Logo: to the web site of Uppsala University

In cold-climate regions, snow can reduce photovoltaic (PV) system power output by up to 100% during individual months and as much as 34 % annually, making accurate snow loss modeling essential for reliable PV power generation simulations. To address this, we present a generalizable snow loss model, inspired by the Marion model and designed to estimate hourly snow-induced PV power losses using remote-sensing derived PV system data, including aerial imagery, Light Detection and Ranging (LiDAR), and satellite-derived irradiance and weather data. The model is integrated into a simplified gridded PV simulation framework that reduces the computational time by 99% while matching the accuracy of the standard IV-model. The snow loss model shows consistent improvements in coefficient of determination, mean absolute error and root mean squared error, and results on par with previous snow loss studies. When incorporated into a full remote sensing-based PV power simulation pipeline, the resulting average percentage root mean square error was 5.7 % when simulating the hourly power output across 40 systems. This demonstrates that individual PV power generation from multiple distributed PV systems can be assessed at scale in cold climates, even without access to system specific technical data.