1 Introduction

Shanghai region covers a total area of 6340 km2 and is located on the deltaic deposit of Yangtze River with an average altitude of 4.0 m above the sea level. With the dramatically quick development and utilization of underground space, many underground infrastructures have been built, such as cross-river tunnels, deep basement of high-rise buildings and metro lines (Chen et al. 2011; Gao et al. 2012; Wu et al. 2016). This trend of underground constructions is easing the space insufficiency caused by the limited ground space; nevertheless, it constitutes a big challenge from a practical point of view because of the detrimental hydrogeological and geological conditions (Huang et al. 2015; Wu et al. 2015b, 2017; Xu et al. 2009).

Indeed, the widely distributed Quaternary deposit contains abundant underground water; most of which are confined water that may lead to problems such as water inrush, leakage, foundation corrosion and floatation of underground structures. Moreover, sand strata with different thicknesses are also extensively distributed in these shallow stratum (Shen and Xu 2011; Shen et al. 2013). As a result, quicksand often occurs resulting from the interaction between the confined water and sand strata, leading to serious problems such as foundation pit instability, bottom hydraulic heave and even collapse of underground structures (Shi 2010). For example, a serious accident causing huge economic loss of approximately 227 million dollars (USD) occurred on July 1, 2003, during the construction of Shanghai Metro Line 4, due to the confined water inrush and accompanied quicksand at the connected aisle of the tunnel (Shen et al. 2014; Wu et al. 2015a; Elbaz et al. 2016). Thereby, the two geological hazards—groundwater inrush and quicksand—have become the main concern in exploitation of underground space in Shanghai, and it is of paramount importance to understand the characteristics and distribution of the confined aquifers and soil stratum to manage such hazards.

Many studies have been conducted to evaluate hazards induced by confined aquifer, such as water inrush and land subsidence (Shen and Xu 2011; Wu et al. 2015a, b, c, 2014; Xu et al. 2012a, b). For Shanghai city, a series of investigations were conducted dealing with the problem caused by underground water. For example, Luo et al. (2015; Shen et al. 2014; Xu et al. 2009) elaborated on the hydrogeological characteristics of Shanghai geology and pointed out that Quaternary deposit in Shanghai region is composed of a layered multi-aquifer-aquitard system (MAAS). Xu et al. (2009), Huang et al. (2015) and Elbaz et al. (2016) studied the potential hazards during the underground construction and mentioned that MAAS with soft soil layers, sand strata and high water content aquifers would be more easily disturbed by underground engineering construction, thus causing water inrush and quicksand disasters. Chai et al. (2004), Shen et al. (2013) and Xu et al. (2008, 2012a, b, 2015) investigated the land subsidence hazard caused by underground water pumping and pointed out that local hazards were more significant than overall hazards so that zoning is required (Shen et al. 2014; Wu et al. 2015a).

To study the geological hazards such as landslide and flood, zoning maps are often adopted (Akbar and Ha 2011; Fell et al. 2008; Ganapathy and Rajawat 2015; Liu et al. 2017; Stevens and Hanschka 2014; Thierry et al. 2008). The zoning studies to groundwater are crucial to prevent the occurrence of accidents caused by water inrush (Béjar-Pizarro et al. 2017; Wu et al. 2014). Owing to the significant regional characteristics of aquifers distribution in Shanghai region, zoning of aquifers was also conducted over the past years. For instance, Shi (2010) analyzed the characteristics and groundwater levels of the feeble confined aquifer and Aquifer I and proposed a zoning map of water inrush. Sun et al. (2010) developed a zoning map for Aquifers I and II with consideration of the roof depths of confined aquifers. Nevertheless, the two aforementioned studies were limited to the feeble confined aquifers, Aquifers I and II. Moreover, the zoning of aquifers was mainly carried out for the central district of Shanghai, with separate consideration of the influences of groundwater level and aquifers roof depths.

Quicksand, as a local geological hazard, is often disastrous. To reduce the damage caused by geological hazards, risk assessment was widely used as an effective method (Buttrick et al. 2001; Hsu et al. 2013; Kijko et al. 2002, 2003; Michael-Leiba and Baynes 2002; Peng and Wang 2015; Qi et al. 2017; Zhou and Chen 2008). According to the Asian Disaster Reduction Center for disaster risk management (ADRC 2005), the risk level of geological hazard is evaluated by two factors, i.e., the hazard and the vulnerability of disaster bearing bodies. The former component mainly focuses on the natural attributes of hazard risk, while the latter emphasizes the comprehensive social and economic impacts of the hazard and the resilience capability of the community (Bednarik et al. 2012; Cutter et al. 2008a, b; Dewan 2013; Fuchs et al. 2012, 2016; Westen et al 2008; Yoon 2012). This point of view is also verified by many other researchers (Schneiderbauer and Ehrlich 2004; Wolf 2012; Zhou and Chen 2008). In addition, the vulnerability of geological hazard is concerned by more and more scholars (Ahmed and Dewan 2017; Fuchs et al. 2012; Huang et al. 2011; Lummen and Yamada 2014; Usha et al. 2012; Westen et al. 2008; Wisner et al. 2004). And vulnerability assessment has been becoming widely applied in the prevention of geological disasters, including landslide, flood and seismic (Ahmed and Dewan 2017; Alam et al. 2012; Bender 2002; Huang et al. 2011; Sowmya et al. 2015; Nie et al. 2017).

To date, study on quicksand risk is rare. Lots studies concerning quicksand were aimed to reveal its mechanism (e.g., El Shamy and Zeghal 2005; Fujisawa et al. 2013; Sun 2016). To mitigate the local quicksand risk in Shanghai region, Shi (2010) estimated the occurrence probability of quicksand in central urban area, based on the regional characteristics of sand strata. However, this study was limited to the natural attributes of quicksand, while the vulnerability assessment was not considered. As a typical geological hazard, the vulnerability assessment of quicksand is essential in the quicksand risk assessment without exception. Therefore, a reasonable assessment and zoning study of the quicksand risk considering both its hazard and vulnerability are quite necessary for the Shanghai region.

The objective of this study therefore is to develop a zoning map of confined aquifers distributed throughout the entire Shanghai region in terms of its hydrogeological characteristics and evaluating the quicksand hazard, vulnerability and risk. In terms of large test data obtained from laboratory and field investigations, aquifer zoning maps, quicksand risk and its management zoning maps are developed, which give an overall view about the hydrogeological conditions in Shanghai and provide helpful information for the future urban planning and underground constructions, thereby mitigating the disasters caused by confined water inrush and quicksand.

2 Hydrogeological characteristics of confined aquifers in Shanghai region

The confined water in Shanghai region is located in six aquifers which are separated by five clayey aquitards. The typical profile and geological characteristics of soil layers are shown in Table S1 in ESM (Chen et al. 2011; SUCTC 2012). For the underground constructions in this region, the maximum excavation depth has reached 41 m (Xu et al. 2009). At this depth, it involves the feeble confined Aquifer ⑤2, Aquifers I and II corresponding to Layer No. ⑦ and ⑨ (note that labels ⑤2, ⑦ and ⑨ are the same as the standard hierarchical labels in the SUCTC 2012, as specified in Table S1 in ESM). In the following sections, the distribution and characteristics of these aquifers are presented and discussed.

2.1 Characteristics of the feeble confined ⑤2 aquifer

The artesian aquifer above soil layer ⑦ is often dubbed the feeble confined aquifer and mainly buried in layer ⑤2. This aquifer is in silty soil, fine sand, silty sand and silty sand imbedding in the thin silty clay and has a wide variability in the region of paleochannel, the three islands of Chongming District (namely Chongming Island, Changxing Island and Hengsha Island) and the southern part of the center city (SIGS 2008).

The roof depth of this ⑤2 aquifer varies from – 10 to − 40 m. (The standard Wusong elevation (SLRMB 2010) is used as the base level.) The buried depth is up to – 30 to − 40 m and – 40 to − 50 m in some parts in the three islands of Chongming District (SIGS 2008; Wei et al. 2010).

This aquifer has a poor water quality (Shi 2010; SLRMB 2010). The groundwater level (hereinafter referred to as water level) is seasonally influenced by the meteorology, hydrology and irrigation. Monitoring data show that the water level is generally below − 4 m in the south of the central city (particularly on both sides of the Huangpu River around the Expo site area) and is higher than − 2 m outside the central city and at – 1 to − 3 m in other regions as shown in Table S1 in ESM.

2.2 Characteristics of the ⑦ Aquifer I

The aquifer buried in soil layer ⑦ is called Aquifer I (SUCTC 2012). The soil layer is mainly composed of sandy silt, silty sand and silty-fine sand. The soil type gradually varies from fine to coarse sand with increasing depth (Shi 2010). Aquifer I is widely distributed in Shanghai, but it is absent in the north and west of Shanghai, including Chongming District, some local area of Jiading, Baoshan, Jinshan Districts and the region surrounding Dianshan Lake in Qingpu District. The thickness and depth of Aquifer I have a large difference in different regions due to the cutting effect of paleochannel. Generally, it is thin (about 5–11 m) in the west area of paleochannel, but it is thick (about 11–37 m) in the eastern and the normal sedimentary area. The thickness ranges from 5 to 37 m, and the roof elevation lies between – 11 and − 43 m (− 11 to − 19 m in the north and west, − 19 to − 35 m in the central city, and – 35 to − 43 m in the eastern and southern) (SIGS 2008; Wei et al. 2010).

The water level of Aquifer I is stable with time at one site, but it varies spatially (Shi 2010; SLRMB 2010). Monitoring data indicate that the water level varies in the range from 0 to − 4 m and decreases and then increases from the south to north. The water level is high in Baoshan District, Yangpu District and Gaoqiao area in north Pudong New District, being at 0 to − 2 m, but the water level is low in the Putuo, Jingan Districts and the Jinqiao area in Pudong New District, being at – 3 to − 4 m; it is about − 2 m in south districts including Jinshan, Fengxian and Pudong Districts; the south Tangqiao and the Expo area have the lowest water level at around − 4 m, as shown in Table S1 in ESM.

2.3 Characteristics of the ⑨ Aquifer II

The aquifer buried in the ⑨ soil layer is named Aquifer II (SUCTC 2012). This widely distributed aquifer is absent in the outcrop bedrock areas. This soil layer contains gravelly coarse-medium sand, fine-medium sand and silty-fine sand, with the particle size increasing with depth (Shi 2010). The thickness ranges from 10 to 70 m, and the roof elevation lies at – 60 to − 65 m in the north and the central city, while – 65 to − 70 m in the other regions (SIGS 2008; Wei et al. 2010).

The water level of Aquifer II is stable in different regions (SLRMB 2010). Monitoring data show that the water level lies at 1 m to − 1 m in the three islands of Chongming District, at – 3 to − 6 m in Jiading District and Baoshan District and at − 4 m in central city; it is at − 3 to − 8 m in the center and south part, at – 2 to − 15 m in west part and deep to − 20 m in some areas of northwestward Jinshan District, as shown in Table S1 in ESM.

From the analysis on the above three aquifers, it can be seen that: (a) the soil layers mostly consist of silty sandy soil and sandy soil; (b) the aquifers have regional characteristics in burial conditions and distributions and vary significantly in roof depth and thickness; (c) the water level is stable in general but also varies greatly in different areas; (d) most aquifers have high water content and high water head (higher than 8 m). From a practical point of view, it is of great importance to develop a zoning map of the aquifers in terms of hydrogeological characteristics for guiding the future urban planning and underground construction in Shanghai region. In the next section, aquifer zoning is conducted based on the above characteristics of confined aquifers.

3 Zoning of the confined aquifers in Shanghai region

For the construction projects at their preliminary planning stage (Stage I) and design/construction stage (Stage II), the zoning of the confined aquifers is developed based on two different criteria incorporating different requirements of the projects at different stages. Detailed information is presented in the following section:

3.1 Aquifer zoning for projects at stage I

At stage I, it is expected to have the general distribution of the groundwater in the region. Hence, the criteria for zoning the acquirers only account for the number of the aquifers in the zone. As a result, as shown in Table 1, the confined aquifers are firstly divided into 3 zones I, II and III, based on the number of the aquifers in each zone. Zones I and II are then, respectively, divided into 2 and 3 sub-zones according to the types of aquifers. It is found that the feeble confined aquifer is located at the ⑤2 soil layer, while Aquifer I and Aquifer II are located at the ⑦ and ⑨ soil layer, respectively.

Table 1 Aquifer zoning for projects at preliminary planning stage considering the number of the aquifers
Full size table

Based on the above criteria together with the basic geological data and the distribution maps of the ⑤2, ⑦ and ⑨ soil layers in Shanghai (SIGS 2008; Wei et al. 2010), the zoning map with a measure scale ratio of 1:20 million is developed, as shown in Fig. 1. It can be observed from Fig. 1 that most districts of Shanghai are located in zone II3 with complex hydraulic conditions; it is mainly because Aquifers I and II are widely distributed in Shanghai region. Several local areas (e.g., part of Sanxing Town, Miao Tang and Chenjia Town) in the three islands of Chongming District are located in zone I2; it is mainly because there is a deletion of Aquifer I in Chongming District. The rest areas are mostly located in zone II2. The northern Jiading and Baoshan Districts have many different zones, including zone I2, II2 and II 3. Zone I1 spreads in the area where the bedrock exposed. Although zone III has a scattered distribution, the water inrush risk here is higher since it contains the feeble aquifer ⑤2, Aquifers I and II. Thus, important project constructions are not recommended in zone III. Since the zone II3 contains Aquifers I and II, special attention should be paid to mitigate the water inrush when constructions are conducted.

Fig. 1
figure 1

Zoning map of the confined aquifer in Shanghai for projects at preliminary planning stage considering the number of the aquifers

Full size image

3.2 Aquifer zoning at stage II

At stage II, detailed information about the distribution of groundwater is needed. Therefore, the criteria for zoning the acquirers comprehensively take into account the number of the aquifers in the zone as well as the roof depth and the groundwater level of each aquifer. The detailed zoning criteria of the roof depth and the groundwater level are presented in Table 2. According to the detailed content of national economic and social development plan of Shanghai (SMPG 2011), the three islands of Chongming District would be built as an ecological garden, in which the construction activities would be rare, and these areas are not considered in our study. The confined water head (the vertical distance from the top of the confined aquifer to the groundwater level) for the corresponding stratification point is calculated and listed in Table 2. The aquifers are divided into 3 zones, I, II and III. They are divided into I1, I2, I3, II1, II2, II3, III1 and III2 with respect to different confined water head, respectively. Moreover, zones III1 and III2 are divided into three sub-zones labeled with subscripts a, b and c, respectively.

Table 2 Aquifer zoning for projects at design and construction stages considering the distribution, the roof depth and the groundwater level of the aquifers
Full size table

It can be seen from Table 2 that zones I, II and III have 3, 3 and 6 water heads, respectively, with different combinations. There are 12 (i.e., 3 + 3 + 6) sub-zones for one layer, 45 (i.e., 3 × 3 + 3 × 6 + 3 × 6) sub-zones for two layers and 54 (i.e., 3 × 3 × 6) sub-zones for three layers. That is, there are 111 sub-zones in total for the aquifer zoning. However, because some of the combinations of aquifers actually do not exist according to the distribution map of aquifers, 48 combinations of zones of confined aquifers are proposed as shown in Table S2 in ESM. Accordingly, in terms of the zoning criteria for Stage II together with the related distribution geological maps of the ⑤2, ⑦ and ⑨ soil layers, the roof elevation maps (SIGS 2008; Wei et al. 2010) and the water level distributing maps of the three aquifers (SLRMB 2010), the zoning map of the aquifers with a measure scale ratio of 1:20 million is developed as shown in Fig. 2.

Fig. 2
figure 2

(Reproduced with the permission from SIGS 2008; Wei et al. 2010)

Zoning map of the confined aquifer in Shanghai for projects at design and construction stage considering the distribution, the roof depth and the groundwater level of the aquifers.

Full size image

From Fig. 2 and Table S2 in ESM, it can be found that most regions of Shanghai belong to the zones with two aquifer layers, such as II1 + III1-a, II2 + III1-a, II2 + III2-a, II3 + III1-a and II3 + III2-a. These zones are composed of zone II and the high-pressure area of zone III. Zone II includes aquifer I and Zone III includes aquifer II, and both of them have a high water head, which can be observed in Table 2. The zone III mainly consists of sub-zone III-a, except for sub-zone III-c in the southwest region and III-b in the southwest narrow banded region. Most administrative district (except for Fengxian and Qingpu Districts) contains zones II2+III1-a and II3+III2-a, with complex hydraulic conditions. Effective countermeasures should be taken to mitigate the water inrush when constructions are conducted in these areas.

The two-stage zoning maps give an overall view of the water inrush in this region, which would be helpful to the government, urban planners and designers. However, it should be noted that, for the engineering and technical analysis of a specific project, detail engineering data of the construction site should be referred to obtain the specific hydrogeological information of confined aquifers.

4 Quicksand characteristics and assessment

Quicksand is known as a sudden suspension flow phenomenon of saturated loose soil under water pressure (resulting from the water head difference or hydraulic gradient), which could lead to significant damage to the underground constructions. The critical sate governing the occurrence of quicksand is closely associated with the following three typical aspects (SUCTC 2012):

  1. (i)

    Engineering geological conditions Quicksand occurs mostly in the silty-fine sand layer with uniformly distributed particle size, and sometimes in silty soil;

  2. (ii)

    Groundwater condition The groundwater level higher than the depth of the sand layer makes the soil layer saturated, which is an essential condition for quicksand;

  3. (iii)

    Artificial factor Operations such as pit or tunnel excavation often result in free face and changes in groundwater level.

4.1 Characteristics and distribution of soil with high vulnerability for quicksand

The collected data for the Shanghai region (SUCTC 2012) show that the soils with high vulnerability for quicksand have the following main characteristics: (a) the silty sand, sandy silt, silt and sand layers with a thickness of more than 0.25 m; (b) the clay-size fraction is less than 10% and the silt-size fraction is more than 75%; (c) the coefficient of uniformity is less than 5; (d) the porosity is higher than 43%; (e) the water content is higher than 30%; (f) the soil is in a saturated state; (g) the soil has a low permeability and a poor drainage condition.

On the other hand, the underground constructions are generally conducted at a shallow depth, 60 m beneath the ground surface. For example, the excavation depth of Yishan Road Station, the transfer station of Shanghai Metro Line 3, 4 and 9, was 29.7 m and the retaining wall was constructed to a depth of 62.9 m (Shen et al. 2014). The soil strata in this zone can be divided into 7 layers, i.e., the ①–⑦ layers (SUCTC 2012). Quicksand mainly occurs at the ②0 layer, the ②3 layer, the ③2 layer, the ⑤2 layer and the ⑦ layer. The distribution and characteristics of these layers are presented in Table S3 in ESM (SIGS 2008; Wei et al. 2010).

4.2 Characteristics of the quicksand risk in Shanghai region

Quicksand risk needs to be evaluated in terms of hazard occurrence and damage level to human beings and society. In the Shanghai region, sand strata with high water head can be easily disturbed by underground engineering construction, which is the main cause of quicksand disasters (Elbaz et al. 2016; Luo et al. 2015; Shen et al. 2014; Xu et al. 2009). The recent historic cases indicate that the quicksand risk has the following characteristics (Tang et al. 2016): firstly, most of the quicksand occurred suddenly and the destructive process completed in a very short time (Shen et al. 2014; Xu et al. 2009). Secondly, the hydrogeological conditions as well as the social and economic conditions of the zone where quicksand occurred differ significantly from each other. Finally, the human activities such as foundation pit excavation, tunnel excavation are often the decisive factor to trigger the occurrence of quicksand, especially when the design and construction are improper (Elbaz et al. 2016). However, the quicksand risk can be effectively reduced to an acceptable level through a reasonable design, construction and scientific monitoring.

4.3 The assessment index system of quicksand risk

Although the occurrence probability of quicksand has been investigated for hazard assessment in the central urban area in Shanghai (Shi 2010), little attention has been paid to the vulnerability with respect to quicksand even though it has been reported that vulnerability analysis and assessment can contribute significantly to geohazard risk reduction (Fuchs et al. 2012; Lummen and Yamada 2014; Usha et al. 2012; Westen et al. 2008). To perform a further assessment of quicksand risk, a novel assessment index system considering both the hazard factor and the vulnerability factor is proposed, as illustrated in Fig. 3. The hazard factor mainly refers to dynamic conditions such as hydrogeological conditions, meteorological conditions, construction activities, which would lead to the occurrence of quicksand. In contrast, the vulnerability factor reflects the potential total maximum losses caused by quicksand for the specified area and during a reference period of time. The impact on the built environment may be measured by the population size and density—with a densely populated area being more vulnerable to quicksand than a sparsely populated one. The vulnerability of property mainly attributes from losses due to damage of the physical infrastructure and buildings. The economic vulnerability can be measured in terms of the gross domestic product (GDP), which is an index that reflects the degrees of social prosperity, economic development and the disaster recovery capacity of an area (Schneiderbauer and Ehrlich 2004). Economic vulnerability to quicksand disaster is greater in areas with higher the GDP. With respect to quicksand, land resources are the main contributor to resource and environmental vulnerability and may be approximately measured by the total residential investment. Obviously, an area with high land value should be rated as having higher resource vulnerability because it would suffer a greater loss when quicksand occurs. In the hazard-affected area, the resistance capability of the habitat environment, the life and property to the disaster as well as the resilience ability of post-disaster reconstruction reduce with the increase of population density and property density (Schneiderbauer and Ehrlich 2004).

Fig. 3
figure 3

Diagram of elements and index system for quicksand risk assessment

Full size image

5 Quicksand risk assessment and its management zoning

The level of quicksand risk is the interplay between hazard and vulnerability (see Fig. 3). Therefore, the whole risk assessment and zoning of the quicksand disaster include three aspects: hazard analysis, vulnerability analysis and risk assessment (see Fig. 4), as discussed in the following sections.

Fig. 4
figure 4

Procedures of quicksand risk assessment

Full size image

5.1 Quicksand hazard zoning

The aim of the quicksand hazard zoning is to derive the occurrence probability in different locations in the region. The hazard analysis mainly considers the operation and hydrogeological conditions, meteorological conditions and anthropogenic geological activities mentioned above. According to the literature (Deng et al. 2016) and a series of meteorological survey data provided by the Shanghai City Planning and Land Resources Management Bureau (SLRMB 2010), there is a good correlation between meteorological conditions and hydrogeological conditions in Shanghai region. As for the anthropogenic geological activity, it would vary for different projects and should be analyzed for the specific construction conditions of the project. Therefore, the hazard zoning is conducted mainly based on the hydrogeological conditions (distribution of sand strata and the aquifers).

The distribution of the ①3 layer is quite narrow, and the ②0 layer is only distributed at some shore areas along the Huangpu river, while the ③2 layer scatters in some local area as a small interbed. For the groundwater condition, the construction projects would be mainly affected by the feeble confined aquifer, Aquifers I and II (Xu et al. 2009; Sun et al. 2010). The phreatic water is widely distributed and has a relatively stable feature. It is found that the feeble confined aquifer located at the ⑤2 soil layer, Aquifer I located at the ⑦ layer and Aquifer II at the ⑨ soil layer, respectively. That is, there is a one-to-one correspondence between the quicksand stratum and the aquifers. Because the ⑨ layer is widely distributed and has a stable layer position, this layer can be ignored in the further analysis.

Therefore, in this study the hazard zoning of quicksand in the three quicksand strata, i.e., the ②3, ⑤2 and ⑦ layers in Table S4 in ESM is conducted. Three grade zones, namely the low-, medium- and high-hazard zones of quicksand as well as the corresponding hazard zoning map, are developed as shown in Fig. 5. The zones with different characteristics are presented in Table 3.

Fig. 5
figure 5

(Based on SIGS 2008; Wei et al. 2010)

Hazard zoning map of quicksand in Shanghai.

Full size image
Table 3 Hazard zoning of quicksand and the characteristic of each zone in Shanghai
Full size table

It can be seen from Fig. 5 that Zones II1 (medium-hazard zone) and III2 (high-hazard zone) covered most region of Shanghai with III3 (high-hazard zone) sparsely scattered, and Zone III1 (high-hazard zone) and Zone II2 (medium-hazard zone) covered most area of the three islands of Chongming District. The low hazard zone I is only distributed in some local areas in Songjiang, Jiading and Qingpu Districts. Summarizing, most of Shanghai region has high- and medium-quicksand hazard levels. To avoid the quicksand accidents, particular countermeasures should be taken in these high- and medium-hazard zones.

5.2 Quicksand vulnerability zoning

Vulnerability is a complex, dynamic and multidimensional concept, hazard and scale dependent (Fuchs et al. 2012). As discussed in Sect. 4.3, vulnerability can be mainly considered from four aspects: built environment, property, economic, resource and environment. And vulnerability to community can be evaluated with consideration of the population size and density, the physical infrastructure and buildings, the GDP and residential investment. It may be difficult to access the triggering factors of the risk bearing body for characterizing the regional vulnerability due to the lack of detailed information about the infrastructure and buildings and residential investment. Therefore, a semiquantitative approach was adopted here to estimate the vulnerability of communities. The infrastructure investment, residential investment and GDP of each prefecture in Shanghai region have a strong correlation with its population size and density. That is to say, the size and density of the population in a region can reflect the vulnerability of built environment, economic, property and resources. The area with more developed economy and the higher GDP generally has greater population size and density, thus making greater building and infrastructure densities and higher land values. Therefore, the vulnerability of quicksand was estimated semiquantitatively from the population size and density in the administrative region (including both the central city and suburb) of Shanghai region, which were provided by the Shanghai Municipal Statistics Bureau (SMSB 2010). Table S5 in ESM shows the population size and density of each prefecture of Shanghai at 2010. It is observed that the population size and density have obvious relevance with the administrative unit (city center and suburbs) and have a significant difference in different regions. The population density in the urban center is in the range of 18,031–36,307/km2 (an average of 24,137/km2); the suburban population density is in the range of 3169–7029/km2 (an average of 4684/km2); population density of outer suburban district is in the range of 594–2613/km2 (an average of 1388/km2). And the areas surrounding the administrative center also have a bigger population density. According to the above reasons, we classify the quicksand vulnerability in Shanghai region into three levels: low, medium and high. The scope of administrative units of each vulnerability level is indicated in Table S6 in ESM. Quicksand vulnerability zoning map was finally developed and three zones, namely the low zone I, medium zone II and high zone III are shown in Fig. 6. It can be seen that the central city, Lujiazui area along the Huangpu River in Pudong New District and the surrounding area of each administrative center lie in the high vulnerability zone of quicksand; the original Pudong, Minhang and Baoshan Districts which are adjacent to the central city generally lie in the medium-vulnerability zone, except for some scattered high vulnerability zones; the rest areas which are considerable large mostly lie in low vulnerability zone with some scattered high vulnerability zones. As can be observed, the distribution of quicksand vulnerability has different characteristics with the quicksand hazard. Areas with high hazard level may on the contrary have low vulnerability level, just as the three islands of Chongming District and east region of Shanghai. This would affect the quicksand risk discussed in the next section.

Fig. 6
figure 6

Vulnerability zoning map of quicksand in Shanghai

Full size image

5.3 Assessment and zoning of quicksand risk

The quicksand risk level is determined by multiplying the hazard and vulnerability of the quicksand (ADRC 2005):

$$ {\text{Risk }} = {\text{hazard }} \times {\text{vulnerability}} $$
(1)

To provide a practical reference for the prevention of the quicksand risk in Shanghai region, the assessment and zoning of the quicksand risk are conducted on the basis of the quicksand hazard and vulnerability evaluation. For simplicity, the low, medium to high quicksand hazard and vulnerability levels are quantified by values of 1, 2 and 3, respectively. Then, different risk values of 1, 2, 3, 4, 6 and 9 as shown in Table 4 can be obtained using Eq. (1). With regard to risk levels 1, 2 and 3, it can be inferred that either the hazard or the vulnerability value is equal to 1, which means a low hazard or vulnerability level. In this case, the quicksand risk is consequently low. With regard to the risk values 6 and 9, it can be inferred that both factors are higher than 2, one being equal to 3. That is to say, the hazard or the vulnerability level lies in the medium or high level zone and one of them is certainly at a high level. In this case, the quicksand risk will be consequently high. For the risk value 4, both the hazard and the vulnerability are at a medium level and the quicksand risk consequently lies in the medium zone. The risk zones of the quicksand elaborated above are presented in Table 5.

Table 4 Hazard, vulnerability and risk levels of quicksand
Full size table
Table 5 Risk zoning of quicksand
Full size table

The quicksand risk zoning map is established and shown in Fig. 7. It can be found that the high-risk zones are not completely consistent with high hazard and vulnerability zones of quicksand. In the central city, the riverside area along the Huangpu River in Pudong New District and the surrounding area of each administrative center lie in the high quicksand risk zone. Most area of the Pudong, Minhang and Baoshan Districts lie in the medium-risk zone except for some scattered high-risk zones. The rest of the administrative division generally lies in a low-risk zone except for some high zones around the central district. Three islands of Chongming District and east region of Shanghai at high quicksand hazard level have a low-risk level due to their low vulnerability level. Effective measures must be taken to prevent the quicksand disasters before any underground constructions in such medium- and high-risk zones.

Fig. 7
figure 7

Risk zoning map of quicksand in Shanghai

Full size image

5.4 Management zoning of quicksand risk

According to the detailed content of national economic and social development plan of Shanghai (SMPG 2011), Baoshan and Minhang Districts will make further efforts to promote their urbanization level for becoming the expanding area of central city in the near future. A number of industrial projects such as trade zones, bonded areas and industrial parks will be constructed in Pudong District. A series of construction projects will be carried forward along the coastal areas, including Jiading New Town, Songjiang New Town, Qingpu New Town, Pudong Nanhui New Town, Fengxian Nanqiao New Town, Jinshan New Town, Chongming Chengqiao New Town and Chongming District. Based on the development plan (SMPG 2011) and the quicksand risk zoning proposed above, the management of quicksand risk in Shanghai region can be divided into three parts, namely the most-key zone, the sub-key zone and the general-key zone, as illustrated in Table 6 and Fig. 8. It can be found that most parts of Shanghai lies in the most-key and sub-key areas for risk control and management, in which special attention should be paid to mitigate the quicksand risk.

Table 6 Risk management zoning of quicksand in Shanghai
Full size table
Fig. 8
figure 8

Risk management zoning of quicksand in Shanghai

Full size image

6 Conclusions

The zoning of confined aquifers and quicksand risk in Shanghai region is investigated, and the results allow following conclusions to be drawn:

  1. (1)

    The zoning of the confined aquifers is developed based two different criteria in terms of different requirements of underground constructions at different stages (I: preliminary planning stage; II: design and construction stage). As a result, the situation and characteristics of the confined aquifers can be directly obtained from the corresponding zoning maps.

  2. (2)

    According to the zoning for stage I, most of the districts of Shanghai are located in zone II3 where both Aquifers I and II exist, and the rest areas are mostly located in zone II2 where both the feeble aquifer and Aquifer II exist. According to the zoning for stage II, most regions of Shanghai belong to the zones that have high water heads, and effective measures should be taken to mitigate the water inrush when constructions are conducted in these areas.

  3. (3)

    The quicksand risk zoning by incorporating both the quicksand hazard and vulnerability assessments is conducted in Shanghai region. The vulnerability should be considered due to its significant influence on the quicksand risk assessment. Most areas of Shanghai region have a high and medium-quicksand hazard levels. The central city, the riverside area along the Huangpu River in Pudong New District and the surrounding area of each administrative center lie in the high quicksand risk zone. Most area of the Pudong, Minhang and Baoshan Districts lie in the medium-risk zone. According to the management zoning, most areas lie in the most-key and sub-key zones.

This study has clarified the zoning distribution characteristics of hydraulic and geological conditions in Shanghai region. The obtained zoning maps gave an overall view of the water inrush and quicksand risk and provide valuable decision-making information for the government, urban planners and designers. The proposed method can be also extended to other cities by considering the specific hydraulic and geological situations of each city.