Mesoscale Convective System Development, Synoptic Drivers, and Forecast Challenges of a Catastrophic Coastal Rainfall Event in West Africa

On 18–19 June 2018, the metropolitan area of Abidjan in Côte d’Ivoire experienced one of the most extreme daily rainfall events in its observational history with 302 mm within 24 hours. With severe urban flooding, at least 20 fatalities, and substantial economic losses, this event underscore serious knowledge gaps about extreme rainfall processes along the West African coast. In particular, the mechanisms leading to highly localized, sub-daily rainfall extremes in the humid environment of the Guinea Coast region remain insufficiently understood. At the same time, such events are difficult to anticipate with current numerical weather prediction systems which limits the effectiveness of early warning. These challenges motivated a focused process-oriented case study of the extreme June 2018 Abidjan rainfall event.

Leveraging multi-source observational datasets combining unique, largely non-public, sub-daily to daily rain gauge measurements over the District of Abidjan, satellite-based IMERG precipitation and METEOSAT cloud observations, the extremeness of the event and the temporal evolution of the responsible mesoscale convective system (MCS) was investigated. Large-scale and mesoscale environmental conditions, including moisture, flow patterns, and vorticity tendencies, were characterized with the ECMWF’s reanalysis product ERA5. Finally, with the aim of assessing how well a state-of-the-art global prediction system captures the likelihood and timing of an extreme rainfall event over West Africa, forecast performance and practical predictability were evaluated using ensemble predictions from the ECMWF Integrated Forecasting System.

The present analysis reveals several aspects that characterize the June 2018 Abidjan extreme rainfall event.

• First, rain gauge observations show that the locally recorded total of 302 mm within 24 hours ranks among the most extreme daily rainfall amounts documented along the Ivorian coast while surrounding stations simultaneously experienced widespread totals above 100 mm. This highlights the combined localized and regional nature of the event.
• Second, satellite-based cloud tracking indicates that the extreme rainfall was associated with a long-lived, westward-propagating MCS, whose convective signatures weakened as it approached Abidjan, but yet continued to produce exceptional rainfall accumulations within a moisture-rich coastal environment.
• Third, the event was marked by the development of a pronounced moist low-tropospheric vortex over the Abidjan area, accompanied by unusually strong moisture flux convergence and extreme column-integrated water vapor. A vorticity budget analysis suggests that vortex intensification was supported by tilting and the divergence term which underlines the hypothesis of an active MCS-vortex interaction during the extreme event.
• Finally, evaluating the Extreme Forecast Index, enhanced likelihood of extreme rainfall over Abidjan was only indicated at short lead times where ensemble-based extreme precipitation signals emerging not before 12 hours before onset. This showcases substantial limitations in the current predictability of such events over West Africa.

This study suggests that future work should explore the climatology of such moist vortices, their representation in convection-permitting and global models, and their potential as predictors for extreme West African rainfall in ensemble-based and data-driven forecasting approaches. Advancing these directions holds promise for enhancing early warning capabilities at operational prediction centers and reducing flood risk in rapidly growing coastal cities of West Africa.