Simulation and Control of Water Pollution Load in the Xiaoxingkai Lake Basin Based on a System Dynamics Model

Abstract

With the rapid development of the social economy, human activities have increasingly disrupted water environments, and the continuous input of pollutants poses significant challenges for water environment management. Taking the Xiaoxingkai Lake basin as the study area, this paper develops a social–economic–water environment model based on the system dynamics methodology, incorporating subsystems for population, agriculture, and water pollution. The model focuses on four key indicators of pollution severity, namely, total nitrogen (TN), total phosphorus (TP), chemical oxygen demand (COD), and ammonia nitrogen (NH₃-N), and simulates the changes in pollutant loads entering the river under five different scenarios from 2020 to 2030. The results show that agricultural non-point sources are the primary contributors to TN (79.5%) and TP (73.7%), while COD primarily originates from domestic sources (64.2%). NH₃-N is mainly influenced by urban domestic activities (44.7%) and agricultural cultivation (41.2%). Under the status quo development scenario, pollutant loads continue to rise, with more pronounced increases under the economic development scenario, thus posing significant sustainability risks. The pollution control enhancement scenario is most effective in controlling pollutants, but it does not promote socio-economic development and has high implementation costs, failing to achieve coordinated socio-economic and environmental development in the region. The dual-reinforcement scenario and moderate-reinforcement scenario achieve a balance between pollution control and economic development, with the moderate-reinforcement scenario being more suitable for long-term regional development. The findings can provide a scientific basis for water resource management and planning in the Xiaoxingkai Lake basin.

1. Introduction

Lakes are important reservoirs of surface water resources, playing an irreplaceable role in supplying domestic water, regulating river runoff, providing water sources for industrial and agricultural activities, and improving the surrounding ecological environment [1,2,3,4]. However, with the acceleration of population growth, industrialization, and urbanization, lake ecosystems are facing increasingly severe pollution pressures [5]. A large number of pollutants enter lakes through agricultural runoff, domestic sewage, and industrial emissions [6], not only threatening lake water environmental safety and human health but also inhibiting regional socio-economic development [7], thus forming a vicious cycle in which ecological and economic factors constrain each other [8]. Therefore, balancing economic development with water environmental protection has become a common challenge for sustainable development worldwide [9,10].

Currently, the primary methods for studying the interrelationship between regional socio-economic development and the environment include the Fuzzy Comprehensive Evaluation Method [11], Coupling Coordination Degree Models [12], and the Environmental Kuznets Curve hypothesis [13]. These methods have certain advantages in static assessment and quantitative analysis, but they are typically based on linear or unidirectional causal relationships, making it difficult to reveal the dynamic evolution and nonlinear feedback between variables in complex systems [14]. In particular, in the context where pollutant emissions are influenced by multiple factors, these traditional methods struggle to simulate the dynamic response of the system under the combined effects of policy regulation, economic development, and population changes. System dynamics (SD), as an effective tool for simulating the dynamic behavior and feedback mechanisms of complex systems, models the dynamic changes of a system over time through the causal relationships and feedback loops between internal variables [15]. It can reveal the dynamic behavior and development mechanisms of a system [16] and explore the different behavioral patterns formed by the interaction of causal and feedback structures [17]. It is widely applied to solve practical problems characterized by nonlinearity, uncertainty, and dynamic complexity [18]. Currently, research on system dynamics in water resource management has made significant progress, covering multiple aspects such as water quantity, water environment, and ecology [15,19,20]

Xingkai Lake is the largest freshwater lake in East Asia and a border lake between China and Russia, located in Heilongjiang Province and the Primorsky Krai of Russia. It is divided into two parts: Xingkai Lake and Xiaoxingkai Lake. With the expansion of agricultural activities, the rapid development of tourism, and the acceleration of urbanization in the basin [21], the water environmental issues in Xingkai Lake have become increasingly severe [22]. Xiaoxingkai Lake, the “upstream lake” of Xingkai Lake, has overall poor water quality and severe eutrophication [23]. Xiaoxingkai Lake is connected to Xingkai Lake via a floodgate, and there is a certain degree of hydraulic connection between the two. During periods of connectivity, pollutants from Xiaoxingkai Lake can enter Xingkai Lake [24]. Therefore, understanding the load of major pollutants in the Xiaoxingkai Lake basin is crucial for water environment management in the basin and ecological protection of Xingkai Lake, and it also holds significant importance for promoting China–Russia cooperation in water resource management in the Xingkai Lake basin.

This study focuses on the Xiaoxingkai Lake basin within China, examining the challenges of balancing pollution control with socio-economic development in the context of rapid regional growth. Based on the interactive relationships between basin population, economy, and pollutant emissions, a pollution load model based on system dynamics is constructed. The model simulates the trends of four pollution indicators—total nitrogen (TN), total phosphorus (TP), chemical oxygen demand (COD), and ammonia nitrogen (NH₃-N)—entering the river under different scenarios and assesses the impacts of various development patterns on the water environment. The results aim to provide a scientific basis for water resource management and planning in the Xiaoxingkai Lake basin and offer theoretical and methodological references for the integrated management of similar transboundary water bodies.

2. Materials and Methods

2.1. Study Area

Xingkai Lake is located at the border between the Primorsky Krai of Russia and the Heilongjiang Province of China and is the largest freshwater lake in Asia. The northern part of Xingkai Lake belongs to China, while the southern part belongs to Russia. It consists of Xingkai Lake and Xiaoxngkai Lake. Xiaoxingkai Lake (45°16′ N–45°24′ N, 132°20′ E–132°50′ E) is located in the northern part of Xingkai Lake and is an inland lake in China [25]. Xingkai Lake and Xiaoxingkai Lake are separated by natural sand dunes and are connected via a floodgate. During the connection period, pollutants from Xiaoxingkai Lake may enter Xingkai Lake. The main rivers flowing into Xiaoxingkai Lake are mostly located on the northern shore, with the Muling River being the primary water source. The Xiaoxingkai Lake basin has a temperate continental monsoon climate, characterized by long winters and short summers. The annual average temperature is 3.1 °C, with temperatures below freezing from October to April. Annual precipitation averages 574 mm, primarily concentrated during the rainy season from May to September [26]. The Xiaoxingkai Lake basin encompasses Jixi City and Muling City (Figure 1). The total population of the basin in 2020 was 1,933,700, including 696,800 agricultural residents and 1,236,900 non-agricultural residents, with an overall urbanization rate of 63.97%. The regional gross domestic product (GDP) was 70.357 billion yuan, with a per capita GDP of 36,384 yuan. The added value of the primary, secondary, and tertiary industries was 24.56 billion, 18.355 billion, and 27.433 billion yuan, respectively. Land use is primarily dominated by cropland and forestland [27].

2.2. Data Sources

The basin boundary data were downloaded from the Hydro SHEDS database (https://www.hydrosheds.org/) as HYDROBASIN level 6. Socio-economic data, including population, livestock numbers, and cropland area, were obtained from the “Jixi City Statistical Yearbook (2009–2021)” [29] and the “Mudanjiang City Statistical Yearbook (2009–2021)” [30]. Water environment-related data were obtained from the “Heilongjiang Province Ecological Environment Status Bulletin” [31].

2.3. System Dynamics Modeling

2.3.1. Model Boundaries and Structure

System dynamics is a scientific discipline that integrates system science theory with computer simulation to study the feedback structures and behaviors of systems. Based on nonlinear dynamics theory, it combines qualitative and quantitative analysis and employs systematic reasoning methods to address complex system issues. This study takes the Xiaoxingkai Lake basin as the system spatial boundary, combining administrative divisions to define the study area as the entire city of Jixi and Muling City under the jurisdiction of Mudanjiang City, with a system area of approximately 28,500 km². The system simulation period, from 2010 to 2030, is divided into two phases: the first phase from 2010 to 2020 (historical years) and the second phase from 2021 to 2030 (projected years). The simulation time step is set to one year.

In the study area, TN, TP, COD, and NH₃-N mainly originate from agricultural and domestic sources, while industrial sources account for a relatively small proportion. Based on this, the water pollution load system of the Xiaoxingkai Lake basin is divided into three subsystems: the population subsystem, the agricultural subsystem, and the water pollution subsystem. The causal relationships within the system are shown in Figure 2.

Based on the established causal relationship diagram, the functional relationships between the variables are processed, and the level variables, rate variables, auxiliary variables, constants, and table functions in the model are defined. A pollution load SD stock flow diagram for the Xiaoxingkai Lake basin (using TN as an example) is constructed, as shown in Figure 3.

2.3.2. Model Parameters and Equations

The SD simulation model includes multiple variables. The main relevant parameters and system dynamics equations used in this study are shown in

Table 1. Main model parameters and equation settings.

Subsystem	Variables	Variable	Units	Equations and Values
Population subsystem	Total population	Level	×104 persons	INTEG (Number of births − Number of deaths − Out-migrant Population + In-migrant population, 218)
	Number of births	Auxiliary		Total population × Birth rate
	Number of deaths			Total population × Mortality rate
	Birth rate	Rate	‰	Table (time)
	Out-migrant population	Auxiliary	×104 persons
	Urban population			Total population × Urbanization rate
	Urbanization rate	Rate	%	Table (time)
Water contamination subsystem	Urban domestic sewage generation	Auxiliary	t	Urban population × Pollution reduction coefficient × Per capita water consumption of urban residents × 365/1000
	Urban domestic TN generation			TN generation coefficient of urban residents × Urban domestic sewage discharge volume/100
	Rural domestic TN generation			Rural population × TN generation coefficient of rural residents × 365/1000
	Urban domestic TN output			Urban domestic TN generation × (1 − TN removal rate × Urban domestic sewage centralized treatment rate)
	Rural domestic TN output			Rural domestic TN generation × (1 − TN comprehensive removal rate × rural domestic wastewater treatment rate)
	Urban domestic TN load to river			Urban domestic TN output × Pollutant discharge coefficient of urban residents
	Rural domestic TN load to river			Rural domestic TN output × Pollutant discharge coefficient of rural residents
Agricultural subsystem	Cultivated land area	Level	×104 ha	INTEG (change in cultivated land area, 59.63)
	Change in cultivated land area	Auxiliary		Cultivated land area × change in cultivated land area rate
	Cultivated land area change rate	Rate	%	Table (time)
	Livestock stock	Level	×104 heads	INTEG (change in livestock farming volume, 330.80)
	Change in livestock farming	Auxiliary		Livestock stock × Livestock stock change rate
	Livestock stock change rate	Rate	%	Table (time)
Water contamination subsystem	TN output from livestock rearing	Auxiliary	t	Livestock stock × TN output coefficient of livestock stock
	Livestock rearing TN load to river			TN output from livestock rearing × Pollutant discharge coefficient of livestock rearing
	Total TN load to river			Agricultural TN load to river + Rural domestic TN load to river + Urban domestic TN load to river
	Agricultural TN load to river			Cultivated land TN load to river + Livestock rearing TN load to river

.

2.3.3. Model Validity Testing

After the model is constructed, it must be validated [32] to ensure that the simulation results are consistent with reality and can be used for subsequent scenario simulations [33]. Among these, model error validation is the most direct method for assessing whether the model meets simulation requirements. Historical validation, as a common and effective system dynamics validation method, offers advantages such as strong operability, relatively easy data acquisition, and intuitive validation results. By verifying the consistency between simulated values and actual historical data [34], one can determine whether the model structure is reasonable and whether parameter settings are accurate, thereby further validating the model’s usability and reliability. The calculation formula is shown in (1):

$$C_i=\frac{R_i-r_i}{r_i}\times 100\%$$

(1)

Among them, C_i represents the relative error of the variable; R_i and r_i represent the simulated results and historical data values of the variable, respectively.

2.3.4. Scenario Design

To analyze the trends in pollutant load changes in the Xiaoxingkai Lake basin under different development scenarios, key parameters in the system dynamics (SD) model were adjusted with reference to the “Jixi City’s 14th Five-Year Plan for Environmental Protection” [35], the “Master Plan for the Territorial Space of Jixi City (2021–2035)” [36], the “Mudanjiang City’s 14th Five-Year Plan for Environmental Protection” [37], and the “Master Plan for the Territorial Space of Muling County (2021–2035)” [38], combined with regional development expectations. Five scenario modes were designed: status quo continuation, economic development scenario, pollution control enhancement scenario, dual-reinforcement scenario, and moderate-reinforcement scenario. By comparing the simulation results of the system dynamics model under different schemes, while considering pollutant loads in the Xiaoxingkai Lake basin under different schemes, specific parameter settings are shown in

Table 2. Scenario design parameters.

Parameters	Units	S0	S1	S2	S3	S4
Urbanization rate	%	63.97	68	63.97	68	66
Birth rate	‰	3.46	4.5	3.46	4.5	3.8
In-migrant population	×104 persons	0	5	0	5	2.5
Out-migrant population		1.2	0.2	1.2	0.2	0.5
Urban domestic TN removal rate	%	75	75	77	77	76
Urban domestic TP removal rate		84	84	87	87	86
Urban domestic COD removal rate		84	84	87	87	86
Urban domestic NH3-N removal rate		83	83	85	85	84
Rural domestic TN comprehensive removal rate		46	46	48	48	47
Rural domestic TP comprehensive removal rate		46	46	48	48	47
Rural domestic COD comprehensive removal rate		61	61	64	64	63
Rural domestic NH3-N comprehensive removal rate		50	50	53	53	52
Livestock farming TN output coefficient	kg/piece·a	0.6	0.6	0.56	0.56	0.58
Livestock farming TP output coefficient		0.14	0.14	0.13	0.13	0.14
Livestock farming COD output coefficient		2.66	2.66	2.5	2.5	2.58
Livestock farming NH3-N output coefficient		0.21	0.21	0.2	0.2	0.2
Arable land TN output coefficient	kg/(hm2·a)	26.19	27.76	24.88	26.45	25.93
Arable land TP output coefficient		0.99	1.05	0.94	1	0.98
Arable land COD output coefficient		18.99	20.13	18.04	19.18	18.8
Arable land NH3-N output coefficient		4.33	4.59	4.11	4.37	4.29
Rural domestic sewage treatment rate	%	12.5	12.5	50	50	45
Urban domestic sewage centralized treatment rate		62	62	72	72	70
Livestock farming change rate		0.04	0.15	0.04	0.15	0.095
Per capita water consumption of urban residents	L/person·day	115	150	115	150	130

:

(1)

Status quo scenario (S0): The development of the Xingkai Lake basin will remain unchanged from 2021 to 2030. The parameters of the plan will be based on the 2020 values, and the changes in pollutant loads during the period from 2021 to 2030 will be simulated and used as a reference for other scenarios.

(2)

Economic development scenario (S1): Based on the status quo scenario, this scenario assumes that the study area prioritizes economic development, with economic development at a high level and pollution control measures remaining unchanged. This scenario mainly reflects social and economic development by increasing parameters such as livestock and poultry farming scale, population growth rate, and domestic water consumption and simulates the amount of pollutants entering rivers under conditions of rapid economic development.

(3)

Pollution control enhancement scenario (S2): Building on the status quo type, this approach prioritizes environmental protection as its primary objective. It primarily achieves this by increasing the sewage treatment rates for urban and rural domestic wastewater, as well as the treatment rates for domestic pollutants in urban and rural areas. Additionally, agricultural pollution control is enhanced by employing technological measures to reduce the pollutant output coefficients from livestock farming, thereby maintaining environmental protection at a high level. This approach simulates pollutant inflows into rivers under conditions of stringent pollutant control.

(4)

Dual-reinforcement scenario (S3): This scenario combines high economic growth and strong environmental protection, integrating high-intensity parameter settings from both the economic development and pollution control scenarios. It assumes an increase in industrial scale and population growth while implementing stringent pollutant reduction measures. It simulates pollutant loads under conditions of rapid economic growth combined with intensive pollution control.

(5)

Moderate-reinforcement scenario (S4): This scenario represents a balanced approach with moderate levels of economic growth and environmental protection. Compared to the economic development scenario, the growth rates of livestock and poultry farming and population are moderately reduced; compared to the pollution control scenario, the decline in pollutant output coefficients is slowed down. This scenario simulates pollutant loads under conditions of moderate economic growth combined with moderate pollution control. This scenario aims to simulate the common “economic-environmental trade-off” approach found in real-world basin planning, balancing the sustainability of regional economic development with the costs and feasibility of environmental governance, aligning with the governance pathways frequently adopted in policy-making practices.

3. Results

3.1. Model Validity Verification

Based on statistical data from the statistical yearbooks of Jixi City and Mudanjiang City, the total population, rural population, urban population, and cropland area from 2010 to 2020 were selected as historical verification indicators. Historical data were compared to simulation results, and Formula (1) was used to verify the rationality of the model system structure and the accuracy of the simulation results. The specific results are shown in Figure 4.

According to the test results in Figure 4, both the historical and simulated values of the total population show a declining trend year by year, with a relative error ranging from −1.56% to 0.12%. The historical and simulated values of the rural population also show a declining trend year by year, with a relative error ranging from −1.57% to 0.1%. The urban population shows a fluctuating downward trend, with a relative error ranging from −1.56% to 0.13%; the cultivated land area showed little change and remained at a relatively low level before 2016 but saw a significant increase in 2016. The relative error between the simulated values and historical statistical values is relatively large, ranging from −3.9% to 3.5%, but it remains within the acceptable error range. In summary, the relative error between the simulated values and historical statistical values for the selected variables is less than 5%, and their trends are similar, indicating that the simulation model constructed in this study has high accuracy and reliability.

3.2. Contribution Analysis of Contamination Sources

Based on the constructed system dynamics model, simulations were conducted to estimate the riverine loads of four pollution indicators—TN, TP, COD, and NH₃-N—and the contribution rates of different pollution sources for the year 2020. The results are presented in Figure 5. The riverine loads of TN, TP, COD, and NH₃-N were 4100.7 t, 265.7 t, 10,445 t, and 1157.5 t, respectively. The primary source of TN inflow load is agricultural cultivation, accounting for 70.4% of the contribution rate, followed by urban domestic activities and livestock farming, with contribution rates of 18.6% and 9.1%, respectively. Rural domestic activities contribute the least, at 2.0%. The primary source of TP pollution load is agricultural cultivation, accounting for 41.1%, followed by livestock farming and urban domestic activities, contributing 32.6% and 23.3%, respectively; COD primarily originates from urban domestic activities, accounting for 46.9%, with rural domestic activities, agricultural cultivation, and livestock farming contributing 17.3%, 20.0%, and 15.8%, respectively. NH₃-N primarily originates from urban domestic activities (44.7%) and agricultural cultivation (41.2%), with livestock farming and rural domestic activities contributing 11.2% and 2.8%, respectively.

Overall, there are significant differences in the impact of different pollution sources on these four pollutants. TN and TP pollutants primarily originate from agricultural non-point source pollution, indicating that agricultural activities dominate nitrogen and phosphorus loads. COD primarily originates from domestic sources, indicating that domestic wastewater is the core source of organic pollution in the watershed. NH₃-N is jointly dominated by both sources.

3.3. Changes in Pollution Load Under Different Circumstances

Based on the constructed system dynamics model, the trends in the four types of pollution indicators entering the river under five scenarios for the 2020–2030 period in the Xiaoxingkai Lake basin are shown in Figure 6.

Under the current development scenario (S0), the annual discharge of four pollutants into the river basin has been increasing year by year. Specifically, from 2020 to 2030, the annual discharge of total nitrogen (TN) into the Xiaoxingkai Lake basin increased from 4100.7 tons to 4325.7 tons, with an average annual growth rate of 0.54%. TP will increase from 265.7 tons to 306.6 tons, with an average annual growth rate of 1.44%. COD will increase from 10,450 tons to 10,742.8 tons by 2030, with an average annual growth rate of 0.28%. The river inflow of NH₃-N increased from 1157.5 tons to 1191.6 tons, with an annual growth rate of 0.29%. Among these, the growth rates of TP and TN were relatively faster, while those of COD and NH₃-N were relatively slower. This indicates that future water environmental issues in the basin will primarily focus on agricultural non-point source pollution.

Under the economic development scenario (S1), the river inflow of all pollutants is higher than under the status quo continuation scenario and shows a significant upward trend. The river inflow of TN increased from 4100.7 tons to 5350.4 tons, with an average annual growth rate of 2.7%; the average annual growth rate of TP was 5.27%, ranging from 265.7 to 444 tons; COD shows an annual growth rate of 4.05%, increasing from 10,450 tons in 2020 to 15,544.6 tons in 2030; and NH₃-N shows an annual growth rate of 3.67%, increasing from 1157.5 to 1660.6 tons. The annual growth rates of all four pollutants are relatively high, indicating that under conditions of rapid economic development and unchanged pollution control measures, the burden on water environments will intensify.

Under the pollution control enhancement scenario (S2), the annual discharge of TP into rivers shows a slow upward trend, with an average annual growth rate of 0.44%, increasing from 265.7 t to 277.8 t; meanwhile, TN, COD, and NH₃-N exhibit a decreasing trend year by year, with average annual growth rates of −0.23%, −1.41%, and −0.91%, respectively. The annual TN inflow decreased from 4100.7 t to 4008.5 t, COD decreased from 10,450 t to 9069 t, and NH₃-N inflow decreased from 1157.5 t to 1056.2 t. Under this scenario, maintaining the same economic growth rate, improving domestic wastewater treatment rates, and reducing pollutant output coefficients have a good effect on pollutant control in the Xiaoxingkai Lake basin.

Under the dual-reinforcement scenario (S3), pollutant inputs into rivers fall between the economic development scenario and the status quo scenarios and also show a significant upward trend. Specifically, TN inputs increased from 4100.7 t to 4932.8 t, with an average annual growth rate of 1.86%; TP shows an average annual growth rate of 4.19%, ranging from 265.7 t to 400.6 t; COD increased at an average annual rate of 2.31%, from 10,450 tons in 2020 to 13,127.3 tons in 2030; and NH₃-N increased at an average annual rate of 2.34%, from 1157.5 to 1458.8 tons. Under this scenario, environmental governance measures are significantly strengthened, while the intensity of socio-economic activities also increases significantly. Pollutant inputs into rivers show an upward trend, but the growth rate is slower compared to the economically driven scenario, indicating that pollution control measures can, to some extent, mitigate the growth of pollution loads.

Under the moderate-reinforcement scenario (S4), pollutant inputs into rivers show a slow upward trend, falling between the dual-reinforcement scenario and the status quo scenario. Among these, TN inflows increased to 4515.6 t, with an annual growth rate of 0.97%; TP inflows increased to 342.5 t, with an annual growth rate of 2.57%; COD increased to 11,326.2 t, with an annual growth rate of 0.81%; and NH₃-N increased at an annual growth rate of 0.93%, ranging from 1157.5 to 1269.5. In the early stage of the simulation, the river inflow of COD and NH₃-N was slightly lower than the status quo model, but in the later stage, the river inflow of pollutants showed a significant increase. This indicates that moderate-intensity governance measures have a certain effect in the early stage under moderate-intensity socio-economic development, but further strengthening of pollution control is needed in the later stage.

4. Discussion

4.1. Sources and Countermeasures of Basin Pollution Load

Based on the system dynamics simulation of the riverine input of four pollution indicators (TN, TP, COD, and NH₃-N), this study analyzes the pollution sources in the Xiaoxingkai Lake basin from agricultural cultivation, livestock breeding, urban domestic activities, and rural domestic activities. Agricultural cultivation is the dominant contributor to TN, TP, and NH₃-N input. In China, the average utilization rates of nitrogen and phosphorus fertilizers are approximately 30–35% and 10–20%, respectively [39], indicating relatively low fertilizer use efficiency. The nitrogen and phosphorus not fully absorbed by crops can enter water bodies through surface runoff and infiltration, leading to continuous increases in non-point source pollution loads [40] and, consequently, deteriorating lake water quality. In addition, livestock breeding contributes significantly to TP pollution. Livestock and poultry manure are rich in both organic and inorganic phosphorus [41]. The uncontrolled discharge of manure facilitates the accumulation of phosphorus in the environment, exacerbating TP pollution [42]. Urban domestic activities are major sources of COD and NH₃-N. Urban sewage contains large amounts of organic matter. Although wastewater treatment facilities have been established, they cannot completely remove pollutants from domestic sewage [43], thus posing threats to the water environment of the basin. To mitigate pollution from agricultural cultivation, it is essential to promote scientific fertilization practices, optimize fertilizer application rates, and reduce pesticide use. Designating planting zones and establishing vegetative buffer strips can also help minimize pollution. For livestock-related pollution, a circular agricultural model should be promoted by integrating crop and livestock production. Manure should be collected and processed to convert it into usable resources, such as organic fertilizers, thereby enhancing resource recycling and strengthening the linkage between livestock and crop farming. Regarding pollution from domestic sources, efforts should focus on advancing treatment technologies and improving pollutant removal efficiency. Although rural domestic sources contribute less to overall pollution, the centralized treatment rate of rural sewage in the Xiaoxingkai Lake basin remains low. Measures should be taken to establish effective collection systems, develop targeted treatment standards, and improve solid waste management infrastructure in rural areas.

4.2. Scenario Comparison and Future Regulatory Recommendations

The economic development scenario prioritizes rapid socio-economic development and promotes population concentration through regional economic growth. Under this scenario, livestock and poultry farming numbers increase significantly, and population growth drives an increase in urban domestic water consumption. However, this scenario neglects environmental protection and fails to simultaneously enhance pollutant control measures, leading to a significant increase in pollutant discharge into rivers. While short-term economic benefits are evident, pollutant emissions continue to rise, resulting in a sustained decline in environmental quality. Continuous deterioration of environmental quality will, in turn, inhibit socio-economic development [44]; for example, declining water quality will affect the safety of agricultural irrigation and residential water use [45]. In the long term, environmental degradation will become a key factor limiting regional socio-economic development. Therefore, single-minded economic expansion cannot achieve coordinated development of socio-economic and ecological protection.

The pollution control enhancement scenario focuses on pollutant control and achieves good control of pollutant emissions. Under this scenario, the growth rates of the four categories of pollution severity indicators are low, with some pollution indicators showing a downward trend, resulting in significant environmental benefits. However, this approach has limited promotional effects on socio-economic development, as it continues the economic development model of the base year (2020), which may hinder regional economic growth. Additionally, stringent pollution control measures require substantial fiscal investment [46], resulting in high implementation costs. Therefore, this scenario is more suitable for ecologically fragile regions or areas with urgent environmental restoration needs.

The dual-reinforcement scenario simultaneously promotes economic growth and pollutant control to achieve coordinated development of socio-economic and environmental protection. Under this scenario, while ensuring rapid socio-economic development, strengthening pollution control can, to some extent, achieve a balance between economic growth and pollution control. However, simultaneously strengthening socio-economic and environmental protection may lead to relatively high fiscal pressures and governance costs [47], making it less feasible. Under high-intensity pollution control measures, the regional pollutant emission growth rate remains significant, resulting in low environmental benefits. While initial governance efforts yield significant results, the diminishing marginal benefits of pollution reduction lead to rising governance costs in later stages, resulting in a decline in per-unit environmental benefits [48]. As shown in Figure 6, although the dual-reinforcement scenario demonstrates good pollution control effects in the early stages of simulation, the pollutant load entering rivers significantly increases in later stages, second only to the economically driven scenario. This indicates that despite the implementation of high-intensity environmental governance, the effectiveness of pollutant control is limited in the later stages under high-intensity economic development, and the contradiction between socio-economic development and the environment becomes increasingly prominent in the later stages. Therefore, the overall practical feasibility of this scenario is relatively low.

In contrast, the moderate-reinforcement scenario moderately controls the pace of economic expansion to avoid exacerbating pollution due to rapid socio-economic development while advancing pollution control measures, thereby achieving effective control of pollution loads. Compared to the dual-intensive scenario, this scenario involves fewer pollution control efforts, resulting in lower implementation costs. However, the growth rate of pollution loads entering rivers is significantly lower, and the scenario continues to demonstrate good pollution control effects in the later stages of the simulation. This indicates that under this scenario, socio-economic development maintains orderly growth while also addressing environmental protection needs, with good pollution control outcomes and higher operational feasibility and practical adaptability. Therefore, under conditions of limited resource allocation, this scenario offers relatively higher cost-effectiveness and implementation feasibility, making it an optimal choice for currently coordinating the advancement of economic and environmental objectives in the region.

In conjunction with the “Jixi City National Economic and Social Development 14th Five-Year Plan and 2035 Long-Term Goals Outline” “https://www.jixi.gov.cn/jixi/c100328/202103/c06_1024.shtml (accessed on 30 July 2025)”, the current development planning priorities for the Xiaoxingkai Lake basin focus on promoting economic growth and infrastructure construction while increasing environmental governance investments. The overall development path aligns more closely with the moderate-reinforcement scenario development model. This scenario achieves a dynamic balance between pollutant control and economic growth. However, it is worth noting that, compared to the baseline scenario (S0), pollutant inflows into rivers still show an upward trend in this scenario, and the gap in pollutant inflows between the two scenarios widens over time. Therefore, in the later stages of the simulation, efforts should be further intensified to strengthen pollution control, with a focus on controlling point source pollution while shifting the emphasis to addressing agricultural non-point source pollution, thereby achieving true coordination between socio-economic development and environmental protection.

Further integration of land use status, population density, cropland ratio, and livestock density within the basin is essential for a comprehensive assessment of suitable development scenarios for each region (Figure 1b,c). In the central part of the basin, the urban areas of Jixi City, Jidong County, and Mishan City exhibit high livestock density and population density. Additionally, Mishan City, which is the county where Xiaoxingkai Lake and Xingkai Lake are located, boasts extensive cropland. It is recommended that this region prioritize the implementation of a moderate-reinforcement or pollution control enhancement scenario to achieve synergistic development between pollution control and agricultural economics. Additionally, Muling City, as an upstream region in the basin with high livestock farming density, should focus on controlling livestock farming pollution. Hulin City, as a downstream region in the basin, has a lower overall population density and livestock farming density, and can appropriately implement an economically prioritized scenario model.

The pollution control measures in the scenario settings are all focused on source control. However, due to the unique geographical location of the Xiaoxingkai Lake basin, it is necessary to further integrate measures such as intercepting pollutants along transmission pathways and end-of-pipe treatment. For example, establishing ecological buffer zones around Xingkai Lake, diverting farm drainage into ecological ditches, and constructing artificial wetlands can more comprehensively achieve pollution control in the basin through source control, process interception, and end-of-pipe treatment. As a transboundary lake, to achieve overall ecological security in the basin, it is necessary to promote the establishment of a China–Russia cross-border water body joint governance mechanism. This includes strengthening data sharing between the two sides in water quality monitoring and pollution flux estimation, constructing joint automatic water quality monitoring stations, and jointly establishing cross-border ecological buffer zones based on different land use patterns to promote China–Russia collaborative management of water resources in the Xingkai Lake basin.

5. Conclusions

This study employs system dynamics to construct a “socio-economic-water environment” model for the Xiaoxingkai Lake basin, systematically analyzing the trends in typical pollution indicators entering the river under five different development scenarios. The results indicate that, under the current development model, pollutant levels in the basin continue to rise, and water environmental pressure continues to intensify. In particular, under the economic development scenario, regional economic expansion leads to a significant increase in pollutant emissions, posing a serious threat to the water environment. In contrast, under the pollution control enhancement scenario, pollution control is highly effective, with a significant decrease in typical pollutant loads entering the river and a noticeable improvement in water environmental quality, although this slows economic development. Under the “dual-reinforcement scenario” and “moderate-reinforcement scenario” comprehensive scenarios, a certain balance is achieved between environmental and economic objectives, with the moderate-reinforcement scenario demonstrating stronger feasibility. This study not only reveals the future trends in pollutant loads in the Xiaoxingkai Lake basin but also provides references for regional water environment governance and comprehensive pollution prevention and control.

It should be noted that this study endeavors to incorporate as many factors influencing pollutant inputs into rivers as possible, including population growth and agricultural activities. However, it cannot encompass all elements, leading to certain factors being excluded from the model, such as the impact of climate change on pollution loads, the enforcement strength of local government policies, and farmers’ responses to policies. These behaviors often play a crucial role in basin governance. Additionally, system dynamics models have drawbacks, such as high parameter uncertainty (pollutant output coefficients) and weak spatial representation capabilities. To enhance the accuracy and applicability of future research, machine learning methods can be employed to assist in parameter estimation, thereby improving model precision. Additionally, integrating GIS technology can enhance the spatial simulation capabilities of system dynamics models.