Analysis of the ultimate water resources carrying capacity in Yancheng, China Open Access

Since the 20th century, the global population has expanded rapidly and now exceeds 7.5 billion (7.5 × 10⁹). (Chen et al. 2020). With the development of technology and the economy, the living standard of human beings has been greatly improved. However, with the accompanying unreasonable exploitation and utilization of resources, the ecological environment has been seriously damaged. Major problems, including environmental pollution (Wang et al. 2020), ecological degradation (Shen et al. 2019), and resource shortages (Li et al. 2017) have gradually emerged. As an important part of natural resources, water is also confronted with over-exploitation (Liao et al. 2020). To ensure water security and to promote sustainable development of the social economy, scholars have conducted a wide range of research covering rational utilization and protection of water resources, and have gradually formed the concept of water resources carrying capacity (WRCC).

Until now, a major achievement of WRCC is the evaluation of water resources carrying status (WRCS) (Chi et al. 2019). WRCS refers to the overall status of the water resources system, the socioeconomic system and the ecological system in a particular region (Yang et al. 2020). It is evaluated by a comprehensive assessment system, which includes the degree of water resources development and utilization, the degree of social and economic development, the overall ecological health status of the region and other evaluation indicators. The commonly used evaluation methods range from fuzzy mathematics, to the binary index method, cloud model and ecological footprint theory (Dai et al. 2019; Yang et al. 2019; Wu et al. 2020; Zhang et al. 2020). The calculation method of WRCS is relatively simple and convenient for application (Peng & Deng 2020). However, the selection of evaluation indicators and the division of evaluation criteria have a significant impact on the calculation process, leading to a strong subjectivity on the final research results. Moreover, WRCS evaluation fails to offer the upper limit of population size and economic scale that the water resources can support in a region, which are critical values for any regional development strategy.

To break through the limitations of WRCS, research into ultimate water resources carrying capacity (UWRCC) is gradually increasing. UWRCC refers to the largest population size and Gross Domestic Product (GDP) scale that water resources can support in a region under specific economic and technical conditions, based on the premise of maintaining a healthy aquatic ecosystem (Yang et al. 2020). UWRCC is usually calculated using multi-objective programming models with the constraints of available water resources and water environmental capacity (Yan et al. 2019). Among them, Dou et al. (2015) proposed a distributed model that considers constraints like hydro-economic interactions, water supply, water quality, and socioeconomic development, which was applied to calculate the maximum population that the water resources can support in Henan Province. Li et al. (2016) established an optimization model based on approximate stochastic multiple objective programming, which was put to use for the optimization of industrial structures based on the water environmental carrying capacity in Huai River Basin in Shandong province. Meng et al. (2018) developed an inexact two-stage stochastic programming model to analyze the influence of flow level on the UWRCC of Yinma River Basin. Combined with an artificial neural network, Luo et al. (2019) presented an analysis model to predict maximum supportable population of water resources of the Beijing-Tianjin-Hebei region. These achievements provide scientific guidance for water resources management. However, UWRCC is a specific value under given economic and technical conditions. Indicators used to characterize economic and technical conditions, such as per capita GDP, water supply capacity of water conservancy projects, water use efficiency and sewage treatment level, will change during the development of a region, resulting in continuous dynamic changes of UWRCC. Therefore, instead of employing a single value, the UWRCC result sets under different economic and technical levels can provide better reference for the determination of regional development goals in the future.

Yancheng is located on the east coast of China. It is one of the major cities involved in the ‘Great Development Strategy of Jiangsu Coast’ and is under a plan to further expand its socioeconomic scale. However, as agriculture is the leading industry in Yancheng, accounting for 80% of the total water consumption in 2015, the water use efficiency of Yancheng is relatively low and the contradiction between water supply and demand is relatively sharp. To achieve the goal of sustainable development, it is necessary to analyze the rationality of the planned socioeconomic scale and to put forward suggestions on the socioeconomic development of Yancheng from the perspective of water resources utilization.

To obtain the maximum socioeconomic scale that the water resources can support long term under different conditions in Yancheng, this study constructs a UWRCC calculation model that takes the maximum population size and the maximum GDP scale of a region as the objectives and takes water quantity and water quality as the main constraints. Then, combined with 16 computing scenarios, this study first quantitatively analyzes the influence of technical level (water use efficiency and sewage treatment) and economic level (GDP per capita) for UWRCC. Meanwhile, combined with different UWRCC results under different conditions, a diagram of the UWRCC result sets of Yancheng is obtained. The information it provides can help decision makers set and adjust the sustainable social economic development mode from the perspective of water resources without tedious calculations.

This study is divided into four steps, as is shown in Figure 1. The first step is identifying major influence factors for UWRCC for the study area. The second step is constructing the UWRCC calculation model. The third step is setting computing scenarios and calculating. The last step is analyzing the UWRCC results and drawing a diagram of the UWRCC result sets.

The main influencing factors of UWRCC are divided into three categories: natural conditions, technical level, and economic status, as is shown in Figure 2. Natural conditions mainly include precipitation, underlying surface, and the number of rivers, and channel morphology, determining the carrying capacity of the carrying subject. In this study, natural conditions are specifically reflected in the amount of water resources and the self-purification capacity of the rivers. The amount of water resources affects the amount of available water supply in a region which refers to the maximum amount of water resources that is supplied by water conservancy projects from rivers on the premise of maintaining the ecological flow of the rivers. At the same time, to protect the river ecosystem and to maintain the sustainable development of a region, the amount of pollutant emissions must be within the rivers' self-purification capacity. The rivers' self-purification capacity is represented by water environmental capacity. For two similar regions, UWRCC is larger in the one with a larger available water supply and a stronger rivers' self-purification capacity. Technical level refers to the level of water resources development, utilization and protection. The development level reflects the water supply capacity of water conservancy projects, including the ability to intake local water resources and transit water resources. The utilization level reflects the water use efficiency of social and economic sectors. The protection level refers to the reutilization and treatment capacity of tail water. Generally speaking, the higher the technical level, the larger the UWRCC. Economic status mainly refers to the industrial structure and the per capita GDP level of a region. For the industrial structure, compared with regions dominated by agriculture, regions dominated by industry and services cause less water consumption per unit benefit and a smaller pollution emission amount per unit benefit, and correspondingly a greater UWRCC value. In addition, UWRCC is characterized by population size and GDP scale, through which per capita GDP can be obtained. The calculated per capita GDP needs to meet the requirements of the corresponding economic development stage. The above factors will directly affect the UWRCC of a region. When constructing a UWRCC calculation model, natural conditions are considered as constraints, while technical and economic factors are considered by setting scenarios with different technical and economic levels to analyze the impact of their changes on UWRCC.

As is shown in Figure 3, the UWRCC model constructed in this study is a multi-objective optimization model. To realize the optimal allocation of water resources in space, an intact region is divided into different calculation units before model construction. The model sets two objectives: maximum sum of population size of each calculation unit and maximum sum of GDP scale of each calculation unit. The model constraints include water quantity constraint, water quality constraint and socioeconomic constraint. (i) Water quantity constraint predominantly indicates that total water use of all water users must be less than the available water supply in each calculation unit. Water users are divided into three categories: domestic, production, and ecological environment, with a further seven sub-categories (urban domestic, rural domestic, primary industry, secondary industry, tertiary industry, environmental sanitation, and landscape plants). The water supply sources are divided into local water and transit water. Transit water is the public water source, and the amount of transit water allocated to each calculation unit is obtained through the model. Contrary to transit water, local water is supplied to its own calculation unit. (ii) Water quality constraint indicates that pollutant emissions from production and domestic sources into rivers must be less than the water environmental capacity of the rivers in each calculation unit. According to the requirements of the ‘Three Red Lines Policy’ formulated by China's Ministry of Water Resources, chemical oxygen demand (COD) and ammonia nitrogen (NH₃-N) are selected as the pollutant indicators in this model. Water environmental capacity here refers to the maximum amount of COD and NH₃-N that can be contained in a water body on the premise of not affecting its function. The pollution sources are divided into point source and non-point source. Point source pollution includes the wastewater from domestic, secondary industry, and tertiary industry sources, which is discharged after being treated by sewage treatment plants. The pollution from agriculture is considered as non-point source pollution. (iii) Socioeconomic constraint includes three sub-constraints: industrial structure constraint, food supply constraint and economic level constraint. Industrial structure constraint is used to ensure that the calculated industrial structure of each calculation unit is reasonable. Food supply constraint is used to guarantee that local grain production meets the demand of the residents in each calculation unit. Economic level constraint refers to a requirement that GDP per capita is reaches a certain level to ensure the fine economic condition of residents in each calculation unit.

Water use of domestic sources ⁠, water use of primary industry ⁠, water use of secondary industry ⁠, and water use of tertiary industry in the calculation unit k are decision variables of the model.

This constraint includes three aspects: industrial structure constraint, food supply constraint and economic level constraint.

The third step is to use the initial population and the fitness function to finish the genetic operations (i.e., selection, crossover and mutation) of the GA method. When the difference in fitness between neighboring generations is <0.1, the iteration is terminated and the population with the maximum fitness is selected as the solution of the decision variables. Finally, the UWRCC value of the region is figured out by the calculated decision variables.

Yancheng, Jiangsu Province, China, was selected as the research area in this study. Yancheng is located in the lower reaches of the Huai River, in the middle east of North Jiangsu Plain, between 32°34′–34°38′ N and 119°27′–120°54′ E. The mean annual precipitation is 1,016.6 mm and the value decreases from south to north inside Yancheng. The mean annual runoff is 4,317 million m³ and the equivalent runoff depth is 254.0 mm. Yancheng has jurisdiction over 9 districts and counties. By 2015, the total population of the city was 8.28 million, of which 4.56 million were an agricultural population, accounting for 55.13%, and 3.72 million were a non-agricultural population, accounting for 44.87%. The GDP of the whole city was 380 billion yuan, and the industrial structure ratio was 13.6: 47.0: 39.4 (primary to secondary to tertiary industry). Yancheng is one of the major cities involved in the ‘Great Development Strategy of Jiangsu Coast’ and its socioeconomic scale is planned to be further expanded.

This study divided the research area into 9 calculation units based on the county-level administrative districts, namely Tinghu, Yandu, Dongtai, Dafeng, Xiangshui, Binghai, Funing, Sheyang and Jianhu. The geographical location of Yancheng is shown in Figure 5.

To make the calculated UWRCC meet the concept of long-term and stable support, the available water supply in an extremely dry year (after frequency analysis of long series data of annual water resources using the P-III distribution curve, the water quantities of an extremely dry year correspond to the 90% frequency on the P-III distribution curve) of Yancheng was selected as the water quantity input of the model. To compare the influence of different economic levels and technical levels on the UWRCC results, different computing scenarios were set in this study. The influencing factors of technical level included three types of water supply capacity of water conservancy projects, water use efficiency and sewage treatment level. In this study, water supply capacity was not considered in scenario analysis because the water supply capacity of water conservancy projects exceeds the amount of water resources in Yancheng. The technical level in this study was considered from two aspects: water use efficiency and sewage treatment level. It was characterized by a series of parameters and they could also be divided into four grades: basic, medium, relatively high, and high. They referred to the level that Yancheng can reach in 2015, 2020, 2025 and 2030, respectively. Economic level was characterized by the indicators of per capita GDP and industrial structure, which were both divided into four grades: basic, medium, relatively high, and high. The above four economic levels correspond to the economic situation of Yancheng in 2015, 2020, 2025 and 2030, respectively. Combining the above different technical levels and economic levels, 16 different scenarios were obtained. The scenario numbers and the corresponding technical levels and economic levels are shown in the Table 1. Among them, Scenarios I, V, IX, and XIII represented the actual scenario in 2015 and the planned scenarios in 2020, 2025, and 2030 of Yancheng, respectively.

The data for this study were mainly collected from three sources. Water Resources Bulletin of Yancheng offered water supply and water use data. Water Resources Integrated Planning of Yancheng included the data of the existing and planning water supply projects. Socio-economic data, such as current and planning population, economic level of different years and available cultivated area were collected from Statistical Yearbooks and related socioeconomic development planning. The parameter values of the model which could reflect different technical and economic levels were calculated on the basis of the collected data, as is shown in Table 2.

The UWRCC results of 16 computing scenarios for Yancheng are shown in Table 3. Scenario XIII (economic level: basic, technical level: high) has the largest population of each calculation unit, with a total population of 14.17 million, and a GDP of 712.39 billion yuan. Scenario IV (technical level: basic, economic level: high) has the smallest population of each calculation unit, with total population of 4.37 million, only one third of scenario XIII, and a GDP of 864.77 billion yuan. Meanwhile, Scenario XVI (economic level: high, technical level: high) has the largest GDP of each calculation, with total GDP of 1,756.81 billion yuan, and a population of 8.87 million. Scenario I (technical level: basic, economic level: basic) has the smallest GDP of each calculation unit, with total GDP of 313.59 billion yuan, and a population of 6.87 million. Overall, for four different economic scenarios with the same technical level, the higher the economic level, the smaller the population size and the larger the GDP scale in UWRCC. For four different technical scenarios with the same economic level, the higher the technical level, the larger the population size and the GDP scale in UWRCC.

The technical level and economic level of Scenarios I, VI, XI, and XVI correspond to that of 2015, 2020, 2025 and 2030, respectively. The populations of the four scenarios in UWRCC for Yancheng are 6.87 million, 7.50 million, 8.86 million and 8.87 million. The GDP of the four scenarios in UWRCC are 313.59 billion yuan, 733.28 billion yuan, 1,324.66 billion yuan and 1,756.82 billion yuan. It is thus clear that the population size and economic scale are on the rise.

The WRCD calculation results are shown in Figure 6. From 2015 to 2030, the WRCD value of Yancheng dropped from 1.05 to 0.9, which meant the water resources carrying status changed from overloaded to sustainable. It showed that in accordance with the government's planning of water resources utilization and socioeconomic development, Yancheng's water resources would be sufficient to support the population and GDP in the future. From the perspective of districts, the variations of WRCD values were different. The WRCD values of Binghai and Tinghu showed a clear upward trend. Binghai was in a sustainable state in 2015 and then deteriorated into an overloaded state in subsequent years, while Tinghu remained overloaded for all four years. The districts showing a downward trend in WRCD values were Xiangshui, Jianhu, Yandu, Dongtai, and Dafeng. In these districts, except Yandu, water resources were in a sustainable state in future years. Although different in the changing trend of WRCD value, water resources carrying condition of Yandu and Tinghu were similar, overloaded for all four years. For Funing and Sheyang, WRCD values changed insignificantly each year, fluctuating around 1. In order to cope with the possible overload situation in Yandu, Tinghu, and Binghai in the future, further measures should be taken on the basis of existing plans. Specifically, for Yandu and Tinghu, because of the concentration of industry and the relatively large economic scale, the population and major industries should be guided to move to other sustainable districts by adjusting the industrial distribution in the future. For Binghai, water use efficiency and the reutilization and treatment capacity of wastewater should be improved.

Points in Figure 7 represent the calculated population size of different computing scenarios. The points of the same technical level were fitted into a curve and finally four curves were obtained.

The second characteristic to be noted is that the higher the technical level is, the higher the position of the corresponding curve in the diagram. This shows that for the same economic level, when the technical level improves, the population that water resources can carry in Yancheng will increase. Moreover, multiple curves representing different technical levels are parallel to each other. Assuming that two curves representing different technical levels intersect, there will inevitably be a situation that a smaller population can be carried under a higher technical level. That is to say, the increase in water use efficiency and sewage treatment level will lead to a decrease in the population, which is obviously wrong. That is why the curves in the diagram cannot intersect.

The diagram of UWRCC result sets for Yancheng reflects the relationship between the three variables of technical level, economic level and the largest population size that can be carried by water resources. In the case that any two variables are known, the unknown variable can be quickly and directly derived from this diagram. Specifically, when the future population size of Yancheng is determined, a straight line representing the determined population size parallel to the abscissa can be drawn in the diagram of UWRCC result sets. The points where the line intersects the curves represent several different socioeconomic development modes for the determined population size. Each development mode corresponds to a certain economic level and technical level. On the other hand, if the economic level and technical level of Yancheng are determined, the optimal population size can be found using in the diagram.

The diagram is an intuitive presentation of the UWRCC result sets of Yancheng, which provides a concise and straightforward assistant for decision makers to set and adjust the socioeconomic development mode from the perspective of water resources without tedious calculations. Certainly, the diagram can be further refined from the following two aspects: firstly, for each technical level, more economic level scenarios can be set up to increase the point density to better fit the curve, so as to make the fitting results more credible; secondly, more technical level scenarios can be added to increase the number of curves and expand the applicability of the diagram.

The method proposed in the study seeks the largest socio-economic scale of a region under different economic and technical levels from the perspective of the sustainable utilization of water resources. Two aspects should be paid attention to when applying this method to other areas. The first aspect is parameter determination of the UWRCC calculation model, such as values of available water supply and water environment capacity, which need lots of preliminary work. The second is that, before applying the scenario analysis method, it is necessary to identify the major factors of UWRCC to reduce calculation.

The UWRCC research can be improved in the future as follows. Firstly, the calculation step length in the model is one year, which does not take into account the uneven distribution of precipitation (Zhan et al. 2018) and socioeconomic water demand during the year. Especially for grain planting, there are some differences in the water demand in different periods of the planting cycle (Cao et al. 2019). In later research, the calculation step length could be set as a month to reflect the above influence. Secondly, the water exchange brought by grain trade between regions is not considered. Yancheng is the main grain base of Jiangsu Province, whose production fully meets the local grain demand, so its impact is not considered. In regions with high socioeconomic development and dominated by industry and services, local grain yield is low. These regions meet the residents' grain demand mainly by purchasing grain from other regions and the virtual water contained in the purchased grain reduces the agricultural water demand (Zhao et al. 2020), which has an impact on the UWRCC values. This issue can be studied combining with virtual water theory in future UWRCC research.

To help decision makers set and adjust the sustainable socioeconomic development mode from the perspective of water resources, this paper proposed an UWRCC calculation method combined with multi-programming model and scenario analysis. By applying this method to Yancheng, the UWRCC results under different technical levels and economic levels are obtained. Generally, for different economic scenarios with the same technical level, the higher the economic level, the smaller the population size and the larger the GDP scale in UWRCC, for different technical scenarios with the same economic level, the higher the technical level, the larger the population size and the GDP scale in UWRCC. In addition, the calculation results of Scenarios I, VI, XI, and XVI corresponded to the UWRCC values of Yancheng in years 2015, 2020, 2025, and 2030, which are 6.87 million persons and 313.59 billion yuan, 7.50 million persons and 733.28 billion yuan, 8.86 million persons and 1,324.66 billion yuan and 8.87 million persons and 1,756.82 billion yuan, respectively. Analyzing the WRCD of these four years showed that Yancheng's water resources would be sufficient to support the population and GDP in the future. However, as far as districts are concerned, Yandu, Tinghu, and Binghai would experience population and GDP overloading in future years. Based on the UWRCC result, a diagram of the UWRCC result sets under different technical and economic levels in Yancheng was drawn in this study. This diagram reflected the quantitative relationship between the three variables of technical level, economic level and the maximum sustainable population size. Decision makers can get information about UWRCC directly and conveniently from this diagram. Some aspects could still be improved. Firstly, the calculation step length in the model can be changed from year to month. Secondly, virtual water theory can be introduced to analyze the impact of grain trade of UWRCC.

This work was supported by the National Key Research and Development Program [grant number 2016YFC0401306], the Water Resources Department of Jiangsu Province Key Research Program [grant number 2016003, 2019003], and the Fundamental Research Funds for the Central Universities of China [grant number 2014B1605318].

All relevant data are included in the paper or its Supplementary Information.