Standard test methods to assess the degree of movement capability of sealants are currently based on extreme values of amplitude, rate of movement and temperature change. The selection and derivation of these test parameters on the basis of field observations has yet to be fully explained. It appears that these parameters are not based on an understanding of the degree to which sealants undergo cyclic movement on buildings nor the likelihood of occurrence of extreme events. Although joint failure due to cyclic movement has been investigated, there is a need to consider how such models of deterioration can be used to establish test criteria for assessing the long-term performance of sealant products. To address some of these issues, a study was conducted to determine the nature of temperature fluctuations in a northerly climate. Hourly temperature data from the Ottawa international airport for a period of forty years was subjected to probability analysis. The analysis considered the likelihood of occurrence of periods of increasing or decreasing temperature. The occurrence of particular amplitudes ranges and rates of temperature change were also determined on both a seasonal and a yearly basis. Moreover, the annual and multi-year probability of temperature fluctuations at or above extreme values of either rate or amplitude were also assessed. In addition, the maximum and minimum temperatures of the 40 yearly periods were subjected to extremal analysis to determine the likelihood of annual periods covering larger than average temperature ranges. This analysis provides a basis from which to select test parameters.