Future Changes in the Indian Ocean Walker Circulation and Links to Kenyan Rainfall

Abstract

East Africa is vulnerable to hydroclimatic variability and change, and therefore reliable projections of future rainfall are important for climate change adaptation planning. However, the region’s climate is affected by complex multi-scalar processes and poorly represented in climate models, leading to uncertainty surrounding rainfall change. The importance of circulation features in controlling long-term rainfall variability provides an opportunity to constrain projections. We use a process-based climate model evaluation methodology to demonstrate links between Coupled Model Intercomparison Project phase 5 (CMIP5) rainfall biases over Kenya and circulation biases in the Indian Ocean Walker Circulation (IOWC). During both the long and short rains, models with wet biases in historical runs continue to be wet in future. Wet future projections are associated with enhanced easterly winds over the equatorial Indian Ocean, as well as decreasing vertical velocity over Kenya and increasing vertical velocity over the Maritime Continent. We demonstrate that models with a simulated IOWC which is close to reanalysis in historical runs project different changes to Kenyan rainfall than those which do not. In particular, the projected rainfall increase during the long rains is confined to a single month (April) in these models. We call for a renewed focus on the Walker Circulation as a way to constrain uncertain rainfall projections elsewhere in the tropics.