Hydrologic processes in snowmelt dominated systems are traditionally measured and understood at the point scale. However, snowmelt, sublimation, and snowfall rates exhibit significant spatial variability across larger landscapes. These variations are influenced by local atmospheric and terrain characteristics, which control the deviations between small-scale measurements and areally averaged responses to spatially heterogeneous mass and energy inputs. Appropriate upscaling relations are needed to translate small-scale descriptions of snow processes into constitutive relationships that are applicable at larger scales. As a proof of concept, this study examines the temporal and spatial variability of sublimation and snowmelt fluxes in a drainage basin in the Canadian Rockies using machine learning methods. Raven, a hydrological model, generates high-resolution fluxes and state variables used to train a random forest algorithm. The random forest (RF) algorithm is then applied to estimate coarse resolution fluxes. This study involves estimating spatially averaged results from discretized fine-scaled models without explicit knowledge of detailed local response, both with and without low-order statistics of state (e.g., the standard deviation of snow water equivalent). A series of experiments are used to verify that the upscaling methodology can successfully represent the impact of heterogeneity within the system. Spatially averaged forcing estimated from the scale-appropriate machine learning model is then incorporated into a mass balance equation at a coarse resolution to demonstrate the efficacy of this upscaling methodology.