Assessment of the Subseasonal Prediction Performance of the Mozambique Monsoon Rainfall and Its Modulation by the Madden–Julian Oscillation (MJO)

Situated in Southern Africa (Africa south of 10°S), Mozambique is a country with ~ 28 million inhabitants, 65% of whom living in rural areas and whose subsistence heavily relies on agriculture. Besides, most of the energy consumed in the country comes from hydropower generation. The success of both agriculture and hydropower generation, the key sectors to the development of the country, strongly depends on availability of water resources, supplied in the country mainly by the monsoon rainfall, which usually occurs during December-January-February. Thus a skillful sub-seasonal prediction is of great societal and economical value in the highly populated Mozambique provinces, especially during the monsoon season, when frequent extreme events occur, such as persistent heavy rainfalls associated sometimes with tropical cyclones. In this line, this study uses the European Centre for Medium-Range Weather Forecasts (ECMWF) and the National Centers for Environmental Prediction Climate Forecast System version 2 (CFSv2) models hindcasts to examine the performance of multiweek prediction of Mozambique monsoon precipitation during the sub-seasonal to seasonal prediction project (S2S) common period (1999-2010). Preliminary analysis of the results of these models against the Climate Prediction Center gridded precipitation data set and the ECMWF ERA-Interim reanalysis show that the predictive skill of precipitation and its associated large-scale circulation is useful out to a lead time of 2 weeks. Although the ECMWF outscores CFSv2, both models are able to reproduce reasonably well the observed modulation of Mozambique monsoon precipitation by the Madden–Julian Oscillation (MJO). These results suggest that the state-of-the-art operational S2S models have the capability to provide useful monsoon rainfall predictions for Mozambique and neighbouring regions.

KEY WORD: Mozambique monsoon rainfall, agriculture and hydropower, S2S models, MJO.