Use of Segmented Linear Regression Under a Bayesian Approach to Detect Climate Change in Different Regions of the World

In this work we study the behavior of some climate data (annual temperature and precipitation averages) obtained from climate stations in eleven countries in different regions of the world. One of the goals of the study is to determine whether the climate variables have change-points that could indicate the possible beginning of a change in climate. Another goal is to analyze the possible changes detected by the change-points in terms of the linear trends of the climate variables under investigation. Based on the information provided, differences between different regions in terms of the locations of the change-points and the changes they produce may also be inferred. The data sets used in the study consist of the annual averages of the twelve monthly temperature averages and the annual averages of the total rain precipitation observed in each one of the twelve months of the year obtained over a period of time from the end of the 19th century to the end of the 20th century. Segmented linear regression models are used to study the existence of possible changes in the behavior of climatic variables, as well as the types of changes produced.