The Global Land-Atmosphere Coupling Experiment (GLACE; Koster et al., 2004, 2006) provided a method for quantifying the influence of a GCM's soil moisture on its precipitation. In an intercomparison of twelve models under the GLACE framework, three regions were identified as having consistently strong coupling: the Sahel region of Africa, India and the southern United States. Since these are all semi-arid regions, the strong coupling here was attributed to local soil moisture stress, which controls surface evaporation.

We present an analysis of a recent version of the Met Office Hadley Centre's GCM, HadGEM3-A, using the GLACE framework. Unlike its predecessor, HadAM3, this new model displays strong land-atmosphere coupling consistent with the original GLACE consensus, particularly over the Sahel. HadGEM3-A can therefore be used to study the land-atmosphere interactions in this region in more detail. We investigate how the meteorology of this region affects the measured coupling, aiming address two key questions: how does the model's soil moisture affect the monsoon rainfall signal? Is the GLACE diagnostic particularly sensitive to the strong seasonal rainfall signals of monsoon regions?