André Lenouo

Improved predictability of droughts over southern Africa using the standardized precipitation evapotranspiration index and ENSO

The provision of timely and reliable climate information on which to base management decisions remains a critical component in drought planning for southern Africa. In this observational study, we have not only proposed a forecasting scheme which caters for timeliness and reliability but improved relevance of the climate information by using a novel drought index called the standardised precipitation evapotranspiration index (SPEI), instead of the traditional precipitation only based index, the standardised precipitation index (SPI). The SPEI which includes temperature and other climatic factors in its construction has a more robust connection to ENSO than the SPI. Consequently, the developed ENSO-SPEI prediction scheme can provide quantitative information about the spatial extent and severity of predicted drought conditions in a way that reflects more closely the level of risk in the global warming context of the sub region. However, it is established that the ENSO significant regional impact is restricted only to the period December–March, implying a revisit to the traditional ENSO-based forecast scheme which essentially divides the rainfall season into the two periods, October to December and January to March. Although the prediction of ENSO events has increased with the refinement of numerical models, this work has demonstrated that the prediction of drought impacts related to ENSO is also a reality based only on observations. A large temporal lag is observed between the development of ENSO phenomena (typically in May of the previous year) and the identification of regional SPEI defined drought conditions. It has been shown that using the Southern Africa Regional Climate Outlook Forum’s (SARCOF) traditional 3-month averaged Nino 3.4 SST index (June to August) as a predictor does not have an added advantage over using only the May SST index values. In this regard, the extended lead time and improved skill demonstrated in this study could immensely benefit regional decision makers.

Convection activity over the Guinean coast and Central Africa during northern spring from synoptic to intra-seasonal timescales

This study proposes an overview of the main synoptic, medium-range and intraseasonal modes of convection and precipitation in northern spring (March–June 1979–2010) over West and Central Africa, and to understand their atmospheric dynamics. It is based on daily National Oceanic and Atmospheric Administration outgoing longwave radiation and Cloud Archive User Service Tb convection data, daily TRMM and Global Precipitation Climatology Project rainfall products and daily ERA-Interim reanalysis atmospheric fields. It is first shown that mesoscale convective systems can be modulated in terms of occurrences number and intensity at such time scales. Based on empirical orthogonal function analyses on the 2–90-day filtered data it is shown that the main mode of convective and rainfall variability is located along the Guinean coast with a moderate to weak extension over Central Africa. Corresponding regressed deseasonalised atmospheric fields highlight an eastward propagation of patterns consistent with convectively coupled equatorial Kelvin wave dynamics. Then a singular spectrum analysis combined with a Hierarchical Ascendant Classification enable to define objectively the main spectral bands of variability within the 2–90-day band, and highlight three main bands, 2–8-, 8–22- and 20–90-day. Within these three bands, space–time spectral decomposition is used to identify the relative impacts of convectively coupled equatorial Kelvin, Rossby and inertia–gravity waves, as well as Madden–Julian Oscillation (MJO) signal. It confirms that eastward propagating signals (convectively coupled equatorial Kelvin wave and MJO) are highly dominant in these convection and precipitation variability modes over the Guinean coast during northern spring. So, while rain-producing individual systems are moving westward, their activity are highly modulated by sub-regional and regional scales envelops moving to the east. This is a burning issue for operational forecasting centers to be able to monitor and predict such eastward propagating envelops of convective activity.

On convection and static stability during the AMMA SOP3 campaign

Using radiosonde dataset from 15 weather stations over West Africa, this paper investigates the contribution of the couple convection-static stability in the framework of the African monsoon multidisciplinary analyses Special Observing Period 3 (AMMA SOP3) experiment. Within this 31-day period, the boundary layer-winds depictions have revealed the West African monsoon’s (WAM) depth (around 1500 m) is not thick enough to trigger intense convection. However, the midlevel winds distribution (700–600 hPa) has shown the average African easterly jet core strength (15 m s−1) is sufficient to allow the development of African easterly waves (AEWs) necessary for squall lines activities. In return, in the upper levels (200–100 hPa), the speed (below 18 m s−1) of the mean Tropical easterly jet (TEJ) core cannot favor midlevel updrafts. The free tropospheric humidity (FTH) depiction has indicated convective events are more likely in the western Sahel where the highest FTH (FTH >50 %) are recorded. The static stability analysis has testified that convection is stronger in the semi-arid (SA) area during night time (0000 GMT). However, convective activities are inhibited in the wet equatorial (WE) region due to mean low-level stability. We used METEOSAT Second Generation (MSG) infrared (IR10.8) imagery of the 8th September 2006 to confirm that result. Furthermore, a maximum midtropospheric static stability combined with maximum relative humidity (RH) was found on the fringe of the Saharan air layer’s (SAL) top (altitude around 5.3 km) in the WE region.