Dynamically-downscaled probabilistic projections of precipitation changes: A Canadian case study

    loading  Checking for direct PDF access through Ovid


In this study, plausible changes in annual and seasonal precipitation over Ontario, Canada in response to global warming are investigated through a regional climate modeling approach. A high-resolution regional climate model ensemble based upon the Providing REgional Climates for Impacts Studies (PRECIS) model is developed to help explore the possible outcomes of future climate. A Bayesian hierarchical model is then employed to quantify the uncertainties involved in the modeling results and obtain probabilistic projections of precipitation changes at grid point scales. The results show that the projected changes in annual precipitation exhibit a certain degree of spatial variability with the median changes mostly bounded by 0% and 20%, implying that the annual precipitation over Ontario is more likely to increase in the context of global warming. Specifically, the mean changes in annual precipitation for 2030s and 2050s would be ˜7.5%, while the annual precipitation for 2080s is likely to increase by an average of ˜12.5%. By contrast, the spatial variability of seasonal precipitation changes is more significant, especially for the changes in spring precipitation which may vary from −40% in south and 50% in north. It is reported that there would be a continuous increasing trend in winter, spring, and autumn precipitation from 2030s to 2080s by 5–30%, but summer precipitation is likely to decrease by 5% or even higher to the end of this century. Furthermore, our results suggest that the larger the biases in historical simulations, the more uncertain the future projections will be.

Related Topics

    loading  Loading Related Articles