Southern African monsoon: intraseasonal variability and monsoon indices

Southern African (SAF) summer monsoon rainfall varies over a range of timescales. Here the emphasis is placed on its intraseasonal variability (ISV), which is of great importance for agricultural production, water resources management, and subseasonal prediction, but synoptic variability is also considered. Rotated modes of ISV are determined based on daily rainfall gauge data (1979–2005) in eastern SAF. One of the leading modes exhibits strongest factor loadings over the core monsoon region, and is associated with variations in the monsoon circulation. This mode exists in the 10–20, 20–30 and 30–90 day bands of ISV. Spectral analysis, as well as lead-lag composite anomalies keyed to its positive and negative phases in each band confirm oscillations with periods of 12, 24 and 40 days, also present in ISV of South America monsoon. The 12-day oscillation is associated with the quasi-biweekly oscillation originated from Rossby waves in the extratropical westerly belt. The 24 and 40 day oscillations are linked to tropical convection and involve tropical and extratropical teleconnections between South America and SAF, with related convection anomalies over tropical South America/Atlantic Ocean playing a role in the eastward propagation of anomalies towards SAF. In the 30–90 day band, the oscillation is clearly associated with the MJO, with the positive phase more concentrated in MJO phases 1 and 2. A monsoon precipitation index (MPI) is proposed to help characterize, monitor and predict active and break monsoon spells. It is based on the average standardized precipitation anomaly over a selected area with the strongest precipitation variability in the core monsoon region over land, and represents well the monsoon core variability. Monsoon circulation indices associated with the MPI are also defined, since prediction skills are better for circulation than precipitation. Besides, these indices indicate the monsoon circulation features that most affect rainfall variability in different parts of SAF.