Theoretical considerations of day-to-day temperature variability applied to Toronto and Calgary, Canada data

Day-to-day temperature variability has been examined within the context of three theoretical climates: orderly, random and oscillatory. Using highly idealized examples two measures of day-to-day variability were compared. It was found that standard deviation works well to characterize variability for random climates, as expected, but did not perform well for orderly and oscillatory climates. For these climates, a more direct calculation of day-to-day variability was found to be more representative. This method better captures the clustering of temperatures that may result from a number of meteorological and geographic factors which bring order to the local climate. These concepts were applied to two Canadian cities, Toronto, Ontario and Calgary, Alberta. Both cities were found to have a degree of orderliness in their climate. Toronto's orderliness appears to be linked to the temperature mitigating effect of a moister climate and midlatitude cyclones. In addition, the proximity of Lake Ontario produces land/sea breezes which act to reduce thermal variability. Calgary experienced more variability, both on average and in the exceedances of 5 degrees C and 10 degrees C thresholds for day-to-day temperature swings. This is a result of a drier climate with less hydro-climatic inertia and the presence of chinook winds which produce, particularly in winter, swings of temperature frequently exceeding 10 degrees C. Although all the measures of day-to-day temperature variability examined appear to be decreasing in both cities, only the decrease of 5 degrees C exceedances of T(min) at Toronto was statistically significant.