Safer_RAIN: A DEM-Based Hierarchical Filling-&-Spilling Algorithm for Pluvial Flood Hazard Assessment and Mapping across Large Urban Areas

Abstract

The increase in frequency and intensity of extreme precipitation events caused by the changing climate (e.g., cloudbursts, rainstorms, heavy rainfall, hail, heavy snow), combined with the high population density and concentration of assets, makes urban areas particularly vulnerable to pluvial flooding. Hence, assessing their vulnerability under current and future climate scenarios is of paramount importance. Detailed hydrologic-hydraulic numerical modeling is resource intensive and therefore scarcely suitable for performing consistent hazard assessments across large urban settlements. Given the steadily increasing availability of LiDAR (Light Detection And Ranging) high-resolution DEMs (Digital Elevation Models), several studies highlighted the potential of fast-processing DEM-based methods, such as the Hierarchical Filling-&-Spilling or Puddle-to-Puddle Dynamic Filling-&-Spilling Algorithms (abbreviated herein as HFSAs). We develop a fast-processing HFSA, named Safer_RAIN, that enables mapping of pluvial flooding in large urban areas by accounting for spatially distributed rainfall input and infiltration processes through a pixel-based Green-Ampt model. We present the first applications of the algorithm to two case studies in Northern Italy. Safer_RAIN output is compared against ground evidence and detailed output from a two-dimensional (2D) hydrologic and hydraulic numerical model (overall index of agreement between Safer_RAIN and 2D benchmark model: sensitivity and specificity up to 71% and 99%, respectively), highlighting potential and limitations of the proposed algorithm for identifying pluvial flood-hazard hotspots across large urban environments.

Using the present to estimate the future: A simplified approach for the quantification of climate change effects on urban flooding by scenario analysis

“…In the last decades, large efforts have been focused on the modelling of urban floods, with different aims such as hazard mapping, risk communication, probabilistic approaches, design and management of green/grey risk mitigation and climate adaptation solutions (Al-Suhili et al, 2019;Bates et al, 2010;Costabile, Costanzo, De Lorenzo, et al, 2021;Di Baldassarre et al, 2010;Hammond et al, 2015;Nguyen et al, 2006). Many approaches and models exist that are able to tackle those goals with different degrees of detail, ranging from empirical/conceptual black-box models (e.g., Chen et al, 2009;Samela et al, 2020) to rigorous approaches solving mono-, two-or threedimensional problems (Bulti & Abebe, 2020;Costabile, Costanzo, De Lorenzo, & Macchione, 2020;Di Baldassarre et al, 2010;Henonin et al, 2013;Martins et al, 2017;Martins et al, 2018;Taskinen & Bruen, 2007). According to the specific aim of the analysis and to the capabilities of the modelling tools, flood models can account for the rainfall-runoff transformation sub-processes, such as infiltration and interaction between overland flow and the drainage system, with varying degrees of simplification.…”

3D GIS Platform for Flood Wargame: A Case Study of New Taipei City, Taiwan

“…This method takes into account the drained area and local gradient. Samela et al [24] adopted hierarchical filling-and-spilling based on DEM for pluvial flood hazard assessment. The proposed method can handle the flow connectivity among low-lying areas such as pits.…”

Efficient Hazard Assessment for Pluvial Floods in Urban Environments: A Benchmarking Case Study for the City of Berlin, Germany

“…In recent years, the increasing availability of higher resolution DEM has encouraged not only the advanced application of the 2D hydrodynamic models but also the development of quick and simplified methods that are based on DEM as valid alternatives for flood hazard mapping. These simplified methods have been mainly developed and applied for mapping fluvial flood hazards, and their application for pluvial flooding hazard mapping is still sparse in the literature [18]. They are less accurate than using detailed hydrologic and 1D-2D hydrodynamic models, but they are faster and require fewer input data.…”

“…Many of the simplified models are based on the fill-spill-merge (FSM) concept where the excess runoff concentrates in the DEM depressions and when a depression is completely filled with excess runoff, the excess runoff spills, and flows downstream towards lower elevated depressions. References [14,18,19] applied the FSM method to different case studies in China, Italy, and Denmark. FSM can map the pluvial flood hazard in urban watersheds quickly as it only shows the final inundation extent.…”

“…FSM can map the pluvial flood hazard in urban watersheds quickly as it only shows the final inundation extent. Reference [18] evaluated the FSM method performance based on evaluating the estimated inundation area extent with the obtained inundation area extent from a 2D hydrodynamic model. However, they did not evaluate the water depth, which is vital for damage assessment.…”