Wind and snow are major environmental loads that are often considered in the design of buildings and infrastructure. To ensure safety of existing structures and to develop guidelines for future developments, it is important to evaluate how these design loads will be impacted by the anticipated climate change. We evaluated projected changes to selected return levels of wind speed and snow water equivalent (SWE) and associated wind pressure and ground snow loads across Canada for the future 2071-2100 period, using Canadian Regional Climate Model (CRCM5) simulations, driven by two Global Climate Models (GCMs) for two future emission scenarios.
Evaluation of ERA-Interim driven CRCM5 simulation (CRCM5-ERA) indicates that the regional model is able to reproduce, in general, the observed magnitudes and spatial patterns of mean annual wind speed and mean winter SWE (Fig. 1). It should be noted that the CRCM5-ERA underestimates, whereas the ERA-Interim overestimates the observed annual mean wind speeds, particularly for the western mountainous and boreal forest regions. Currently, we are doing some sensitivity experiments to investigate this and to improve simulated wind characteristics. CRCM5-ERA simulation also captures the overall spatial patterns of mean and variance of annual maximum 3-hourly wind speed (Fig. 2). Annual maximum daily SWE is well captured by the model compared to Canadian Meteorological Centre (CMC) data (not shown). The same simulation demonstrates some added value downscaling extreme wind and snow events compared to the driving ERA-Interim reanalysis.
The CRCM5 projections suggest some increases in the future 50-year return levels of wind speed and pressure, mainly due to changes in the inter-annual variability of annual maximum wind speed, particularly for the central and eastern regions. As for SWE loads, results suggest general decreases for southern Canada and increases for northern Canada in the 50-year return levels for RCP 8.5. However, the projections, particularly for wind loads, vary considerably with the driving GCM and the emission scenario, suggesting that larger ensembles including more RCMs and driving GCMs will be required to better quantify uncertainties to support development of climate-resilient design standards and codes.