Accurate snow cover mapping is fundamental to mountain research; however, remotely sensed snow-covered area (SCA) estimates are often unavailable at high temporal and spatial resolutions (< 30 m) and require further validation in forests. Here, we evaluate optical and radar remote sensing approaches for SCA estimation using snow cover maps derived from Airborne Laser Scanning (ALS) in a forested watershed on Vancouver Island, British Columbia. We tested five snow disappearance algorithms applied to observations from the Harmonized Landsat Sentinel collection and Sentinel-1. We included two SAR-primary approaches (the Buchelt et al. [2022] approach and a new method); two optical approaches (both using Normalized Difference Snow Index [NDSI] estimates, but differing in cloud treatment); and one hybrid approach (which constrained NDSI estimates of snow disappearance with SAR estimates of snowmelt onset). We converted snow disappearance estimates to binary SCA maps and evaluated against ALS data for five acquisition dates between March and June in 2020, 2021, and 2023. The most accurate approach was the optical-SAR approach (overall accuracy 84%, F1 = 0.86), which also produced the most reliable estimates of basin SCA (RMSE = 27 km2, basin area = 215 km2). The optical and novel SAR approaches achieved overall accuracies of 81%. While all approaches were less accurate under forest cover, the optical-SAR algorithm minimized this difference, with average accuracies of 88% in open areas and 83% in forested areas. The optical–SAR algorithm for snow disappearance provides a robust, low-cost approach for snow cover estimation in complex, forested terrain.