Non-stationary analysis of the frequency and intensity of heavy precipitation over Canada and their relations to large-scale climate patterns

In recent years, because the frequency and severityof floods have increased across Canada, it is importantto understand the characteristics of Canadian heavy precipitation.Long-term precipitation data of 463 gauging stationsof Canada were analyzed using non-stationary generalizedextreme value distribution (GEV), Poisson distribution andgeneralized Pareto (GP) distribution. Time-varying covariatesthat represent large-scale climate patterns such as ElNiño Southern Oscillation (ENSO), North Atlantic Oscillation(NAO), Pacific decadal oscillation (PDO) and NorthPacific Oscillation (NP) were incorporated to parametersof GEV, Poisson and GP distributions. Results show thatGEV distributions tend to under-estimate annual maximumdaily precipitation (AMP) of western and eastern coastalregions of Canada, compared to GP distributions. Poissonregressions show that temporal clusters of heavy precipitationevents in Canada are related to large-scale climatepatterns. By modeling AMP time series with non-stationaryGEV and heavy precipitation with non-stationary GPdistributions, it is evident that AMP and heavy precipitationof Canada show strong non-stationarities (abrupt andslowly varying changes) likely because of the influence oflarge-scale climate patterns. AMP in southwestern coastalregions, southern Canadian Prairies and the Great Lakes tend to be higher in El Niño than in La Niña years, whileAMP of other regions of Canada tends to be lower in ElNiño than in La Niña years. The influence of ENSO onheavy precipitation was spatially consistent but strongerthan on AMP. The effect of PDO, NAO and NP on extremeprecipitation is also statistically significant at some stationsacross Canada.