High resolution modelling and analysis of urban flooding

Abstract

The predictions on climate change and precipitation patterns in the near future, as well as urbanization, lead to higher volumes of runoff, which in turn challenge the sustainability and functionality of urban drainage pipe networks. In order to understand how these changes will affect the existing infrastructure in urban areas, integrated urban water model software are needed to identify effective solutions to minimize urban flooding. The aim of this thesis is to test and examine a newly developed, vector-based, high resolution software model, the Tokyo Storm Runoff (TSR) model in the small urban catchment of Augustenborg located in the southern part of Malmö, Sweden. It was found that the TSR model demonstrated the capability to simulate high resolution urban flooding and highlight, in extreme detail, areas of the catchment that are more vulnerable to flooding. Precipitation appeared to infiltrate permeable areas until infiltration capacity is reached and excess runoff is observed to flow with respect to Digital Elevation Model (DEM). Due to lack of observed data, no calibration or tuning was performed but standard parameter values from previous studies were used. Although the potential underestimation of inundation depth remains a limitation of this study, the model was able to determine relative performance of drainage outlets within the catchment. It is concluded that TSR software model represents a reliable, effective, and fully customizable tool in terms of urban flood analysis based on comparison of the simulation results between TSR and MIKE Urban and the similarities that both have in common. Despite the TSR model’s minimal requirements for input data, it requires excessive manual effort and therefore a further code development is recommended. This thesis is a part of the research project FloodView coordinated at the Department of Water Resources Engineering, Lund University. FloodView is an international research project aiming to provide a web-based flood control decision support system for various stakeholders, such as municipality managers/decision makers to identify effective solutions to minimize urban flooding.

Abstract (Swedish)

Förutsägande om klimatförändringar och nederbördsmönster i den närmaste framtiden samt urbanisering leder till högre volymer av avrinning, vilket i sin tur utmanar hållbarheten och funktionaliteten hos stadsdräneringsnät. För att förstå hur dessa förändringar kommer att påverka den befintliga infrastrukturen i stadsområden behövs integrerad programvara för urban vattenmodell för att identifiera effektiva lösningar för att minimera urbana översvämningar. Syftet med denna avhandling är att testa och undersöka en nyutvecklad, vektorbaserad mjukvarumodell med hög upplösning, Tokyo Storm Runoff (TSR) modellen i ett urbant område, Augustenborg, beläget i södra delen av Malmö. Det visade sig att TSR-modellen visade förmågan att simulera fin upplösning i städerna och belysa i yttersta detalj områden av avloppet som är mer utsatta för översvämningar. Utfällning verkade infiltrera permeabla områden tills infiltrationskapaciteten uppnåtts och överskridande avrinning observeras flöda med avseende på Digital Elevation Model (DEM). På grund av brist på observerade data utfördes ingen kalibrering eller avstämning men standardparametervärden från tidigare studier användes. Även om den potentiella underskattningen av översvämningsdjupet fortfarande är en begränsning av denna studie, var modellen kapabel att bestämma relativ prestanda av dräneringsutlopp i avloppet. Det konstaterats att TSR-modell representerar ett tillförlitligt, effektivt och fullt anpassningsbart verktyg i termer av översvämningsanalys baserat på jämförelse mellan simuleringsresultaten mellan TSR och MIKE Urban och de likheter som båda har gemensamt. Trots TSR-modellens minimikrav på ingångsdata krävs en omfattande manuell ansträngning och därför rekommenderas ytterligare kodutveckling. Denna avhandling ingår i ett forskningsprojekt FloodView samordnad vid Avdelning för tekniskvattenresurslära, Lunds universitet. FloodView är ett internationellt forskningsprojekt som syftar till att tillhanda-hålla ett webbaserat beslutsfattningssystem för översvämningskontroll för olika intressenter, såsom kommunförvaltare / beslutsfattare för att identifiera effektiva lösningar för att minimera urbana översvämningar.

Popular Abstract

Among all natural disasters in Europe, floods are the most common ones causing enormous damage to environment and society. The predictions on climate change and precipitation patterns in the near future as well urbanization lead to higher volumes of runoff, which in turn challenge the sustainability and functionality of urban drainage pipe networks. In order to understand how these changes will affect the existing infrastructure in urban areas, integrated urban water model software, are needed to identify effective solutions to minimize urban flooding. In this study, a newly developed urban water model software, the Tokyo Storm Runoff (TSR) was tested and examined for the small urban area Augustenborg in Malmö, Sweden.
It was found that the TSR model demonstrated the capability to simulate high resolution urban flooding, determine relative good performance of drainage outlets within the catchment and highlight in extreme detail segments and areas that are more vulnerable to flooding. Precipitation appeared to infiltrate permeable areas until infiltration capacity was reached, and excess runoff was observed to flow with respect to surface elevation. Based on comparison of the simulation results with a well-known commercial software, the MIKE Urban, it was found that TSR model represents a reliable, effective and fully customizable tool in terms of urban flood analysis. In this study, five different rainfall scenarios were simulated, four hypothetical design rainfalls with different recurrence interval periods (10, 20, 50, 100 years) and one real precipitation from 2014 that resulted in major floods within the city of Malmö have been tested and simulation results have been analyzed.
However, TSR as a newly developed software can be categorized as a research tool software with limited access, though. It is still not supported by a graphic user interface, meaning that, as an executive file, it requires excessive manual effort. Although the relative good performance of the model, this study suggests that the model would benefit from further coding developments to improve usability. For instance, assistance to correctly format input data and creation of executive files to read and analyse the exported simulation data, would result in a more user-friendly software. It is further noted that the model requirement for inlet manholes data (e.g. gutter location) was found to be at times absent from the pipe network data package.
As for future research, the simulation results of the study present valuable information in terms of the functionality of the existing drainage pipe network as well as areas of the catchment that appear more vulnerable to flooding. Lastly but not least, the results and the TSR model setup for the small catchment of Augustenborg can be used for further developing of the model. Collection of observed data during a rainfall event in Augustenborg catchment may be used for further calibration and validation of the existing model.

• author: Kosmas, Angelos
• supervisor: Linus Zhang; Ramesh Saagi
• organization: Division of Water Resources Engineering
• course: VVRM01 20172
• year: 2019
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: Tokyo Storm Runoff model, TSR, urban flooding, Augustenborg, stormwater management, pipe networks
• publication/series: TVVR18/5008
• report number: 18/5008
• ISSN: 1101-9824
• language: English
• additional info: Examiner: Magnus Persson
• id: 8976718
• date added to LUP: 2019-09-19 13:53:07
• date last changed: 2019-09-19 13:53:07