Fletcher, J. K., Diop, C. A., Adefisan, E., Ahiataku, M. A., Ansah, S. O., Birch, C. E., Burns, H. L., Clarke, S. J., Gacheru, J., James, T. D., Tuikong, C. K. Ngetich, Koros, D., Indasi, V. S., Lamptey, B. L., Lawal, K. A., Parker, D. J., Roberts, A. J., Stein, T. H. M., Visman, E., and Warner, J.
Testbeds have become integral to advancing the transfer of knowledge and capabilities from research to operational weather forecasting in many parts of the world. The first high-impact weather testbed in tropical Africa was recently carried out through the African Science for Weather Information and Forecasting Techniques (SWIFT) program, with participation from researchers and forecasters from Senegal, Ghana, Nigeria, Kenya, the United Kingdom, and international and pan-African organizations. The testbed aims were to trial new forecasting and nowcasting products with operational forecasters, to inform future research, and to act as a template for future testbeds in the tropics. The African SWIFT testbed integrated users and researchers throughout the process to facilitate development of impact-based forecasting methods and new research ideas driven both by operations and user input. The new products are primarily satellite-based nowcasting systems and ensemble forecasts at global and regional convection-permitting scales. Neither of these was used operationally in the participating African countries prior to the testbed. The testbed received constructive, positive feedback via intense user interaction including fishery, agriculture, aviation, and electricity sectors. After the testbed, a final set of recommended standard operating procedures for satellite-based nowcasting in tropical Africa have been produced. The testbed brought the attention of funding agencies and organizational directors to the immediate benefit of improved forecasts. Delivering the testbed strengthened the partnership between each country's participating university and weather forecasting agency and internationally, which is key to ensuring the longevity of the testbed outcomes. [ABSTRACT FROM AUTHOR]