Trends in Canadian Surface Temperature Variability in the Context of Climate Change

Much of the previous work on trends in temperature extremes has considered anomalies relative to a fixed base period climatology. Calculated in this way, trends toward more extreme warm events and less extreme cold events will be found if the mean temperature is warming. In this study we calculate anomalies relative to a 30-year running mean in order to examine trends in surface temperature variability and extremes separately from changes in the mean. The difference between trends calculated relative to a running mean and those calculated relative to a stationary mean will depend on the magnitude of the trend in the mean. Monthly trends in the positive, negative, and absolute values of daily minimum and maximum temperature anomalies at 158 stations across Canada are presented. The slopes of the trends and their significance are calculated using non-parametric methods. Trends are strongest in winter and early spring. Decreasing variability is found in the west and northeast, with the greater part of this reduction due to less intense cold anomalies. Regions of increased variability exist in the Prairies during winter and in the Atlantic Provinces in early spring. In general, the trends in variability are small compared to the mean temperature trends, implying that while the mean of the temperature distribution at Canadian sites is warming, the variance shows little change.