Assessing the impacts of coastal and riverine urban floods in the future climate, results from the SCORE project.

Assessing the local effects of climate change on coastal areas, and in particular on coastal cities and settlements, is one of the greatest challenges facing our society, aimed at finding innovative and sustainable solutions to increase the resilience of coastal communities to adverse climatic actions. In particular, the use of climate data is crucial for defining a downscaling strategy that starts with climate services on a global scale and goes on to define impacts on a local scale. A comprehensive 'global to local' approach is fundamental to envisage coastal flooding problems. Estimating the effects of CC in coastal cities requires increasing the resolution of urban-scale models to unprecedented levels, to simulate land and coastal flooding conditions for various scenarios and with different return periods, also allowing for the evaluation of financial resilience strategies or ecosystem solutions for adaptation, following a true multidisciplinary approach and fostering, through participatory approaches, the public engagement of citizens, scientists and policy-makers, to identify solutions technically and socially acceptable. We present the results of a full global to local study, to estimate the effects of coastal and riverine floods associated with extreme events at three coastal cities located in the Mediterranean Sea (Massa - Italy and Villanova - Spain) and the Bay of Biscay (Oarsoaldea - Spain), for different climate projections. The present work is part of the H2020-SCORE project and the analyzed cities are organized in a network of Coastal City Living Labs (CCLLs). We implemented a relocatable modeling chain which uses the data from a Regional Circulation Model (RCM) provided by EuroCordex as atmospheric forcing to three different models: i) a WaveWatchIII model for the simulation of wave forcing, ii) a SHYFEM model for the simulation of storm surges and sea-level dynamics iii) a LISFLOOD model for river discharge. The wave and sea-level models are implemented on unstructured grids with increasing resolution at the target cities, whereas the river discharge is determined considering the basin located upstream of the city. The simulations are performed for the evaluation, historical, RCP45 and RCP85 datasets associated with the CMIP5 experiment. Time series of wave height and water level close to the coast, and river discharge are employed in an extreme value analysis procedure to obtain values associated to specific return periods (namely 25, 100, 200 years). These values are employed to simulate floods due to the effect of storm surge and peak river discharge, by means of a hydraulic model built on a high-resolution digital elevation model of the coastal city, including information on buildings, coastal bathymetry and river cross sections. Preliminary results from the calculated hazard maps (water depth associated with a return period event) show interesting differences in the three analyzed coastal cities based on different exposure to coastal or riverine floods.