This study focuses on internal and external forcing factors for climate change in southern Africa. The former are discussed in terms of SST variability, the latter in terms of changing solar activity and volcanic influences.

The monthly Global Sea-Ice and Sea-Surface Temperature (GISST2.3b) data and the Kaplan SST anomalies have been subjected to principal component analyses (PCA) in order to identify main centres of SST variability. The dominant SST pattern refers to El Nino. The t-mode PCA shows it on the first PC with the well known time coefficients, increasing since the mid 1970s. Other features, such as dipole patterns in the Atlantic and Indian Oceans, are rather secondary implying patterns of more regional than global influence. The time coefficients of s-mode PCAs have further been analysed by canonical correlation analyses (CCA) in order to investigate the influence of the Atlantic, Indian and Pacific Oceans on the climate of southern Africa. CCA patterns for October show the coupling of the tropical Indian Ocean Dipole and precipitation variability in East Africa. Additional CCAs were applied to investigate the link of tropical SSTs between the Atlantic, Indian and Pacific ocean basins. The positive correlation between equatorial Atlantic SSTs and the west Indian Ocean shows a marked trend mode. These results are consistent for all months. The strongest couplings are between the eastern Pacific and the Indian Ocean except for the equatorial region. This region is out of phase coupled with both Atlantic and Pacific Ocean SSTs.

Solar variability is a controversial factor of climate change across all time scales. This study tries to establish statistical associations between various solar parameters and southern African climate. Reconstructed solar irradiance data, the number of sunspots, the solar diameter and the solar cycle length are therefore used, in order to identify possible links in the sun-climate relationship. Significant correlations were found between solar irradiance and surface air temperature with highest values in the region of Angola and the central interior of southern Africa. This region reveals the strongest increasing temperature trend in the 20th century. Correlations with precipitation and SLP show very unequal patterns, where as correlations with SSTs are significant in the high southern latitudes. Regarding time series analyses, spectral analyses couldn't detect any conclusive result, so that wavelets are used for more detailed analyses.

Volcanoes are known for their impacts on weather and climate on time scales from days to several years. Exceeding short-term volcanic signals in surface temperature, further work is focussed on possible cumulative effects on climate on decadal time scales.