Delays in the onset of the wet season, or even its failure, can reduce yields and bring food insecurity. Africa is acutely vulnerable to climate change so understanding future changes in the seasonal cycle of African rainfall is crucial to establishing adaptation strategies.

Researchers project future climate using climate models – computer-based numerical simulations that use the equations for fluid dynamics and energy transfer to represent atmospheric weather patterns and ocean circulation. Evaluating these models is an important stage of their improvement.

In recent years, there’s been a shift in thinking on such evaluation. Previously, most climate-model evaluation over Africa compared the mean rainfall amount in model output for today’s conditions to current observations for fixed seasons (June to August, December to February, etc), with limited analysis of the annual cycle. This doesn’t provide enough information for a continent where the seasonal cycle of precipitation has such socio-economic importance. Preparation of the 5th Intergovernmental Panel on Climate Change (IPCC) Assessment report highlighted the uninformative nature of climate-model evaluation for many applications. As scientists move towards the 6th IPCC assessment report, they’re addressing this, placing increasing importance on using climate-model evaluation to assess suitability for a range of applications. The Vulnerability, Impacts, Adaptation, and Climate Services (VIACS) Advisory Board aims to build bridges between the climate modelling community and those working on applications who use model outputs to assess how climate change will affect agriculture, water resources and health, amongst many other sectors. For the next large climate-model intercomparison project (CMIP6), the VIACS Advisory Board has requested that the climate-model simulations produce new variables of particular societal relevance. The board also highlighted its priorities for relevant model evaluation, including assessing the representation of extreme events and the seasonal progression of temperature and rainfall.

Our recent study in Environmental Research Letters (ERL) finds that simulations capture the gross seasonal cycle of African precipitation on a continental scale, yet are deficient over key regions. The Horn of Africa – including Somalia, Ethiopia and Kenya – experiences two wet seasons per year; the “long rains” during March-May and the “short rains” during October-November. Whilst the simulations capture two wet seasons per year, they exhibit significant timing biases with, on average, the long rains around three weeks late and the short rains nearly four weeks too long. Accounting for these biases may be crucial in interpreting contrasts between observations and models. For example, the “long rains” in recent years have seen declining rainfall but models project increasing amounts of “long rains” rainfall in the future.

Missing break

The most notable bias affects the southern coastline of West Africa, a region of complex meteorology with growing population and declining air quality. This area experiences its first wet season from April to June and the second from mid-September to October, separated by a “little dry season” (LDS) in July to August. The LDS can be useful for weeding and spraying crops with pesticides between the two wet seasons but, if it is too long or too pronounced, can damage crop yields. We found that simulations produce an unrealistic single summer wet season, with no mid-summer break in the rains, and this is linked with biases in ocean temperature patterns. Given that climate simulations cannot capture the current seasonality, we should treat future projections of the rains in this region with caution.

Our study highlights important challenges in representing the seasonal cycle of rainfall in climate simulations. This has implications for the reliability of future climate projections and impact assessments, including water availability for hydropower generation, the length of the malaria transmission season, and future crop yields. To address these challenges, we need to understand the physical mechanisms that drive the seasonal cycle of rainfall and trends such as the Sahel drought of the 1980s followed by recovery. As well as the complexity of the West African Monsoon, researchers are looking at the drivers of seasonality for East African rainfall, for example as part of the HyCRISTAL project. We need to assess whether climate-model simulations represent these drivers adequately, and if they don’t, we must develop the models.