Extreme rainfall events such as floods are most destructive with regard to their consequences on economics, infrastructures and humans as well as on ecosystems and wildlife. It is expected that weather extremes will become more and more frequent and more intense for many regions in the world. During the last decades, the metropolitan city of Abidjan (Côte d’Ivoire) has been under severe heavy storm events leading the World Bank in 2018 to rank the nation among the states most threatened by disasters due to global warming. The city has been flooded frequently during the last years with a lot of economic and human losses. In 2014, 23 persons died; in 2015, 16 persons died; in 2017, 15 persons died and in 2018, 17 persons died during the flood events as shown in Figure 1.
The expected changes in magnitude and patterns of extreme precipitation due to climate extremes impose the necessity to reevaluate the significance of extreme events.The aim of this research is to investigate the skill of hydrological and hydraulic model combination and merged satellite rainfall data in forecasting urban floods over a tropical catchment characterized by scarce ground-based observations. The framework will be based on bias-corrected merged satellite data and hydrological model (HEC-HMS) and a robust 2-D shallow water model (hms) under limited past storm events and post-event measurements. The research is applied to the urban catchment of the metropolitan city of Abidjan (Côte d’Ivoire).
The research should answer the following questions?
What are the differences between satellite-based rainfall estimates and rain gauge-based estimates in the city of Abidjan?
The answers to these questions are the optimal way to overcome the challenges of the ungauged urban basins of Abidjan (Côte d’Ivoire).
The study will be carried out over the metropolitan city of Abidjan (Côte d’Ivoire). The district of Abidjan is located in the south of Cote d’Ivoire between latitudes of 5° 10′ and 5° 38′ North and longitudes of 3° 4′ and 5° 21′ West and covers an area of approximately 2,119 km2.
The methodology used in this research will be divided into four parts:
Scenarios of extreme events
From the statistical analysis, three heavy rainfall phases can be classified as shown in Figure 3. High potential (May and June), moderate potential (April, October and November) and low potential (December to March and July to September).
The heavy rainfall periods of May and June will be investigated for hydrological modelling and the corrected satellite rainfall estimates from TRMM 3B42RT and CMORPH will be used in addition to ground rainfall data.
From the ground rainfall daily maximum per year, three distribution functions have been adjusted in Figure 2.
The Chi-square and Kolmogorov-Smirnov tests were used to select the Gumbel distribution function as the best fit function. From the satellite rainfall estimated and the ground rainfall measured, it can be noticed that up to a 400 year return period rainfall has been observed (1979-2019). Therefore, a good investigation of extreme rainfall must take into account up to a 500 year return period.
The delineated catchment which will be used for hydrological modelling is shown in Figure 5.
The next steps will be to finish simulating and calibrating the rainfall-runoff model for extreme rainfall scenarios. Further, hms model will be used for hydraulic simulation to define flood-prone areas and later some adaptation measures and an early warning system will investigate.
Project head:
Scientific assistant:
Gogous Habib Gogous, M.Sc.
Project period:
April 2020 – March 2023
Funding:
Institut National Polytechnic Felix Houphouet Boigny (INP-HB) of Cote d’Ivoire