Improvement of climate modelling for Africa
Improved understanding and modelling of African climate and its drivers.
Improved representations of African climate and its drivers in climate models is a key requirement to improve the physical basis of predictions and to reduce prediction uncertainty. CSRP-1 has undertaken a large body of research to evaluate the performance of state-of-the-art models for Africa, improve understanding of African climate and its drivers and pull through the knowledge gained into development of the Met Office's HadGEM3 climate model, which is the basis of its seasonal, decadal and longer-term climate predictions. The research has contributed to key advances in HadGEM3's performance for Africa, including:
- Improved simulation of seasonal rainfall climatology in 7 of 8 African key rainy seasons assessed - with substantial improvements in 5 seasons;
- Improved simulation of the remote influences of sea surface temperatures (SSTs) on African rainfall variability (teleconnections). HadGEM3 performance now leads that of other current generation models as available from CMIP5. These teleconnections - such as those associated with El Niño and La Niña events - are key drivers of African rainfall variability;
- Improved capabilities for long range (up to 18 months ahead) prediction of El Niño and La Niña events - as well as for rainfall trends up to 5 years ahead in some regions of Africa, including in the Sahel. These advances, achieved with a new monthly-to-decadal system based on HadGEM3, have increased potential for climate early warnings informing resilience and adaptation planning.