Evaluation of water resources carrying capacity in the Yellow River Basin: a Hu Huanyong Line perspective Open Access

As a significant water source area in north and northwest China, the Yellow River Basin has a vital position in China's social economy development and ecological conservation pattern (Liu et al. 2022; Zhou et al. 2022). The ‘high-quality social development and ecological environment conservation in the Yellow River Basin’ was proposed in 2018, indicating that the watershed has ushered in many new development opportunities (Wang et al. 2023). However, there are still many water resource problems in the basin, such as flood control (Ran et al. 2020), decreasing incoming water (Wang & Sun 2021), excessive groundwater exploitation (Lu et al. 2022), and low water-use efficiency (Zhang et al. 2024). The industrial structure within the basin exhibits a high proportion of secondary industry, especially heavy industry. This has resulted in a fragile ecological environment, and the resource utilization does not align with the capacity of the natural environment.

In this study, the Hu Huanyong Line serves as the core demarcation line for the WRCC zoning, mainly based on the integrated divergence law of natural and human elements revealed by the line. The Hu Huanyong Line was proposed in 1935 (Hu 1935) and was initially used to explain the differences in population distribution between the northwest and southeast regions of China. With the deepening of multidisciplinary research, the significance of the Hu Line has gradually expanded from population geography to socio-ecological analysis. There are systematic differences in natural resource endowment, socio-economic conditions, and carrying capacity on both sides of the line (Huang et al. 2017; Zhong & Sun 2018). In China, the 400 mm annual average isodic precipitation line roughly overlaps with the Hu Line. The difference between the arid climate on the west side of the line and the monsoon climate on the east side leads to obvious differences in precipitation and evaporation, which directly influence the regional water resources status. Compared with the traditional hydrological zoning, which focuses on the physical attributes of water balance within the watershed unit, the research perspective based on the Hu Line integrates the spatial characteristics of climate, topography, and human activities and can comprehensively reflect the spatial coupling between the water resources system and socio-economic activities behind the WRCC. Studies have shown that the Hu Huanyong transition zone, as an ecologically sensitive area, has a strong impact on the aridification trend of terrestrial ecosystems and hydrological conditions, which ultimately affects human–land relations (Li et al. 2024), which illustrates the unique value of this framework in explaining the complex interactions between human–land systems. However, most of the existing studies focus on the role of the Hu Line in population distribution (Gao et al. 2017), urbanization (Hu et al. 2016), economic development (Sha 2023), and environmental impacts (Lou et al. 2024), but the explanation of spatial differentiation of WRCC is still insufficient. Therefore, this study builds a framework based on the Hu Line, aiming to reveal the spatial heterogeneity of human–water relations that is difficult to show by traditional hydrological boundaries.

The concept of ‘carrying capacity’ was first proposed in the study of community ecology (Park & Burgess 1924). Since the middle of the 20th century, people's disorderly exploitation of fossil fuels and natural resources has triggered various ecological problems (Mou et al. 2020; Shao et al. 2024). In the background, social development sustainability and environmental carrying capacity have received increasing attention. In the 1980s, the definition of ‘resource-carrying capacity’ was proposed (UNESCO 1985). With the further development of carrying capacity research, water resources carrying capacity (WRCC)-related research has gradually emerged. A study in Xinjiang, China, first applied the concept of WRCC in analyzing the water supply–demand relationship (Ye 1992). In recent decades, many scholars have put forward their opinions on the concept and connotation of WRCC. Most of them emphasize the maximum population that regional water resources can support, the maximum amount of regional water resources that can be developed, and the ability of water resources to support the economy and society (Wu et al. 2020; Dai et al. 2022). This study defines WRCC as the maximum scale of the water resources system that can support the sustainable development of a region's population, economy, and ecology based on the regional natural environment and socio-economic development.

WRCC assessment frameworks generally follow two methodological paradigms. The first adopts a human–environment nexus perspective, integrating hydrological, socio-economic, and ecological dimensions. Studies include Lv et al. (2023) applying enhanced Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) modeling in Heilongjiang's eastern region, Zhang et al. (2019) combining principal component analysis with nature-based solutions in the Xiangjiang Basin, and Yang et al. (2019) utilizing analytic hierarchy process (AHP)-SD (system dynamics) hybrid modeling in Xi'an. The second paradigm employs pressure–state–response (PSR) derivatives, exemplified by Zhang & Duan's (2024) driver–pressure–impact framework for the Pearl–Xijiang Economic Belt and Zhao, Wang et al.'s (2021) pressure–support–degradation matrix for the Beijing–Tianjin–Hebei region. Despite methodological advancements, current systems demonstrate regional specificity without establishing unified analytical standards.

Existing quantitative methods for WRCC evaluation primarily include the TOPSIS method (Deng et al. 2021), the entropy weight method (Wang et al. 2018), the analytic hierarchy process (Yang et al. 2019; Ren et al. 2020), neural networks (Sun et al. 2023; Zhang et al. 2022), and system dynamics modeling (Tian et al. 2021). While prior studies in the Yellow River Basin have focused on provincial-level WRCC comparisons, two critical gaps persist: long-term temporal evaluations and main obstacle factor identification. Addressing these limitations, this study establishes a novel evaluation framework through three methodological innovations: first, we develop a 14-indicator WRCC evaluation system based on Hu Line zoning, encompassing water resources, socio-economic factors, residential life, and ecological dimensions. Second, an integrated weighting approach combining entropy weight and AHP methods is employed to enhance indicator weighting accuracy. Third, the TOPSIS method with combined weights is applied to systematically analyze spatiotemporal patterns and quantitatively identify key obstacle factors. These methodological advancements provide critical insights for optimizing water resource allocation strategies and enhancing WRCC in the Yellow River Basin.

Taking into account the natural boundaries of the basin, the correlation between regional economic development and the Yellow River, as well as the results of existing research (Zhao, Hou et al. 2021; Zhou et al. 2022), the planning water-receiving area of the basin was identified as the region of interest. The region includes 70 cities across nine provinces through which the Yellow River passes, with a gross area of about 2,154,000 km² and a population of 216 million in 2020. The climate within the watershed is influenced by monsoons and altitude, with four climate types: arid, semi-arid, humid, and semi-humid. Most of the study area is in arid regions. The spatial distribution of water varies markedly: under the influence of the monsoon, water resources are more abundant in the south western and easternmost parts of the region, while they are scarce in the northwestern area. The mismatch between water demand and supply in the basin has long been prominent. When it comes to agricultural resources, the planning water-receiving area is well endowed with sunlight and heat and has a good foundation for agricultural development. The Hetao Plain and the Yellow Huaihai Plain in the region are important agricultural areas in China. Constrained by climatic conditions, the agriculture in the region is mainly dry farming, and most of the irrigation water comes from groundwater and the Yellow River. In terms of mineral resources, the study area is rich in coal, petroleum, rock gas, and various metals. In 2013, China identified 262 resource-oriented cities, and nearly half of the cities in the watershed were listed (Xu & Shu 2023).

The datasets are selected from authoritative national data, such as the Urban Statistical Yearbook of China, the Urban Construction Statistical Yearbook of China, the Environmental Statistical Yearbook, Water Resources Bulletin, and National Economic and Social Development Statistical Bulletin of the provinces and municipalities in the watershed. The study period is from 2011 to 2020, and the minimum computational unit of this study is the municipal administrative region.

A synthetic analysis was conducted on the natural conditions, water utilization, and ecological conditions of the planning water-receiving areas in the watershed. Referring to existing achievements (Lv et al. 2023; Wei et al. 2023), based on the four criteria of water status, economic and social development capacity, residential life, and eco-environment construction, 14 indicators were selected to build an assessment indicator system for the WRCC. A positive attribute indicates that the indicator promotes the WRCC, while the opposite suggests that it hinders it. A indicators and their attributes are shown in Table 1.

The AHP is a multi-objective decision analysis method. The method regards a complex decision problem as a system, which is decomposed into multiple objectives, criteria, or indicators. It then forms a structured model with several levels to realize the systematic and structured analysis of the problem (Rios & Duarte 2021).

The entropy weighting method is an objective weighting method based on information theory, which is mainly used to determine the weight of each index in the multi-index comprehensive evaluation problem. Its core lies in the use of information entropy to measure the uncertainty or dispersion of index information and assign weights accordingly (Mon et al. 1994).

To verify the robustness of the method, Spearman correlation analysis is used to calculate the weighting results of the four weighting schemes (pure AHP, pure entropy weighting, additive integration, and multiplicative integration); and the Kendall test is used to validate the ranking of the city's WRCC value and the obstacle factors. The results show that there is a correlation between the weight calculation results under different weighting schemes; both city and barrier factor rankings have a strong consistency (statistical correlation >0.70). Therefore, despite the theoretical differences in the weighting methods, the core conclusions are not sensitive to changes in the weights, which verify the reliability of the research results.

The TOPSIS method is a systematic evaluation method of multi-index and multi-scheme. It constructs positive ideal solutions and negative ideal solutions in order, calculates the distance between each evaluation scheme and the two ideal solutions, and then calculates the proximity (C_i) (Chen et al. 2019). The C_i value can reflect the proximity of each program to the optimal program; the closer the value is to 1, the higher the proximity. Referring to the previous relevant research results (Yan et al. 2014; Lu et al. 2022), the C_i value is categorized into five levels: severely overloaded, overloaded, critical state, modest carrying capacity, and excellent carrying capacity, as shown in Table 2.

In the formula, ⁠,⁠, and ⁠, respectively, represent the contribution of the jth indicator to the WRCC, and the deviation and obstacle degrees of the indicators.

Using the AHP and entropy evaluation method, the weight of each index was obtained. Equation (2) was then utilized to derive the combined weights corresponding to each indicator, as shown in Table 3. In terms of the subsystem level, the water resources subsystem has the largest weight of 0.455, followed by the eco-environment meta system (0.413) and socio-economic meta system (0.100); and the residents' life subsystem has the smallest weight, accounting for 0.032 only. The top three indicators are available water resources per capita (C1), per capita consumption of water (C2), and ecological water use rate (C14), indicating that they have the most obvious influence on the WRCC. This coincides with the shortage of water resources, the large population, and the problem of per capita water shortage in the basin.

After calculating the weights of the evaluation indicators, we evaluate the WRCC of the three regions divided based on the Hu Line.

The WRCC of the three regions is ranked as follows: the region west of the line > the region on the line > the region east of the line. The annual average C_i values are 0.096, 0.070 and 0.068, respectively. The C_i value of the area west of the line is more than twice the value of the other two areas in most years. In terms of changing trends, the C_i value in the area west of the line also fluctuated upward, reaching its lowest value in 2013. The C_i values of the area on the line and the area east of the line were relatively close. In 2011, the C_i values of the area east of the line were higher than those of the area on the Hu Line, but in all other years, the C_i values of the region on the line were higher. From 2015 to 2020, the C_i value of the zone on the Hu Line increased, while the C_i value of the zone east of the line decreased, leading to a greater gap between the WRCC of the two areas. When it comes to the carrying capacity of the subregion, the WRCC of the zone west of the line was in Class V from 2011 to 2015; the WRCC of the area increased and was in Class IV from 2016 to 2020. In contrast, the WRCC both of the zone on the Hu Line and of the zone east of the Hu Line were in Class V from 2011 to 2020.

In terms of the variation, for cities in the west of the line whose WRCC is at the top of the Yellow River Basin Planning Water-Receiving Area, their WRCC has increased. Among them, Yushu and Guoluo have significantly improved their WRCC in 2020 compared with 2011, becoming the only two cities in the study area with a WRCC of level I. The WRCC level of Alxa has decreased from level II in 2011 to level III in 2020. The WRCC level of Haixi and Haibei has increased from level IV in 2011 to level III in 2020. In 2011 and 2020, most cities in the middle and eastern regions had a WRCC level of V, which indicated severe overload. In the Hu Line area, only Aba and Ganzi have WRCC that do not belong to level V, and their levels have risen from level III in 2011 to level II in 2020. In the zone east of the Hu Line, only Shangluo, Anyang, and Zhengzhou had the WRCC of Class IV in 2011, while for the rest of the cities it was of Class V. By 2020, the WRCC of these three cities has fallen to Class V.

From 2011 to 2020, the WRCC exhibited distinct spatial patterns: the region west of the Hu Line (0.153) outperformed those on the line (0.067) and east of the line (0.057), with a higher concentration of cities achieving superior WRCC ratings. Notably, areas on and east of the Hu Line faced severe WRCC challenges, with most cities classified as level V. Subregionally, Alxa, Yushu, and Guoluo in the west consistently led in WRCC, while Haixi and Haibei also ranked above most regional peers. On the Hu Line, excluding Aba and Ganzi in the south, all cities it predominantly fell into level V. East of the line, only Zhengzhou reached level IV, with the remaining cities classified as level V.

The obstacle degree analysis of 14 indicators across 70 cities in the Yellow River Basin's water-receiving area (2011–2020) reveals critical constraints: per capita water resources (C1, 43.75) and ecological water use rate (C14, 40.54) emerge as dominant barriers (Table 4). These constraints primarily stem from regional water scarcity and industrial water demand encroaching on ecological allocations. At the guideline-layer level, water resource endowment (B1, 49.03) and ecological protection capacity (B4, 43.03) exerted predominant influences, while social economy (B2, 6.74) and residents' life (B3, 1.20) showed limited impacts. Notably, the spatial–temporal stability of indicator impacts suggests structural deficiencies in water allocation systems and ecological compensation mechanisms, indicating systemic challenges in coordinating basin-wide water–ecology–economic balances.

To identify key constraints on WRCC, an obstacle factor diagnostic model was employed to analyze 14 indicators in the planning water-receiving area. Using a descending order method, the top six indicators (obstacle value >1) were identified as primary barrier factors, reflecting the most significant regional challenges.

Under the combined impacts of global climate change and human activities, the Yellow River Basin has experienced intensifying water resource imbalances and ecosystem degradation. For example, distinct ecological challenges manifest across regions: upper reaches show habitat deterioration, midstream areas suffer severe soil erosion, and downstream regions face compound environmental stresses (Huang et al. 2023).

Given the low WRCC grades of resource-dependent cities, region-specific development policies are essential. West of the Hu Line, the Yellow River's source region should leverage its hydropower resources and ecological security advantages to enhance green development while ensuring sustainable water use and reducing environmental costs. Along the Hu Line, with its dense population and development potential, efforts could focus on soil conservation, industrial upgrading, green mining, and wastewater control, alongside promoting water-saving technologies and optimizing water resource utilization. East of the Hu Line, economically advanced cities should prioritize water efficiency, pollution remediation, and cleaner production technologies, supported by increased investment in water management and conservation initiatives. Additionally, wetland restoration and ecological water allocation should be expanded to balance economic growth with environmental protection.

To address ecological challenges in the Yellow River's Planning Water-Receiving Area, a multi-level regional cooperation mechanism should be established. For example, cities with diverse resource types should promote resource circulation and complementary utilization within their vicinity. For cities sharing similar resource endowments, enhancing information sharing, developing specialized processing industries, and forming distinctive industrial systems are crucial to leveraging their unique strengths and fostering differentiated competitive strategies.

As a crucial energy base, agricultural hub, and ecological barrier in China, the Yellow River Basin holds significant strategic importance. However, its fragile environment, high population density, and water scarcity have constrained the basin's WRCC at a low level. China has prioritized the basin's ecological protection and sustainable development, implementing measures such as water conservation systems, ecological restoration projects, and efficiency initiatives. Recent studies highlight the success of restoration efforts, including the Grain for Green Program and the Three-North Shelter Forest Program, which have significantly improved vegetation cover and water use efficiency in the Loess Plateau (Fan et al. 2024; Ma et al. 2022). The 2019 strategy for ecological protection and high-quality development, followed by the 2020 Yellow River Basin Protection Law, has further advanced coordinated water resource management and water-saving society construction. Moving forward, regional collaboration and context-specific strategies are essential to consolidate governance achievements and enhance the basin's WRCC.

Quantifying the response mechanism of WRCC to climate fluctuations is crucial to understanding the dynamic balance of human water systems in the basin.

Based on the Hu Huanyong Line, this study constructed a four-part evaluation index system of water resources, social economy, residents' life and ecological environment, systematically evaluated the WRCC of 70 cities in the Yellow River Basin Planning Water-Receiving Area from 2011 to 2020, and identified the main obstacle factors by using the obstacle degree model. The main conclusions are as follows:

• (1) Temporal evolution characteristics: The WRCC proximity (C_i) of the study area showed a slightly upward trend in the range of 0.076–0.092, and the carrying capacity index increased by 11.25% in ten years, but the overall situation was in Grade V (severe overload), indicating that the regional water resources overload problem was still serious.

• (2) Spatial differentiation law: There are obvious regional differences in WRCC in the study area. The C_i mean showed the characteristics of the region west of the Hu Line (0.096) > the region on the Hu Line (0.070) > the region east of the Hu Line (0.068).

• (3) System disorder diagnosis: The order of criterion level disorder was water resource system (49.03) > ecosystem (43.03) > economic system (6.74) > living system (1.20). Among them, water resources per capita (C1, 43.75) and ecological water use rate (C14, 40.54) constitute the main obstacle factors, and their combined contribution is 84.29.

• (4) Regional barrier differences: Key barrier factors present spatial heterogeneity. The barrier degree of per capita water resources decreased from 51.87 in the eastern district of the line to 46.43 in the western district of the line, while the barrier degree of the ecological water use rate showed the opposite trend (40.72) in the eastern district and (44.74) in the western district of the line, reflecting the different resource constraints faced by different regions.

The results show that the Yellow River Basin Planning Water-Receiving Area is facing the complex constraints of insufficient water resources endowment and ecosystem pressure. It is suggested to adopt a regional governance strategy: the east region of the line could focus on improving water use efficiency and optimizing water resource allocation. The western region should strengthen the ecological flow guarantee and promote the coordination of human and water systems. The region along the line should try to break the coupling constraint of water resources, economy, and ecology. The ‘dual-core drive’ obstacle factor identification framework proposed in this study can provide a decision-making basis for watershed water resources management.

This study has the following limitations and directions for improvement:

• (1) The data for WRCC indicators primarily come from statistical yearbooks, and the variation in the filling standard of indicator data across different provinces may lead to bias in the assessment. The reliability of the data can be further improved by cross-checking the special bulletins issued by provincial and municipal water resources departments, extending the statistical cycle, and increasing field research in key areas.

• (2) The current use of an entire city as the assessment unit may conceal internal differences. For example, the WRCCs of different counties within the same city may not be consistent, and the water stress in industrial clusters and ecological reserves may be very different. In the future, on the one hand, we will consider obtaining finer spatial-scale data for further WRCC evaluation; on the other hand, we can select typical cities in the basin to make a zoning comparison of ‘urban – suburban’ and carry out a detailed analysis by combining the characteristics of the industrial structure and population distribution of different functional areas.

This study was jointly funded by the China Geology and Mineral Resources Survey Project(DD20230563 and DD20243224), the National Natural Science Foundation of China (42377354), and the Natural Science Foundation of Hubei Province, China (2024AFB951).

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.