Global Modeling of Seasonal Mortality Rates From River Floods

Abstract

AbstractIn spite of their recurrent seasonal patterns, every year river floods affect millions of people and kill thousands. Knowing the average timing and magnitude of floods can help to better prepare for disasters by improving the humanitarian assistance and the management of resources, to ultimately reduce the overall flood impacts. This work presents a physically based modeling framework to estimate population exposed and mortality rates from river floods in all the world countries, as well as their temporal distribution within the average year. The methodology is based on a modeling cascade involving hydrologic‐hydraulic simulations at spatial resolution up to 1 km. Inundation areas are combined with population density to estimate human exposure and then aggregated at the country level. Mortality rates per region and income class are inferred from two global disaster data sets and linked through flood exposure to estimate country average annual figures. Estimates of flood impacts are then disaggregated according to the seasonal variability of floods, derived statistically from a global hydrological reanalysis for the years 1980–2018. Global human exposure from river floods is here estimated at 54 million people per year. Summer floods in Asia are the main cause of mortality from river floods, accounting for 63% of the 6,120 casualties per year globally. In contrast, the 6 months between November and April contribute to only 13% of global mortality, mostly due to floods in South America and Africa.