Evaluating the impact and risk of pluvial flash flood on intra-urban road network: A case study in the city center of Shanghai, China
Introduction
Pluvial flash flood (PFF) is among the most common and destructive natural hazards, resulting in considerable direct losses (e.g. personal injury and property damage) and increasing indirect impacts (e.g. interruption of public services and economic activities), especially in the urbanized areas around the world (e.g. Ramos et al., 2005, Mejia and Moglen, 2010, Smith et al., 2012, Wright et al., 2012). PFF is a rapid flood caused by heavy rain and can be distinguished from a regular pluvial flood by a short timescale, generally less than six hours. It is usually by their very fast evolution and occurs within minutes or a few hours of excessive rainfall (Naulin et al., 2013). There exists a broad consensus that the combined effect of climate change and rapid urbanization is generally recognized as the primary cause for more frequent, heavier rainfall–runoff (IPCC, 2013, Du et al., 2012, Suriya and Mudgal, 2012, Zhou et al., 2012). Furthermore, the lack of anticipation of PFF events such as the unavailability of short-term forecasting and warning, combined with insufficient, postponed adaption measures (e.g. inadequate drainage capacity) largely limit the efficiency of urban flood risk management, leading to the enhanced consequences of these events in most cities.
In all the flood prone areas, urban road network may be the major assets affected first by inundations, which cause not only infrastructure damage but also transportation disruption, due to its low-lying nature compared to the neighborhood and high-density throughout urban territory. More seriously, almost half of the flash flood casualties involve people on flooded roads by trapping in their cars or by escaping in the rapid rise of open water (Drobot et al., 2007, Versini et al., 2010a). Over the past several decades, the significant disasters associated with street networks and PFF events have been frequently occurring in different urban environments such as New York and London in developed countries as well as Beijing and Bangkok in developing countries (Pitt, 2008, Hung et al., 2009, NYC Emergency Management, 2014). Road networks of China’s cities are particularly vulnerable to PFF events as the conflict between rapid urbanization and the lagging urban (infrastructure) planning is emerging onto the surface. This is evidenced by the recent PFF events in major Chinese cities, including Beijing, Shanghai, Guangzhou, Shenzhen, Nanjing, and Hangzhou (Zhang et al., 2012). For example, the July 2012 PFF event in Beijing city led to widespread transportation disruptions and claimed 79 lives, among which a large number of fatalities were caused by road inundations (Yin et al., 2015).
Although there is a huge demand for understanding urban road network risk to PFFs, until now, very few work have attempted to systematically examine the potential impacts of flash flood events on urban road systems largely due to the lack of adequate data and observations concerning road flooding, difficulty in numerically modeling the PFF dynamics and the complexity of cascading effects resulting from temporary flooding on transportation systems (Suarez et al., 2005). Of the existing studies, a number of studies are often conducted for the purpose of flood emergency evacuation (e.g. Chang et al., 2007), accessibility loss (e.g. Sohn, 2006) or traffic delay (e.g. Zhou et al., 2012), while some others address short-term (and even real-time) forecasting or warning of road inundation risk (RIR) at specific intersections of streams and roads. In terms of road flood risk assessment, the extension of distributed hydro-meteorological model to road network scan approach have been usually applied or tested in a number of regions. A representative methodology is Road Inundation Warning System (RIWS) which combines a rainfall–runoff model with a road inundation susceptibility analysis (Versini et al., 2010b). Chang et al. (2010) integrated a hydrologic model (PRMS), a 1D hydraulic model (HEC-RAS) and a travel forecast model into an assessment framework of urban flooding and transportation systems. Moreover, a LRSRM (Local Regional Scale Risk Model) was designed to identify the interruption of a road network due to a hazardous event from a multi-scale perspective by using the biclustering technique (Freiria et al., 2014). The previous studies, however, mainly focus on regional (basin) scale and intercity level, paying little attention to PFF-induced RIR at the intra-urban level. Also, they can not explicitly provide detailed information of 2D surface flood routing on urban road network in time and space to allow a real-time risk management. In recent years, the development of simplified raster-based flood models (e.g LISFLOOD and FloodMap) combined with the existing high resolution LiDAR (i.e. airborne and terrestrial laser scanners) techniques makes it possible to accurately simulate PFF processes on intra-urban roads and the output inundation maps (both flood depth and spatial extent) could further be used to identify the road network at risk and to take the appropriate risk mitigation measures (Bates and De Roo, 2000, Yu and Lane, 2006a, Yu and Lane, 2006b, Sampson et al., 2012).
Within this study, an integrated methodology is proposed by incorporating flood inundation modeling, transportation disruption (i.e. road closure) detection and risk assessment to measure the impact and risk of PFF on intra-urban road networks. This is the first attempt to apply high resolution 2D hydrodynamic model and flood depth-dependent measure for evaluating the impact of pluvial flooding on road disturbance at the intra-urban level. The city center of Shanghai, China, where roads are in particular prone to pluvial flash flooding, is selected as a case study area. The road inundation risk analysis is focused on transportation disruption rather than on infrastructure damage due to PFFs. The main objectives in this paper are to: (a) develop a step-by-step approach to investigate the transportation impact from temporary pluvial flooding for intra-urban road network; (b) explore the spatio-temporal characteristics of pluvial flash flooding on intra-urban roadways; (c) quantify intra-urban transportation disruption risk under a range of PFF magnitudes. The rest of this paper is organized as follows: Section 2 is devoted to the presentation of the materials and methods, including the introduction of study area, the available data, the description of flood modeling, road impact and risk assessment; Section 3 presents the results and discussions; the conclusions are given in Section 4.
Section snippets
Study area
The city center of Shanghai, China has been chosen as study site because it exhibits high risk of pluvial flash flooding and also associated impact of transportation disruption. The study area located at the junction of Huangpu River and Suzhou Creek, bordering with South–North Elevated Highway and Yan’an Elevated Highway (Fig. 1). It covers about 3.25 km2 on the north part of Huangpu District with a mild and low-lying topography (about 3 m above Wusong Datum). The region belongs to a northern
Road flooding
The predicted spatial and statistical results of maximum inundations (extent and depth) throughout the simulations for different flood return periods are presented in Fig. 2 and Table 1 respectively. In general, the derived inundation patterns are characterized by a high degree of consistency among the five scenarios, albeit with various magnitudes corresponding to different return periods. Almost total inundations (above 5 cm) are observed in the road networks except scattered ponding in
Conclusions
This paper presents a novel approach to identify the impact and risk of pluvial flash flooding on intra-urban road networks through an integration of pluvial flood numerical modeling, GIS-based road failure analysis and risk assessment model. The methodology proposed here could be adopted for applications in other PFF-prone urban areas facing similar road problem. The results obtained are promising and indicate that given the flash nature of rainfall, extensive pluvial flood mainly occurred in
Acknowledgments
This work was supported by the Open Research Fund of State Key Laboratory of Estuarine and Coastal Research (Grant No: SKLEC-KF201407), the National Natural Science Foundation in China (Grant Nos: 41201550, 41371493, 41401176, 41401600, 71373084), Project of Joint Center for Shanghai Meteorological Science and Technology (Grant No: 2015-03), Innovation Program of Shanghai Municipal Education Commission (Grant No: 13YZ061, 13ZZ035).
References (41)
- et al.
A simple raster-based model for flood inundation simulation
J. Hydrol.
(2000) - et al.
A simple inertial formulation of the shallow water equations for efficient two-dimensional flood inundation modelling
J. Hydrol.
(2010) - et al.
An objective assessment of soil moisture deficit models
J. Hydrol.
(1983) - et al.
A scenario planning approach for the flood emergency logistics preparation problem under uncertainty
Transport. Res. E
(2007) - et al.
Assessing the effects of urbanization on annual runoff and flood events using an integrated hydrological modeling system for Qinhuai River basin, China
J. Hydrol.
(2012) - et al.
Risk factors for driving into flooded roads
Environ. Hazards
(2007) - et al.
Spatially distributed flood forecasting in flash flood prone areas: application to road network supervision in Southern France
J. Hydrol.
(2013) - et al.
Visualization of storm severity
J. Hydrol.
(2005) - et al.
Use of terrestrial laser scanning data to drive decimetric resolution urban inundation models
Adv. Water Resour.
(2012) Evaluating the significance of highway network links under the flood damage: an accessibility approach
Transport. Res. A
(2006)