Dynamic simulation of water resource management focused on water allocation and water reclamation in Chinese mining cities

In Ordos, the resource industries occupy more than 60% of the total output. Tracing back to the last decade, Ordos has experienced a rapid development with an average growth rate of 33.4% annually. However, Ordos will encounter a bottleneck of water resource. In 2012, the sewage treatment rate was 71%, the ratio of water reclamation was 57%, and the removal rate of water contaminants was just 26%. According to the current water allocation and utilizing strategy, the water deficit will reach 809 million m³ in 2020 (Li, 2011). Water resources cannot guarantee the sustainable development of socio-economics in Ordos (The Twelfth Five-year Plan, 2011). Therefore, water resource management should be put forward to ensure water supply, enhance water utilization efficiency and improve water quality without damaging economic development of mining cities.

Regarding water resource management, the previous studies mainly focus on the management of water allocation, water reclamation and water quality improvement. First, since the total water resource is limited, a portion of scholars have built linear programing models to discuss how to allocate the limited water resource efficiently among multiple water users to achieve economic, social and environmental benefits with equilibrium constraints (Andrew et al., 2008; Mahdi, 2010; Wolfgang et al., 2013). Secondly, since water reclamation is an effective approach for water exploitation, a group of scholars have utilized a systematic framework to quantify the water reclamation potential to ensure water supply for socio-economic activities (Angelakis et al., 2003; Hong & Abbaspour, 2007; Apostolidis et al., 2011). Thirdly, since the significance of water quality has aroused residents’ attention, a series of studies have focused on wastewater treatment with adoption of innovative sewage disposal technology to reduce water contaminants and provide high-quality water to residents (Helen & James, 2008; Zhang et al., 2013). However, even the increase in wastewater treatment will not prevent water companies from experiencing the effects of adverse events; several studies have analyzed the risks and reliability linked to operation of water supply both on quality and quantity (Tchorzewska-Cieslak, 2007; Boryczko & Tchorzewska-Cieslak, 2013; Boryczko & Tchorzewska-Cieslak, 2014). Most of these researches lack the dynamic linkage among socio-economic development, population growth, environmental impact and the intervention of technologies and policies. Furthermore, all of these researches mainly focus on one aspect of water management. Research on integrated water resource management is relatively rare.

Water resource management contains many variables and constraints. In order to achieve more reliable results, it is necessary to take the related factors into full consideration. In recent years, several scholars have applied dynamic models and a simulation approach to conduct studies on integrated urban water resource management. These studies consider the connection of water resource, water pollution and economic development, and evaluate the economic and environmental impacts of integrated water management including wastewater treatment, water utilization and water reclamation (Mizunoya et al., 2006; Akmam & Higano, 2007; Xiang et al., 2014). However, most of these researches concentrate on the developed regions with a focus on water quality; there is a lack of early-warning studies for developing mining cities with consideration of both water allocation and water reclamation.

Therefore, based on these considerations, this paper aims to achieve sustainable economic development under the constraints on water quantity and water quality, with an integrated water resource management focusing on water allocation and water reclamation. There are three main goals to achieve in this paper. (1) This study selects Ordos as a case study, attempting to construct and analyze a comprehensive model with integrated water resource management, which is tailored to suit the characteristics of Chinese growing mining cities. (2) This study tries to explore the optimal water resource management measures, which include water allocation and water reclamation with the introduction of new sewage treatment technologies under the constraints of water resource and water pollutants. (3) This study tries to evaluate and predict the influences of integrated water resource management on economic growth, water utilization and water environment. In this way, the result offers the potential for growing mining cities to avoid development dilemmas and provides references for local policy-makers in other mining cities.

Considering the interactions among water resources, water pollutants, multiple industrial sectors and households, as well as the impacts caused by the implementation of water resource management on macroeconomic variables, industrial output levels, long-term economic growth and other factors (Moffatt & Hanley, 2001), a comprehensive methodology is adopted in this article. The input–output model extended to the field of environment and natural resources was first adopted to evaluate environmental carrying capacity, and later was applied to study the environmental influences of industrial development since it is a very suitable method for conducting environmental assessment (McKenzie & Durango-Cohen, 2010). However, the assessments can provide valuable quantitative information on the anthropogenic environmental loads of economic activities, but the resource and environmental loads can't serve as restrictions to react upon for industrial activities. Linear programing is a useful optimization approach for synthetic decision-making since it can help with identifying an optimal strategy to control the environmental impact and resource consumption related to economic activities. The linear programing method has been already mixed with input–output analysis for solving the problems of environmental impacts and nature resources' utilization (Vogstad, 2009). Therefore, based on the input–output model extended to the field of environment and natural resources (Leontief, 1977; Lee, 1982; Forsund, 1985; Perrings, 1987) with the principles of material balance and value balance (Freeman et al., 1973), the model in our paper uses linear equations and laws of linear algebra (Jin et al., 2003) to set the links between economic activities and related water consumption and pollutant emission, so as to provide an optimal water resource management that can simultaneously ensure water supply and improve water quality without damaging economic development.

To implement comprehensive modeling and simulation, a systematic model is formulated as a linear optimization program, which uses the mathematical software package LINGO (Lindo, Chicago, IL) (Wagner, 1959; Hadley, 1962; Hillier & Gerald, 1967). LINGO, as a linear programing procedure, is always an efficient way to solve optimization dynamic simulation problems (Beroggi, 1999), which enables the simulation of the synthetic system by creating as many linear functions as possible that are close to reality. Finally, a global optimal solution will be obtained through computer operation.

As outlined in Figure 2, water resource management focuses on water allocation and water reclamation. On the one hand, optimal water allocation considers the intensities of both water resource utilization and water pollutants' emission connected to industrial activities, which can improve the utilizing efficiency of limited water resources. On the other hand, water reclamation with the introduction of new sewage treatment technologies, according to the treatment demand of different districts, the treatment capacity and construction costs of different technologies, can improve water quality and gain water supply. Through this integrated water resource management approach, this paper can address the detailed water allocation plan among industries and the water reclamation plan with technologies' selection, arrangement and subsidies, so as to maximize the gross regional product (GRP) of the target area under the particular constraints of water supply limitation and water environmental carrying capacity.

The data collected in this paper consist of national and local published data, survey data and calculated data. The published data include the Ordos Yearbook 2009–2013, Ordos Environment Bulletin in 2013, Emission Coefficient Manuals of Industrial and Urban Residents in 2010, Input–output table of Inner Mongolia in 2007, Water Resource Bulletin in 2013, and Environmental protection ‘Twelfth Five-year plan’ in Ordos. The survey data are collected from relevant government agencies, such as the Municipal water resources bureau, Environmental protection agency, and Land and resources bureau. Based on the collected data, we calculated the water demand coefficient, sewage water discharge coefficient, chemical oxygen demand (COD) emission coefficient, and input–output coefficients for simulation modeling. All of the coefficients remain unchanged since we assume the technical level would stay the same during the simulation period. The exogenous variables in this model include industrial production, water supply and different kinds of coefficients, all of which are listed in the Appendix (available with the online version of this paper).

We divided Ordos city into eight administrative regions (Dongsheng District, Dalad Qi, Jungar Qi, Etuokeqian Qi, Etuoke Qi, Haggin Qi, Uxin Qi, Ejin Horo Qi) and 10 industries in the simulation formulation. In addition, we set two types of decision variables, one endogenous (en) and the other exogenous (ex). The exogenous variables are calculated based on current data sources (listed in the Appendix, available online), whereas the endogenous variables will be derived from simulation. The target term (t) is 13 years, from 2012 to 2025.

In this section, the industrial structure, population development and city urbanization work together to construct the socio-economic development of Ordos. The flow balance of market commodities is the foundation of its industrial production. Due to the limited space, we simply demonstrated the formulas for industrial production and adjustment of industrial structure below.

To ensure water supply and improve the percentage of disposed sewage, we encourage the treatment of more sewage water with new sewage treatment plants. In Ordos, there are a total of 18 sewage treatment plants (The List of National Urban Sewage Treatment Facilities Operation, 2013), and these mainly adopt the traditional Activated Sludge treatment technology (technology A in Table 1), which is widely used in China at present (Xiang et al., 2014). Water treated by technology A is suitable for agricultural irrigation and environmental maintenance due to its lower removal ability and its operating and maintaining cost. Aiming at the characteristics of emission in mining cities, there are two advanced technologies expected to be introduced. One is the Double Membrane Bio-Reactor (technology B), which is applied in developed areas such as Beijing (Song et al., 2010). Water treated by technology B can be used for industry. The other one is the Ceramic Membrane Bio-Reactor (technology C), which is imported from Japan (Peng, 2010). The reclaimed water produced by technology C can satisfy all kinds of water demands due to its high standard. As demonstrated in Table 1, technology C holds the highest output ratio for reclaimed water production and the best capacity for water pollutant removal, whereas technologies A and B are more efficient on investment.

In this study, we set a series of constraints for simulation. According to Urban Overall Planning of Ordos (2011–2030): (1) the upper limitation of groundwater, local surface water, transfer water and other water resources supply remains at 2,341 million m³; (2) the aimed reduced rate of water pollutant COD is 2% each year to fulfill the emission reduction target; (3) the reclaimed water productivity increases year by year; and (4) industry development constraints defined by input–output model and the scope of industrial output will vary around the average growth rate of the last decade.

Based on the built model and constraints' setting, actual data of Ordos are used to conduct simulation experiments. After repeated trials, this simulation model has passed a consistency test, validity test and sensitivity test. In this section, the optimal water resource management, which focuses on water allocation and water reclamation under the constraints of water quantity and water quality, is explored. The influences of water resource management on economic development, water utilization and water quality are predicted accordingly.

Integrated water resource management includes water allocation and water reclamation. Optimizing water allocation and increasing the water reclamation ratio with the introduction of new sewage treatment technologies can improve water supply, water utilization efficiency and water pollutant emission intensity without damaging the economic growth.

According to Figure 4, water demand of the following industries should be curtailed after considering their large amount of water consumption, water pollutant emissions and lower value-added ratios: the primary industry from 73 to 58%; the non-coal mining industry from 0.4 to 0.1%; and the other manufacturing industries from 3.5 to 0.6%. On the contrary, water demand of the following industries should be increased since their higher economic benefits can cover their water pollution loss: the petroleum and chemical industry from 2.5 to 5.9%; the metal, nonmetal manufacturing industries from 0.7 to 1.6%; the equipment manufacturing industry from 0.1 to 1.2%; the electricity and water production industry from 3.8 to 17.2%; and the construction industry from 0.7 to 1.7%.

In order to increase the water reclamation ratio, an optimal selection of sewage treatment technologies and installation plan is derived from the simulation. As shown in Table 2, a total of nine instances of new sewage plants with technologies A, B and C are suggested to be built, mainly in Dongsheng District and Jungar Qi with fiscal subsidy of US$512.6 million including US$107.5 million for construction and US$405.1 million for operation and maintenance.

The Dongsheng District, as the administrative center of Ordos, is undergoing rapid urbanization. The natural growth rate of the population is 9.66%, and the proportion of the construction industry and the tertiary industry is over 70%, which will boost the pressure on both water demand and water contamination. Therefore, one instance of technology C is suitable to be introduced in Dongsheng District because technology C holds the advantages of a lower construction cost, a smaller sewage treatment amount, a higher removal rate and better water quality.

In addition, Jungar Qi is the important manufacturing district in Ordos, the scale of the coal mining industry and the electricity and water production industry accounts for 58%. These industries require large water consumption. Combined with the growing population and rapid industrialization and urbanization, Jungar Qi will face great pressure on water quality. To address the shortage, eight sewage plants with one instance of technology A, one instance of technology B, and six instances of technology C are suggested to be built in Jungar Qi. Technology C is chosen at the beginning of the target period due to its high removal ratio. To guarantee further economic development, the mathematical modeling elects technologies A and B due to their high efficiency of investment.

Based on this industrial structure, GRP in Ordos continues to rise steadily from US$68,220 million in 2012 to US$170,674 million in 2025, and the average annual growth rate remains at 7.6%. We can see from Figure 6 that in 2013, the fiscal subsidy for water resource management is not executed, so the economic growth rate reaches 9.5%. From then, the industrial structure is influenced by water allocation and the subsidy used for sewage water treatment, making the economic growth rate drop back to a minimum percentage of 6.2% in 2015. After 2015, due to the further optimization of industrial structure, the growth rate rises solidly to 8.8% in 2023 but ultimately falls to 6.6% in 2025. The reason for this is that the introduction of new sewage plants does not mean a win–win situation for economic growth and environmental improvement simultaneously in the later stage of the target period under the restriction of water supply and water pollutant carrying capacity.

The simulation result demonstrates that with an annual GRP growth rate of 7.6%, the implementation of water resource management focused on water allocation and water reclamation can realize the trade-off between supply and demand and fulfill the task of 2% emissions' reduction each year during the target period.

To ensure water supply, enhance water utilization efficiency and improve water quality without damaging economic development in mining cities, this research constructs a comprehensive dynamic linear optimization model, including a socio-economic mode, water supply–demand mode and water quality control mode with interaction among the three sub-modes, to explore optimal water resource management that focuses on water allocation and water reclamation, and observe the consequences of water resource management on economic development, water utilization and water quality.

After modeling and simulating a typical area, Ordos, the results contain detailed water resource management measures, including a water allocation plan, water reclamation plan with technologies' selection, arrangement and subsidies. During the target term (from 2012 to 2025), water demand of the primary industry, the non-coal mining industry, and the other manufacturing industries is recommended to be cut by 15%, 0.3%, and 2.9%, respectively. Nine wastewater treatment plants of three technologies are suggested to be built mainly in Dongsheng District and Jungar Qi with fiscal subsidy of US$512.6 million including US$107.5 million for construction and US$405.1 million for operation and maintenance. After implementation of water resource management, the following objectives are foreseen simultaneously by 2025: the ratio of water reclamation reaches 73.9%, reclaimed water production increases by 93.95 million m³, the emissions of water pollutant COD decrease by 23%, and the economic growth rate remains at 7.6% annually.

However, there are still some limitations that need to be considered for future extensions. For example, the classification of water pollutant materials may not be sufficient to manifest the unique characteristics of mining cities. Furthermore, water price can also influence the decisions of water policy, and we will continue this study and try to manipulate the Computable General Equilibrium (CGE) model to investigate the impact on water price caused by integrated water resource management (Llop & Ponce-Alifonso, 2012) as well as the response of different economic factors according to the variation of relative price, subsidy and progress of water-using technology (Li et al., 2014), to put forward a more specific regional plan for policy-makers.

The research concludes that the proposed comprehensive dynamic linear optimization model can elaborate the interrelations among industries with water resource utilization and water pollutant emission by drawing linear equations, and can precisely simulate and predict the policy effects of applicable approaches and technology introductions under the precondition of satisfying multiple goals of socio-economic development, resource saving and environmental conservation. Therefore, this synthetic methodology can serve as a pre-evaluation approach for conducting early-warning research for mining cities in a growing stage to avoid a series of inharmonious problems, offering feasible suggestions for government decision-making. Moreover, on account of the openness of modeling, this model can be replicated by adding substance to solve coordinated development issues in socio-economics, resources and the environment in other regions.

The authors wish to thank Hua Zhang, Jingsongdi Yu and Nan Xiang for their critical comments and suggestions. Furthermore, this study is supported by grants from Survey and Evaluation on Prospecting Breakthrough of Important Mineralization Zones (No. 1212011220851), National Natural Science Foundation of China (No. 41101559), Beijing Social Science Foundation (No. 14JGC116), Fundamental Research Funds for the Central Universities (No. 2652015156) and Dr. Scientific Research Foundation of Shijiazhuang University of Economics (No. BQ201504). The authors would also like to thank the editor of Water Policy and the anonymous reviewers for their precious time and constructive comments on the paper.

The authors declare no conflict of interest.