The challenges of implementing early warning systems in a context of climate change: The case of the poorly gauged or ungauged watersheds of Abidjan, Côte d'Ivoire.

Early warning is a key element in flood risk management. It primarily relies on a thorough understanding of hazards and available observations. However, in areas where hydrological data are scarce or unavailable, the lack or rarity of records poses a significant obstacle to the development of Early Warning Systems (EWS). This challenge is particularly common in tropical developing regions, where ungauged or poorly gauged basins are frequent. Over the past decades, major scientific initiatives, such as the "Decade of Predictions in Ungauged Basins (PUB)" by IAHS, have sought to address this scientific bottleneck.

How, then, can the limited available data be leveraged for EWS under non-stationary conditions? Our study focuses on floods affecting small, dynamic urban or peri-urban basins that are poorly gauged in Abidjan, Côte d’Ivoire. We specifically examine the evolving dynamics of hydrological responses in relation to changes in the physical characteristics of these basins. To achieve this, we selected four representative catchments, analyzed historical hydrological data, and traced the urbanization dynamics in these areas.

Our results reveal that the evolution of hydrological behaviors can be correlated with urbanization dynamics. These findings offer promising opportunities to inform flood forecasting models within the framework of EWS. This work was conducted as part of a scientific project aimed at updating hydrological standards in West Africa, under the IRN ActNAO network of the Institut de Recherche pour le Développement (IRD) and UNESCO's Friend-Water program.