- Email: [email protected]
Application of system dynamics in analyzing the carrying capacity of water resources in Yiwu City, China
Available online at www.sciencedirect.com
Mathematics and Computers in Simulation 79 (2008) 269–278
Application of system dynamics in analyzing the carrying capacity of water resources in Yiwu City, China Li-Hua Feng ∗ , Xing-Cai Zhang, Gao-Yuan Luo Department of Geography, Zhejiang Normal University, No. 688 Yingbin Road, Jinhua 321004, China Received 31 March 2007; received in revised form 14 October 2007; accepted 28 November 2007 Available online 4 December 2007
Abstract A risk assessment model for water shortage is constructed using a risk analysis method based on the information diffusion theory. The application of this model is demonstrated in the city of Yiwu in Zhejiang Province, China. Based on the analytical results from a small sample, this study indicates that the present model is more stable and effective than the traditional model. Risk assessment results are used to analyze the carrying capacity of water resources from an ecological angle. For this study, the carrying capacity of water resources is defined as the maximum volume of water suitable for supporting human activity in certain stages of social development that can be borne by water resources under favorable ecological conditions. Further study on Yiwu is also performed, with results indicating that water shortages in this city are not related to types of water source, but can be classified in terms of water quality and conservation. To verify the results of theoretical investigation in this paper, the authors simulate changes in the carrying capacity of water resources under the conditions of future water management policies. This simulation uses the system dynamics (SD) model, based on the historical data collected by the city over the past 20 years and governmental plans to raise inhabitants’ living standards between the present and 2020. The paper simultaneously indicates that both singularly pursuing fast economic development at the expense of the environment and promoting environmental protection via reduced economic development are undesirable for Yiwu. Simultaneously giving consideration to both the economic development and environmental protection is likely to produce better overall results. However, if the present water supply level is maintained but does not increase in the near future, Yiwu’s water supply will be unable to satisfy requirements even under this scheme. In this case, the carrying capacity of water resources in the region can only be effectively improved by promoting more efficient use of water and water conservation schemes, as well as strengthening long-term investment in environmental protection. © 2007 IMACS. Published by Elsevier B.V. All rights reserved. Keywords: Water shortage; Information diffusion; Risk assessment; Carrying capacity of water resources; Guaranteed rate
1. Introduction The concept of water resource carrying capacity was first suggested by the China Xinjiang Water Resource SoftScience Research Panel. This is a new concept [20] that has yet to be clearly defined either at home or abroad. Some people consider water resource carrying capacity to be the capacity to sustain a society with a good standard of living; others consider it a threshold value; the capability of supporting the activities of human beings [15]. Internationally, very few breakthroughs have been achieved in a single project related to researching water resource carrying capacity, having been ∗
Corresponding author. Tel.: +86 579 82281008. E-mail address: [email protected] (L.-H. Feng).
0378-4754/$32.00 © 2007 IMACS. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.matcom.2007.11.018
270
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
considered briefly only in theories of sustainable development [19]. Some scholars use such terms as “sustainable water utilization” [14] and “the ecological limits of the water resource” or “the natural system limits of the water resource” [5] to express a similar meaning to water resource carrying capacity. URS Corporation performed a study on the carrying capacity of water resources in the Florida Keys valley [4] and Haris focused on water resource carrying capacities in areas of agricultural production [9]. Research performed by scholars such as Falkenmark also dealt with the limits of water resource carrying capacity [5]. Studies focusing exclusively on water resource carrying capacity have been performed mainly in China. The concept of water resource carrying capacity was first suggested by the Xinjiang Water Resource Soft-Science Research Panel in 1989, and 1990–2000 are considered to be the initial phase of water resource carrying capacity research. Water resource carrying capacity has become a topic of great debate since 2001 and, at present, represents a new academic frontier [17]. Preliminary studies on water resource carrying capacity began in areas where droughts are common. Shi and other researchers performed a quantitative study of water resource carrying capacity along the Urumchi valley [20], Xu probed into water resource carrying capacity along the Hei River valley [24], Li and other researchers identified an indicative system for water resource carrying capacity in the Chaidamu Basin [15] and Chen and other researchers performed a systematic analysis of water resource carrying schemes in the Chaidamu Basin [3]. However, a systematic and scientific theory of water resource carrying capacity itself has not yet been proposed [16]. In fact, the carrying capacity of water resources is a concept with attributes related to both nature and society. This means that it is a complex large-scale system, involving numerous factors including but not limited to population, resource availability, the environment, ecology, society, economics and technology [8]. These factors interact as both cause and effect, restrict each other and act as both positive and negative feedback. The answers to several important questions about the exact population level that can be supported by water resources, whether the sustainable development of a social economy can be successfully achieved, and whether a favorable ecological system can be smoothly realized, depend exclusively upon policy parameters such as economic policy, development speed, strategic policy, etc. [7]. The choice of policy parameters is a difficult problem, which can be effectively solved through a mathematical method derived from system dynamics (SD) [18]. Risk is the possibility of being subject to harm or losses. Similarly, water shortage risk is defined as the potential for losses incurred due to water shortages. The carrying capacity of water resources is the basic standard of measurement for water safety [23] and so is closely related to the risk of water shortages. However, the calculated result of system dynamics is usually not related to risk factor. Therefore, in this paper, the water shortage risk factor is added into equations of water resource carrying capacity based on system dynamics. It is obvious that water shortages are closely related to the sharp increases in water consumption caused by human activity, and this phenomenon first appeared at the beginning of the 1980s. Therefore, only limited data are available for use in water shortage risk assessment. This makes water shortage risk assessment an issue with very little sample data for evaluation. One method of dealing with the limited volume of sampling data available is to regard the small sample as fuzzy information, and then optimize the data using information diffusion technology [12] to produce more reliable results for risk assessment [10]. 2. Methods 2.1. Risk assessment Information diffusion is a processing method from abstract mathematics that provides a set numerical method for dealing with samples [11], transforming a single-valued sample into a set numerical sample. The simplest model is to use the normal diffusion model. If the water shortage index field can be represented as U = {u1 , u2 , . . . , um }, then the information carried by a single-valued observation sample of xi can be diffused into each term in field U according to the following equation: 1 2 2 fi (uj ) = √ e−(xi −uj ) /2h , h 2π
j = 1, 2, . . . , m
(1)
where h is the diffusion coefficient, which can be determined according to the maximum and minimum values of the samples, and the sample number is in the set [2,13]. Let Ci =
m j=1
fi (uj )
(2)
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
271
The related attachment function of the fuzzy subset can be represented as follows: μxi (uj ) =
fi (uj ) Ci
(3)
The function μxi (uj ) is called the normalized information distribution of sample xi . A good result for risk analysis can be obtained by resolving the function of μxi (uj ). Let x1 , x2 , . . . , xn be the observation values specified by n, then the function is the information quantum diffused from the sample X = {x1 , x2 , . . . , xn } to the observation point μj . This can be represented as follows: q(uj ) =
n
μxi (uj )
(4)
i=1
The above function shows that if the observation value of water shortage can only be chosen as one of the values in the series u1 , u2 , . . . , um , then the sample number with the observation value uj is q(uj ) through the information diffusion of the observation set {x1 , x2 , . . . , xn }, in regard to all values of xi , which represent the samples. The value q(uj ) is usually not a positive integer, but is never a number less than zero. Let Q=
m
q(uj )
(5)
j=1
Q is the sum of the values of each sample uj . The function is the frequency value of the sample appearing on the point uj , and its value is the estimated probability value. This can be represented as follows: p(uj ) =
q(uj ) Q
(6)
It is also obvious that probability values exceeding the value uj are as follows: p(u ≥ uj ) =
m
p(uk )
(7)
k=j
The value of p(u ≥ uj ) is the required value for risk assessment. 2.2. System dynamics Although water resources are limited, the exact amount of water resources that can be supplied by the environment is unknown at present [1]. Water resources supplied by any body of water, including rivers, lakes and groundwater, have a threshold value. If this threshold is exceeded, ecological conditions will enter a deteriorative cycle [5,6]. Therefore, the carrying capacity of water resources should be defined as the maximum volume of water resources that can be provided to support human activity in certain stages of social development in a favorable ecological system. The carrying capacity of water resources can be calculated according to the following equation, derived from system dynamics [18]: BW(K) = BW(J) + DT × BWR(JK)
(8)
where BW represents the volume of usable water that can be carried, and J, K and JK denote the preceding time, current time and adjacent time periods, respectively. DT is the simulation step length; BWR represents the rate of change in volume of water carried, which includes increased volume due to newly built water retention works and improvements in water recycling facilities. Let TBW represent the total carrying capacity of water resources (the total volume of usable water borne by a water body) in a particular region. TBW can be divided into three parts: the volume of water suitable for agricultural use that can be carried (ABW), the volume of water suitable for industrial use that can be carried (IBW) and the volume of water suitable for consumption by human beings and domestic animals (PSBW), and can be represented as follows: TBW = ABW + IBW + PSBW
(9)
272
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
From this formula it can be seen that the relative reliability of the rate of change of usable water carrying capacity (BWR) is the key factor in calculating the carrying capacity of water resources. However, BWR also contains nondeterminacy owing to the unpredictability of future water shortages [15]. To simplify analysis, the non-determinacy of water shortages is assumed that it only related to the non-determinacy of natural precipitation. Therefore, estimation of the probability distribution of annual precipitation in the studied region becomes the key to risk assessment of potential water shortages. It is also assumed that natural precipitation during the time interval in which the study is performed is a stationary Markov process. Therefore, the probability distribution over the studied time interval can be regarded as unchanged. Let x denotes annual precipitation and the probability distribution of p(x) represent the risk of water shortages. It can be estimated that the discrete expression of p(uj ) for p(x) using the normal diffusion model with a small sample of data (Eq. (6)). The expected value for this distribution is also the average risk of water shortages and can be calculated as follows: m
μ=
1 μj p(uj ) m
(10)
j=1
The rate of change of usable water carrying volume (BWR) is solely composed of μ, defined above, and other water sources. Under the assumption of the stationary state of the Markov process, μ remains unchanged during the studied time interval, and therefore variations in BWR represent only changes in other water sources [19]. Therefore, only by considering the uncertainty factors of other water sources can the risk of water shortage be fully assessed. 3. Results 3.1. Water shortage risk assessment in Yiwu [0, 2000] defined over the set of real numbers can be regarded as the field of xi according to actual annual rainfall statistics collected over the 27 years from 1980 to 2006, as measured by the Yiwu weather station, China. The continuous field [0, 2000] can be transformed into a discrete field through equidistant selection of points. Considering accuracy requirements, 101 points were selected to form the discrete field, represented as follows: U = {u1 , u2 , . . . , um } = {0, 20, 40, . . . , 2000} Risk assessment of water shortages in Yiwu can be obtained using Eqs. (1)–(7), as shown in Table 1. When performing calculations, years were selected as the unit of measurement. Line 1400 in the table shows that the probability of conditions in which precipitation is greater than 1400 mm per year occurring in Yiwu is p = 0.6615. Thus, droughts occur in Yiwu at an average of once in every 3 years (recurrence interval is equal to 1/(1 − p)). The volume of rainfall each year is an important influence on the utilization of water resources in Yiwu. Precipitation in the driest year over the next 5 years is calculated to be 1213 mm, and precipitation in the driest year over the next 10 years to be 1076 mm. Both of these statistics are calculated through interpolation using the risk analysis model, according to the actual information collected over the 27 years from 1980 to 2006, as measured by the Yiwu weather station. Table 1 Risk assessment for water shortage in Yiwu Annual rainfall (mm)
Surpass probability
Annual rainfall (mm)
Surpass probability
800 900 1000 1100 1200 1300 1400
1 1 0.9922 0.9299 0.8749 0.7831 0.6615
1500 1600 1700 1800 1900 2000 2100
0.5514 0.3861 0.2322 0.1189 0.0351 0.0009 0
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
273
3.2. Analysis of the carrying capacity of water resources in Yiwu Yiwu is located in a humid area and has an average annual precipitation of 1403 mm. In contrast to cities located in dry areas [22], water shortages in Yiwu are related not to the type of water source but to water quality and types of water conservation projects. This includes water shortages caused by pollution, and the lack of water storage projects. Therefore, protection of water sources and water storage projects such as large or middle scale reservoirs are very important factors [21]. Several indicators that have a large influence on the carrying capacity of water resources in Yiwu were selected according to the principle of simplicity. These indexes include population, industrial and agricultural output, gross domestic product (GDP), investment in environmental protection, volume of sewage discharge, lengths of polluted stretches of rivers, total required volume of usable water, water conservation schemes, water supply volume, and the volume of usable water carried by water resources. The indicator system used in this study on the carrying capacity of water resources in Yiwu can be divided into five subsystems—population, agriculture, industry, environmental protection and water resources. As can be calculated using system dynamics, there exists a mutual relationship of cause and effect between one subsystem and another, as well as between the different factors of each subsystem. Moreover, this relationship forms a closed feedback structure. Therefore, cause and effect diagrams and system flow diagrams can be created according to the indicator system and the feedback structure of the system. Such diagrams explain the logical relationships between each variable in the system, but fail to demonstrate the quantitative relation between variables. Thus, the language DYNAMO is required to build system dynamics equations. Variables used in system dynamics include level variables, rate variables and auxiliary variables and their equivalent equations, level equations, rate equations and the auxiliary equations, respectively. System dynamics equations are organically grouped by the above dynamic equations to fully reflect dynamic variation in the carrying capacity of water resources. The most important of these equations is the level equation, as it quantitatively describes changes in dynamic systematic variables over time. In this paper, field investigation work was conducted along rivers in Yiwu. Data related to water resources and social economic systems since 1980 were comprehensively collected by the author according to the actual situation in Yiwu and the requirements of the SD model, with ecology as the guiding ideology. System dynamics equations for the five subsystems of population, agriculture, industry, environmental protection and water resources were built according to the characteristics of water resources in Yiwu. More than 100 variables and parameters were included in the model, which also includes nine level equations, nine rate equations and numerous auxiliary equations. The model was created using Vensim. The testing time for the historical review is 27 years (1980–2006) and the terminal time for the simulation is the year 2020, with a step length of 1 year. The structure of the model was proven to be reasonable through analysis of parameter error and sensitivity, and reflects the actual characteristics of the carrying capacity of water resources in Yiwu. Therefore, it can be used to accurately forecast the dynamic development of the system after future policy parameters have been implemented. The total volume of water resources in Yiwu is 6.03 × 108 m3 . In order to model the development of water resource carrying capacity and economics in the region during the next 10–15 years, several indicators were selected as policy parameters, according to historical data and standards for building a society with improved standards of living. These indicators include agriculture, industry, GDP, accelerated investment in environmental protection, irrigation quotas for agriculture, water consumption per unit of output with the value of the 1000 RMB by industry, and volume of sewage treated. Furthermore, three development schemes were further simulated using the SD model. The three schemes can be represented as a primary scheme with economic development as the main goal, a secondary scheme with environmental protection as the main goal and a third scheme balancing economic development and environmental protection. Detailed analysis of water resource carrying capacity under these three schemes is shown below. 3.3. Scheme emphasizing economic development (high scheme) The objective of quadrupling Yiwu’s GDP can be easily realized by 2009 if this scheme is selected. GDP would reach 125.4 billion RMB by 2020 (as shown in Table 2 and Fig. 1). However, unilateral pursuit of fast economic development would lead to increases in sewage discharge and decreased investment in environmental protection. Sewage discharge would increase to 67.23 million tonnes by 2020 and result in all sections of all rivers being rated below class IV PPR (polluted percentage of river length). Sewage would be clearly visible everywhere, and cause severe environmental damage, as shown in Fig. 2. At the same time, the total value of industrial output would reach 247.8 billion RMB owing
274
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
Table 2 Variation of the main indexes in the three schemes for carrying capacity of water resources in Yiwu 1
2
3
4
5
6
7
8
9
10
High scheme
2005 2010 2015 2020 2005 2010 2015 2020 2005 2010 2015 2020
13.82 22.26 35.85 57.73 13.82 19.84 28.48 40.89 13.82 17.64 22.51 28.73
593.20 955.36 1538.61 2477.95 593.20 851.62 1222.61 1755.21 593.20 757.09 966.26 1233.22
300.30 483.64 778.91 1254.44 300.30 431.12 618.93 888.56 300.30 383.27 489.16 624.31
9.86 12.59 16.07 20.50 9.86 14.16 20.33 29.18 9.86 15.88 25.58 41.20
3208 4133 4684 6723 3208 3100 1262 100 3208 2104 0 0
50.1 64.5 73.0 104.7 50.1 48.3 19.5 1.1 50.1 32.7 0 0
2.36 2.60 2.88 3.57 2.36 2.51 2.63 2.99 2.36 2.43 2.42 2.58
2.42 2.61 2.76 2.94 2.42 2.66 2.86 3.09 2.42 2.71 2.96 3.24
Middle scheme
Low scheme
1, scheme; 2, year; 3, gross agricultural output value (108 yuan); 4, gross industrial output value (108 yuan); 5, gross domestic product (108 yuan); 6, investment of environmental protection (108 yuan); 7, sewage discharge (104 tonnes); 8, PPR (%); 9, total water demand (108 m3 ); 10, bearing volume of water use (108 m3 ).
Fig. 1. Simulation curve of gross domestic product (GDP).
Fig. 2. Simulation curve of polluted percentage of river (PPR) length.
to rapid industrial development. A series of measures including adjustments to industrial structure, separately supplying water of different qualities for different uses and increasing the price of water were executed to decrease industrial water consumption for each unit of output valued at 10,000 RMB from 9 to 7 m3 could be implemented. However, industrial water requirements would increase to 1.98 × 108 m3 , exceeding water consumption by agriculture. This increases total water requirements to 3.57 × 108 m3 . Due to lack of investment, implementation of water conservation and recycling measures would also be limited and the ability to supply water to the city would be reduced dramatically. Fig. 3 shows that the simulation curves would intersect in 2014, indicating that the total volume of usable water carried
Fig. 3. Simulation curves of total requirement for water supply (TWD) and total carrying volume of water use (TBW).
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
275
would already be insufficient to satisfy total water requirements. Supply of water resources would be unable to meet demand at this time. The total volume of usable water carried would be 2.94 × 108 m3 by 2020, with a massive water shortage of 0.63 × 108 m3 . 3.4. Scheme emphasizing environmental protection (low scheme) The primary goal of this scheme is environmental protection. Therefore, investment in environmental protection increases year on year and causes sewage discharge to decrease gradually. Sewage treatment rates would reach 100% by 2013 and the percentage of river length rated below class IV PPR would also gradually decrease, from 50.1% in 2005 to 32.7% in 2010 and would finally reach 0% in 2013. Lengths of river with inferior water quality would disappear owing to the city’s emphasis on environmental protection. Picturesque scenery with green hills and clear waters would appear around the city and its ecologic environment would be changed completely, as shown in Fig. 2. Due to the slowing of industrial development, industrial water requirements would be merely 0.99 × 108 m3 in 2020, which is 0.99 × 108 m3 less than that of the primary scheme. Therefore, total water requirements would be only 2.58 × 108 m3 , far below the maximum usable water carrying capacity of water resources, as shown in Fig. 3. Although this scheme has an outstanding effect on balancing the demand for water against supply, it will also result in reduced resource exploitation and restricted development of industry and agriculture. This is owing to decreased investment in industrial fixed assets and agriculture. Therefore, under this scheme the objective of quadrupling GDP will only be achieved in 2013, as shown in Table 2 and Fig. 1. 3.5. Scheme balancing economic development and environmental protection (middle scheme) It is obvious that pollution is unavoidable during economic development. However, the economy cannot be developed blindly at the expense of the environment. In particular, the consumption of resources cannot exceed the regenerative ability of ecological systems. Concurrent economic development and environmental protection should be the final objective in Yiwu. The relationship between economic development and other factors, such as population, resource availability and the environment are carefully considered in the middle scheme. Resource availability, the environment, the needs of industry and market conditions must all be taken into account through the weighing of advantages and disadvantages from an ecological angle. Under this scheme, the objective of quadrupling the city’s GDP by the year 2011 could easily be realized if several measures are properly executed. These include inviting outside investment, accelerating construction of the International Trade City and timely adjustment of industrial structure, as shown in Table 2 and Fig. 1. At the same time, sewage discharge would also decrease and the percentage of river length polluted would be reduced to 1.1% by 2020, creating a comfortable living environment (Fig. 2). The total volume of usable water borne could potentially reach 3.09 × 108 m3 by the same time, which can fully satisfy the total annual water requirement of 2.99 × 108 m3 . Equilibrium between supply and demand for water resources would be achieved, as shown in Fig. 3. 4. Discussion 4.1. Economic growth and water resources in Yiwu Yiwu is located in the center of Zhejiang Province, China. The total area of the city is 1109 km2 and the total population is 6.974 million. Yiwu has a mid-subtropical monsoon climate; it is hot and receives a lot of rain. In recent years, the city’s government has decided upon the strategic objective of becoming a key city in central-western Zhejiang Province. The three strategic policies of promoting industrialization, accelerating urbanization and advancing integration of urban and rural areas have been strongly emphasized. Distinct improvements in economic growth have already been achieved by the city. GDP reached 35.200 billion RMB in 2006. GDP per capita also increased to 6300 US dollars. Yiwu city was ranked number 12 by the National Statistics Bureau in a list of the 100 top counties in terms of national comprehensive strength in 2005. Unfortunately, precipitation distribution in Yiwu is not homogeneous in space and time. Moreover, there also exists the phenomenon of river pollution, and water storage projects in the city are very limited. This situation is not ideal, and makes water shortages a likely possibility. This possibility was realized in the summer and autumn of 2003 when
276
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
Table 3 Irrigation quotas for agriculture and water consumption amount per unit output value of 10,000 yuan for industry in different years 1
2
3
2007–2010 2011–2020
32–30 30–28
20–9 8–7
1, year; 2, irrigation quotas for agriculture (m3 /hm2 ); 3, water consumption amount per unit output value of 10,000 yuan for industry (m3 /104 yuan).
the area was hit by record high temperatures and drought. Precipitation from July to October was just 124 mm, only 30% of average precipitation at that time of year. Demand for water was high, but the affluent residents of Yiwu could only sit and watch as polluted water from the river flow past their doors. The Yiwu county government provided funds of about 200 million RMB in November 2000 for purchasing fresh water resources of 5 × 107 m3 annually from a reservoir in Dongyang county on a permanent basis. This became the first instance of trading water rights after the Ministry of Water Conservancy set forth the theory of water rights and the water market. Even with this policy, water usage in Yiwu still requires restrictions in terms of time-sharing, sectioning and water pressure-reducing. It is obvious that the problem of water shortage has become a bottleneck for comprehensively building a comfortable society by 2020. Therefore, it is of great importance to sustainable economic development in Yiwu to make an assessment of the risk of water shortage and water resource carrying capacities. 4.2. Suggestions for future policy parameters The primary scheme of unilaterally pursuing fast economic development at the expense of the environment and the secondary scheme with environmental protection as the main goal, achieved via slowing of the economy, are both undesirable for Yiwu. The third scheme balancing both economic development and environmental protection should be the chosen scheme. To realize the mutual benefits of economic development and environmental protection, a development speed of 7.5% annually in terms of agriculture, industry and GDP is suggested. 7.5% per year is also suggested as the rate at which environmental protection should be accelerated. Irrigation quotas for agriculture and water consumption per unit of industrial output valued a 10,000 RMB in different years are shown in Table 3. Transcendental probability can be described not only as a recurrence interval but also as a guaranteed rate, which implies the ability to reach a certain required percentage. If the level of water supply from 2000 to 2006 is maintained in the near future, then the probability of satisfying total estimated water requirements of 2.51 × 108 m3 in 2010 would be just 0.5053, according to Eqs. (1)–(7) of the water shortage risk assessment. This is still true when the third scheme, encompassing the mutual benefits of economic development and environmental protection is implemented, which means that on average water supply will only be sufficient to meet demand once in every 2 years. Furthermore, the probability of satisfying the estimated demand for water of 2.99 × 108 m3 by 2020 would be almost zero. Therefore, three powerful measures must be adopted in order to attain equilibrium of supply and demand for water resources in Yiwu. The three measures are (1) increased water storage: the amount of reservoir storage for newly built retaining works should reach 0.03 × 108 m3 per year by 2010 and 0.05 × 108 m3 per year by 2016; (2) reduction of water usage: the amount of water resources saved each year should be 0.8% of the total carrying capacity of usable water, accounted for by changing industrial procedures requiring water use, reducing water usage, separating supplies of different water qualities, and increasing the price of water; and (3) improvements in water recycling: the amount of water that can be recycled and used again should increase 0.04 × 108 m3 each year, facilitated by the use of measures such as advanced technology and the recycling of treated sewage. 5. Conclusions Several conclusions can be drawn from the above simulation results and theoretical analysis. These are as follows: (1) Information diffusion technology is an effective method of water shortage risk assessment. (2) The carrying capacity of water resources is non-deterministic and therefore must be calculated on the basis of water shortage risk assessment.
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
277
(3) Precipitation in future drought years in each 5-year cycle will be just 1213 mm and 1076 mm in each 10-year cycle. (4) The primary scheme of unilaterally pursuing fast economic development at the expense of the environment and the secondary scheme focusing on environmental protection as the primary goal by slowing economic development are both undesirable for Yiwu. The third scheme balancing economic development with environmental protection is the most effective scheme. (5) If current water supply levels are maintained in the near future, then water requirements cannot be easily satisfied according to the water shortage risk assessment, even when the third scheme is selected. Water supply will be insufficient to meet demand. (6) At present, water shortages in Yiwu are not due to types of water source, but are caused by low water quality and a lack of water conservation projects. The carrying capacity of water resources of the region can only be improved through the long-term strengthening of investment in environmental protection and the undertaking of new water storage projects. These projects include the building of large or medium scale reservoirs, the adjustment of industrial procedures and the creation of a society that naturally saves water. Although economic development and water resources may be different from each other in many areas, this contradiction will exist forever. Only when the risk of water shortage and water resource carrying capacity are assessed in a reasonable way, is it possible for this contraction to be resolved in a beneficial way, resulting in harmony between nature and the human race. Acknowledgments We are very grateful to the referees and the editors for their helpful suggestions. This work was supported by National Natural Science Foundation of China (No. 40771044) and Zhejiang Provincial Science and Technology Foundation of China (No. 2006C23066). References [1] M. Beuhler, Potential impacts of global warming on water resources in southern California, Water Sci. Technol. 47 (2003) 165–168. [2] E.A. Chatman, Diffusion theory: a review and test of a conceptual model in information diffusion, J. Am. Soc. Inf. Sci. 37 (1986) 377– 386. [3] B. Chen, L.J. Li, H.C. Guo, System analysis on water resources supporting alternatives for Chaidamu basin, Environ. Sci. 21 (2000) 16– 21. [4] A.L. Clarke, Assessing the carrying capacity of the Florida Keys, Popul. Environ. 23 (2002) 405–418. [5] M. Falkenmark, J. Lundqvist, Towards water security: political determination and human adaptation crucial, Nat. Resour. Forum 21 (1998) 37–51. [6] F. Ghassemi, A. Close, J.R. Kellett, Numerical models for the management of land and water resources salinisation, Math. Comput. Simul. 43 (1997) 323–329. [7] J.K. Gilmour, R.A. Letcher, A.J. Jakeman, Analysis of an integrated model for assessing land and water policy options, Math. Comput. Simul. 69 (2005) 57–77. [8] C. Giupponi, J. Mysiak, A. Fassio, V. Cogan, MULINO-DSS: a computer tool for sustainable use of water resources at the catchment scale, Math. Comput. Simul. 64 (2004) 13–24. [9] J.M. Harris, S. Kennedy, Carrying capacity in agriculture: globe and regional issue, Ecol. Econ. 29 (1999) 443–461. [10] C.F. Huang, An application of calculated fuzzy risk, Inf. Sci. 142 (2002) 37–56. [11] C.F. Huang, Demonstration of benefit of information distribution for probability estimation, Signal Process. 80 (2000) 1037–1048. [12] C.F. Huang, Information matrix and application, Int. J. Gen. Syst. 30 (2001) 603–622. [13] C.F. Huang, Principle of information diffusion, Fuzzy Sets Syst. 91 (1997) 69–90. [14] C. Hunter, Perception of the sustainable city and implications for fresh water resources management, Int. J. Environ. Pollut. 10 (1998) 84–103. [15] L.J. Li, H.C. Guo, B. Chen, Water resources supporting capacity of Chaidamu Basin, Environ. Sci. 21 (2000) 20–23. [16] S.F. Liu, J.H. Chen, Water resources carrying capacity based on the theory of ANN, Resour. Sci. 29 (2007) 99–105. [17] T.R. Long, W.C. Jiang, Q. He, Water resources carrying capacity: new perspectives based on eco-economic analysis and sustainable development, J. Hydraulic Eng. 35 (2004) 38–45. [18] Y. Motohashi, S. Nishi, Prediction of end-stage renal disease patient population in Japan by system dynamics model, Int. J. Epidemiol. 20 (1991) 1032–1036. [19] I.E. Ofoezie, Human health and sustainable water resources development in Nigeria: schistosomiasis inartificial lakes, Nat. Resour. Forum 26 (2002) 150–160.
278
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
[20] Y.F. Shi, Y.G. Qu, Carrying Capacity of Water Resources in Urumqi River Basin and its Rational Utilization, Science Press, Beijing, 1992, pp. 128–139. [21] C.H. Toledo, Water resources planning and management in Mexico, Int. J. Hydropower Dams 13 (2006) 53–55. [22] D. Verschuren, K.R. Laird, B.F. Cumming, Rainfall and drought in equatorial east Africa during the past 1100 years, Nature 403 (2000) 410–414. [23] J. Xia, Y.Z. Zhu, The measurement of water resources security: a study and challenge on water resources carrying capacity, J. Nat. Resour. 17 (2002) 263–269. [24] Z.M. Xu, A scenario-based framework for multicriteria decision analysis in water carrying capacity, J. Glaciol. Geocryol. 21 (1999) 100– 106.
Mathematics and Computers in Simulation 79 (2008) 269–278
Application of system dynamics in analyzing the carrying capacity of water resources in Yiwu City, China Li-Hua Feng ∗ , Xing-Cai Zhang, Gao-Yuan Luo Department of Geography, Zhejiang Normal University, No. 688 Yingbin Road, Jinhua 321004, China Received 31 March 2007; received in revised form 14 October 2007; accepted 28 November 2007 Available online 4 December 2007
Abstract A risk assessment model for water shortage is constructed using a risk analysis method based on the information diffusion theory. The application of this model is demonstrated in the city of Yiwu in Zhejiang Province, China. Based on the analytical results from a small sample, this study indicates that the present model is more stable and effective than the traditional model. Risk assessment results are used to analyze the carrying capacity of water resources from an ecological angle. For this study, the carrying capacity of water resources is defined as the maximum volume of water suitable for supporting human activity in certain stages of social development that can be borne by water resources under favorable ecological conditions. Further study on Yiwu is also performed, with results indicating that water shortages in this city are not related to types of water source, but can be classified in terms of water quality and conservation. To verify the results of theoretical investigation in this paper, the authors simulate changes in the carrying capacity of water resources under the conditions of future water management policies. This simulation uses the system dynamics (SD) model, based on the historical data collected by the city over the past 20 years and governmental plans to raise inhabitants’ living standards between the present and 2020. The paper simultaneously indicates that both singularly pursuing fast economic development at the expense of the environment and promoting environmental protection via reduced economic development are undesirable for Yiwu. Simultaneously giving consideration to both the economic development and environmental protection is likely to produce better overall results. However, if the present water supply level is maintained but does not increase in the near future, Yiwu’s water supply will be unable to satisfy requirements even under this scheme. In this case, the carrying capacity of water resources in the region can only be effectively improved by promoting more efficient use of water and water conservation schemes, as well as strengthening long-term investment in environmental protection. © 2007 IMACS. Published by Elsevier B.V. All rights reserved. Keywords: Water shortage; Information diffusion; Risk assessment; Carrying capacity of water resources; Guaranteed rate
1. Introduction The concept of water resource carrying capacity was first suggested by the China Xinjiang Water Resource SoftScience Research Panel. This is a new concept [20] that has yet to be clearly defined either at home or abroad. Some people consider water resource carrying capacity to be the capacity to sustain a society with a good standard of living; others consider it a threshold value; the capability of supporting the activities of human beings [15]. Internationally, very few breakthroughs have been achieved in a single project related to researching water resource carrying capacity, having been ∗
Corresponding author. Tel.: +86 579 82281008. E-mail address: [email protected] (L.-H. Feng).
0378-4754/$32.00 © 2007 IMACS. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.matcom.2007.11.018
270
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
considered briefly only in theories of sustainable development [19]. Some scholars use such terms as “sustainable water utilization” [14] and “the ecological limits of the water resource” or “the natural system limits of the water resource” [5] to express a similar meaning to water resource carrying capacity. URS Corporation performed a study on the carrying capacity of water resources in the Florida Keys valley [4] and Haris focused on water resource carrying capacities in areas of agricultural production [9]. Research performed by scholars such as Falkenmark also dealt with the limits of water resource carrying capacity [5]. Studies focusing exclusively on water resource carrying capacity have been performed mainly in China. The concept of water resource carrying capacity was first suggested by the Xinjiang Water Resource Soft-Science Research Panel in 1989, and 1990–2000 are considered to be the initial phase of water resource carrying capacity research. Water resource carrying capacity has become a topic of great debate since 2001 and, at present, represents a new academic frontier [17]. Preliminary studies on water resource carrying capacity began in areas where droughts are common. Shi and other researchers performed a quantitative study of water resource carrying capacity along the Urumchi valley [20], Xu probed into water resource carrying capacity along the Hei River valley [24], Li and other researchers identified an indicative system for water resource carrying capacity in the Chaidamu Basin [15] and Chen and other researchers performed a systematic analysis of water resource carrying schemes in the Chaidamu Basin [3]. However, a systematic and scientific theory of water resource carrying capacity itself has not yet been proposed [16]. In fact, the carrying capacity of water resources is a concept with attributes related to both nature and society. This means that it is a complex large-scale system, involving numerous factors including but not limited to population, resource availability, the environment, ecology, society, economics and technology [8]. These factors interact as both cause and effect, restrict each other and act as both positive and negative feedback. The answers to several important questions about the exact population level that can be supported by water resources, whether the sustainable development of a social economy can be successfully achieved, and whether a favorable ecological system can be smoothly realized, depend exclusively upon policy parameters such as economic policy, development speed, strategic policy, etc. [7]. The choice of policy parameters is a difficult problem, which can be effectively solved through a mathematical method derived from system dynamics (SD) [18]. Risk is the possibility of being subject to harm or losses. Similarly, water shortage risk is defined as the potential for losses incurred due to water shortages. The carrying capacity of water resources is the basic standard of measurement for water safety [23] and so is closely related to the risk of water shortages. However, the calculated result of system dynamics is usually not related to risk factor. Therefore, in this paper, the water shortage risk factor is added into equations of water resource carrying capacity based on system dynamics. It is obvious that water shortages are closely related to the sharp increases in water consumption caused by human activity, and this phenomenon first appeared at the beginning of the 1980s. Therefore, only limited data are available for use in water shortage risk assessment. This makes water shortage risk assessment an issue with very little sample data for evaluation. One method of dealing with the limited volume of sampling data available is to regard the small sample as fuzzy information, and then optimize the data using information diffusion technology [12] to produce more reliable results for risk assessment [10]. 2. Methods 2.1. Risk assessment Information diffusion is a processing method from abstract mathematics that provides a set numerical method for dealing with samples [11], transforming a single-valued sample into a set numerical sample. The simplest model is to use the normal diffusion model. If the water shortage index field can be represented as U = {u1 , u2 , . . . , um }, then the information carried by a single-valued observation sample of xi can be diffused into each term in field U according to the following equation: 1 2 2 fi (uj ) = √ e−(xi −uj ) /2h , h 2π
j = 1, 2, . . . , m
(1)
where h is the diffusion coefficient, which can be determined according to the maximum and minimum values of the samples, and the sample number is in the set [2,13]. Let Ci =
m j=1
fi (uj )
(2)
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
271
The related attachment function of the fuzzy subset can be represented as follows: μxi (uj ) =
fi (uj ) Ci
(3)
The function μxi (uj ) is called the normalized information distribution of sample xi . A good result for risk analysis can be obtained by resolving the function of μxi (uj ). Let x1 , x2 , . . . , xn be the observation values specified by n, then the function is the information quantum diffused from the sample X = {x1 , x2 , . . . , xn } to the observation point μj . This can be represented as follows: q(uj ) =
n
μxi (uj )
(4)
i=1
The above function shows that if the observation value of water shortage can only be chosen as one of the values in the series u1 , u2 , . . . , um , then the sample number with the observation value uj is q(uj ) through the information diffusion of the observation set {x1 , x2 , . . . , xn }, in regard to all values of xi , which represent the samples. The value q(uj ) is usually not a positive integer, but is never a number less than zero. Let Q=
m
q(uj )
(5)
j=1
Q is the sum of the values of each sample uj . The function is the frequency value of the sample appearing on the point uj , and its value is the estimated probability value. This can be represented as follows: p(uj ) =
q(uj ) Q
(6)
It is also obvious that probability values exceeding the value uj are as follows: p(u ≥ uj ) =
m
p(uk )
(7)
k=j
The value of p(u ≥ uj ) is the required value for risk assessment. 2.2. System dynamics Although water resources are limited, the exact amount of water resources that can be supplied by the environment is unknown at present [1]. Water resources supplied by any body of water, including rivers, lakes and groundwater, have a threshold value. If this threshold is exceeded, ecological conditions will enter a deteriorative cycle [5,6]. Therefore, the carrying capacity of water resources should be defined as the maximum volume of water resources that can be provided to support human activity in certain stages of social development in a favorable ecological system. The carrying capacity of water resources can be calculated according to the following equation, derived from system dynamics [18]: BW(K) = BW(J) + DT × BWR(JK)
(8)
where BW represents the volume of usable water that can be carried, and J, K and JK denote the preceding time, current time and adjacent time periods, respectively. DT is the simulation step length; BWR represents the rate of change in volume of water carried, which includes increased volume due to newly built water retention works and improvements in water recycling facilities. Let TBW represent the total carrying capacity of water resources (the total volume of usable water borne by a water body) in a particular region. TBW can be divided into three parts: the volume of water suitable for agricultural use that can be carried (ABW), the volume of water suitable for industrial use that can be carried (IBW) and the volume of water suitable for consumption by human beings and domestic animals (PSBW), and can be represented as follows: TBW = ABW + IBW + PSBW
(9)
272
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
From this formula it can be seen that the relative reliability of the rate of change of usable water carrying capacity (BWR) is the key factor in calculating the carrying capacity of water resources. However, BWR also contains nondeterminacy owing to the unpredictability of future water shortages [15]. To simplify analysis, the non-determinacy of water shortages is assumed that it only related to the non-determinacy of natural precipitation. Therefore, estimation of the probability distribution of annual precipitation in the studied region becomes the key to risk assessment of potential water shortages. It is also assumed that natural precipitation during the time interval in which the study is performed is a stationary Markov process. Therefore, the probability distribution over the studied time interval can be regarded as unchanged. Let x denotes annual precipitation and the probability distribution of p(x) represent the risk of water shortages. It can be estimated that the discrete expression of p(uj ) for p(x) using the normal diffusion model with a small sample of data (Eq. (6)). The expected value for this distribution is also the average risk of water shortages and can be calculated as follows: m
μ=
1 μj p(uj ) m
(10)
j=1
The rate of change of usable water carrying volume (BWR) is solely composed of μ, defined above, and other water sources. Under the assumption of the stationary state of the Markov process, μ remains unchanged during the studied time interval, and therefore variations in BWR represent only changes in other water sources [19]. Therefore, only by considering the uncertainty factors of other water sources can the risk of water shortage be fully assessed. 3. Results 3.1. Water shortage risk assessment in Yiwu [0, 2000] defined over the set of real numbers can be regarded as the field of xi according to actual annual rainfall statistics collected over the 27 years from 1980 to 2006, as measured by the Yiwu weather station, China. The continuous field [0, 2000] can be transformed into a discrete field through equidistant selection of points. Considering accuracy requirements, 101 points were selected to form the discrete field, represented as follows: U = {u1 , u2 , . . . , um } = {0, 20, 40, . . . , 2000} Risk assessment of water shortages in Yiwu can be obtained using Eqs. (1)–(7), as shown in Table 1. When performing calculations, years were selected as the unit of measurement. Line 1400 in the table shows that the probability of conditions in which precipitation is greater than 1400 mm per year occurring in Yiwu is p = 0.6615. Thus, droughts occur in Yiwu at an average of once in every 3 years (recurrence interval is equal to 1/(1 − p)). The volume of rainfall each year is an important influence on the utilization of water resources in Yiwu. Precipitation in the driest year over the next 5 years is calculated to be 1213 mm, and precipitation in the driest year over the next 10 years to be 1076 mm. Both of these statistics are calculated through interpolation using the risk analysis model, according to the actual information collected over the 27 years from 1980 to 2006, as measured by the Yiwu weather station. Table 1 Risk assessment for water shortage in Yiwu Annual rainfall (mm)
Surpass probability
Annual rainfall (mm)
Surpass probability
800 900 1000 1100 1200 1300 1400
1 1 0.9922 0.9299 0.8749 0.7831 0.6615
1500 1600 1700 1800 1900 2000 2100
0.5514 0.3861 0.2322 0.1189 0.0351 0.0009 0
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
273
3.2. Analysis of the carrying capacity of water resources in Yiwu Yiwu is located in a humid area and has an average annual precipitation of 1403 mm. In contrast to cities located in dry areas [22], water shortages in Yiwu are related not to the type of water source but to water quality and types of water conservation projects. This includes water shortages caused by pollution, and the lack of water storage projects. Therefore, protection of water sources and water storage projects such as large or middle scale reservoirs are very important factors [21]. Several indicators that have a large influence on the carrying capacity of water resources in Yiwu were selected according to the principle of simplicity. These indexes include population, industrial and agricultural output, gross domestic product (GDP), investment in environmental protection, volume of sewage discharge, lengths of polluted stretches of rivers, total required volume of usable water, water conservation schemes, water supply volume, and the volume of usable water carried by water resources. The indicator system used in this study on the carrying capacity of water resources in Yiwu can be divided into five subsystems—population, agriculture, industry, environmental protection and water resources. As can be calculated using system dynamics, there exists a mutual relationship of cause and effect between one subsystem and another, as well as between the different factors of each subsystem. Moreover, this relationship forms a closed feedback structure. Therefore, cause and effect diagrams and system flow diagrams can be created according to the indicator system and the feedback structure of the system. Such diagrams explain the logical relationships between each variable in the system, but fail to demonstrate the quantitative relation between variables. Thus, the language DYNAMO is required to build system dynamics equations. Variables used in system dynamics include level variables, rate variables and auxiliary variables and their equivalent equations, level equations, rate equations and the auxiliary equations, respectively. System dynamics equations are organically grouped by the above dynamic equations to fully reflect dynamic variation in the carrying capacity of water resources. The most important of these equations is the level equation, as it quantitatively describes changes in dynamic systematic variables over time. In this paper, field investigation work was conducted along rivers in Yiwu. Data related to water resources and social economic systems since 1980 were comprehensively collected by the author according to the actual situation in Yiwu and the requirements of the SD model, with ecology as the guiding ideology. System dynamics equations for the five subsystems of population, agriculture, industry, environmental protection and water resources were built according to the characteristics of water resources in Yiwu. More than 100 variables and parameters were included in the model, which also includes nine level equations, nine rate equations and numerous auxiliary equations. The model was created using Vensim. The testing time for the historical review is 27 years (1980–2006) and the terminal time for the simulation is the year 2020, with a step length of 1 year. The structure of the model was proven to be reasonable through analysis of parameter error and sensitivity, and reflects the actual characteristics of the carrying capacity of water resources in Yiwu. Therefore, it can be used to accurately forecast the dynamic development of the system after future policy parameters have been implemented. The total volume of water resources in Yiwu is 6.03 × 108 m3 . In order to model the development of water resource carrying capacity and economics in the region during the next 10–15 years, several indicators were selected as policy parameters, according to historical data and standards for building a society with improved standards of living. These indicators include agriculture, industry, GDP, accelerated investment in environmental protection, irrigation quotas for agriculture, water consumption per unit of output with the value of the 1000 RMB by industry, and volume of sewage treated. Furthermore, three development schemes were further simulated using the SD model. The three schemes can be represented as a primary scheme with economic development as the main goal, a secondary scheme with environmental protection as the main goal and a third scheme balancing economic development and environmental protection. Detailed analysis of water resource carrying capacity under these three schemes is shown below. 3.3. Scheme emphasizing economic development (high scheme) The objective of quadrupling Yiwu’s GDP can be easily realized by 2009 if this scheme is selected. GDP would reach 125.4 billion RMB by 2020 (as shown in Table 2 and Fig. 1). However, unilateral pursuit of fast economic development would lead to increases in sewage discharge and decreased investment in environmental protection. Sewage discharge would increase to 67.23 million tonnes by 2020 and result in all sections of all rivers being rated below class IV PPR (polluted percentage of river length). Sewage would be clearly visible everywhere, and cause severe environmental damage, as shown in Fig. 2. At the same time, the total value of industrial output would reach 247.8 billion RMB owing
274
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
Table 2 Variation of the main indexes in the three schemes for carrying capacity of water resources in Yiwu 1
2
3
4
5
6
7
8
9
10
High scheme
2005 2010 2015 2020 2005 2010 2015 2020 2005 2010 2015 2020
13.82 22.26 35.85 57.73 13.82 19.84 28.48 40.89 13.82 17.64 22.51 28.73
593.20 955.36 1538.61 2477.95 593.20 851.62 1222.61 1755.21 593.20 757.09 966.26 1233.22
300.30 483.64 778.91 1254.44 300.30 431.12 618.93 888.56 300.30 383.27 489.16 624.31
9.86 12.59 16.07 20.50 9.86 14.16 20.33 29.18 9.86 15.88 25.58 41.20
3208 4133 4684 6723 3208 3100 1262 100 3208 2104 0 0
50.1 64.5 73.0 104.7 50.1 48.3 19.5 1.1 50.1 32.7 0 0
2.36 2.60 2.88 3.57 2.36 2.51 2.63 2.99 2.36 2.43 2.42 2.58
2.42 2.61 2.76 2.94 2.42 2.66 2.86 3.09 2.42 2.71 2.96 3.24
Middle scheme
Low scheme
1, scheme; 2, year; 3, gross agricultural output value (108 yuan); 4, gross industrial output value (108 yuan); 5, gross domestic product (108 yuan); 6, investment of environmental protection (108 yuan); 7, sewage discharge (104 tonnes); 8, PPR (%); 9, total water demand (108 m3 ); 10, bearing volume of water use (108 m3 ).
Fig. 1. Simulation curve of gross domestic product (GDP).
Fig. 2. Simulation curve of polluted percentage of river (PPR) length.
to rapid industrial development. A series of measures including adjustments to industrial structure, separately supplying water of different qualities for different uses and increasing the price of water were executed to decrease industrial water consumption for each unit of output valued at 10,000 RMB from 9 to 7 m3 could be implemented. However, industrial water requirements would increase to 1.98 × 108 m3 , exceeding water consumption by agriculture. This increases total water requirements to 3.57 × 108 m3 . Due to lack of investment, implementation of water conservation and recycling measures would also be limited and the ability to supply water to the city would be reduced dramatically. Fig. 3 shows that the simulation curves would intersect in 2014, indicating that the total volume of usable water carried
Fig. 3. Simulation curves of total requirement for water supply (TWD) and total carrying volume of water use (TBW).
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
275
would already be insufficient to satisfy total water requirements. Supply of water resources would be unable to meet demand at this time. The total volume of usable water carried would be 2.94 × 108 m3 by 2020, with a massive water shortage of 0.63 × 108 m3 . 3.4. Scheme emphasizing environmental protection (low scheme) The primary goal of this scheme is environmental protection. Therefore, investment in environmental protection increases year on year and causes sewage discharge to decrease gradually. Sewage treatment rates would reach 100% by 2013 and the percentage of river length rated below class IV PPR would also gradually decrease, from 50.1% in 2005 to 32.7% in 2010 and would finally reach 0% in 2013. Lengths of river with inferior water quality would disappear owing to the city’s emphasis on environmental protection. Picturesque scenery with green hills and clear waters would appear around the city and its ecologic environment would be changed completely, as shown in Fig. 2. Due to the slowing of industrial development, industrial water requirements would be merely 0.99 × 108 m3 in 2020, which is 0.99 × 108 m3 less than that of the primary scheme. Therefore, total water requirements would be only 2.58 × 108 m3 , far below the maximum usable water carrying capacity of water resources, as shown in Fig. 3. Although this scheme has an outstanding effect on balancing the demand for water against supply, it will also result in reduced resource exploitation and restricted development of industry and agriculture. This is owing to decreased investment in industrial fixed assets and agriculture. Therefore, under this scheme the objective of quadrupling GDP will only be achieved in 2013, as shown in Table 2 and Fig. 1. 3.5. Scheme balancing economic development and environmental protection (middle scheme) It is obvious that pollution is unavoidable during economic development. However, the economy cannot be developed blindly at the expense of the environment. In particular, the consumption of resources cannot exceed the regenerative ability of ecological systems. Concurrent economic development and environmental protection should be the final objective in Yiwu. The relationship between economic development and other factors, such as population, resource availability and the environment are carefully considered in the middle scheme. Resource availability, the environment, the needs of industry and market conditions must all be taken into account through the weighing of advantages and disadvantages from an ecological angle. Under this scheme, the objective of quadrupling the city’s GDP by the year 2011 could easily be realized if several measures are properly executed. These include inviting outside investment, accelerating construction of the International Trade City and timely adjustment of industrial structure, as shown in Table 2 and Fig. 1. At the same time, sewage discharge would also decrease and the percentage of river length polluted would be reduced to 1.1% by 2020, creating a comfortable living environment (Fig. 2). The total volume of usable water borne could potentially reach 3.09 × 108 m3 by the same time, which can fully satisfy the total annual water requirement of 2.99 × 108 m3 . Equilibrium between supply and demand for water resources would be achieved, as shown in Fig. 3. 4. Discussion 4.1. Economic growth and water resources in Yiwu Yiwu is located in the center of Zhejiang Province, China. The total area of the city is 1109 km2 and the total population is 6.974 million. Yiwu has a mid-subtropical monsoon climate; it is hot and receives a lot of rain. In recent years, the city’s government has decided upon the strategic objective of becoming a key city in central-western Zhejiang Province. The three strategic policies of promoting industrialization, accelerating urbanization and advancing integration of urban and rural areas have been strongly emphasized. Distinct improvements in economic growth have already been achieved by the city. GDP reached 35.200 billion RMB in 2006. GDP per capita also increased to 6300 US dollars. Yiwu city was ranked number 12 by the National Statistics Bureau in a list of the 100 top counties in terms of national comprehensive strength in 2005. Unfortunately, precipitation distribution in Yiwu is not homogeneous in space and time. Moreover, there also exists the phenomenon of river pollution, and water storage projects in the city are very limited. This situation is not ideal, and makes water shortages a likely possibility. This possibility was realized in the summer and autumn of 2003 when
276
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
Table 3 Irrigation quotas for agriculture and water consumption amount per unit output value of 10,000 yuan for industry in different years 1
2
3
2007–2010 2011–2020
32–30 30–28
20–9 8–7
1, year; 2, irrigation quotas for agriculture (m3 /hm2 ); 3, water consumption amount per unit output value of 10,000 yuan for industry (m3 /104 yuan).
the area was hit by record high temperatures and drought. Precipitation from July to October was just 124 mm, only 30% of average precipitation at that time of year. Demand for water was high, but the affluent residents of Yiwu could only sit and watch as polluted water from the river flow past their doors. The Yiwu county government provided funds of about 200 million RMB in November 2000 for purchasing fresh water resources of 5 × 107 m3 annually from a reservoir in Dongyang county on a permanent basis. This became the first instance of trading water rights after the Ministry of Water Conservancy set forth the theory of water rights and the water market. Even with this policy, water usage in Yiwu still requires restrictions in terms of time-sharing, sectioning and water pressure-reducing. It is obvious that the problem of water shortage has become a bottleneck for comprehensively building a comfortable society by 2020. Therefore, it is of great importance to sustainable economic development in Yiwu to make an assessment of the risk of water shortage and water resource carrying capacities. 4.2. Suggestions for future policy parameters The primary scheme of unilaterally pursuing fast economic development at the expense of the environment and the secondary scheme with environmental protection as the main goal, achieved via slowing of the economy, are both undesirable for Yiwu. The third scheme balancing both economic development and environmental protection should be the chosen scheme. To realize the mutual benefits of economic development and environmental protection, a development speed of 7.5% annually in terms of agriculture, industry and GDP is suggested. 7.5% per year is also suggested as the rate at which environmental protection should be accelerated. Irrigation quotas for agriculture and water consumption per unit of industrial output valued a 10,000 RMB in different years are shown in Table 3. Transcendental probability can be described not only as a recurrence interval but also as a guaranteed rate, which implies the ability to reach a certain required percentage. If the level of water supply from 2000 to 2006 is maintained in the near future, then the probability of satisfying total estimated water requirements of 2.51 × 108 m3 in 2010 would be just 0.5053, according to Eqs. (1)–(7) of the water shortage risk assessment. This is still true when the third scheme, encompassing the mutual benefits of economic development and environmental protection is implemented, which means that on average water supply will only be sufficient to meet demand once in every 2 years. Furthermore, the probability of satisfying the estimated demand for water of 2.99 × 108 m3 by 2020 would be almost zero. Therefore, three powerful measures must be adopted in order to attain equilibrium of supply and demand for water resources in Yiwu. The three measures are (1) increased water storage: the amount of reservoir storage for newly built retaining works should reach 0.03 × 108 m3 per year by 2010 and 0.05 × 108 m3 per year by 2016; (2) reduction of water usage: the amount of water resources saved each year should be 0.8% of the total carrying capacity of usable water, accounted for by changing industrial procedures requiring water use, reducing water usage, separating supplies of different water qualities, and increasing the price of water; and (3) improvements in water recycling: the amount of water that can be recycled and used again should increase 0.04 × 108 m3 each year, facilitated by the use of measures such as advanced technology and the recycling of treated sewage. 5. Conclusions Several conclusions can be drawn from the above simulation results and theoretical analysis. These are as follows: (1) Information diffusion technology is an effective method of water shortage risk assessment. (2) The carrying capacity of water resources is non-deterministic and therefore must be calculated on the basis of water shortage risk assessment.
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
277
(3) Precipitation in future drought years in each 5-year cycle will be just 1213 mm and 1076 mm in each 10-year cycle. (4) The primary scheme of unilaterally pursuing fast economic development at the expense of the environment and the secondary scheme focusing on environmental protection as the primary goal by slowing economic development are both undesirable for Yiwu. The third scheme balancing economic development with environmental protection is the most effective scheme. (5) If current water supply levels are maintained in the near future, then water requirements cannot be easily satisfied according to the water shortage risk assessment, even when the third scheme is selected. Water supply will be insufficient to meet demand. (6) At present, water shortages in Yiwu are not due to types of water source, but are caused by low water quality and a lack of water conservation projects. The carrying capacity of water resources of the region can only be improved through the long-term strengthening of investment in environmental protection and the undertaking of new water storage projects. These projects include the building of large or medium scale reservoirs, the adjustment of industrial procedures and the creation of a society that naturally saves water. Although economic development and water resources may be different from each other in many areas, this contradiction will exist forever. Only when the risk of water shortage and water resource carrying capacity are assessed in a reasonable way, is it possible for this contraction to be resolved in a beneficial way, resulting in harmony between nature and the human race. Acknowledgments We are very grateful to the referees and the editors for their helpful suggestions. This work was supported by National Natural Science Foundation of China (No. 40771044) and Zhejiang Provincial Science and Technology Foundation of China (No. 2006C23066). References [1] M. Beuhler, Potential impacts of global warming on water resources in southern California, Water Sci. Technol. 47 (2003) 165–168. [2] E.A. Chatman, Diffusion theory: a review and test of a conceptual model in information diffusion, J. Am. Soc. Inf. Sci. 37 (1986) 377– 386. [3] B. Chen, L.J. Li, H.C. Guo, System analysis on water resources supporting alternatives for Chaidamu basin, Environ. Sci. 21 (2000) 16– 21. [4] A.L. Clarke, Assessing the carrying capacity of the Florida Keys, Popul. Environ. 23 (2002) 405–418. [5] M. Falkenmark, J. Lundqvist, Towards water security: political determination and human adaptation crucial, Nat. Resour. Forum 21 (1998) 37–51. [6] F. Ghassemi, A. Close, J.R. Kellett, Numerical models for the management of land and water resources salinisation, Math. Comput. Simul. 43 (1997) 323–329. [7] J.K. Gilmour, R.A. Letcher, A.J. Jakeman, Analysis of an integrated model for assessing land and water policy options, Math. Comput. Simul. 69 (2005) 57–77. [8] C. Giupponi, J. Mysiak, A. Fassio, V. Cogan, MULINO-DSS: a computer tool for sustainable use of water resources at the catchment scale, Math. Comput. Simul. 64 (2004) 13–24. [9] J.M. Harris, S. Kennedy, Carrying capacity in agriculture: globe and regional issue, Ecol. Econ. 29 (1999) 443–461. [10] C.F. Huang, An application of calculated fuzzy risk, Inf. Sci. 142 (2002) 37–56. [11] C.F. Huang, Demonstration of benefit of information distribution for probability estimation, Signal Process. 80 (2000) 1037–1048. [12] C.F. Huang, Information matrix and application, Int. J. Gen. Syst. 30 (2001) 603–622. [13] C.F. Huang, Principle of information diffusion, Fuzzy Sets Syst. 91 (1997) 69–90. [14] C. Hunter, Perception of the sustainable city and implications for fresh water resources management, Int. J. Environ. Pollut. 10 (1998) 84–103. [15] L.J. Li, H.C. Guo, B. Chen, Water resources supporting capacity of Chaidamu Basin, Environ. Sci. 21 (2000) 20–23. [16] S.F. Liu, J.H. Chen, Water resources carrying capacity based on the theory of ANN, Resour. Sci. 29 (2007) 99–105. [17] T.R. Long, W.C. Jiang, Q. He, Water resources carrying capacity: new perspectives based on eco-economic analysis and sustainable development, J. Hydraulic Eng. 35 (2004) 38–45. [18] Y. Motohashi, S. Nishi, Prediction of end-stage renal disease patient population in Japan by system dynamics model, Int. J. Epidemiol. 20 (1991) 1032–1036. [19] I.E. Ofoezie, Human health and sustainable water resources development in Nigeria: schistosomiasis inartificial lakes, Nat. Resour. Forum 26 (2002) 150–160.
278
L.-H. Feng et al. / Mathematics and Computers in Simulation 79 (2008) 269–278
[20] Y.F. Shi, Y.G. Qu, Carrying Capacity of Water Resources in Urumqi River Basin and its Rational Utilization, Science Press, Beijing, 1992, pp. 128–139. [21] C.H. Toledo, Water resources planning and management in Mexico, Int. J. Hydropower Dams 13 (2006) 53–55. [22] D. Verschuren, K.R. Laird, B.F. Cumming, Rainfall and drought in equatorial east Africa during the past 1100 years, Nature 403 (2000) 410–414. [23] J. Xia, Y.Z. Zhu, The measurement of water resources security: a study and challenge on water resources carrying capacity, J. Nat. Resour. 17 (2002) 263–269. [24] Z.M. Xu, A scenario-based framework for multicriteria decision analysis in water carrying capacity, J. Glaciol. Geocryol. 21 (1999) 100– 106.