A multivariate classification and regionalization of west African climates

Abstract

Abstract Cluster analysis is used to group into regions over West Africa, a climatic data set which had been previously reduced to four factors. The regions thus derived are more firmly climate based than is the case with previous classifications. Various degrees of generality are also achieved by the methods, ranging from five regions at the primary level to twenty regions for more detailed work. Overall, a ten‐region climatic cluster is found to be optimal for most purposes.