In this paper, a numerical stochastic model of the time series of the average daily bioclimatic index of severity of climatic regime is proposed and validated. This model is based on an assumption that real weather processes are non-stationary random processes on a year-long interval. In this assumption, the model takes into account the seasonal variation of the real meteorological processes. The input parameters of the model are determined from the data of long-term real observations at weather stations. It is shown that the trajectories of the model proposed are close in their statistical properties to the real time series of the bioclimatic index under consideration. The results related to studying the influence of a climate change on the time series of the average daily bioclimatic index of severity of climatic regime are given.