Using the snow-day fraction to measure climatic change in southern Ontario (Canada): historical trends in winter season precipitation phase

Abstract

Global temperatures are increasing, and regional precipitation patterns are changing. Snow is an excellent indicator of regional climate change; 50 years of temperature and precipitation data were analysed from weather stations located within the five most populated cities of Ontario (Canada). Recorded measurements for temperature and precipitation were converted into binary values to indicate the frequency of rain days, snow days, wet days (when total precipitation is greater than 0.2 mm) and freezing days (when the average temperature is less than 0 °C); then, these values were summed over each winter season from 1970/71 to 2019/20. The snow-day fraction was calculated from the seasonal totals by dividing the total number of snow days by the total number of wet days. Historical trends were detected using Pearson’s R, Kendall’s Tau and Spearman’s Rho. Differences in mean values between the first decade (1971–1980) and the last decade (2011–2020) within the time series for the snow-day fraction and total freezing days were determined using Student’s t-tests. During the winter season in southern Ontario (December 1 to March 31), total snow days, total wet days, the snow-day fraction and freezing days were all decreasing at statistically significant rates (90 to 99% confidence levels) across four of the five cities studied (Toronto, Ottawa, Hamilton and London). Mississauga was the exception, being the only city where rain days were increasing, but no trends were detected for snow days or wet days. The snow-day fraction was decreasing in Mississauga but not at a statistically significant rate, despite freezing days decreasing at the greatest rate compared to the other four cities. Total freezing days were highly correlated with the snow-day fraction during the winter season, being able to explain 61 to 76 percent of the observed variability, where Mississauga recorded the weakest correlation and London recorded the strongest correlation.