GIS-based approach to identify climatic zoning: A hierarchical clustering on principal component analysis

Abstract

In tropical environments, the design of bioclimatic houses adapted to their environment represents a crucial issue when considering thermal comfort and limiting energy consumption. A preliminary aspect of such design endeavors is the acquisition of accurate knowledge regarding the climatic conditions in each region of the studied territory. The objective of this paper is to propose a climatic zoning for Madagascar from a database of 47 meteorological stations by performing hierarchical clustering on principal components (HCPC). The results are then combined with spatial interpolation using geographic information system (GIS) tools, enabling us to define three climate zones corresponding to dry, humid and highland areas. These results make it possible to define standard meteorological files to evaluate the thermal performance of traditional Malagasy houses. Regardless of the type of house and the areas considered, the percentage of thermal comfort according to the Givoni bioclimatic chart varies from average values of 20% without ventilation to 70% with an air velocity of 1 m/s. In summary, Madagascar's traditional habitat typologies have adapted over time to the constraints of their environment.