Urban Pluvial Flood Risk Assessment: Challenges and Opportunities for Improvement Using a Community-Based Approach

Although there are different types of flood risk assessments, methodologies, and goals, the framework behind them follows a similar standard, from defining a scenario and acquiring specific basic data to obtaining key flood hazard parameters. Ground measurements, remote sensing analyses, and hydrological modeling are the current tools applied in flood hazard assessment. These classical tools present some specific constraints related to their intrinsic characteristics of the context of the application. For ground data acquired by different instruments, spatial and temporal coverage is a great challenge, both in terms of data processing costs and a representative's maintenance. As for hydrological modeling, it is very sensitive to the quality and quantity of input data, affecting the results' acceptability and accuracy. Moreover, the absence of comprehensive watershed data such as details of drainage systems can complicate the watershed modeling process, especially when trying to simulate or predict extreme events in unique locations. In this study, we modeled an urban watershed response to 2- and 5-year return period rainfall events by using basic, existing data and applying a physically based, spatially distributed urban watershed model, PCSWMM. To pursue these questions, participatory research was applied in a low-resourced community in the City of Paterno, in Sicily, Italy. We collected as unbiased as possible information about pluvial flooding events through 207 surveys. The locally gathered data were compared to the advanced hydrodynamic model results to explore this type of model's challenges and shortcomings at a large city-scale. Finally, we discuss the opportunities to improve urban pluvial flood modeling at the small-scale through community engagement in data collection and characterization of urban watersheds. Instead of handling the modeling at a large scale, we can identify priority areas and, consequently, have better modeling by focusing on small scale watershed modeling and capture related details.