Intermittent snow depth observations can be leveraged with data assimilation (DA) to improve model estimates of snow water equivalent (SWE) at the point scale. A key consideration for scaling a DA system to the basin scale is its performance at locations with forest cover – where canopy-snow interactions affect snow accumulation and melt, yet are difficult to model and parameterize. We implement a particle filter (PF) assimilation technique to assimilate intermittent depth observations into the Flexible Snow Model (FSM2), and validate the output against snow density and SWE measurements across paired forest and open sites, at two locations with different climates and forest structures. Assimilation reduces depth error by 70-90%, density error by 5-30%, and SWE error by 50-70% at forest locations (relative to control model runs) and brings errors in-line with adjacent open sites. The PF correctly simulates the seasonal evolution of the snowpack under forest canopy, including cases where interception lowers SWE in the forest during accumulation, and shading reduces melt during the ablation season (relative to open sites). The snow model outputs are sensitive to canopy-related parameters, but DA reduces the range in depth and SWE estimates resulting from spatial variations or uncertainties in these parameters by more than 50%. The results demonstrate that the challenge of accurately measuring, estimating, or calibrating canopy-related parameters is reduced when snow depth observations are assimilated to improve SWE estimates.