A 1961-90 climatology for Africa south of the Equator and a comparison of potential evapotranspiration estimates

A 1961-90 mean monthly climatology for Africa south of the Equator was constructed at a resolution of 0.5° latitude/longitude for a suite of eight surface climate variables: minimum, maximum and mean air temperature; rainfall; sunshine hours; vapour pressure; wind speed; and rain day (>0.1 mm)frequencies. This climatology was constructed from observed station data distributed across the region with station frequencies ranging from 288 (wind speed) to 916 (rainfall). Over 92% of these data is based on observations between 1961-90 and over 85% has been supplied by national meteorological agencies from the region. Additionally, for mean temperature and rainfall, monthly anomalies with respect to the 1961-90 average have been calculated at the same spatial resolution for each month from January 1961 to December 1994. This paper describes the dataset which has been compiled for this work, the interpolation methods which have been used, together with some assessment of the accuracy of the resulting climate surfaces. The interpolation of the 1961-90 normals uses elevation, as well as longitude and latitude, as predictor variables and this enables three climate surfaces to be constructed for each variable, reflecting the 'minimum', mean and 'maximum' elevation within each 0.5° cell. The interpolation of the anomaly fields uses a simpler method in which elevation is not considered. These anomaly time series enable fields of interannual variability to be established for mean temperature and rainfall. The second part of the paper uses this climatology to construct fields of mean monthly potential evapotranspiration (PET) for the region using a number of different calculation methods. A comparison of the resulting PET estimates highlights significant spatial and seasonal biases due to differences in climate input variables and in the theoretical representation of PET. The climatology, described and analysed here is available from the authors for use in climate modelling and the study of climate change in Africa south of the Equator.