Real-time mapping of flood risk in Mali based on rainfall forecasts, remote sensing, and deep learning

As a landlocked country, Mali is one of the most vulnerable to climate stress due to its socio-economic status, geographical location, and climate-sensitive economy. Two-thirds of the country lies in the arid Sahara and the semi-arid Sahel. Mali is exposed to recurrent extreme events, including severe droughts, variable rainfall, and catastrophic floods.

Floods are among the world's most devastating natural disasters, causing thousands of deaths, affecting billions of people, and costing the world billions of dollars every year. In Mali, fluvial and pluvial floods cause loss of life and property almost every year. Throughout the Sahel, there has been a marked increase in the frequency and severity of floods in Mali. Climate projections also suggest that extreme rainfall (and hence flooding) will become more frequent in the future. 

The overall objective of this technical assistance will be to strengthen the existing early warning system for the risk of flooding in Mali based on rainfall and water level forecasts, with the help of remote sensing and deep learning. The specific objectives of this technical assistance are as follows:

- To overcome the lack of accurate data for developing hydrological models using deep learning models in a municipality in Mali (proposed municipality to be confirmed: Sébékoro) through the use of satellite and UAV data. 

- Precise characterization of the types of infrastructure in at-risk areas 

- Integrate the PGRCI's hydrological models and flood warning system in the selected rural commune. 

- Implement a low-cost hazard monitoring technology based on the use of microcontrollers connected to a pressure sensor and a GSM card to transmit water levels in the selected area.

Final geographical scope
Final objective
Countries
Mali
Implementation scale
Response project