Introduction

With the rise of superlarge and urban areas around the world, the total economic volume of cities continues to rise, while their economic density gradually increases. According to the National Bureau of Statistics “Statistical Chart of Economic and Social Development: Basic Information on the Population of Megacities in the Seventh National Census,” there are seven megacities in China: Shanghai, Bei**g, Shenzhen, Chongqing, Guangzhou, Chengdu, and Tian**. All these megacities are sorted by population.Footnote 1 As a result of the population and economic agglomeration characteristics of cities, the human casualties and economic losses caused by natural disasters are becoming increasingly serious (Naef 2022). As a result, the losses caused by waterlogging disasters caused by rainstorms are constantly expanding. The ensuing typhoon “fireworks” have affected a total of 2.711 million people in Zhejiang, Shanghai, Jiangsu, and Anhui provinces and cities, causing a total of 2.74 million people to be displaced. More than 1100 houses were damaged, and the direct economic loss was 3.35 billion yuan. Due to the continuous improvement of the urban economic system, the strengthening of infrastructure, the continuous improvement of the social system, and the strengthening of the links between systems, sudden disasters are expected to cause more serious losses in the future (Chen and Quan 2021). Shanghai is a megacity with a total area of 6340.5 km2. It is generally low-lying and flat, with an average elevation of less than 5 m. Through regional consolidation, there are now 16 districts in Shanghai. As the financial center of China, Shanghai has an important strategic location in terms of economic development and social contribution. Its economic development benefits from its regional distribution, which plays an important role in the economy of Shanghai. However, Shanghai belongs to the subtropical monsoon climate zone and is located in a mid-latitude region that is significantly affected by the ocean. In addition, the average elevation of Shanghai is low, and the river network is densely distributed, causing the city to be extremely vulnerable to waterlogging. According to historical data, Shanghai has suffered many natural disasters, including drought, urban waterlogging, and storm surges that have affected public safety. With the continuous changes in the global climate, the increasingly serious rainfall disasters occurring in Shanghai impact the security of the resilient system of this megacity and have increasingly serious impacts on the urban economy, production, life, and urban public security management.

The significance of studying the resilience of Shanghai lies in the following considerations: first, supporting the overall national development strategy: Shanghai is an important engine and window of national economic development. Secondly, it can provide a reference for national policies: the study of Shanghai’s urban elasticity can provide an important reference basis for national policies. By understanding Shanghai’s responses and performances in the face of various pressures and challenges, the country can better formulate policies to support sustainable urban development and improve urban resilience and resilience. Finally, promote urban transformation and upgrading: by studying the urban resilience of Shanghai, the advantages and disadvantages of the region can be identified in terms of the economy, society, and environment to provide a basis and direction for urban transformation and upgrading.

Literature review

Connotation and evolution of urban resilience

“Resilience” comes from the Latin “resilio,” meaning “rebound”; that is, resilience reflects the recovery of the original state of affairs after an external impact (Godschalk 2003). In 1973, Canadian ecologist Holling introduced the concept of resilience to ecology research for the first time and interpreted resilience as the ability of the ecosystem itself to adapt, maintain, resist, and restore balance after a disaster or other short-term impact (Folke et al. 1996).

In modern studies on the theory of resilience, based on the different fundamental cognitions of the system ontology, the concept of resilience can be divided into two processes (Comfort 2006). The first process is the ideological core of engineering resilience and ecological resilience; that is, a system eventually returns to the equilibrium steady state after being disturbed. The second process is the view of evolutionary resilience, which comes from adaptive cycle theory (Cutter et al. 2008); this theory abandons the traditional system’s pursuit of equilibrium and the steady-state, believes that systems are in a dynamic process of constant self-adaptation and change, and emphasizes that resilience is a system’s ability to deal with interference and adapt and change internally. This theory is consistent with the basic concept of the third-generation system theory: the theory of complex adaptive systems. Compared to resilience, urban resilience, as the name implies, refers to the ability of urban systems and regions to realize the normal operation of public safety, social order, and economic construction through reasonable preparation, buffering and co** with uncertain disturbances. Godschalk believes that resilient cities should be a combination of sustainable physical systems and human communities, and the planning of physical systems should work through the construction of human communities (Godschalk 2003). Gotham and Campanella, by contrast, value the power of the human community (Gotham and Campanella 2011). By assessing New Orleans’ performance after Hurricane Katrina, one researcher argued that urban resilience essentially depends on a more resilient and resourceful population (Comfort 2006). Shari further argued that there are four main components of urban resilience, namely, infrastructure resilience, institutional resilience, economic resilience, and social resilience (Shari et al. 2021).

To date, scholars have mostly discussed the characteristics of urban resilience from the perspectives of “capacity” (disaster warning, emergency response, disaster resistance and reduction, and recovery/reconstruction) and “process” (resistance, resilience, and adaptability). Resilient systems can withstand a series of changes and still maintain functional and structural control; resilient systems have the ability to self-organize; and resilient systems have the ability to establish and promote learning and self-adaptation (Desouza and Flanery 2013). To solve the economic crisis, environmental damage, natural disasters, and other uncertain problems in urban development and gradually build the core system of modern urban resilience theory, urban resilience connotations have been analyzed mainly from the perspectives of specific resilience (disaster resilience) and generalized resilience (urban system) in studies on the intentions of human activities and the boundaries of resilient systems highlighting the ability of resilient cities to resist, absorb, and effectively respond to disasters in a timely manner (Jabareen 2013). In addition, scholars have mostly focused on analyzing the impacts of natural disasters and measuring the composition of urban resilience.

By sorting out the above concepts of resilience, it can be seen that the evolution of the concept of resilience has experienced a development process from monostable to poly-stable to dynamic thought, from restoring the initial state to finding a new state to constantly adapting and learning, and from definite order to complex disorder to uncertainty and chaos; this development context is basically the same as the logical evolution pace of the concept of urban disaster prevention. However, in new disaster prevention planning, the concept of urban resilience has not been integrated into the resilience assessment or management processes applied to specific urban areas.

Types and methods of urban waterlogging disaster assessments

As complex social ecosystems, urban areas cover all aspects, including the development of subsystems such as economy, society, infrastructure, and ecology, from the community scale to the regional and even whole-country scales. Since the construction of self-resilient cities has received wide attention in practice, Chinese scholars have studied the characteristics of such cities at all levels.

In terms of model construction, some scholars have explored resilience evaluation models of cities in response to waterlogging disasters. Ruan et al. (2021) established a resilience evaluation model under flood conditions based on the three attributes of resistance, resilience, and adaptability. Huang et al. (2022) constructed a resilience index to evaluate the decoupling relationship between urban resilience and environmental pressure for China’s top ten urban agglomerations. Xu and Xue (2017) analyzed subway waterlogging exposure in the central urban area of Shanghai from the perspective of multiple scenarios and selected the indicators most directly related to waterlogging to construct an evaluation system for the central urban area of Shanghai. Malalgoda et al. (2016) focused on evaluating urban waterlogging resilience by studying waterlogging prediction models. Mehmood (2016) built a resilience index evaluation model containing 7 aspects. In terms of the scope of assessment, Hatuka et al. (2018) determined community flood disaster resilience indicators at the social, economic, institutional, and physical levels. Mell (2018) analyzed the theoretical evolution and characteristics of resilient cities. Zhao et al. (2022) evaluated the resilience of prefectural cities in the Yangtze River Delta from the perspective of the social ecosystem based on the regional perspective. Other studies focused on the urban resilience at different scales including communities, cities, urban agglomerations, regions, and the whole country (Li et al. 2020). In terms of the research contents, urban resilience was mainly first used in urban research performed at the natural disaster resistance and postdisaster recovery levels (Boulos et al. 2015) and then gradually expanded to research in the field of public security (Croese et al. 2020). In recent years, the literature has begun to shift from single-discipline research to cross-discipline research. Wang et al. (2018) tried to apply a BP neural network to study urban resilience. Zhang et al. (2019) studied the impact of the economic development level and environmental pressure on urban resilience. The literature review of the barriers to urban resilience highlights the complexities of cities and various factors, such as society, economy, politics, culture, and the environment (Chen and Quan 2021). These factors may affect the resilience of cities and critically impact the sustainable development of cities. To measure the urban resilience barrier factors, some scholars have proposed various research methods, such as investigation studies, case studies, qualitative analyses, and quantitative analyses (Hangebruch and Othengrafen 2022). Quantitative analyses can quantify the influence degree of urban resilience barrier factors, so this paper adopts a quantitative analysis research method (Huang et al. 2022). Finally, the research methods are diversified. From the initial literature review, concept analyses have transitioned from simple quantitative analyses gradually to higher-level quantitative analyses. From normal panel regression to analyses of spatial and temporal characteristics, spatial factors have been added into the econometric model.

While the research scopes of the past studies were relatively wide, they lacked detailed analyses of urban districts based on specific disasters, paid more attention to assessing urban resilience, and rarely analyzed the key indicators hindering the development of urban resilience. Therefore, based on an analysis of the literature on this concept, including its composition and index construction for urban resilience, this paper analyzes urban resilience construction, builds a resilience evaluation system, analyzes the barriers to obtaining evaluation results, and proposes suggestions to improve urban resilience, providing some theoretical support for urban resilience construction.

Research methods

Determination of indicators

Indicator selection

Starting from four dimensions (urban economic resilience, urban ecological resilience, urban infrastructure resilience, and social resilience), this paper constructs a resilience assessment system framework according to those four dimensions and selects indicators (see Table 1). The selection of indicators follows the principles of accessibility, comprehensiveness, and scientific urban operability to ensure that the degree of urban resilience in Shanghai is effectively and truly reflected. The reasons for the selection of indicators are explained as follows.

Table 1 Shanghai resilience evaluation index system and index weights after AHP empowerment
Full size table

In the selection of economic resilience indicators, urban economic resilience refers to the ability of the regional economy to resist external shocks. When an urban economic system is affected by disasters, it can quickly recover and develop to reach the original level or a higher level than the original economic level (Godschalk 2003; Anh et al. 2014). That is, only with resilience can an urban system achieve sustainable growth and development. When waterlogging occurs in a region, it is manifested as the economic strength to withstand the damage caused by waterlogging. The per capita GDP of an urban area is a reflection of the comprehensive capacity of the urban economy. Increasing employment opportunities is conducive to improving people’s livelihood and contributes to the construction of a harmonious society and economic development. The more the general public budget expenditure is, the more conducive the system is to improving infrastructure and the further development of the urban economy; the total industrial output value of each district is also an important measure of urban economic development, and the economic strength of a region is also a reflection of the comprehensive strength of the region.

In terms of the selection of ecological resilience indicators, according to the definition of urban resilience, the resilience of an urban area in the face of waterlogging is the best measure of a resilient urban area (Restemeyer et al. 2015). People and cities are mutually promoting, and the healthy development of an urban area cannot be separated from the construction or planning of the people. An urban area can improve its risk adaptability in the face of waterlogging disasters by building a good ecosystem, a developed urban drainage system, and a new water resource recovery system. Among them, a good natural ecosystem can promote the development of an urban area in all aspects, including the economy and infrastructure. A good ecosystem can automatically regulate the urban water system and ecological diversity. Examples include the ability of a system to adjust the sewage, road section water, urban garbage, and post-disaster disease concerns caused by waterlogging disasters and its ability to restore the normal order of the urban area (Wu and Chiang 2018). The number, area, green area, and regional green area of regional parks reflect the distribution and area of green ecology in urban ecosystems. The river channel (lake) area reflects the urban water conservation and disaster capacity when waterlogging occurs.

In terms of the selection of infrastructure resilience indices, we take drainage pipes as an example. Traditional urban waterlogging prevention projects focus on urban drainage pipes, but resilient cities need not only urban flood prevention and waterlogging projects but also high standards for facility construction and complete inspection and assessment systems. At the same time, the more sludge there is in drainage pipes, the more unfavorable the conditions are to flood discharge and excessive-rainfall discharge. This leads to urban waterlogging. When waterlogging occurs, drainage pumps serve as important facilities for drainage and are very important for the operation, maintenance inspection, and assessment of drainage pump stations. When such assessments are not up to the standards, timely maintenance is needed to avoid equipment problems and failures when waterlogging occurs. Developed urban drainage systems have certain drainage capacities and make full use of rainwater resources by means of water storage and water purification to promote the harmonious development of urban people and nature (Kiener et al. 2018). In the face of disasters, people are not helpless. In the event of human injury caused by disasters, certain medical institutions and relevant technical personnel are needed to provide support. Moreover, hospital beds are a great reflection of the ability of rescue personnel to bear disasters. As shown via the number of COVID-19 patients treated in 2020, the number of hospital beds and the capacity to accommodate people are extremely important. When waterlogging disasters occur, all kinds of urban domestic sewage are often produced, which is not conducive to the health of residents and easily causes all kinds of pollution and diseases. To some extent, the number of sewage treatment plants can reflect the regional treatment of sewage generated by waterlogging disasters, providing an important reflection of the resilience of urban infrastructure.

When selecting social resilience indicators, residents’ awareness of waterlogging disasters is among the important factors to consider in urban adaptability measurements (Abubakar and Aina 2019). In fact, demographic data can be used to inform and guide urban planning efforts to better meet the needs of diverse populations, including older adults and children. By understanding the age structure of a population, urban planners can develop more inclusive and resilient cities that prioritize the well-being of all individuals (Folke et al. 1996). Overall, it is important to consider multiple factors when planning for resilient cities, including demographic trends and the well-being of individuals. By taking a holistic approach to urban planning, we can work toward creating cities that are both resilient and equitable for all members of society. When waterlogging occurs, the population above 60 years old and the population below 17 years old tend to be more vulnerable to disasters than other age groups. The more vulnerable groups there are in a region, including regions with relatively large population densities and large administrative areas, the more concentrated the affected areas are, and the weaker the regional social resilience level is (Ilgen et al. 2019). A well-developed volunteer service system can remind residents to take effective measures to avoid the adverse effects of waterlogging disasters and ensure that residents can work and live normally in case of rain and flood disasters. In addition, residents can cultivate a good awareness of and ability to respond to flood control to ensure flood-control preparations can be made before the arrival of waterlogging. Among them, factors such as the number of health technicians and volunteer groups affect the speed and degree of urban recovery and are thus important factors for measuring urban social resilience (Liang 2018).

Data sources

The latest relevant data of the Shanghai Statistical Yearbook were collected and collated in various ways, and some data were obtained from the Shanghai Statistical Yearbooks of previous years, as shown in Table 2.

Table 2 Data sources of each indicator
Full size table

Analytical methods

The evaluation includes two aspects. One aspect involves using the AHP-TOPSIS method to comprehensively evaluate the resilience of cities in the middle and lower Yangtze River regions (Liu et al. 2022). By comprehensively evaluating the spatial urban resilience pattern, the comprehensive levels of urban resilience in the middle and lower reaches of the Yangtze River are analyzed. The second involves analyzing the restrictive factors impeding the improvement of urban resilience, including factors such as the contribution degree, index deviation degree, and obstacle degree. Combined with a comprehensive comparison with high urban resilience, the obstacle degrees of urban resilience development in the middle and lower reaches of the Yangtze River are calculated to provide certain reference information for the improvement of the urban resilience level (Liao et al. 2022).

AHP-TOPSIS model (Han et al. 2022)

To eliminate the unit variations among different indicators and unify the indicator dimensions, this paper adopts the extreme value standardization method to conduct a standardization processing for the indicators. Using the TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) evaluation method, the evaluation index system is divided into a specific point of n spaces. In addition, the indices within the urban resilience evaluation index system are ranked according to the distance between the certain point in space where each evaluation index is located and the ideal solution. The relativity among the evaluation indices is reflected according to the different distances between the index indices. This method not only conforms to the relativity between the urban resilience indices and the resilience level but also makes full, comprehensive use of the original data, thus avoiding index information losses and providing other advantages (Yang et al. 2018).

  • Step 1: Transform all indicators into extremely large indicators using the positive formula of indicators as follows:

    $$\mathrm{max}-x$$
    (1)

    where max is the maximum value of the indicator item and x is the item that needs to be positive.

  • Step 2: Standardize indicators. The standardization formula is as follows:

    $${z}_{ij}={x}_{ij}/\sqrt{{\sum }_{i=1}^{n}{x}_{ij}^{2}}$$
    (2)

    where zij is the normalized matrix, x is an index, i is the number of rows, and j is the number of columns.

  • Step 3: When the matrix distance analysis method (TOPSIS evaluation method) is combined with multiple index factors, the weight determination accuracy is poor. The analytic hierarchy process is a systematic analysis method that can divide a system into different levels, determine the importance of each level index factor, evaluate the importance of each index through expert talks and questionnaire surveys, and then, calculate the weight of each index. The principle of the analytic hierarchy process (AHP) is that before analyzing a phenomenon or problem, the phenomenon or problem is first decomposed into relevant factors according to their nature, and a multilevel structural model is formed according to the classification of the relationships between the factors (Macharis et al. 2004). Then, the relative importance of low-level factors to high-level factors can be judged and measured through experience or experts, and the weight ranking can be obtained according to the degree of importance to make quantitative analyses and comparisons. The relative importance degrees (between two factors) of N indicators within the same level are scored by experts. The relative importance scale is between 1 and 9 (Jawad et al. 2023). The vector of each line in the matrix is geometrically averaged (root method) and then normalized; that is, the weight of each evaluation index and the feature vector \(\omega\) are obtained. In this paper, the weight is calculated by using Yaahp (Version 12.9.8167) software. However, because the AHP method is relatively superior, the TOPSIS evaluation method and AHP are complementary to some extent. In view of this, the author combines the two methods to construct a new comprehensive evaluation model of urban resilience with the following expression formula:

    $$Z=\omega {z}_{ij}$$
    (3)

    Therefore, the weighted standardized decision matrix is constructed. Suppose there are n objects to be evaluated, and the weighted standardized decision matrix of m evaluation indicators is expressed as follows:

    $$Z=\left[\begin{array}{cccc}{z}_{11}& {z}_{12}& \cdots & {z}_{1m}\\ {z}_{21}& {z}_{22}& \cdots & {z}_{2m}\\ \vdots & \vdots & \ddots & \vdots \\ {z}_{n1}& {z}_{n2}& \cdots & {z}_{nm}\end{array}\right]$$
    (4)

  • Step 4: Define the maximum and minimum values of the matrix and calculate the maximum and minimum distances of the evaluation object.

    Define the maximum value:

    $$\begin{array}{c}Z^+=\left(Z_1^+,Z_2^+,\cdots,Z_m^+\right)\\=\left(\max\left\{z_{11},z_{21},\cdots,z_{n1}\right\},\max\left\{z_{12},z_{22},\cdots,z_{n2}\right\},\cdots,\max\left\{z_{1m},z_{2m},\cdots,z_{nm}\right\}\right)\end{array}$$
    (5)

    Define the minimum value∶

    $$\begin{array}{c}Z^-=\left(Z_1^-,Z_2^-,\cdots,Z_m^-\right)\\=\left(\min\left\{z_{11},z_{21},\cdots,z_{n1}\right\},\min\left\{z_{12},z_{22},\cdots,z_{n2}\right\},\cdots,\min\left\{z_{1m},z_{2m},\cdots,z_{nm}\right\}\right)\end{array}$$
    (6)

    Define the distance between object i(i = 1,2,…,n) and the maximum value:

    $${D}_{i}^{+}=\sqrt{{\sum }_{j=1}^{m}{\left({Z}_{j}^{+}-{z}_{ij}\right)}^{2}}$$
    (7)

    Define the distance between object i(i=1,2,...,n) and the minimum value:

    $${D}_{i}^{-}=\sqrt{{\sum }_{j=1}^{m}{\left({Z}_{j}^{-}-{z}_{ij}\right)}^{2}}$$
    (8)

  • Step 5: Calculate the nonnormalized score of object \(\it \it i(i=\mathrm{1,2},\cdots ,n)\):

    $${S}_{i}=\frac{{D}_{i}^{-}}{{D}_{i}^{+}+{D}_{i}^{-}}$$
    (9)

    where \(0\le {S}_{i}\le 1\); the larger \({S}_{i}\) is, the smaller \({D}_{i}^{+}\) is, that is, the closer the result is to the idealized target.

  • Step 6: From four dimensions, the resilience assessment of 16 districts in Shanghai was conducted to analyze the urban resilience characteristics under waterlogging scenarios. The optimized quantitative model is expressed as follows:

    $$R=\frac{1}{4}*{[S}_{i}(A){W}_{A}+{S}_{i}(B){W}_{B}+{S}_{i}(C){W}_{C}+{S}_{i}(D){W}_{D}]$$
    (10)

    where R stands for the resilience level of comprehensive cities; \({S}_{i}(A)\) represents the resilience level of the urban economy; \({S}_{i}(B)\) represents the level of urban ecological resilience; \({S}_{i}(C)\) represents the resilience level of urban infrastructure; \({S}_{i}(D)\) represents the level of urban social resilience; \({S}_{i}\) \(i(i=\mathrm{1,2},\cdots ,n)\) is the fit degree between evaluation indicators and positive idealized targets, namely, the resilience index; \({W}_{A}\) represents the weight of the economic resilience index; \({W}_{B}\) represents the weight of the ecological resilience index; \({W}_{C}\) represents the weight of the infrastructure resilience index; and \({W}_{D}\) represents the weight of the social resilience index.

Analysis of obstacle factors

On the basis of sorting out and calculating the urban resilience index values for various districts in Shanghai in order to formulate reasonable policies for scientifically improving the urban resilience of Shanghai, this study introduces an obstacle factor analysis to analyze and summarize the key factors hindering the improvement of urban resilience in various districts in Shanghai, allowing corresponding policy suggestions to be made for the improving the urban resilience of various districts in Shanghai. In this way, the current regional urban resilience situation can be improved more effectively, and prefecture-level cities can cope with the urban problems caused by waterlogging disasters, enhance their regional comprehensive competitiveness, and achieve sustainable development in all districts in Shanghai. The formula used to calculate the obstacle degree of urban resilience is expressed as follows:

$$\begin{array}{cc}\mathrm F\mathrm a\mathrm c\mathrm t\mathrm o\mathrm r\;\mathrm c\mathrm o\mathrm n\mathrm t\mathrm r\mathrm i\mathrm b\mathrm u\mathrm t\mathrm i\mathrm o\mathrm n\;\mathrm d\mathrm e\mathrm g\mathrm r\mathrm e\mathrm e:&U_j=W_j\end{array}$$
(11)
$$\begin{array}{cc}\mathrm I\mathrm n\mathrm d\mathrm e\mathrm x\;\mathrm d\mathrm e\mathrm v\mathrm i\mathrm a\mathrm t\mathrm i\mathrm o\mathrm n\;\mathrm d\mathrm e\mathrm g\mathrm r\mathrm e\mathrm e:&V_j=1-p_{ij}\end{array}$$
(12)
$$\begin{array}{cc}\mathrm D\mathrm e\mathrm g\mathrm r\mathrm e\mathrm e\;\mathrm o\mathrm f\;\mathrm o\mathrm b\mathrm s\mathrm t\mathrm a\mathrm c\mathrm l\mathrm e:&M_j=\left(U_j\cdot V_j\right)/\sum\nolimits_{j=1}^{22}\left(U_j\cdot V_j\right)\end{array}$$
(13)

where the factor contribution degree (Uj) represents the influence degree of an index factor in the urban resilience evaluation system on the total urban resilience index, that is, the weight of the index factor in the urban resilience evaluation on the total urban resilience index (Wj). The index deviation degree (Vj) represents the difference between an urban area’s resilience evaluation index factor and the total resilience of the regional urban area and is set as the difference between the singe resilience index value \({p}_{ij}\) and the ideal value (100%). The obstacle degree refers to the influence value of a certain urban resilience evaluation index factor on the regional urban resilience level. This index is the main purpose and result of the obstacle analysis performed to improve urban resilience in each district of Shanghai. According to the calculation formula derived for the obstacle degree analysis, the obstacle degree factors corresponding to each district are calculated.

Result analysis

Resilience analysis of all dimensions

Analysis of economic resilience

From the calculation results of the economic data, to analyze the reasons for the differences in the economic resilience levels of various districts in Shanghai, we calculated the district-scale values of each economic resilience index (Fig. 1 a). GDP is calculated at the area level, while the per-capita GDP reflects the economic output per person. Both measures are useful for understanding the economic performance of an area. From the statistical results, the higher the regional GDP of each region is, the greater the resilience level index value will be in the face of waterlogging disasters. The economic force is reflected in the resistance, recovery and adaptation of an area during disasters, and its role is particularly prominent in the resistance, resilience, and adaptability of urban systems. Therefore, the low economic resilience levels identified in some regions, such as Hongkou District and Putuo District, are importantly correlated with the level of regional GDP development, resulting in a low regional economic resilience index. In addition, the main reason why Jiading District and Songjiang District rank high in economic resilience is related to the population densities and total industrial output values of these two regions. Songjiang District has a large area, a large population, and a wide distribution of industrial factories, so its regional GDP is relatively high. Similarly, Jiading District has a relatively high economic resilience index due to its small population and developed regional industry. In addition, the higher the general public budget expenditure is, the more likely the resilience of the economic system is to be affected by waterlogging. The public budget expenditure has a great impact in Pudong New Area, where a high resilience index was obtained, and in Fengxian District and **shan District, where moderate resilience levels were identified; in contrast, this index had little impact on Putuo District or Hongkou District, where low resilience index values were obtained. This was because the regional GDP and per capita GDP of Putuo District and Hongkou District were much different from those of the other districts, and their public budget expenditures were not sufficient to significantly impact these two districts.

Fig. 1
figure 1

Distributions of the economic resilience index (a), ecological resilience index (b), infrastructure resilience index (c), and social resilience index (d) in Shanghai

Full size image

Ecological resilience analysis

The value of the ecological resilience index in each district was obtained through calculations, and the reasons for the differences in ecological resilience among the districts in Shanghai were analyzed in depth (Fig. 1 b). According to the index values of ecological resilience, the regional ecological resilience index values of Pudong New Area and Chongming District were highest. Compared to the other districts, these two regions have the following characteristics. First, the regional green areas and river and lake areas of Pudong New Area and Chongming District are significantly higher than those of other districts, and these green areas had great contributions to the ecological resilience of Pudong New Area and Chongming District. Second, the area of river channels and lakes in Qingpu District is higher than those in the other 16 districts, imposing a certain influence on the regulation of waterlogging and influencing the urban ecological resilience of Qingpu District. Third, by comparing the data between Hongkou District and other regions, we found that the regional green area of Hongkou District was much lower than those of other regions in terms of the river and lake areas, and the ecological adjustment ability of Hongkou District to rain and waterlogging was poor, which was another reason for the low urban ecological resilience index of Hongkou District obtained in the face of waterlogging. Fourth, the number of parks did not differ among the districts, and the impact of parks on the urban ecological resilience index was relatively small, but their role was relatively large; parks serve a refuge, provide climate regulation service, and adjust of the regional ecosystem. Since the number of parks in each district did not differ, the factor that truly contributed to the differences in the regional ecological resilience index was the regional green rate. The higher the regional green rate was, the higher the regional ecological resilience index level to waterlogging disasters was.

Infrastructure resilience analysis

To further analyze the reasons for the differences in infrastructure resilience among different districts in Shanghai, the infrastructure resilience index values were calculated (Fig. 1 c). From the perspective of the district-level infrastructure resilience index results, the reason why little difference was found was that the districts within Shanghai had relatively little differences in terms of the infrastructure supporting facilities, and all had relatively excellent infrastructure. For example, the number of sewage treatment plants and health service institutions were relatively evenly distributed among all districts except the Pudong New Area. In addition, for example, the assessment scores of drainage pipe maintenance management were relatively consistent in the assessment results of the policies and standards of all districts and in the assessment scores of drainage pump operation and maintenance. These differences were small compared to the economic and ecological differences but are important indicators of whether an urban area can quickly and effectively divert flood waters during waterlogging scenarios. The infrastructure construction of each district is subject to the unified planning and management of the Shanghai Municipal Planning Bureau, and infrastructure is constructed according to different regional needs, such as the number of regional health institutions, the number of regional sewage treatment plants, the number of hospital beds in health institutions, and the output of pipeline sludge per unit length. Since all districts belong to Shanghai and the terrain and landform are similar among each district, the data were not very different. However, there were some areas with relatively obvious differences.

Social resilience analysis

To analyze the reasons for the differences in social resilience among districts in Shanghai, the social resilience index values were calculated (Fig. 1 d). First, the social resilience index results showed that among the 16 districts of Shanghai, the population of people in Pudong New District over 60 and under 17 years old (D1) was larger than those in other districts, and the administrative area (D5) was also larger than those in other districts. As the index shows a negative correlation, the larger the original data were, the lower the relative level of the resilience index was. Compared to Minhang District and Songjiang District, the Pudong New Area had obvious advantages in terms of its population density and number of volunteer groups. However, the social resilience gap between Pudong and Xuhui was relatively small because there are more vulnerable groups in the Pudong New Area (people over 60 years old and under 17 years old), it is more affected by disasters, the social population system is fragile, and the administrative area is relatively large. With a wide range of disasters and a high degree of exposure, short-term recovery requires extensive human and financial inputs. Therefore, the overall social resilience level was higher here than in Xuhui District, and there was little difference in the final index calculation results.

Analysis of comprehensive urban resilience

The economic resilience index, ecological resilience index, infrastructure resilience index, and social resilience index values of each district in Shanghai were processed using the resilience measurement model shown in formula 10, and the overall resilience level index R of each district was calculated (Table 3). According to the ArcGIS natural breakpoint method, a relatively high resilience level was defined as an R value greater than 0.3811 and less than 0.8154. An R value greater than 0.3207 and equal to or below 0.3464 referred to the moderate resilience level, and an R value greater than or equal to 0.1841 and less than or equal to 0.2066 corresponded to the low resilience level. According to the R value calculation results, Pudong New Area (0.8154) was the only area among the 16 districts of Shanghai with a high resilience level. The regions in the moderate and upper-moderate resilience levels were Qingpu (0.3811) > Songjiang (0.3716) > Chongming (0.3576) > Jiading (0.3567) > Xuhui (0.3464) > Minhang (0.3421) > **shan (0.3357) > Huangpu (0.3280), while the areas with moderate-low and low resilience levels were mainly located in the center of Shanghai, namely, Fengxian (0.3207) > Yangpu (0.3054) > Baoshan (0.3043) > **g’an (0.3022) > Putuo (0.2066) > Hongkou (0.2047) > Changning (0.1841), due to the dense population in the downtown area, smaller green park area, wide coverage of ground-hardening areas, small rainwater penetration, and slow penetration rate. According to the R value, Pudong New Area had the highest social resilience level in the face of waterlogging disasters among all districts of Shanghai, with an R value of 0.8154. In the other 15 districts, the resilience levels were similar, indicating that the social resilience levels of natural disasters in these 15 districts were roughly the same, between 0.2 and 0.4.

Table 3 Main obstacles affecting the improvement of resilience in Shanghai districts
Full size table

According to the evaluation indicators and final results (Fig. 2), we found a large relative difference between Pudong New Area and Changning District. The main reasons for the resilience difference between these two districts were the large differences in the economic resilience and ecological resilience indicators between the two areas, while the difference in the social policy resilience indicator between the two areas was not large. According to the urban economic indicator results, we compared the positive resilience-level indicators A2 (per capita GDP), A3 (general public budget expenditure), A4 (regional industrial output value), and A5 (regional GDP). The Pudong New Area is larger than Changning District. From the perspective of ecological resilience, the green area, river area, and lake area in Pudong New Area are much larger than those in Changning District. From the impact indicators of infrastructure, C2 (output per unit length of pipeline sludge, operation and maintenance inspection and assessment scores of drainage pump stations), C4 (number of health institutions), and C5 (number of hospital beds) were much higher in Pudong New Area than in Changning District. To some extent, the improvement in the regional GDP and general public budget expenditure in the economic resilience indicator, in the regional green area in the ecological resilience indicator, and in the number of health institutions and beds in the infrastructure indicator significantly impacted the level of regional resilience.

Fig. 2
figure 2

Horizontal regional map of resilience under a waterlogging scenario in Shanghai

Full size image

Factor analysis of the urban resilience barrier degree

Based on our assessment of the resilience level of each district in Shanghai, to determine the main factors affecting regional resilience under a waterlogging disaster scenario in each district, this study introduced the obstacle degree model. According to the formula used to calculate the obstacle degree, the calculation results are shown in Table 3. As shown in Table 3, from the perspective of a waterlogging disaster, the main obstacle factors affecting the resilience of each district in Shanghai are discussed as follows (Fig. 3).

Fig. 3
figure 3

The main obstacle factors and obstacle degree in each district of Shanghai

Full size image

First, regarding economic resilience, the per capita GDP (A2) was found to be the most important obstacle factor; this factor was distributed mainly among the first three obstacle factors and appeared most frequently as the third obstacle factor. The second factor was the total industrial output value of each district (A4), which was distributed mainly as the third and fourth obstacle factors. In terms of ecological resilience construction, the park area (B3) was the main obstacle factor in **shan District, the urban green area (B4) was the main obstacle factor in Yangpu District, and the river area (B5) was the main obstacle factor in Putuo District; these obstacles occurred 3 times, 6 times, and 7 times, respectively, in the analyzed dimensions and districts. Based on the perspective of flood disasters, these three indices were the most important in affecting the ecological construction of Shanghai city.

Second, in terms of infrastructure resilience, the number of hospital beds (C5) was the main obstacle factor in Jiading District and Songjiang District, and the number of sewage treatment plants (C6) in the service area was the main obstacle factor in Baoshan District; all these factors occurred 6 times. The sludge yield per unit length of pipeline ditching (C2) in Changning District was the main obstacle to improving regional toughness in the face of flood disasters. Sewage treatment systems are indispensable and important municipal infrastructures in modern cities facing flood disasters. They also form the backbone of urban water pollution prevention and urban drainage and flood control projects. Their function is to treat domestic sewage, industrial wastewater, and precipitation generated in urban areas in a timely and reliable manner so that the city is free from sewage and storm water pollution disasters to create a comfortable and safe living and production environment for people, make the energy flow and material circulation of the urban ecosystem normal, maintain ecological balance, and ensure sustainable development. Increasing the number of sewage treatment plants is the main infrastructure construction and development direction for improving regional toughness in each district based on the waterlogging disaster scenario. At the same time, the number of hospital beds plays an important role in the treatment of postdisaster injuries and epidemics and in improvements to urban infrastructure.

Finally, regarding social policy, the population density (D2) was the main obstacle factor faced in **g’an District and Hongkou District, health technicians (D4) were the main obstacle factor in Qingpu District, Huangpu District, and Fengxian District, the administrative area (D5) was the main obstacle factor in Chongming District, and the per capita living housing area (D6) was the main obstacle factor in Minhang District and Xuhui District; these factors occurred 4 times, 7 times, 5 times, and 4 times, respectively. Among them, the population over 60 years old and under 17 years old (D1) was the most important obstacle in the Pudong New Area. The details are shown in Table 3.

Conclusion and discussion

With the acceleration of urbanization, people’s awareness of safety and risk is constantly improving, and urban waterlogging disasters caused by heavy rain cause vast economic losses and other indirect losses. Based on authoritative data, this paper scientifically selects evaluation indicators, constructs an evaluation index system, and uses the AHP and TOPSIS evaluation methods to evaluate the regional resilience levels of 16 districts in Shanghai city. Based on the evaluation results, an obstacle analysis is conducted on the factors hindering the development of resilience in each district. The main research results are summarized as follows.

  1. 1.

    By summarizing the relevant theories of urban resilience in China and abroad and the construction methods of the evaluation index system, an index system for evaluating urban resilience under the waterlogging disaster scenario in the urban areas of Shanghai was constructed, and key points were applied to evaluate the urban resilience levels of various districts of Shanghai. The overall basic situations of urban resilience of the various districts of Shanghai were obtained. The resilience levels of the urban center and remote areas such as Chongming and Fengxian were relatively low, and the urban resilience levels of all districts in Shanghai need further improvement.

  2. 2.

    According to the calculation formula of the obstacle degree, the leading factors of urban resilience in all regions under the waterlogging disaster scenario tended to be consistent. The results were focused mainly on the total industrial output value of each district, the areas of rivers (lakes) and urban green space, the number of hospital beds, the number of sewage treatment plants in service areas, and the number of health technicians. It is thus necessary to further strengthen the construction of the comprehensive resilience of each district in Shanghai in terms of these aspects.

  3. 3.

    Under the influence of global climate change, waterlogging disasters caused by rainstorms are becoming increasingly frequent. Shanghai needs to start from the perspective of each district, adapt to local conditions, and improve the urban resilience level of each district. This paper took Shanghai as the object of empirical research and conducted scientific research to provide a reference for the future construction of resilient cities.

Based on urban resilience research performed under disaster conditions, this study has made some achievements, but it still has certain limitations. First, the index system used herein was not perfect. In the indicator selection process, the unavailability of some direct indicators caused the index system to be imperfect. For example, rainfall, one of the key factors of urban waterlogging caused by rainstorms, may lead to different disaster degrees when different rainfall amounts and rainfall durations occur. Therefore, in future research, it will be necessary to constantly update the evaluation index system according to official releases and to big data characterizing urban areas and to further improve the evaluation index system. Second, the selection of the analytic hierarchy process (AHP) method for determining the index weights was subjective to a certain extent, although scholars and experts were consulted many times and a large number of questionnaires were completed. However, the scientific determination of the indicators and the direction of the indicators need to be further strengthened. Finally, in the construction method of urban resilience in Shanghai, the countermeasures proposed in this paper still had many shortcomings. To improve the level of urban resilience in Shanghai, build a safe and harmonious society in Shanghai, and improve the capacity of public safety management in Shanghai, more in-depth research will be needed in future assessments of urban resilience to better benefit the people of Shanghai, seek development methods for Shanghai, and find a way to construct resilient urban structures in China.