Geography Department, Bindura University of Science, Bindura, Zimbabwe; Oceanography Department, University of Cape Town, Cape Town, South Africa; Geography Department, National University of Lesotho, Maseru, Lesotho; Postgraduate Research Division, University of the Western Cape, South Africa
Manatsa, D., Geography Department, Bindura University of Science, Bindura, Zimbabwe, Oceanography Department, University of Cape Town, Cape Town, South Africa; Matarira, C.H., Geography Department, National University of Lesotho, Maseru, Lesotho; Mukwada, G., Postgraduate Research Division, University of the Western Cape, South Africa
The current paper is an observational study that investigates the October to December (OND) rainfall variability over the east Southern African Development Community (SADC) mainland region in relation to El Niño-Southern Oscillation (ENSO) and the Indian Ocean dipole zonal mode (IODZM) for the period 1950-1999. An empirical orthogonal function (EOF) analysis of OND rainfall field revealed that the north-south aligned areas of the eastern SADC are located in different covariability regions. This meridionally aligned dipole rainfall anomaly configuration is captured only in the dominant principal component (PC1), making it possible for the opposing rainfall anomalies of the two regions to have a common trigger. However, ENSO which is the standard attribute for regional rainfall variability failed dismally to adequately explain this dipole rainfall anomaly pattern. Instead, there appears to be consistent evidence through statistical techniques which strongly indicates the likelihood of the participation of only the positive IODZM phase events in the creation of the positive dipole rainfall phase (i.e. simultaneous floods over the northeast and droughts over the southeast of the SADC region). Since the negative IODZM phase events can hardly be linked to the reverse rainfall pattern, it implies that the positive and negative rainfall dipoles have fundamentally different causes. Thus, contrary to convectional knowledge, the ENSO association to this dipole rainfall anomaly pattern is by no means robust and could be symptomatic of the well-known ENSO-IODZM connection. Interestingly, however, when analysed over the 31-year overlapping segments, IODZM's once significant independent influence on this dipole rainfall seems to be diminishing gradually as from the early 1990s, whereas that of ENSO is correspondingly being reinforced. © 2010 Royal Meteorological Society.
ENSO; IODZM; Rainfall anomaly; SADC; Wet season; Atmospheric pressure; Drought; Financial data processing; Fourier analysis; Nickel compounds; Orthogonal functions; Principal component analysis; Rain; air-sea interaction; drought; El Nino-Southern Oscillation; empirical analysis; rainfall; sea surface temperature; wet season; Indian Ocean