Premium
Reginalization of daily rainfall in Kenya by means of common factor analysis
Author(s) -
Bärring Lars
Publication year - 1988
Publication title -
journal of climatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.58
H-Index - 166
eISSN - 1097-0088
pISSN - 0196-1748
DOI - 10.1002/joc.3370080405
Subject(s) - orographic lift , geography , climatology , subdivision , environmental science , physical geography , statistics , mathematics , precipitation , meteorology , geology , archaeology
Daily rainfall data from 73 stations distributed throughout Kenya were subjected to maximum likelihood factor analysis. Data for the period January 1971 to December 1985 were used. A brief analysis of data quality and representativity and the problem of missing data are included. The data were transformed by natural logarithms. Scree and LEV plots together with the Kendall/North criterion for eigenvalue separation indicated break points at 5, 11 and 15 factors. The 11‐factor model was judged as being the most parsimonious, for both statistical and climatological reasons. It explained 38 per cent of the total variance. The general features of the regionalization are found to be reasonably consistent with other studies. The major regions (the coast, the Highlands, the Lake Victoria area, and the dry northern/eastern areas) appear in this regionalization but are divided into several subregions. It can be concluded that this subdivision is partly linked to physiographical regions and orographic influence. In addition, possible links to regional circulation patterns are discussed.