“…However, this result was not replicated by fully coupled atmosphere–ocean–vegetation models participating in PMIP2 and PMIP3, including HadCM3 with TRIFFID (Braconnot et al., 2012; Harrison et al., 2014). Recent climate model simulations suggest that changes in dust could also be important (Pausata, Messori, & Zhang, 2016), although this result may have been an artefact of that particular model (Hopcroft & Valdes, 2019; Thompson, Skinner, Poulsen, & Zhu, 2019). A lack of sensitivity of modelled vegetation to rainfall in sub‐Saharan Africa may also contribute to the problem (Hopcroft, Valdes, Harper, & Beerling, 2017).…”