Using land surface information to improve nowcasts of mesoscale convective systems in West Africa

<p>Mesoscale convective systems (MCSs) dominate rainfall and its extremes in most parts of West Africa, frequently producing flash floods that results in major damage and loss of life. As West African storms are already intensifying, these effects are expected to become more frequent and severe under climate-change and rapid urban expansion. To help mitigate these impacts, we co-developed a prototype nowcasting system with West African meteorological services based on conditioned climatologies of MCSs as seen from the Meteosat Second Generation satellites since 2004. The system is now being trialled for improving storm nowcasts for up to 6 hours ahead and for the first time incorporates near real-time remotely-sensed land surface information (land surface temperature, soil moisture) to increase storm predictability due to surface effects on convective instability, moisture convergence and regions of high wind shear. We will present the workflow of this nowcasting system and will discuss our current understanding of the land surface effects that play a role for storm development and prediction in the region. Crucially, the developed nowcasting system is computationally inexpensive to run operationally once set up, can be combined with existing forecast workflows, and achieves skill in the absence of rainfall radar, which is not available over most of Africa. Lessons learnt on co-developing nowcasting tools and further development steps will be discussed.</p>