation, shifts from positive to negative sensible heat flux with distance downstream, and greater influences of lateral inflows in the headwater region. Heat fluxes that differed from many temperate systems included consistently negative net longwave radiation and small average latent heat fluxes. Radiative heat fluxes comprised 88% of total absolute heat flux while all other heat fluxes contributed less than 5% on average. Periodic significance was seen for lateral inflows (up to 26%) and latent heat flux (up to 18%) in the lower and higher stream order portions of the watershed, respectively. Evenly distributed lateral inflows from large scale flow differencing and temperatures from representative tributaries provided a data efficient method for estimating the associated heat loads. Poor model performance under low flows demonstrated need for further testing and data collection to support the inclusion of additional heat fluxes.