Modeling West African Monsoon Dacadal Variability and Feedbacks: Second West African Monsoon Modeling and Evaluation Project Experiment (WAMME II)

It is a great challenge to understand the impact of external forcings: oceans, land, and aerosols, on west African monsoon (WAM) variability, especially their roles in the Sahel drought. The West African Monsoon Modeling and Evaluation (WAMME) is a project comprised of both general circulation models (GCMs) and regional climate models (RCMs) with the objective to collectively provide best estimation of the relative importance of all those external forcing on WAM on seasonal to multi-decadal time scales. Observational evidence has shown strong decadal climate variabilities in the Sahel from the 1950s to the 2000s, not only in precipitation, but also in SST, vegetation cover, land use and land cover (LULC) change, and aerosols. In WAMME-2, multi-model intercomparison experiments are designed to test how seasonal and decadal variabilities of WAM precipitation are associated with these forcings, and assess their relative contributions in producing/amplifying the WAM seasonal and decadal climate variability. The WAMME-2 strategy is to apply observational data-based anomaly forcing of SST, land surface and aerosols, i.e., "idealized but realistic" forcing, in GCM and RCM simulations with the specific purpose of estimating the relative impacts of each forcing and feedback mechanisms. Eight GCMs and five RCMs are participating in this experiment. This present reports the most up-to-date progress in understand and simulate the WAM with the state-of-the-art GCMs and RCMs.

In the SST experiment, in addition to the global SST effect, each ocean's role is also evaluated. To test this, anomalies of SST forcing in each ocean is removed sequentially from the global SST anomalies, which differs from common practice. The preliminary results from most GCMs consistently indicate that SST has a maximum impact on the WAM decadal variability compared with other forcings, and that the effect of the Pacific Ocean is most dominant. The models, however, differ in producing other oceans' contribution to WAM decadal variability. Moreover, the models with specified maximum SST forcing are still unable to produce full Sahel drought. In the LULC change experiment, a newly available land use change map is applied. To reduce the uncertainty in model simulation, a consistent change in the vegetation maps is imposed for each modeling group. The simulated LULC change impact is also substantial, compatible to but less than the SST forcing. In the dust experiment, the direct impact of dust on the radiation budget and its influence to the Sahel rainfall are evaluated using GOCART dust data and are compared with other external forcings. This is the first attempt to use multi-GCMs and RCMs to collectively explore the roles of multiple external forcing in WAM variability. WAMME2's achievement will provide better understanding of relative importance of various forcing and possible feedback mechanisms, complementary to experiments under IPCC, which are focused more on impacts of emission control scenarios, and CORDEX, which is focus on RCM downscaling ability. WAMME also provides a useful prototype multidisciplinary approach for other monsoon regions' researches.