Assessing expected changes in flood losses produced by fluvial floods in urban areas in climate change

Floods are the natural hazard that causes the highest economic and human damages in Europe. Flood patterns will be modified in the future due to climate change. In addition, extent and density of urban areas have increased during the last decades. We propose a methodology to quantify the impact of climate change on river flood losses in urban areas. The methodology is applied to the metropolitan area of Pamplona in northern Spain. In this study, climate change projections are considered, a distributed hydrological model is used to obtain flood hydrographs, a two-dimensional (2D) hydrodynamic model to obtain flood extents, and finally a flood loss model to quantify direct flood damages.

The effect of climate change on flooding in the Arga river catchment has been estimated by combining the distributed hydrological model RIBS (Real time Interactive Basing Simulator) with delta changes in daily precipitation quantiles obtained from climate change projections in previous studies. 12 climate models, seven return periods, two representative concentration pathways (RCP 4.5 and RCP 8.5), and three periods (2011–2040, 2041–2070, and 2070–2100) are considered (Garijo and Mediero, 2019; Lompi et al., 2021). 33 %, 50 %, and 67 % percentiles of flood quantiles in climate change are considered.

The 2D hydrodynamic IBER model has been calibrated by using 15-minute streamflow data recorded at four gauging stations that belong to the real-time SAIH system of the River Ebro Basin Authority. Flood extents simulated with IBER are compared with flood extents in real flood events supplied by the Regional Government of Navarre. A high resolution digital terrain model (DTM) with a cell size of 1 meter has been used. as input data of the IBER model. For each climate change scenario, flood extent and water depth in each DTM cell in the metropolitan area of Pamplona are obtained with the calibrated IBER hydrodynamic model.

The Safer_DAMAGE algorithm developed by the SaferPlaces project has been used to estimate direct flood losses to residential and commercial buildings in urban areas (Paprotny et al., 2021). Safer_DAMAGE uses the occupancy data provided by OpenStreetMap Buildings and average water depths in buildings to estimate flood losses at the building scale. The Safer_DAMAGE algorithm has been benchmarked in the Pamplona metropolitan area by using the insurance database of observed flood losses in the period 1996-2019 supplied by the Spanish Consorcio de Compensación de Seguros. In this database, data are aggregated by postal codes. Flood losses are estimated for each synthetic flood event with the calibrated Safer_DAMAGE. Total direct damages have been obtained in selected neighbourhoods that are prone to flooding. In addition, expected changes in direct damages due to climate change have been assessed in terms of building type. Finally flood losses are expected to be smaller in the future for low return periods (2-50 years). However, an increase in flood losses is expected for high return periods (50-1000 years).

The methodology proposed in this study can be useful for assessing the impact of climate change on flood losses in urban areas.