Bias-corrected estimates of glacier thickness in the Columbia River Basin, Canada

Several global datasets of glacier thickness exist, but the number of observations from western Canada are sparse and spatially biased. To supplement these limited observations, we measured ice thickness with ice penetrating radar on five glaciers in the Columbia Mountains, Canada. Our radar surveys, when combined with previous surveys for two glaciers in the Rocky Mountains, total 182 km of transects that represent 34 672 point measurements of ice thickness. Our measurements are, on average, 38% thicker than previous surface inversion model estimates of glacier thickness. Using our measurements within a cross-validation scheme, we model ice thickness with the Open Global Glacier Model (OGGM) driven with recent observations of surface mass balance and glacier elevation. We calibrated OGGM ice thickness by optimizing the ice creep parameter in the model. The optimized OGGM yields an ice volume for Columbia Basin of 122.5 ± 22.4 km3 for the year 2000, which is 23% greater than the range of previous estimates. At current rates of glacier mass loss for this region, glaciers would disappear from the basin in about 65–80 years. Disappearance of these glaciers will negatively affect the basin's surface hydrology, freshwater availability and aquatic ecosystems.

Meteorological, elevation, and slope effects on surface hoar formation

Failure in layers of buried surface hoar crystals (frost) can cause hazardous snow slab avalanches. Surface hoar crystals form on the snow surface and are sensitive to micro-meteorological conditions. In this study, the role of meteorological and terrain factors was investigated for three layers of surface hoar in the Columbia Mountains of Canada. The distribution of crystals over different elevations and aspects was observed on 20 days of field observations during a period of high pressure. The same layers were modelled over simplified terrain on a 2.5 km horizontal grid by forcing the snow cover model SNOWPACK with forecast weather data from a numerical weather prediction model. Modelled surface hoar growth was associated with warm air temperatures, high humidity, cold surface temperatures, and low wind speeds. Surface hoar was most developed in regions and elevation bands where these conditions existed, although strong winds at high elevations caused some model discrepancies. SNOWPACK simulations on virtual slopes systematically predicted smaller surface hoar on south-facing slopes. In the field, a complex combination of surface hoar and sun crusts were observed, suggesting the simplified model did not adequately resolve the surface energy balance on slopes. Overall, a coupled weather–snow cover model could benefit avalanche forecasters by predicting surface hoar layers on a regional scale over different elevation bands.

Meteorological, elevation, and slope effects on surface hoar formation

Failure in layers of buried surface hoar crystals (frost) can cause hazardous snow slab avalanches. Surface hoar crystals form on the snow surface and are sensitive to micro-meteorological conditions. In this study, the role of meteorological and terrain factors were investigated for three surface hoar layers in the Columbia Mountains of Canada. The distribution of crystals was observed over different elevations and aspects during 20 days of field observations. The same layers were modelled on a 2.5 km horizontal grid by forcing the snow cover model SNOWPACK with forecast weather data from a numerical weather prediction model. The moisture content of the air (i.e. absolute humidity) had the largest impact on modelled surface hoar growth, with warm and moist air being favourable. Surface hoar was most developed at certain elevation bands, usually corresponding to elevations with warm humid air, light winds, and cold surface temperatures. SNOWPACK simulations on virtual slopes systematically predicted smaller surface hoar on south-facing slopes. In the field, a complex combination of surface hoar and sun crusts were observed, suggesting the model did not adequately resolve the surface energy balance on slopes. Overall, a coupled weather–snow cover model could benefit avalanche forecasters by predicting surface hoar layers on a regional scale over different elevation bands.