Global flood hazard mapping using statistical peak flow estimates

Abstract

Our aim is to produce a world map of flooded areas for a 100 year return period, using a method based on large rivers peak flow estimates derived from mean monthly discharge time-series. Therefore, the map is supposed to represent flooding that affects large river floodplains, but not events triggered by specific conditions like coastal or flash flooding for instance. We first generate for each basin a set of hydromorphometric, land cover and climatic variables. In case of an available discharge record station at the basin outlet, we base the hundred year peak flow estimate on the corresponding time-series. Peak flow magnitude for basin outlets without gauging stations is estimated by statistical means, performing several regressions on the basin variables. These peak flow estimates enable the computation of corresponding flooded areas using hydrologic GIS processing on digital elevation model.