ABSTRACT: We used multi-year satellite observations to study aerosol effects on the large-scale variability in precipitation of the West African Monsoon (WAM) region, which is often impacted by high concentrations of desert dust and biomass-burning smoke. We find a statistically significant precipitation reduction associated with high aerosol concentration near the coast of the Gulf of Guinea from late boreal autumn to winter. The largest aerosol-related precipitation reduction (∼1.5 mm d−1) is about 50 % of the climatological mean precipitation in the region and occurs mainly at rain rates in the range of 2–17 mm d−1 off the northern coast of the Gulf of Guinea. This reduction cannot be linearly attributed to known climate and weather factors such as El Niño–Southern Oscillation, North Atlantic Oscillation, Atlantic sea-surface temperature, or water vapour. The fractional precipitation variance related to aerosol is about 13%, a value comparable to those related to the known climate factors. Based on the spatial pattern and seasonality of the observed precipitation reduction and its dependence on the rain rate, the observed negative correlation cannot be readily attributed to precipitation effects on aerosol by wet deposition or to rain and cloud contamination of satellite aerosol retrievals. We therefore suggest that our results can be taken as observational evidence of aerosol effects on precipitation. The aerosol associated with the observed precipitation reduction can be traced back to various African sources where large quantities of desert dust and biomass-burning smoke are emitted during much of the year. Given that the emissions of dust and smoke have varied considerably over the past several decades, in part attributable to human activities, our observed rainfall reduction may

DYNAMO is a US research program motivated by two outstanding problems: (i) Current prediction skill for the Madden-Julian Oscillation (MJO) is very limited; particularly, it is the lowest for the MJO initiation phase over the Indian Ocean. (ii) The inability of state-of-the-art global models to produce the MJO degrades their seasonal to interannual prediction and lessens our confidence in their ability to project future climate. The overarching goal of DYNAMO is to expedite our understanding of processes key to MJO initiation over the Indian Ocean and our efforts to improve simulation and prediction of the MJO. DYNAMO consists of four integrated components: a field campaign, data analysis, modeling, and forecasting.