Development and application of a technique for projecting novel and disappearing climates using cluster analysis

We present an objective climate classification method based on cluster analysis, and its application for the assessment of local impact of the projected global climate change in the RCP8.5 IPCC AR5 scenario. The novelty of the method is in the use of cluster analysis to identify novel and disappearing climates. We detect the absence of a climate type by classificating the local climates of two different time periods in a single cluster analysis. If any of the resulting clusters contains only local climates taken from one of the time periods, the climate type represented by that cluster is not present in the other time period. We present the application of this method to the ends of the 20th and 21st centuries on a global scale, and the results of the equivalent analysis restricted to a circle of 500 km radius around each location. They show that, in large areas, many current ecosystems would not find 20th century conditions in a near place at the end of 21st century.