Because of rapid economic development and urbanization, water shortage has become a serious problem in the arid region of China. To investigate urban water resource security, the supply demand pressure of water resources and the urban expansion index were analyzed under different developing scenarios in this paper. Based on the economic data of Urumqi, a typical inland city in the arid area, under the present development scenario from 2011 to 2030, a system dynamics model was constructed to simulate the water resource security. The results show that there will be great influence of urban expansion on water resource security in Urumqi in the future. Water resources are projected to become increasingly scarce if the urban expansion is left unchanged in terms of population, economic growth and water-use efficiency. To find a sustainable method for water resource use, four scenarios of urban expansion were set up based on the sensitive variables. Based on comparison of water consumption under the different scenarios, the harmonize scheme for urban water resource security is the best choice for the development of Urumqi. If the impact of urban expansion on urban water resource security alleviates in the future, the main parameters would have to reach a new standard of water use. Reducing the sewage and increasing the reuse proportion of wastewater are also very important for relieving the stress of water shortage. This research can serve as a reference for water resource allocation and urban planning in arid areas.

INTRODUCTION

In recent years, because of rapid economic development and urbanization, water shortage has become a serious problem in the arid region of China (Peters & Meybeck 2000; Vörösmarty et al. 2000; Palmer et al. 2004). Water resource security has received considerable attention because of its vital role in the ecology and environment, as well as in economic and social development (Falkenmark & Widstrand 1992; Niemczynowicz 1999; Bogardi et al. 2012). As one of the core measuring indices, the carrying capacity of water resources could evaluate the risk of water resources (Rijsberman & Van De Ven 2000; Xia & Zhu 2002). This has recently become a focus of research (Al-Otaibi & Abdel-Jawad 2007; McDonald 2008). Numerous evaluations of water resource security have been conducted in the irrigation area of an oasis (Ji et al. 2006; Cook & Bakker 2012). However, the stress of urban water use has also increased because of rapid urbanization in the oasis (Chen et al. 2011). Research on evaluating the water resource security and obtaining a sustainable water-use plan in the arid area is not sufficient.

Many methods have been used to evaluate water resource security, such as the Analysis Hierarchy Process (Jaber & Mohsen 2001), fuzzy clustering (Han et al. 2003), neural networks and system dynamics (SD) (Sun & Chi 2008). System dynamics is an approach to describing the behavior of complex systems with time (Simonovic & Fahmy 1999; Ahmad & Simonovic 2000). The internal feedback loops and time delays in the entire system could be integrated (Forrester et al. 1976; Sterman 2000). A complex water resource system usually includes many types of subjective variables (Danielson 1979). Many methods are still not fully satisfactory, partly because most deal with over-simplified systems (Yeh 1985; Ahmad & Simonovic 2000). Thus, SD are well suited for the modeling of and application to water resources problems (Keyes & Palmer 1993; Fletcher 1998; Li & Simonovic 2002). This method was first used for evaluating the carrying capacity of water resources in the 1990s (Motohashi & Nishi 1991; Feng et al. 2008), although the SD model cannot adequately represent changing processes in space (Ahmad & Simonovic 2004). An SD model is less useful in predicting exact future system states than for specifying how alternative choices would alter the tendency to move towards each of those conditions (Vennix 1996; Li & Simonovic 2002). Therefore, coupling development scenarios are very necessary in an SD model for predicting water resources security.

Urumqi, a megalopolis in the arid area of China, is facing the dilemma of water shortage for maintaining domestic life, agriculture and environmental requirements because of rapid urbanization of the last 20 years. A quantitative study of water resource carrying capacity was conducted along the Urumqi valley (Shi & Qu 1992), but an evaluation of future water resource security has not yet been reported.

To evaluate the water resource security of Urumqi, an SD model was constructed based on an urban expansion index for describing the development of Urumqi; then the water resource carrying capacity during the period of 2006–2030 was evaluated with the consideration of climate change, population growth and industrial development; lastly, an optimal scheme was proposed for the sustainable utilization of water resources based on the SD model. This research could serve as a reference for water resource allocation and urban planning in the arid area of China.

DATA AND METHODS

Description of the study area

The study area, Urumqi, is located in the arid area of northwestern China. It covers a surface area of approximately 14,216.3 km2 between latitudes 42°45′32″–44°08′00″N and longitudes 86°37′33″–88°58′24 E (Figure 1). The climate in this area is dry. The average annual precipitation is 286.3 mm, and the average annual temperature is 6.9 °C, with a maximum annual temperature of approximately 23.7 °C and minimum annual temperature of approximately −12.6 °C; the annual average water resource is 11.38 × 108 m3. The amount of water available to the average person decreased from 729.54 to 377.8 m3/year during 1995–2012. The population increased by 2 million, and the economy increased six-fold in this period. The urban water resource security was badly affected.
Figure 1

Relief map of the studied area.

Figure 1

Relief map of the studied area.

Construction of the system dynamics model

The SD method was introduced in this paper for evaluating the urban water resource security. A water resource system is a complex system that affects both the socio-economic and ecological systems (Hunter 1998; Rijsberman & Van de Ven 2000). Thus, the factors in the SD model for water resource carrying capacity of Urumqi are closely correlated with the economy, eco-environment and population. The system was divided into two components, water demand (WD) and water supply, and five subsystems: economic WD subsystem, social life WD subsystem, ecology and environmental WD subsystem, water resources subsystem and recycled water subsystem. They consist of three feedback loops. The economy, social life and environment belong to the WD system, and the water resources and recycled water belong to the water supply system. The relationship of these subsystems are expressed in Figure 2.
Figure 2

Analysis of urban water resources system structure.

Figure 2

Analysis of urban water resources system structure.

Figure 3

Subsystem diagram for urban water resources system.

Figure 3

Subsystem diagram for urban water resources system.

The elements of the subsystems influence significantly and interact with each other (Figure 3). Their relationships form multiple feedback loops.

Generally, WD is determined by the ecological water, life water and economic water. Life water is determined by the population and water consumption quota. Ecological water includes urban afforestation water and street flushing water. Economic water depends on the primary, secondary and tertiary industry. Water supply depends on the water resource quantity and reclaimed water. The gap between supply and demand could reflect the degree of water shortage, and vice versa.

The SD model in this study was developed within The Vensim Personal Learning Edition (VENSIM PLE) software, a visual modeling tool (Shiflet & Shiflet 2011). VENSIM PLE provides a simple and flexible way of building simulation models from causal loops or stock and flow diagrams (Figure 4). Variables are listed in Table 1.
Table 1

Variables description

System Subsystem Abbreviation Description Unit 
Water demand Economic IA Irrigation area 103 hm2 
  GDP Gross domestic product 108 Yuan 
  GAOGDP Growth amount of GDP 108 Yuan/year 
  GAOIA Growth amount of irrigation area 103 hm2/year 
  PWGDP Primary industrial GDP 108 Yuan 
  PIWD Primary industrial water demand 108 m3 
  SIGDP Secondary industrial GDP 108 Yuan 
  SIWD Secondary industrial water demand 108 m3 
  TIGDP Tertiary industrial GDP 108 Yuan 
  TIWD Tertiary industrial water demand 108 m3 
  POPI Proportion of primary industrial GDP – 
  POSI Proportion of secondary industrial GDP – 
  POTI Proportion of tertiary industrial GDP – 
  IS Industrial structure – 
  GROPIGDP Growth rate of primary industrial GDP 1/year 
  GROSIGDP Growth rate of secondary industrial GDP 1/year 
  GROTIGDP Growth rate of tertiary industrial GDP 1/year 
  GROIA Growth rate of irrigation area 1/year 
  RUROIA Repetitive use rate of industrial water 1/year 
  AIQ Agricultural irrigation quota m3/mu 
  IWUP Industrial water used per 104 Yuan 108 m3 
  TIWUP Tertiary industrial water used per 104 Yuan 108 m3 
 Social life TP Total population 104 persons 
  GAOTP Growth amount of total population 104 persons/year 
  RP Rural population 104 persons 
  UP Urban population 104 persons 
  RDWQ Rural domestic water quota L/person/day 
  UDWQ Urban domestic water quota L/person/day 
  GROTP Growth rate of total population 1/year 
  UR Urbanization rate 1/year 
  RDWD Rural domestic water demand 108 m3 
  UDWD Urban domestic water demand 108 m3 
  DWD Domestic water demand 108 m3 
 Ecology and environment EWD Ecology and environmental water demand 108 m3 
  SFW Street flushing water 108 m3 
  WCOUL Water consumption of urban landscaping 108 m3 
  SFWQ Street flushing water quota m3/m2/year 
  IQ Irrigation quota m3/hm2/year 
  GROGLA Growth rate of green land area 1/year 
  GRORA Growth rate of road area 1/year 
  GLA Green land area hm2 
  RA Road area 104 m2 
  GAOGA Growth amount of green area hm2/year 
  GAORA Growth amount of road area 104 m2/year 
  TWD Total water demand 108 m3 
Water supply Water resources CWQ Conventional water quantity 108 m3 
 Recycled water IWC Industrial water consumption 108 m3 
  IWEA Industrial wastewater emission amount 108 m3 
  TAOS Total amount of sewage 108 m3 
  DS Domestic sewage 108 m3 
  DOTIW Drainage of tertiary industrial wastewater 108 m3 
  IWEC Industrial wastewater emission coefficient – 
  TIWWEC Tertiary industrial wastewater emission coefficient – 
  DSDC Domestic sewage discharge coefficient – 
  RORWU Rate of reclaimed water use 1/year 
  TWS Total water supply 108 m3 
  DSP Demand–supply pressure – 
  SOSAD Shortfall of supply and demand 108 m3 
System Subsystem Abbreviation Description Unit 
Water demand Economic IA Irrigation area 103 hm2 
  GDP Gross domestic product 108 Yuan 
  GAOGDP Growth amount of GDP 108 Yuan/year 
  GAOIA Growth amount of irrigation area 103 hm2/year 
  PWGDP Primary industrial GDP 108 Yuan 
  PIWD Primary industrial water demand 108 m3 
  SIGDP Secondary industrial GDP 108 Yuan 
  SIWD Secondary industrial water demand 108 m3 
  TIGDP Tertiary industrial GDP 108 Yuan 
  TIWD Tertiary industrial water demand 108 m3 
  POPI Proportion of primary industrial GDP – 
  POSI Proportion of secondary industrial GDP – 
  POTI Proportion of tertiary industrial GDP – 
  IS Industrial structure – 
  GROPIGDP Growth rate of primary industrial GDP 1/year 
  GROSIGDP Growth rate of secondary industrial GDP 1/year 
  GROTIGDP Growth rate of tertiary industrial GDP 1/year 
  GROIA Growth rate of irrigation area 1/year 
  RUROIA Repetitive use rate of industrial water 1/year 
  AIQ Agricultural irrigation quota m3/mu 
  IWUP Industrial water used per 104 Yuan 108 m3 
  TIWUP Tertiary industrial water used per 104 Yuan 108 m3 
 Social life TP Total population 104 persons 
  GAOTP Growth amount of total population 104 persons/year 
  RP Rural population 104 persons 
  UP Urban population 104 persons 
  RDWQ Rural domestic water quota L/person/day 
  UDWQ Urban domestic water quota L/person/day 
  GROTP Growth rate of total population 1/year 
  UR Urbanization rate 1/year 
  RDWD Rural domestic water demand 108 m3 
  UDWD Urban domestic water demand 108 m3 
  DWD Domestic water demand 108 m3 
 Ecology and environment EWD Ecology and environmental water demand 108 m3 
  SFW Street flushing water 108 m3 
  WCOUL Water consumption of urban landscaping 108 m3 
  SFWQ Street flushing water quota m3/m2/year 
  IQ Irrigation quota m3/hm2/year 
  GROGLA Growth rate of green land area 1/year 
  GRORA Growth rate of road area 1/year 
  GLA Green land area hm2 
  RA Road area 104 m2 
  GAOGA Growth amount of green area hm2/year 
  GAORA Growth amount of road area 104 m2/year 
  TWD Total water demand 108 m3 
Water supply Water resources CWQ Conventional water quantity 108 m3 
 Recycled water IWC Industrial water consumption 108 m3 
  IWEA Industrial wastewater emission amount 108 m3 
  TAOS Total amount of sewage 108 m3 
  DS Domestic sewage 108 m3 
  DOTIW Drainage of tertiary industrial wastewater 108 m3 
  IWEC Industrial wastewater emission coefficient – 
  TIWWEC Tertiary industrial wastewater emission coefficient – 
  DSDC Domestic sewage discharge coefficient – 
  RORWU Rate of reclaimed water use 1/year 
  TWS Total water supply 108 m3 
  DSP Demand–supply pressure – 
  SOSAD Shortfall of supply and demand 108 m3 
Figure 4

The flow diagram for urban water resource system.

Figure 4

The flow diagram for urban water resource system.

Available data

We obtained data on water resources, including the water consumption of each industry and the environment and the water supply from the Xinjiang water resources bulletin (2006–2011) edited by the Xinjiang Department of Water Resources. Population, gross domestic product (GDP) of each industry and land use area data (2006–2011) were obtained from the Urumqi Statistical Yearbook edited by the Xinjiang Statistics Department (http://tongji.cnki.net/kns55/Navi/NaviSearch.aspx). The population, GDP of each industry and land use area (2011–2020) were acquired from the urban master planning of Urumqi and the 12th 5-year plan of Urumqi edited by the Urumqi Planning and Management Bureau (http://www.ans.com.cn).

Determination of model variables and parameters

The variables in the SD model are mainly categorized into flow variables, flow rate variables, constants, auxiliary variables and table functions. These variables required relational inputs into the VENSIM software. The flow variables express the cumulative quantities, the flow rate variables express the rate of change to cumulative quantities, and the constants do not change over time in a time interval. The table functions are used to express the non-linear relationship between some variables in the model. There are 58 variables and parameters (Table 1).

The causal and logical relationships among the variables, flow rates and table functions were constructed into mathematical relations (i.e., state and auxiliary equations) using linear regression; more details about the constructing equations can be found in Mohapatra et al. (1994).

The parameters were obtained based on the urban master planning of Urumqi (2011–2020), the 12th 5-Year Plan of Urumqi, the water resources bulletin (2006–2011) and the statistical yearbook (2006–2011). The results are listed in Table 2. Owing to the uncertainty of climate factors, the annual average water resources (11.38 × 108 m3) was selected as the reference value in the future. The area of cultivated land was assumed to remain at a constant level of 67.23 km2, as in 2011, and irrigated with water saving irrigation methods.

Table 2

Variables and parameters in the urban water resources SD model

Variables 2006 2011 2015 2020 2030 
GDP 683.68 1,338.52 2,700.00 4,200.00 8,500.00 
POPI 1.54 1.49 1.20 1.00 1.00 
POSI 42.84 44.86 45.30 40.00 29.00 
POTI 55.62 53.65 53.00 59.00 70.00 
GROPIGDP 14.25 9.02 4.17 10.23 10.23 
GROSIGDP 18.90 12.15 6.28 4.25 4.25 
GROTIGDP 18.49 11.80 11.94 12.72 12.72 
IA 56.80 61.97 64.00 65.00 67.00 
GAOIA 1.25 0.66 0.26 0.28 
TP 221.03 243.03 365.85 430.33 506.18 
GAOTP 4.64 3.79 6.54 4.69 4.69 
UR 76.10 73.80 84.50 87.00 91.50 
AIQ 697.00 730.00 650.00 580.00 500.00 
IWUP 48.96 40.81 35.00 30.00 25.00 
TIWUP 6.70 2.92 2.75 2.50 2.30 
UDWQ 213.00 156.90 156.90 156.90 156.90 
RDWQ 61.00 69.30 69.30 69.30 69.30 
CWQ 9.52 9.85 10.09 10.59 11.09 
GLA 14,415.00 17,316.00 53,115.30 215,345.00 754,225.00 
GROGLA 8.05 29.35 25.03 20.43 17.68 
RA 1,474.00 2,005.00 2,684.28 3,886.30 8,223.96 
GRORA 1.24 7.53 7.63 7.75 7.86 
IQ 30.00 30.00 30.00 30.00 30.00 
SFWQ 0.73 0.73 0.73 0.73 0.73 
RORWU 37.00 27.00 30.00 35.00 40.00 
RUROLA 23.56 34.89 36.00 38.00 40.00 
DSDC 56.39 53.25 53.25 53.25 53.25 
IWEC 41.84 58.68 58.68 58.68 58.68 
TIWWEC 59.26 53.33 53.33 53.33 53.33 
Variables 2006 2011 2015 2020 2030 
GDP 683.68 1,338.52 2,700.00 4,200.00 8,500.00 
POPI 1.54 1.49 1.20 1.00 1.00 
POSI 42.84 44.86 45.30 40.00 29.00 
POTI 55.62 53.65 53.00 59.00 70.00 
GROPIGDP 14.25 9.02 4.17 10.23 10.23 
GROSIGDP 18.90 12.15 6.28 4.25 4.25 
GROTIGDP 18.49 11.80 11.94 12.72 12.72 
IA 56.80 61.97 64.00 65.00 67.00 
GAOIA 1.25 0.66 0.26 0.28 
TP 221.03 243.03 365.85 430.33 506.18 
GAOTP 4.64 3.79 6.54 4.69 4.69 
UR 76.10 73.80 84.50 87.00 91.50 
AIQ 697.00 730.00 650.00 580.00 500.00 
IWUP 48.96 40.81 35.00 30.00 25.00 
TIWUP 6.70 2.92 2.75 2.50 2.30 
UDWQ 213.00 156.90 156.90 156.90 156.90 
RDWQ 61.00 69.30 69.30 69.30 69.30 
CWQ 9.52 9.85 10.09 10.59 11.09 
GLA 14,415.00 17,316.00 53,115.30 215,345.00 754,225.00 
GROGLA 8.05 29.35 25.03 20.43 17.68 
RA 1,474.00 2,005.00 2,684.28 3,886.30 8,223.96 
GRORA 1.24 7.53 7.63 7.75 7.86 
IQ 30.00 30.00 30.00 30.00 30.00 
SFWQ 0.73 0.73 0.73 0.73 0.73 
RORWU 37.00 27.00 30.00 35.00 40.00 
RUROLA 23.56 34.89 36.00 38.00 40.00 
DSDC 56.39 53.25 53.25 53.25 53.25 
IWEC 41.84 58.68 58.68 58.68 58.68 
TIWWEC 59.26 53.33 53.33 53.33 53.33 

Model verification

The changes of WD, GDP and population during 2007–2011 were simulated using the above model based on the base year of 2006. The results are shown in Table 3.

Table 3

Verification results in the urban water resources SD model

Item Type 2007 2008 2009 2010 2011 
TWD (108 m3Historical data 9.93 12.13 11.12 10.68 
Simulated data 8.92 9.95 12.12 10.92 10.55 
Relative error (%) −0.60 −0.86 0.28 −0.11 −1.25 
GDP (108 Yuan) Historical data 810.57 982.37 1,087.50 1,338.52 1,690.03 
Simulated data 811.08 990.67 1,096.26 1,349.42 1,703.26 
Relative error (%) −0.06 −0.85 −0.81 −0.81 −0.78 
TP (104 persons) Historical data 231.29 236.05 241.20 243.03 249.35 
Simulated data 231.30 236.08 241.22 243.06 249.37 
Relative error (%) 0.01 0.01 0.01 0.01 0.01 
Item Type 2007 2008 2009 2010 2011 
TWD (108 m3Historical data 9.93 12.13 11.12 10.68 
Simulated data 8.92 9.95 12.12 10.92 10.55 
Relative error (%) −0.60 −0.86 0.28 −0.11 −1.25 
GDP (108 Yuan) Historical data 810.57 982.37 1,087.50 1,338.52 1,690.03 
Simulated data 811.08 990.67 1,096.26 1,349.42 1,703.26 
Relative error (%) −0.06 −0.85 −0.81 −0.81 −0.78 
TP (104 persons) Historical data 231.29 236.05 241.20 243.03 249.35 
Simulated data 231.30 236.08 241.22 243.06 249.37 
Relative error (%) 0.01 0.01 0.01 0.01 0.01 

Compared with the original data during the same periods, Table 3 shows that the relative error is less than 5%, so it is precise and available to evaluate the water resource security.

Sensitivity degree analysis

To analyze the influence of system variables on the urban water resource security, the sensitivity analysis method was introduced. The sensitivity degree can be defined as follows (Guo et al. 2001): 
formula
1
where SQ is the sensitivity degree of state Q to variable X; t is time; Q(t) denotes the system state at time t; X(t) represents the system variable affecting the system state at time t; ΔQ(t) and ΔX(t) denote increments of state Q and variable X at time t, respectively.
For n state variables (Q1, Q2Qn), the general sensitivity degree of a variable at time t can be defined as follows: 
formula
2
where n denotes the number of state variables; SQi is the sensitivity degree of state Qi; S is the general sensitivity degree of the n states over the parameter X.

For the study system, 19 variables are identified for representing system states of population growth, industrial and agricultural development, wastewater reuse and urbanization level, and seven parameters are analyzed to examine their impacts on the system states. To examine the sensitivity degree of each state variable, each parameter is increased (or decreased) by 10% per 1 year for the study horizon of 2007–2030.

Urban expansion index

Urbanization is a long and complicated process, which has many influencing factors. The population, economy, industrial structure, infrastructure construction, living standards and environment usually were used to represent the urbanization (Camagni et al. 2002; Deng et al. 2008). Based on the results of the sensitivity degree analysis (Table 4), an urban expansion index was set up to describe the urban development and analyze the relationship with water resource security. The factors include the urbanization level, industrial structure, living standards and the environment. The urban expansion index is expressed as: Y = log (y1y2y3y4), where y1 is the urbanization level, which is equal to the ratio of the non-agricultural population to the total population; y2 is the industrial structure, which is equal to the ratio of the third industrial production to the GDP; y3 is the living standard, which is equal to the GDP per person; y4 is the environment which, is equal to the green area per person.

Table 4

Results of sensitivity analysis for the urban water resources SD model

No. Var./Para. TP GDP TWD TWS POWSAD TAOS SIGDP Average 
GROTP 0.0891 0.0001 0.0129 0.0017 1.2106 0.0436 0.0001 0.1940 
UR 0.0001 0.7413 0.1030 0.0136 3.5859 0.3410 0.0008 0.6837 
UDWQ 0.0002 0.0010 0.1437 0.0190 3.1160 0.4720 0.0011 0.5361 
RDWQ 0.0001 0.0147 0.0019 1.3803 0.0496 0.0001 0.2067 
AIQ 0.0010 0.0059 0.6486 0.0001 15.8827 0.0035 0.0059 2.3640 
GROIA 0.0002 0.0564 4.6222 0.0001 0.0002 0.6685 
IWUP 0.0002 0.0014 0.2059 0.0184 1.7276 0.4552 0.0014 0.3443 
TIWUP 0.0002 0.0223 0.0031 0.1256 0.0798 0.0002 0.0330 
GROPIGDP 0.0051 0.0012 0.0001 2.7065 0.0026 0.0051 0.3886 
10 GROSIGDP 0.2634 0.0608 0.0058 2.5839 0.1377 0.2634 0.4736 
11 GROTIGDP 0.2471 0.0566 0.0054 2.4765 0.1286 0.2471 0.4516 
12 POPI 0.0002 0.2606 0.1732 0.0159 15.7601 0.4176 0.8533 2.4973 
13 POSI 0.0002 0.1381 0.2284 0.0201 4.9111 0.4904 1.1256 0.9877 
14 POTI 0.0001 0.1433 0.0488 0.0028 3.4326 0.0762 0.4280 0.5903 
15 DSDC 0.0066 0.0002 0.0001 0.0209 7.7949 0.5181 0.0002 1.1916 
16 IWEC 0.2463 0.0002 0.0001 0.0185 9.5457 0.4561 0.0002 1.4667 
17 RORWU 0.0001 0.0004 0.0001 0.0424 4.1440 0.0002 0.0004 0.5982 
18 RUROIA 0.0001 0.0075 0.6273 0.1835 0.0001 0.1169 
19 GROGLA 0.0001 0.0146 0.0001 0.0021 
20 IQ 0.0006 0.1076 0.0155 
21 GRORA 0.0037 0.5506 0.0792 
22 SFWQ 0.0001 0.0147 1.3570 0.0001 0.0001 0.1960 
No. Var./Para. TP GDP TWD TWS POWSAD TAOS SIGDP Average 
GROTP 0.0891 0.0001 0.0129 0.0017 1.2106 0.0436 0.0001 0.1940 
UR 0.0001 0.7413 0.1030 0.0136 3.5859 0.3410 0.0008 0.6837 
UDWQ 0.0002 0.0010 0.1437 0.0190 3.1160 0.4720 0.0011 0.5361 
RDWQ 0.0001 0.0147 0.0019 1.3803 0.0496 0.0001 0.2067 
AIQ 0.0010 0.0059 0.6486 0.0001 15.8827 0.0035 0.0059 2.3640 
GROIA 0.0002 0.0564 4.6222 0.0001 0.0002 0.6685 
IWUP 0.0002 0.0014 0.2059 0.0184 1.7276 0.4552 0.0014 0.3443 
TIWUP 0.0002 0.0223 0.0031 0.1256 0.0798 0.0002 0.0330 
GROPIGDP 0.0051 0.0012 0.0001 2.7065 0.0026 0.0051 0.3886 
10 GROSIGDP 0.2634 0.0608 0.0058 2.5839 0.1377 0.2634 0.4736 
11 GROTIGDP 0.2471 0.0566 0.0054 2.4765 0.1286 0.2471 0.4516 
12 POPI 0.0002 0.2606 0.1732 0.0159 15.7601 0.4176 0.8533 2.4973 
13 POSI 0.0002 0.1381 0.2284 0.0201 4.9111 0.4904 1.1256 0.9877 
14 POTI 0.0001 0.1433 0.0488 0.0028 3.4326 0.0762 0.4280 0.5903 
15 DSDC 0.0066 0.0002 0.0001 0.0209 7.7949 0.5181 0.0002 1.1916 
16 IWEC 0.2463 0.0002 0.0001 0.0185 9.5457 0.4561 0.0002 1.4667 
17 RORWU 0.0001 0.0004 0.0001 0.0424 4.1440 0.0002 0.0004 0.5982 
18 RUROIA 0.0001 0.0075 0.6273 0.1835 0.0001 0.1169 
19 GROGLA 0.0001 0.0146 0.0001 0.0021 
20 IQ 0.0006 0.1076 0.0155 
21 GRORA 0.0037 0.5506 0.0792 
22 SFWQ 0.0001 0.0147 1.3570 0.0001 0.0001 0.1960 

RESULTS AND DISCUSSION

Sensitivity analysis

Based on Equation (1), five sensitivity degree values can be obtained for each parameter–variable pair. Their average represents the general sensitivity degree of the parameter to the variable. Furthermore, according to Equation (2), an average can be obtained for all 19 system states to the parameter.

Table 4 shows the results of the sensitivity degree analyses. The results indicate that the study system responds to most of the parameters with a low degree of sensitivity. This fact demonstrates that the developed SD model can undertake effective prediction of the system's behaviors.

Effect of urban expansion on water resource security

Using year 2011 as the current year, the water resource balance and development of Urumqi from 2012 to 2030 was simulated.

Prediction of urban expansion

To quantitatively analyze the urban development, four variables of the urban expansion index were calculated according to the data (Table 5).

Table 5

The urban expansion index of Urumqi city

Year y1 y2 y3 y4 Y 
2011 0.7350 0.5365 55,076.3280 71.2505 6.1896 
2012 0.7625 0.5361 60,073.3923 89.8145 6.3435 
2013 0.7900 0.5358 64,521.4247 111.2023 6.4824 
2014 0.8175 0.5354 68,240.4562 135.2383 6.6063 
2015 0.8450 0.5350 71,074.7600 161.5511 6.7152 
2016 0.8500 0.5460 72,902.7542 189.5625 6.8071 
2017 0.8550 0.5570 75,026.5971 221.5652 6.8985 
2018 0.8600 0.5680 77,477.7548 257.9510 6.9896 
2019 0.8650 0.5790 80,292.7787 299.1165 7.0802 
2020 0.8700 0.5900 83,515.0145 345.4554 7.1705 
2021 0.8745 0.6010 87,195.2037 397.3510 7.2603 
2022 0.8790 0.6120 91,113.9023 455.9814 7.3493 
2023 0.8835 0.6230 95,288.4522 522.0452 7.4374 
2024 0.8880 0.6340 99,737.4744 596.2903 7.5248 
2025 0.8925 0.6450 104,481.3551 679.5016 7.6114 
2026 0.8970 0.6560 109,541.6170 772.5123 7.6972 
2027 0.9015 0.6670 114,942.2615 876.1903 7.7822 
2028 0.9060 0.6780 120,709.2173 991.4428 7.8664 
2029 0.9105 0.6890 126,869.6482 1119.2018 7.9498 
2030 0.9150 0.7000 133,454.2085 1260.4368 8.0324 
Year y1 y2 y3 y4 Y 
2011 0.7350 0.5365 55,076.3280 71.2505 6.1896 
2012 0.7625 0.5361 60,073.3923 89.8145 6.3435 
2013 0.7900 0.5358 64,521.4247 111.2023 6.4824 
2014 0.8175 0.5354 68,240.4562 135.2383 6.6063 
2015 0.8450 0.5350 71,074.7600 161.5511 6.7152 
2016 0.8500 0.5460 72,902.7542 189.5625 6.8071 
2017 0.8550 0.5570 75,026.5971 221.5652 6.8985 
2018 0.8600 0.5680 77,477.7548 257.9510 6.9896 
2019 0.8650 0.5790 80,292.7787 299.1165 7.0802 
2020 0.8700 0.5900 83,515.0145 345.4554 7.1705 
2021 0.8745 0.6010 87,195.2037 397.3510 7.2603 
2022 0.8790 0.6120 91,113.9023 455.9814 7.3493 
2023 0.8835 0.6230 95,288.4522 522.0452 7.4374 
2024 0.8880 0.6340 99,737.4744 596.2903 7.5248 
2025 0.8925 0.6450 104,481.3551 679.5016 7.6114 
2026 0.8970 0.6560 109,541.6170 772.5123 7.6972 
2027 0.9015 0.6670 114,942.2615 876.1903 7.7822 
2028 0.9060 0.6780 120,709.2173 991.4428 7.8664 
2029 0.9105 0.6890 126,869.6482 1119.2018 7.9498 
2030 0.9150 0.7000 133,454.2085 1260.4368 8.0324 

Table 5 shows that the urban expansion index would increase with time, i.e. the urban expansion would continue into the future. In addition, the urbanization level, industrial structure, living standards and environment would all increase gradually.

Prediction of urban water supply and demand

Based on the SD model, the trend of urban water supply and demand from 2012 to 2030 was simulated. The demand–supply pressure index, the ratio of the difference between the demand and supply to the total water supply, was calculated. The results are shown in Figure 5.
Figure 5

The prediction of WD in Urumqi.

Figure 5

The prediction of WD in Urumqi.

Figure 5 shows that the total water demand (TWD) of Urumqi increases with time. The proportion of agricultural water decreases, while the industrial water, living water and environmental water increase. The greatest water consumption is from agricultural water, followed by industrial water. However, industrial water exceeds agricultural water after 2027 with economic development.

Figure 6(a) indicates that the TWD is greater than the water supply. The reuse water remains at a low level. According to the 12th 5-year plan of Urumqi, no other water source could be added and the exploitation amount has reached its limits. However, one million Yuan of industrial output value approximately requires 32.7 m3 of water at present, approximately five times that of developed countries; the industrial water recycling rate is 37%, far less than the 75–80% in developed countries. Thus, water conservation is the key factor. If the current water-use efficiency in Urumqi during 2011–2030 remains constant, the water resource security would be damaged. Figure 6(b) shows that the demand–supply pressure index is positive from 2011, which means that water shortage is currently occurring. The projection of the demand–supply pressure shows a sharp increase in 2020. Then, the increasing trend continues during 2020–2030.
Figure 6

Relationship between the urban expansion index and pressure of water supply and demand from 2011 to 2030. (a) The prediction of TWD in Urumqi. (b) The prediction about pressure of water supply and demand in Urumqi.

Figure 6

Relationship between the urban expansion index and pressure of water supply and demand from 2011 to 2030. (a) The prediction of TWD in Urumqi. (b) The prediction about pressure of water supply and demand in Urumqi.

In all, there is great influence of urban expansion on the future water resource security of Urumqi. The change of water-use structure is mainly due to the urbanization process. The water resources would become increasingly scarce if the urban expansion remains unchanged in terms of population, economic growth and water-use efficiency. The regional development is currently unsustainable with serious ecological, environmental and social economic problems.

Assessment of water resource security under different scenes of urban expansion

We analyzed the impact of urban development on the urban water resources system to find a sustainable method for water resource use, according to the actual situation of Urumqi and other authors' experience (Feng et al. 2008; Cheng 2010). In this study the base year was 2006, the first target year was 2015, and long-term target years were 2020 and 2030. To find a sustainable method for water resource use, four scenarios of urban expansion were set up based on the sensitive variables.

Design of the urban expansion scenes

Scheme 1 (present development scenario): this scenario is also known as the trend-based scenario by assuming that the development policies and system structure do not change much in the forecasting period. The constant parameters are the same as in the model calibration stage, while the table-valued parameters are hypothesized with a moderate trend. The parameters are based on the analyzed data of Urumqi.

Scheme 2 (economic priority scenario): emphasize the importance of economic growth and improve the growth rate of each industry and optimize industrial structure. The economic parameters are referenced by the level of developed countries in the same period.

Scheme 3 (water-saving scenario): the aim of this scenario is to improve the efficiency of water and increase the source of water supply. The parameters of water-use efficiency are referenced by the level of developed countries in the same period.

Scheme 4 (harmonize scenario): this scenario is also known as a sustainable development scenario by emphasizing economic development and the protection of water resources at the same time. The parameters are referenced economic priority scenario and water-saving scenario.

Regarding the domestic and foreign information in these parameters, we obtained the values of 18 sensitive parameters (Table 6).

Table 6

Schemes design by changing parameters

 Scheme 1
 
Scheme 2
 
Scheme 3
 
Scheme 4
 
Variables 2015 2020 2030 2015 2020 2030 2015 2020 2030 2015 2020 2030 
RUROIA 36 38 40 36 38 40 39.6 41.8 44 39.6 41.8 44 
IWUP 35 30 25 35 30 25 31.5 27 22.5 31.5 27 22.5 
TIWUP 2.75 2.5 2.3 2.75 2.5 2.3 2.48 2.25 2.07 2.48 2.25 2.07 
AIQ 650 580 500 650 580 500 585 522 450 585 522 450 
POPI 1.2 1.2 
POSI 45.3 40 29 40.77 36 26.1 45.3 40 29 40.77 36 26.1 
POTI 53.5 59 70 58.23 63 72.9 53.5 59 70 58.23 63 72.9 
GROPIGDP 4.17 10.23 10.23 4.59 11.25 11.25 4.17 10.23 10.23 10 10 
GROSIGDP 6.28 4.25 4.25 6.91 4.68 4.68 6.28 4.25 4.25 
GROTIGDP 11.94 12.72 12.72 13.14 13.99 13.99 11.94 12.72 12.72 14 15 15 
RORWU 30 35 40 30 35 40 30 35 40 33 38.5 44 
DSDC 53.25 53.25 53.25 53.25 53.25 53.25 55 55 55 55 55 55 
IWEC 58.68 58.68 58.68 58.68 58.68 58.68 60 60 60 60 60 60 
TIWWEC 53.33 53.33 53.33 53.33 53.33 53.33 56 56 56 56 56 56 
IQ 30 30 30 30 30 30 30 30 30 33 33 33 
SFWQ 0.73 0.73 0.73 0.73 0.73 0.73 0.73 0.73 0.73 0.81 0.81 0.81 
UDWQ 156.9 156.9 156.9 172.59 172.59 172.59 141.21 141.21 141.21 141.21 141.21 141.21 
RDWQ 69.3 69.3 69.3 76.23 76.23 76.23 62.37 62.37 62.37 62.37 62.37 62.37 
 Scheme 1
 
Scheme 2
 
Scheme 3
 
Scheme 4
 
Variables 2015 2020 2030 2015 2020 2030 2015 2020 2030 2015 2020 2030 
RUROIA 36 38 40 36 38 40 39.6 41.8 44 39.6 41.8 44 
IWUP 35 30 25 35 30 25 31.5 27 22.5 31.5 27 22.5 
TIWUP 2.75 2.5 2.3 2.75 2.5 2.3 2.48 2.25 2.07 2.48 2.25 2.07 
AIQ 650 580 500 650 580 500 585 522 450 585 522 450 
POPI 1.2 1.2 
POSI 45.3 40 29 40.77 36 26.1 45.3 40 29 40.77 36 26.1 
POTI 53.5 59 70 58.23 63 72.9 53.5 59 70 58.23 63 72.9 
GROPIGDP 4.17 10.23 10.23 4.59 11.25 11.25 4.17 10.23 10.23 10 10 
GROSIGDP 6.28 4.25 4.25 6.91 4.68 4.68 6.28 4.25 4.25 
GROTIGDP 11.94 12.72 12.72 13.14 13.99 13.99 11.94 12.72 12.72 14 15 15 
RORWU 30 35 40 30 35 40 30 35 40 33 38.5 44 
DSDC 53.25 53.25 53.25 53.25 53.25 53.25 55 55 55 55 55 55 
IWEC 58.68 58.68 58.68 58.68 58.68 58.68 60 60 60 60 60 60 
TIWWEC 53.33 53.33 53.33 53.33 53.33 53.33 56 56 56 56 56 56 
IQ 30 30 30 30 30 30 30 30 30 33 33 33 
SFWQ 0.73 0.73 0.73 0.73 0.73 0.73 0.73 0.73 0.73 0.81 0.81 0.81 
UDWQ 156.9 156.9 156.9 172.59 172.59 172.59 141.21 141.21 141.21 141.21 141.21 141.21 
RDWQ 69.3 69.3 69.3 76.23 76.23 76.23 62.37 62.37 62.37 62.37 62.37 62.37 

Change of water resource security under different scenarios of urban expansion

The change of WD and demand–supply pressure index under different scenarios of urban expansion was simulated (Table 7).

Table 7

Simulation results of major variables under different scenarios of urban expansion

 Scheme 1
 
Scheme 2
 
Scheme 3
 
Scheme 4
 
Variables 2020 2030 2020 2030 2020 2030 2020 2030 
TP 377.77 596.76 377.77 596.59 378.02 597.67 378.07 597.63 
GDP 3,154.93 7,964.00 3,391.05 9,620.60 3,158.48 7,986.49 3,435.16 10,173.10 
TWD 12.05 15.92 12.19 17.08 10.88 14.44 10.91 14.59 
PIWD 5.47 4.79 5.47 4.79 4.92 4.31 4.92 4.31 
SIWD 3.79 5.77 3.66 6.28 3.98 8.18 3.34 5.97 
TIWD 0.47 1.28 0.53 1.61 0.42 1.16 0.49 1.54 
DWD 2.01 3.26 2.21 3.58 1.81 2.93 1.81 2.93 
EWD 0.32 0.83 0.32 0.82 0.32 0.83 0.36 0.92 
TWS 11.53 12.87 11.57 13.08 11.43 12.68 11.52 13.04 
RWQ 0.94 1.78 0.98 1.99 0.84 1.59 0.93 1.95 
SOSAD 0.52 3.05 0.63 4.00 −0.55 1.76 −0.61 1.54 
DSP 0.04 0.24 0.05 0.31 0.04 0.14 −0.05 0.13 
TAOS 1.75 2.67 1.82 2.99 1.57 2.59 1.49 2.49 
IS 1:40:59 1:29:70 1:36:63 1:26.1:72.9 1:40:59 1:29:70 1:36:63 1:26.1:72.9 
 Scheme 1
 
Scheme 2
 
Scheme 3
 
Scheme 4
 
Variables 2020 2030 2020 2030 2020 2030 2020 2030 
TP 377.77 596.76 377.77 596.59 378.02 597.67 378.07 597.63 
GDP 3,154.93 7,964.00 3,391.05 9,620.60 3,158.48 7,986.49 3,435.16 10,173.10 
TWD 12.05 15.92 12.19 17.08 10.88 14.44 10.91 14.59 
PIWD 5.47 4.79 5.47 4.79 4.92 4.31 4.92 4.31 
SIWD 3.79 5.77 3.66 6.28 3.98 8.18 3.34 5.97 
TIWD 0.47 1.28 0.53 1.61 0.42 1.16 0.49 1.54 
DWD 2.01 3.26 2.21 3.58 1.81 2.93 1.81 2.93 
EWD 0.32 0.83 0.32 0.82 0.32 0.83 0.36 0.92 
TWS 11.53 12.87 11.57 13.08 11.43 12.68 11.52 13.04 
RWQ 0.94 1.78 0.98 1.99 0.84 1.59 0.93 1.95 
SOSAD 0.52 3.05 0.63 4.00 −0.55 1.76 −0.61 1.54 
DSP 0.04 0.24 0.05 0.31 0.04 0.14 −0.05 0.13 
TAOS 1.75 2.67 1.82 2.99 1.57 2.59 1.49 2.49 
IS 1:40:59 1:29:70 1:36:63 1:26.1:72.9 1:40:59 1:29:70 1:36:63 1:26.1:72.9 

Table 7 shows that there is a different status with different development scenarios. Under the present development scenario, the total required water is projected to reach 15.92 × 108 m3 in 2030. The gap between supply and demand is 3.05 × 108 m3. Under the economic priority scenario, the total required water is increased further. The gap between supply and demand becomes larger than the first scenario. Under the water-saving scenario, the water-use pressure decreases to a certain extent, but the sewage (2.59 × 108 m3) is a problem. In contrast, the harmonize scenario would have a high GDP of 10,173.1 × 108 Yuan, a small difference between demand and supply of 1.54 × 108 m3 and a low sewage discharge of 2.39 × 108 m3. Thus, the harmonize scenario with minimal consumption of water resources has the highest GDP and water-use efficiency. The effect of water shortage and pollution level on the social and economic development is also lower than in the other three development scenarios. This would be the best scheme of water resource development and utilization in Urumqi city.

Selection of urban expansion scenario based on water resources security

Figure 7 shows that the demand–supply pressure swiftly increases with urban expansion. The order of demand–supply pressure is the following: economic priority scenario > present development scenario > water-saving scenario > harmonize scenario. The order of urban expansion level is the following: economic priority scenario > harmonize scenario > water-saving scenario > present development scenario. From a water resource security perspective, the harmonize scenario is the best choice for the development of Urumqi. To achieve the goal, the main parameters would have to reach a new standard, i.e. the water-use quota for living, irrigation quota, water-use quota for industry and the service industry. The ratio of industrial structure would optimize to 1:26.1:72.9 for three industries. Reducing the sewage and increasing the reuse proportion of wastewater are also very important to relieve the stress of water shortage.
Figure 7

Results of urban expansion in different schemes simulated by urban water resources SD model. (a) Simulation of demand–supply pressure in different schemes. (b) Simulation of urban expansion index (Y) in different schemes.

Figure 7

Results of urban expansion in different schemes simulated by urban water resources SD model. (a) Simulation of demand–supply pressure in different schemes. (b) Simulation of urban expansion index (Y) in different schemes.

CONCLUSIONS

Through simulating the water resource security under different development scenarios using the SD model, the result shows that the WD increases with urban expansion. The water shortage has become more and more serious in Urumqi and has begun to restrict the development of the city.

Regarding water use, the largest demand is water consumption due to agriculture, followed by industry. However, the industrial water exceeds the agricultural water after 2027 with economic development.

The harmonize scenario is the best choice for the development of Urumqi. There is great influence of urban expansion on the future water resource security in Urumqi. The demand–supply pressure swiftly increases with urban expansion. The order of demand–supply pressure is the following: economic priority scenario > present development scenario > water-saving scenario > harmonize scenario. The order of urban expansion level is the following: economic priority scenario > harmonize scenario > water-saving scenario > present development scenario.

The contradiction between supply and demand of water resources in Urumqi city is very prominent. To keep water resources safe, scenarios planning to realize the economic, living and environmental goals are essential.

ACKNOWLEDGEMENTS

The authors would like to thank the anonymous reviewers for suggesting improvements to the manuscript. This work was supported by the Natural Foundation of China (41161008, 41371419, U1203282) and the Natural Science and Technological Innovation Project of Shihezi University (2011ZRKXTD-0304). We also thank the help of Elsevier Language Editing Services.

REFERENCES

REFERENCES
Ahmad
S.
Simonovic
S. P.
2000
System dynamics modeling of reservoir operations for flood management
.
Journal of Computing in Civil Engineering
14
,
190
198
.
Al-Otaibi
A.
Abdel-Jawad
M.
2007
Water security for Kuwait
.
Desalination
214
,
299
305
.
Bogardi
J. J.
Dudgeon
D.
Lawford
R.
Flinkerbusch
E.
Meyn
A.
Pahl-Wostl
C.
Vielhauer
K.
Vörösmarty
C.
2012
Water security for a planet under pressure: interconnected challenges of a changing world call for sustainable solutions
.
Current Opinion in Environmental Sustainability
4
,
35
43
.
Chen
J.
Xia
L.
Wang
H.
2011
Research on Urban Water Security Evaluation Based on Technique for Order Preference by Similarity to Ideal Solution Model
. In:
Modeling Risk Management for Resources and Environment in China
(
Zhou
Y.
Wu
D. D.
, eds).
Springer
,
Berlin
, pp.
341
348
.
Cheng
L. I.
2010
System dynamics model of Suzhou water resources carrying capacity and its application
.
Water Science and Engineering
3
(
2
),
144
155
.
Cook
C.
Bakker
K.
2012
Water security: debating an emerging paradigm
.
Global Environmental Change
22
,
94
102
.
Deng
X.
Huang
J.
Rozelle
S.
Uchida
E.
2008
Growth, population and industrialization, and urban land expansion of China
.
Journal of Urban Economics
63
,
96
115
.
Falkenmark
M.
Widstrand
C.
1992
Population and water resources: a delicate balance
.
Population Bulletin
47
,
1
36
.
Feng
L. H.
Zhang
X. C.
Luo
G. Y.
2008
Application of system dynamics in analyzing the carrying capacity of water resources in Yiwu City, China
.
Mathematics and Computers in Simulation
79
,
269
278
.
Fletcher
E.
1998
The use of system dynamics as a decision support tool for the management of surface water resources
.
First International Conference on New Information Technologies for Decision Making in Civil Engineering
, pp.
909
920
.
Forrester
J. W.
Mass
N. J.
Ryan
C. J.
1976
The system dynamics national model: understanding socio-economic behavior and policy alternatives
.
Technological Forecasting and Social Change
9
,
51
68
.
Guo
H. C.
Liu
L.
Huang
G. H.
Fuller
G. A.
Zou
R.
Yin
Y. Y.
2001
A system dynamics approach for regional environmental planning and management: a study for the Lake Erhai Basin
.
Journal of Environmental Management
61
,
93
111
.
Han
Y.
Ruan
B.
Xie
J.
2003
Multi-objective and multilevel fuzzy optimization model and its application in water security evaluation
.
Resources Science
25
,
37
42
.
Hunter
C.
1998
Perceptions of the sustainable city and implications for fresh water resources management
.
International Journal of Environment and Pollution
10
,
84
103
.
Keyes
A. M.
Palmer
R. N.
1993
The role of object oriented simulation models in the drought preparedness studies
.
Water Management in the '90s: A Time for Innovation
.
ASCE
,
New York
, pp.
479
482
.
McDonald
R. I.
2008
Global urbanization: can ecologists identify a sustainable way forward?
Frontiers in Ecology and the Environment
6
,
99
104
.
Mohapatra
P. K.
Mandal
P.
Bora
M. C.
1994
Introduction to System Dynamics Modeling
.
University of Nevada Press
,
Reno
pp.
132
156
.
Motohashi
Y.
Nishi
S.
1991
Prediction of end-stage renal disease patient population in Japan by system dynamics model
.
International Journal of Epidemiology
20
,
1032
1036
.
Palmer
M. A.
Bemhard
E. S.
Chomesky
E. A.
Collins
S. L.
Dobson
A. P.
Duke
C. S.
Gold
B. J.
Jacobson
R.
Kingsland
S.
Kranz
R.
Mappin
M. J.
Martinez
M. L.
Micheil
F.
Morse
J. L.
Pace
M. L.
Pascual
M.
Palumbi
S.
Reichman
O. J.
Townsend
A.
Tumer
M. G.
2004
Ecological science and sustainability for a crowded planet
.
Report from the Ecological Society of America, Ecological Visions Committee, USA
, pp.
17
26
.
Rijsberman
M. A.
Van De Ven
F. H.
2000
Different approaches to assessment of design and management of sustainable urban water systems
.
Environmental Impact Assessment Review
20
,
333
345
.
Shi
Y.
Qu
Y.
1992
Carrying Capacity of Water Resources in Urumqi River Basin and its Rational Utilization
.
Science Press
,
Beijing
, pp.
128
139
.
Shiflet
A.
Shiflet
G.
2011
System Dynamics Tool: Vensim PLE Tutorial 1. Introduction to Computational Science: Modeling and Simulation for the Sciences
.
Whitney Darrow
,
Princeton
, pp.
12
43
.
Simonovic
S. P.
Fahmy
H.
1999
A new modeling approach for water resources policy analysis
.
Water Resources Research
35
,
295
304
.
Sterman
J. D.
2000
Business Dynamics: Systems Thinking and Modeling for a Complex World
.
Irwin/McGraw-Hill
,
Boston
,
19
.
Sun
C. Z.
Chi
K. X.
2008
Establishment and application of the assessment model for water resources safety in Dalian
.
Journal of Safety and Environment
1
,
33
.
Vennix
J. A.
1996
Group model building: Facilitating team learning using system dynamics
.
Wiley
,
Chichester
.
Vörösmarty
C. J.
Green
P.
Salisbury
J.
Lammers
R. B.
2000
Global water resources: vulnerability from climate change and population growth
.
Science
289
,
284
288
.
Xia
J.
Zhu
Y. Z.
2002
The measurement of water resources security: a study and challenge on water resources carrying capacity
.
Journal of Natural Resources
3
,
262
269
.