Multivariate Statistical Analysis of the Seasonal Rainfall Regimes of the Guinea-Fouta Djallon Mountains of West Africa.

A study of the seasonal structure of rainfall organization in a humid tropical mountain region of West Africa using multivariate classification techniques revealed considerable variations in rainfall regimes from ultra-humid to sub-humid conditions. Principal Component Analysis produced a three-component model which was used to characterize and explain the seasonal rainfall variations within the region. These are the Dry Season component which accounts for most of the variance in the rainfall data and represents the December-April period; the Wet Season component (June-September) and the Wet/Dry and Dry/Wet transitions component in October-November and May, respectively. Using the three components and a hierarchical clustering procedure, seven clusters representing the rainfall regimes were identified and characterized. A better understanding of the regional rainfall climate is important to West African countries whose economies depend largely on agriculture and related activities governed by the reliability of rainfall.