AOPP Seminar - African weather forecasting needs to be done differently

Meteorologists have always known that tropical weather is different to midlatitude weather, and we have understood for at least 50 years that its predictability is also different. But this understanding has not percolated through to policy, which still generally sees the improvement of African climate services as a process of knowledge transfer, importing methods from the Global North. The presentation will compare the differing patterns of weather prediction skill for rainfall in Europe and Africa according to lead time and spatial resolution. It is probably no surprise to meteorologists that the short-range (24h) high-resolution (O(10km)) skill for Europe is higher than that for tropical Africa. These differences are consistent with our understanding of the predictive stability of midlatitude cyclones on the daily timescale, and the relative unpredictability of the deep convection which is a signature of African weather. For these reasons, it will be argued that African forecasting systems for high-impact storms will have limited value without the provision of nowcasting: current progress on the implementation of nowcasting in Africa will be presented. In contrast to the short-range prediction, tropical Africa enjoys greater forecast skill than Europe on sub-seasonal timescales, which we understand to be due to the slower variability of ocean and probably also land surface systems. This sub-seasonal skill exists for conditions aggregated over larger spatial and/or longer temporal scales, and can benefit short-range, daily forecasting by establishing the prevailing conditions in terms of enhanced or suppressed statistical likelihood of rainfall.

It will be argued that we need to look afresh at the basic principles of how African forecasting should be done. This process of reinventing the forecasting industry needs to be led by African organisations. Some of the leading African national weather centres are already taking innovative steps, but there is a widespread need for greater scientific capacity in the African meteorological community, in order to take ownership of the path to a new approach to weather prediction for their communities. The risks and opportunities brought by private-sector commercialisation of improved services will be discussed.