https://doi.org/10.1002/hyp.70455

Snowmelt in the Southern Andes is a key water source for ecosystems, agriculture, hydropower and general water supply. Snow–vegetation interactions in this region remain poorly understood, especially under Mediterranean climates dominated by deciduous forests. We investigated how forest canopy, elevation, and aspect influence snowpack processes in a 3.7 km2 headwater catchment in south-central Chile (36.9° S). The physically based Cold Regions Hydrological Modelling platform (CRHM) was implemented to represent the impact of forest under different canopy cover, aspects and elevations, on the snowpack dynamics between 2015 and 2024. Snowpack simulations were evaluated against snow transects and pits (2022–2024) across under canopy and open conditions, and over north- and south-facing slopes. The model shows a reasonable agreement against snow water equivalent (R2 = 0.88; RMSE = 75 mm; mean bias of 23 mm) and snow depth (R2 = 0.68; RMSE = 0.37 m; mean bias of −0.10 cm) under the different canopy cover condition and slopes, suggesting its suitability for further analyses. On average, forest canopy reduces peak SWE by 36%, primarily due to sublimation losses associated with canopy interception. Snow duration under canopy is between 10 and 30 days shorter than in the open depending on aspect and elevation, and mean melt rates are slower (10–15 mm d−1) under canopy than in open areas (16–18 mm d−1) due to shading by the canopy. Interannual variability strongly modulated the influence of the canopy on snow accumulation, with SWE differences exceeding 200 mm (> 50%) in cold, snowy years and below 50 mm (< 20%) in warm, dry years. This study provides new insights into snow–forest interactions in the Southern Hemisphere and validates the use of parameterizations developed for the Northen Hemisphere under different trees species (typically needleleaf evergreens) despite identified limitations, highlighting the importance of explicitly incorporating the effect of vegetation on snow accumulation and melt in hydrological models in the Southern Andes.

meteorological forcing data and model configuration files used in this study are publicly available in the Zenodo repository at:
https://doi.org/10.5281/zenodo.18409912