Remote sensing snowfall retrievals are powerful tools for advancing our understanding of global snow accumulation patterns. However, current satellite-based snowfall retrievals rely on assumptions about snowfall particle shape, size and distribution which contribute to uncertainty and biases in their estimates. Vertical radar reflectivity profiles provided by the VertiX X-band radar instrument in Egbert, Ontario are compared with in situ surface snow accumulation measurements from January-March 2012 as a part of the Global Precipitation Measurement (GPM) Cold Season Precipitation Experiment (GCPEx). In this work, we train a random forest (RF) machine learning model on VertiX radar profiles and European Centre for Medium-RangeWeather Forecasts (ECMWF) Reanalysis version 5 (ERA-5) atmospheric temperature estimates, to derive a surface snow accumulation regression model. Using event-based training-testing sets, the RF model demonstrates high predictive skill in estimating surface snow accumulation at 5-minute intervals with a low mean square error (MSE) of approximately 1.8×10−3 mm2 when compared to collocated in situ measurements. The machine learning model outperformed other common radar-based snowfall retrievals (Ze − S relationships) which were unable to accurately capture the magnitudes of peaks and troughs in observed snow accumulation. The RF model also displayed consistent skill when applied to unseen data at a separate experimental site in South Korea. An estimate of predictor importance from the RF model reveals that combinations of multiple reflectivity measurement bins in the boundary layer below 2 km were the most significant features in predicting snow accumulation. Nonlinear machine learning-based retrievals like those explored in this work can offer new, important insights into global snow accumulation patterns and overcome traditional challenges resulting from sparse in situ observational networks.