A geomatic method for flood damage analysis at national scale

Floods are among the most threatening natural hazards worldwide and it is expected that, due to climatic changes, the increasing urbanization and the intensification of economic activities in flood-prone areas, the consequent damages are going to significantly escalate. In this view, a large-scale flood risk analysis that properly evaluates and quantifies the three components of risk (hazard, exposure and vulnerability) is essential in order to support national and global policies, emergency operations, land use management and expeditious damages assessment. Nevertheless, limits in time and data represent significant hurdles to the application of a correct evaluation of flood risk management at the national scale. In this context, the present study proposes a cost-efficient geomatic method for a large-scale analysis of direct economic flood damage for the entire Romanian Country at 30-m resolution. The proposed methodologi-cal framework consists of three main stages: (i) deriving a flood-extent and water depth map for the full study area with a DEM-based method. Specifically, the map is obtained by performing a supervised linear binary classification based on the Geomorphic Flood Index, which combines the geomorphic information extracted from the DEM along with existing detailed information about flood hazard exposure, typically available for small parts of a basin (a reference flood hazard map of at least 2% of the basin of interest is required); (ii) generating an exposed land-use map developed from multi-spectral Landsat 8 satellite images using machine-learning classification algorithms trained with existing detailed land-use map, i.e. Urban Atlas (available for few large cities in Europe); (iii) applying the free and open source FloodRisk GIS tool to carry out the flood damage estima-tion based on the vulnerability (depth-damage) curves method. The demonstrative application of the proposed method over the entire territory of Romania (including minor order basins) for the 100-year return time, showed how the description of flood risk may particularly benefit from the integrated use of geomorphic methods, machine learning algorithms and Earth Observation freely available data. The ability of the proposed cost-efficient model to carry out medium-resolution and large-scale analyses in data-scarce environments could make it possible to perform future risk assessments and manage-ment, keeping abreast also of temporal and spatial changes in hazard, exposure, and vulnerability