Theme A : Specific weather and climate phenomena permitted by high-resolution

Past studies of added value afforded by RCMs have proceeded by separating spatial scales into those resolved by driving GCMs and those additional scales permitted by the higher resolution RCM, by using either Fourier (Denis et al., 2002) or digital (Feser, 2006) filtering, or with a perfect-prognosis approach based on aggregation (Di Luca et al., 2011). These approaches have all shown limitations.

We propose to approach the quantification of added value from a different perspective based on an understanding of the physical mechanisms that contribute to temperature and precipitation extremes. Our working assumption is that for each region of Canada, the most significant local weather and climate phenomena (such as freezing rain, downslope windstorms, valley drainage winds) can be related to characteristic signatures in large-scale atmospheric circulation or specific environmental precursors related to recent past anomalies (such as a wet or dry previous season). Monitoring the presence of such signatures in GCM climate simulations would allow focussing the efforts on specific episodes for performing expensive very high-resolution simulations of the anticipated significant phenomena with RCMs, in order to identify and best exploit the added value afforded by the use of high-resolution in dynamical downscaling.

This Theme will try to answer pressing questions such as: How will the intensity of future storms and their trajectories be modified under a warmer climate? What will be the consequences of such changes for local climate conditions? What modelling tools are required to adequately represent these physical processes?

Team Leader(s): 

Research Projects

Extreme precipitation events represent a significant economic and societal threat to the well being of Canadians. Specific examples include the 1998 freezing-rain storm in the St. Lawrence River Valley (SLRV) that imposed more than $3 billion in economic losses. Warm-season extreme rainfalls, particularly those associated with the transition of hurricanes to extratropical cyclones, can also inflict large impacts on the citizens of eastern Canada. For example, the extratropical transitions of Hurricanes Katrina (2005), Rita (2005), and Ike (2008), each contributed more than half of the...

The many types of precipitation that occur in Canada during winter storms may lead to major disruptions to the society by affecting power networks, the aviation industry and ground transportation. Winter precipitation commonly observed in Eastern Canada can be in the form of rain, freezing rain, ice pellets and wet snow. The prediction of the type of winter precipitation is challenging when it is formed at temperatures near freezing, as their formation and evolution imply phase changes that impact the dynamics and thermodynamics of storms.

Projected changes in weather storminess over Canada are expected to contribute to changes in weather extremes and hydro-meteorological hazards, through an interrelation of large-scale and regional-scale physical processes. The main objective of this project is to improve our understanding of the regional features of storm changes and the links between storm activities, including blocking events and weather extremes, and assessment of the effects of storminess changes on various timescales, from seasonal to decadal scales, for several regions across Canada.

The working assumption underlying regional climate modelling is that RCMs’ fine mesh permits the development of fine-scale structures that are required for an accurate description of local climates (e.g. Laprise et al., 2008). To this day however the quantification of the added value of RCMs remains a central issue. Frequency distribution analysis has confirmed the benefits of fine-mesh simulations in better reproducing the extreme values of key weather elements such as precipitation intensity (Mladjic et al., 2011).

Climate models are subject to structural uncertainty resulting from poorly constrained parameters of their parameterized physical processes. These parameters are usually tuned using expert judgment in a way that more often than not lacks planning, an established method and a clear aim. Recently, a promising objective optimization method has been developed and successfully tested on a European regional domain with the COSMO-CLM model (Bellprat et al., 2012a and b). The method has also been recently adapted at Ouranos within the same model for the North American continent.