INCORPORATION OF FLOW VELOCITY IN FLOOD DAMAGE ESTIMATION: AHR RIVER VALLEY 2021 STUDY, GERMANY

• Published in: WIT Transactions on the Built Environment Vol. 208; p. 15
• Main Authors: PAŽOURKOVÁ, EVA, Salaj, Martin, Wong Montoya, Ricardo E
• Format: Journal Article
• Language: English
• Published: Southampton W I T Press 01.01.2022
• Subjects: Catastrophic events; Damage assessment; Damage detection; Flood damage; Flooding; Floods; Flow velocity; Impact damage; Loss probability; Rainfall; Rainfall intensity; Simulation; Structural damage; Valleys; Velocity; Water depth
• ISSN: 1746-4498; 1743-3509
• DOI: 10.2495/FRIAR220021

A common practice when catastrophe model vulnerability component is being developed is the use of inundation depth as a hazard metric. However, other quantities associated with inundated waters could also yield a significant effect on estimated flood loss being widely published. The key aspect of the study is to assess the relevance of the water depth, flow velocity, and their combinations in the damage estimates done by catastrophe models. Could the flow velocity in the loss calculation process provide some benefits to the catastrophe loss modelling process? In July 2021 the western part of Germany was affected by serious flooding. The physical conditions of the inundated buildings were evaluated in the frame of the Copernicus EMS satellite-based damage assessment. The two heavily impacted areas, the towns of Schuld and Altenahr, were selected for further re-simulation aiming to test the effect of depth and velocity. In the first step, horizontal velocity and flood depth were calculated for each building using a 2D hydraulic simulation. Subsequently, structural damage models were analysed with focus on their predictive skills and variability. A force-based threshold was selected to calibrate the total loss probability, as a feature of the new vulnerability component. As a result, a hazard intensity metric expressed as combined parameter of water depth and flow velocity is finally implemented in the catastrophe model. When comparing the proposed solution with traditional depth-based approach, one can see a slight increment in the modelled monetary damage and a significantly better correlation with the observed damage identified in the study area. The embedded effect of velocity could therefore improve the accuracy and sensitivity of catastrophe flood models, particularly in high-slope areas and in events with extreme and short rainfall intensities to sudden increments in the building damage level, as assessed for the flood in the Ahr valley.