城市暴雨内涝情景模拟与灾害风险评估_英文_（全文完整）

下面是小编为大家整理的城市暴雨内涝情景模拟与灾害风险评估_英文_（全文完整）,供大家参考。

　DOI: 10.1007/s11442-011-0844-7 © 2011 Science Press Springer-Verlag Community-based scenario modelling and disaster risk assessment of urban rainstorm waterlogging

　 YIN Zhan’e 1 , YIN Jie 2 , * XU Shiyuan 2 , WEN Jiahong 1

　 1. Geography Dept. of Shanghai Normal University, Shanghai 200234, China; 2. Geography Dept. of East China Normal University, Shanghai 200062, China Abstract: Scenario modelling and the risk assessment of natural disasters is one of the hot- spots in disaster research. However, up until now, urban natural disaster risk assessments lack common procedures and programmes. This paper selects rainstorm waterlogging as a disaster to research, which is one of the most frequently occurring hazards for most cities in China. As an example, we used a small-scale integrated methodology to assess risks relating to rainstorm waterlogging hazards in the Jing’an District of Shanghai. Based on the basic concept

　of

　disaster

　risk,

　this

　paper

　applies

　scenario

　modelling

　to

　express

　the

　risk

　of small-scale urban rainstorm waterlogging disasters in different return periods. Through this analysis of vulnerability and exposure, we simulate different disaster scenarios and propose a comprehensive analysis method and procedure for small-scale urban storm waterlogging disaster risk assessments. A grid-based Geographical Information System (GIS) approach, including an urban terrain model, an urban rainfall model and an urban drainage model, was applied to simulate inundation area and depth. Stage-damage curves for residential buildings and contents were then generated by the loss data of waterlogging from field surveys, which were further applied to analyse vulnerability, exposure and loss assessment. Finally, the ex-ceedance probability curve for disaster damage was constructed using the damage of each simulated event and the respective exceedance probabilities. A framework was also devel-oped for coupling the waterlogging risk with the risk planning and management through the exceedance probability curve and annual average waterlogging loss. This is a new explora-tion for small-scale urban natural disaster scenario simulation and risk assessment. Keywords: scenario modelling; small-scale; rainstorm waterlogging; disaster risk assessment; Shanghai 1 Introduction With global climate change and rapid urbanization, the intensity and frequency of urban natural disasters and associated losses are increasing (Xu et al., 2006; Shi, 2006; Zhou and Received: 2010-10-20

　 Accepted: 2010-11-28 Foundation: National Nature Science Foundation of China, No.41071324; No.40730526; Key Subject Developing Project by Shanghai Municipal Education Commission, No.J50402; Science and Technology Commission of Shang-hai Municipality, No.08240514000; Leading Academic Discipline Project of Shanghai Normal University, No.DZL809 Author: Yin Zhan’e (1963–), Ph.D and Associate Professor, specialized in natural hazards and remote sensing. E-mail: zhaneyin@126

　 * Corresponding author: Xu Shiyuan, Professor, E-mail: syxu@ecnu.edu

　geogsci

　 springerlink

　/content/1009-637X

　 Yuan, 1993). Waterlogging is one of the most serious natural disaster phenomena, not only in Chinese cities but also around the world, causing considerable personal injury and prop-erty damage. For example, more than 200 roads and 50,000 buildings and houses were in-undated by waterlogging from Typhoon Matsa in 2005 in Shanghai and the direct economic loss was about RMB 1358 million yuan. Thus, it is important to take into account disaster management to reduce economic loss, social impact and casualties from natural hazards (Su et al., 2005; Hays, 1991). In disaster management, risk assessments have become a research hotspot. There are different methods to conduct large-scale disaster risk assessments. How-ever, few methodologies are available for small-scale disaster risk assessments. The main objectives of this study are to provide a framework and an alternative method for small-scale risk assessments of urban waterlogging. Within the framework, the Jing’an District (an ur-ban centre and an area in Shanghai prone to waterlogging) is taken as a case study to dem-onstrate the proposed methodology. 2 Materials and methods 2.1

　Study area and datasets

　The Jing’an District, with its centre located at 31.13ºN and 121.26ºE, is one of the 19 ad-ministrative districts in Shanghai, China (Figure 1). With an area of 7.62 km 2 , it is com-prised of five blocks, with a population reaching 248,400 and a density of 32,598 peo-ple/km 2

　in 2008 (Shanghai Municipal Statistics Bureau, 2009). Most of the region is very flat and low lying, with an average altitude of 3 m above Wusong Datum. Due to a decrease in the number of rivers in the city centre and a subsequent weakening of excess-water stor-age capacity (Cheng et al., 2007), surface runoff has increased. As one of the traditional ur-ban centres, this area has the typical characteristics of an urban landscape, such as crowded buildings, heavy traffic, an “urban heat island” and “urban rain island”, thus making the Figure

　1

　Location

　of

　the

　study

　area

　(1-Jing’an;

　2-Changning;

　3-Putuo;

　4-Zhabei;

　5-Hongkou;

　6-Yangpu; 7-Huangpu; 8-Luwan; 9-Xuhui)

　276 Journal of Geographical Sciences District more vulnerable to waterlogging. The data sources used in this study include the administrative boundary data of Shanghai in 2006, the census data of Shanghai in 2006, land use and land cover maps of Shanghai in 2006, the topographic contours (0.5 m contour intervals) data relative to the Wusong Vertical Datum of Shanghai in 2005 and the building footprints of the Jing’an District obtained by manual visual interpreting from aerial photographs (0.36 m horizontal resolution) in 2006. Waterlogging loss data was obtained from in-situ investigation on a house-by-house basis using traditional paper questionnaires. 2.2 Methods Risk can be defined as the possibility of expected losses (e.g., casualties, injuries, property damage and disruption of economic activities or environmental degradation) due to hazard-ous events in a given area during a specific reference period (UNDRO, 1991; Helm, 1996). It can be expressed as follows: R = H∩V∩E (1) where R is risk; H is hazard, the probability of a potentially damaging physical event, phe- nomenon or human activity that may cause an expected loss within a determined period and region; V is vulnerability, the degree of loss to a given element at risk or a set of elements at risk resulting from the occurrence of a natural phenomenon of a given magnitude; E is ex-posure, the element at risk including buildings, population, property or other human activities. As a major and frequently occurring natural hazard, waterlogging risk can also be as-sessed using the above equation. So far, common experimental hydrological models and hy-draulics numerical simulations have only been used to analyse the characteristics of water-logging hazard. The experimental hydrological models, such as STORM and SCS (U.S. Soil Conservation Service, 1972) models (Yang and Wu, 2009) are simple but it can hardly ex-press the dynamics and the detailed spatial information. Conversely, the 1D and 2D hydrau-lics

　numerical simulations are

　useful in

　depicting the

　spatial and

　temporal distribution (Zheng and Hu, 2003; Cong et al., 2006; Qiu et al., 2000; Li et al., 2009). However, these methods require much specific data, which cannot be acquired easily and demand massive data processing; thus, it is difficult to use them for small-scale assessments (Yen and Akan, 1999).

　Other than hazard assessment

　data, vulnerability information remains limited in China. Less attention has been paid to the construction of vulnerability assessment models and the investigation of disaster damage. Based on previous studies (Wang et al., 2004; Zhao et al., 2004; Jonkman et al., 2008; Dutta et al., 2003), a comprehensive urban waterlogging risk assessment model was devel-oped

　to

　simulate

　different

　inundated

　scenarios

　and

　build

　vulnerability

　curves

　(or stage-damage curves/loss function); we then calculated the expected loss. Considering the complexity of underlying surface and human-induced disturbances, grid-based scenario analytical approaches (including urban terrain modelling, urban rainfall modelling and urban drainage modelling) were employed to derive the hazardous potential in hazard analysis. Several vulnerability curves were built for different human activities based on field surveys after waterlogging in Shanghai. Human activities that are submerged beneath waterlogging can be revealed in exposure analysis. Finally, the risk can be expressed as an exceedance

　 Figure 2

　Framework of rainstorm waterlogging risk assessment based on a GIS-grid probability curve, a risk curve and an average annual waterlogging loss. Figure 2 exhibits the framework for the urban waterlogging risk assessment. 3 Hazard analysis As the first step of risk assessment, hazard analysis is to identify some parameters of haz- ardous potential in different return periods (Plate, 2002). In this paper, water depth was per- haps the best indicator for the impact from waterlogging. Based on previous studies, the ur- ban terrain model, the urban rainfall-runoff model, the urban drainage model and the “hy- drology” tool in the “Spatial Analyst” module of ArcGIS were integrated within a compre- hensive analysis procedure to detect the depth of inundated water.

　3.1 Catchment area

　A topographic dataset in the form of a digital elevation model (DEM) is the basis for hydro- logic and flood inundation modelling. The elevation in the topographic contours was inter- polated to generate a DEM with a grid cell resolution of 30 m using the ArcGIS 9.2 software. Considering that the direction of flow is affected by gravity, the flow direction map was generated using the Shanghai DEM; the catchments were then obtained using the GIS ap- plication. Some catchments, which were involved in the Jing’an District, were merged into a watershed. Furthermore, the watershed boundary was modified according to actual urban terrain characteristics, such as river banks and roads....