Modelling the formation of surface hoar layers and tracking post-burial changes for avalanche forecasting

•Surface hoar size predicted from the latent heat flux•Empirical mass-size conversion for surface hoar based on field observations•Used a numerical weather prediction (NWP) model to predict surface hoar formation•High fracture propagation was a persistent property for layers with large surface hoar

Predicting the spatial distribution and persistence of buried surface hoar layers is important when evaluating avalanche hazard. This study used weather-based models to predict the formation of surface hoar and investigated how buried layers change over time. Seven years of study plot observations from the Columbia Mountains of British Columbia were used to calibrate models for surface hoar formation. The latent heat flux was modelled with weather station data and forecasted data from the Canadian numerical weather prediction model (GEM15). A linear relationship was found between vapour mass flux and observed surface hoar crystal size (r2 of 0.84 with weather station data and 0.70 with GEM15 data), and was used to predict crystal size over seven winters. Crystal size predictions had root mean square errors of 2.4 and 4.1 mm with weather station and GEM15 data, respectively. The model was compared with other empirical weather-based models. Layers of buried surface hoar were tracked with shear frame tests, compression tests (CT) and propagation saw tests (PST). PSTs and fracture character in CTs indicated that the propensity for propagation in layers of surface hoar remained high for up to six weeks. Layers with large crystals were found to weakly indicate low stability. Results from this study could be used to improve the representation of surface hoar layers in snow cover models and make spatial predictions with NWP data.