Analysis of the spatial and temporal evolution of water and soil resource carrying capacity in arid region of northwest China Open Access

The carrying capacity of water and land resources refers to the maximum capacity of water and soil resources to support social development and ecosystems in a specific region. The arid desert areas of northwest China are rich in land resources and have sufficient light and heat conditions, but water resources are relatively scarce and unevenly located for space. Disharmony of water and soil resources in arid desert areas has been greatly alleviated by water-lifting irrigation, and on this basis, artificial oases have gradually evolved, but the ecological background of the arid desert areas is fragile and very sensitive to the response of water and soil resources development in the context of irrigation. The injection of external water resources has changed the original land resource evolution process in the arid desert areas. The unique climate of low precipitation, high evaporation and large temperature difference have a long-term and stereoscopic driving impact on the background of regional water and soil resources, which drives the evolution of the water and soil resource-bearing state of the artificial oasis in a complex system (Xu 2010; Angelovicová & Fazekašová 2014; Felix et al. 2018; Xu et al. 2019, 2021). As a result, long-term monitoring data, spatial remote sensing data, and other data are needed to reveal the influence process and driving mechanism of each driving factor for the water and soil resource carrying capacity on spatial and temporal scales, and to contribute to the study of soil and water resource carrying capacity evolution in arid areas.

The carrying state of water and soil resources is affected by the multilayer coupling of natural climate, hydrogeology and human activities (Setegn et al. 2009; Lv 2015; Vogeler et al. 2016; Lamsal et al. 2017). Accurately revealing its spatial and temporal evolution patterns, dynamics mechanisms of driving processes and coupling relationships are the hot spots and difficulties in the field of soil and water environment. In recent years, numerous people have conducted extensive research on the carrying capacity of water and soil resources on the regional scale. Among them, Ait-Aoudia & Berezowska-Azzag (2014) analyzed the correlation between water resources and human production activities in Algiers from the perspective of water resources carrying capacity for the population; Motoshita et al. (2020) assessed the sustainability of freshwater water supply on a global scale from the depletion value of the freshwater water carrying capacity in a single regional context; Williams et al. (2017) combined a population growth model and a land resource to food supply model to synthesize an analytical model for quantitative analysis of land use change and population carrying support capacity. Jayanthi et al. (2020) used hierarchical analysis of a multicriteria decision method to conduct a systematic analysis of land resource carrying characteristics in four broad categories: land type, water quality, soil characteristics and infrastructure availability. Jiang et al. (2011) established a projection tracing evaluation model based on a particle swarm optimization algorithm to evaluate the carrying capacity of each water resource unit within the Three Rivers Plain; Ren et al. (2011) technically integrated the principal component analysis method with the projection tracing transformation model and used it in the comprehensive evaluation of the carrying capacity of water and soil resources in the Three Rivers Plain. The above studies have achieved many results, but there are still shortcomings. Firstly, in terms of research objects, the current research mainly takes administrative regions (such as provinces, cities and counties) as evaluation units, and there is little research on the long series of water and soil resources carrying capacity under raster cells, while the visualization of data on raster cells is more obvious; secondly, in terms of research methods, the existing research mainly focuses on the study of water resources and land resources, and the details of their coupling and driving mechanism are not sufficiently revealed. In addition, the exploration of the spatial and temporal evolution of water and soil resources bearing capacity at the regional scale is not systematic enough, and the spatial changes of water and soil resources bearing status and evolutionary flow direction are not sufficiently considered. The cloud model has a unique advantage in portraying the fuzzy and uncertainty problems of fuzzy evaluation systems, and can quantitatively reveal the actual water and soil resource bearing state of the artificial oasis in arid desert areas.

In order to clarify the driving process and spatial and temporal evolution of water and soil resource carrying capacity of the artificial oasis in the arid zone, this paper uses the Jingtaichuan Pumping Irrigation District (hereinafter referred to as Jingdian Irrigation District) in northwestern China as the research object. Realizing the dynamic prediction of water and soil resource carrying capacity under future scenarios by studying the spatial and temporal evolution of water and soil resource carrying capacity. The system dynamics principle and multilevel fuzzy theory are introduced to analyze the driving factors of water and soil resource carrying systems. The evaluation model is constructed based on the cloud model, the golden division method and the combined assignment method, combining long series monitoring data, spatial telemetry and geographic information data to ensure the reliability and comprehensiveness of the evaluation results. Based on the above methods and data, the water and soil resource carrying capacity assessment and spatial and temporal evolution analysis of the study area were carried out. The objectives of this study were: (1) to study the spatial and temporal evolution process and characteristics of the drivers; (2) to evaluate the bearing capacity of water and soil resources and their evolution pattern from 2002 to 2018; (3) to reveal the changed pattern and evolution flow of the bearing state of water and soil resource; (4) to enrich and promote the optimal allocation of water and soil resource in the irrigation area.

This paper analyzes the relevant data since the operation of the irrigation district and investigates the key nodes of the change of water and soil resources carrying capacity, the years 2002, 2010 and 2018 were finally selected as the representative research years. In 2002, the first phase of the irrigation district was completed, and the irrigated area was 38,500 hm², with an annual water-lifting volume of 322 million m³; in 2010, the second phase of the irrigation district was completed, the irrigated area and the annual water-lifting volume increased to 49,200 hm² and 386 million m³; in 2018, the irrigation area was basically the same as its current situation, the irrigated area increased to 60,500 hm² and the annual water-lifting volume increased to 460 million m³ (Liu 2018). In this paper, the data required for the study were obtained by means of long series monitoring, spatial data download and regional information survey. Monitoring data were collected through field surveys and information such as land survey reports were obtained from the irrigation district management. The nature and sources of evaluation indicator data are listed in Table 1.

In order to realize the spatial faceting of long series monitoring data, the source data need to be applied to ArcGIS for spatial interpolation, and the commonly used interpolation methods are divided into several categories such as inverse distance weight method, natural neighbor method, and ordinary kriging method. Using the normal QQ test of Origin9.1 to test the normality of the sample source data, and the results showed that the above index data satisfied the normal distribution. The cross-validation method and error matrix were introduced to verify and analyze the interpolation accuracy of the results (Sawaya et al. 2003; Ma et al. 2018), and the optimal interpolation method for each element index was finally optimized (Table 2).

Water and soil resources carrying capacity change pattern mapping adopts the principle of land class spatial flow change analysis. Based on the statistics of different water and soil resources carrying class mapping, characterizing the ‘space–property–process’ integrated evolution trend and trend of water and soil resources carrying status among different research nodes, which can reveal the pattern of water and soil resources carrying evolution and pattern under the scenario of spatial and temporal dynamic changes at the regional scale. Wang et al. (2011) and Gong et al. (2012) took into account the actual water and soil resources status in each study year in the irrigation area, we used S₁, S₂, S₃, S₄ and S₅ to classify ‘severe bearing zone V₁’, ‘slight bearing zone V₂’, ‘critical bearing zone V₃’, ‘safe bearing zone V₄’ and ‘good bearing zone V₅’ respectively. The spatial and temporal evolution patterns of the irrigation area water and soil resources carrying state mapping are defined into five types (Table 3).

Based on Landsat series remote sensing image data, the vegetation coverage of the irrigation area was calculated using the image dichotomous model. Considering the land use types in the study area, the vegetation coverage was divided into four classes, as listed in Table 4.

From the results (Figure 15) of spatial interpretation of vegetation coverage, it can be seen that the vegetation coverage of the irrigation area increased spatially significantly from 2002 to 2018, while due to the continuous improvement of the structure of irrigation distribution facilities in the Phase II irrigation area, the spatial and temporal trend of the gradual evolution from Phase I to Phase II irrigation area is characterized. The statistical vegetation coverage using the analysis and statistics module of ArcGIS10.2 obtained, we found that the percentage of very low vegetation coverage in the irrigation area decreased by 20.93% and the percentage of high vegetation coverage increased by 24.17% from 2002 to 2018, indicating that the overall trend of vegetation coverage in the irrigation area showed a transition from low cover to high cover.

Consult experts in this field, according to the importance of indicators to determine the index queuing level, by Equation (1) to calculate the process layer, factor layer and state layer index weight (Tables 5–7). According to Table 8, the driving effect of each factor index on the evolution process of water and soil resources carrying the state in irrigation area is as follows: soil–water coordination > groundwater depth > mineralization of groundwater > soil salinity > surface irrigation volume > surface temperature > land contamination load > average annual evaporation > average annual precipitation > vegetation coverage > soil electrical conductivity > elevation > slope. This shows that soil salinity, groundwater depth, groundwater salinity, surface irrigation amount and water and soil coordination degree are the key factors driving the evolution process of water and soil resources carrying state in the irrigation area.

Three experts in this field, four daily staff in the irrigation area and three management personnel are invited to make a combined evaluation meeting. Based on the evaluation layer, the corresponding element indexes of different research nodes are objectively evaluated. According to Figure 7, the cloud model characteristic parameters of the state layer, factor layer and process layer are calculated. The cloud model parameters of compre–hensive evaluation are calculated using Equation (2). The characteristic parameters of the process layer and comprehensive evaluation the cloud model of each research node are listed in Table 9.

From Figure 19, it can be seen that the carrying state of water and soil resources in the irrigation area in 2002, 2010 and 2018 is ‘critical bearing’, ‘critical bearing–safe bearing’ and ‘critical bearing–safe bearing’, respectively. The whole process is in a healthy and sustainable evolution. Referring to the research results done by Liu (2018) and Zhu (2020), the evaluation results are consistent with the actual water and soil resource carrying status of the Jingdian irrigation area in China. It indicates that the carrying capacity of water and soil resources in the Jingdian irrigation area has been improved, but the improvement rate is gradually weakening. The reason is that the regional water transfer has greatly alleviated the problem of poor adaptability and coordination between water resources and land resources in irrigation areas, and has largely realized the optimal allocation of regional water and land resources. Human predatory development of water and soil resources in the region resulted in increased land reclamation rate and increased pollutant load; unscientific irrigation mode leads to the increase of groundwater depth and groundwater salinity, which makes soil salinization gradually become the key problem restricting the improvement of water and soil resources in the irrigation area.

To further verify the reliability of the evaluation results, the fuzzy comprehensive evaluation method (Xu & Li 2006) and the cloud center of gravity evaluation method (Yang et al. 2014) were introduced to analyze the results. The results are compared with those obtained from the comprehensive evaluation cloud model, and the results are shown in Table 10.

From the comparative analysis of the three evaluation results, the evaluation results obtained by the comprehensive evaluation cloud are closer to those obtained by the other two models, with the maximum error value of 2.67% and the minimum value of only 0.42%, indicating a higher evaluation accuracy. The evaluation results obtained by using the comprehensive evaluation cloud method have three numerical characteristics, which are expectation (Ex), entropy (En), and super entropy (He), and the above three numerical characteristics represent the average level, reliability, and stability of the evaluation results, respectively. The other two evaluation methods, whose evaluation results are characterized by only one numerical feature, can represent the correctness of the evaluation results to a certain extent, but the reliability of their results cannot be characterized. In contrast, the evaluation results obtained by using the comprehensive evaluation cloud are richer and more reliable, and have more advantages in terms of information advantages and flexibility.

As can be seen from Figure 20 and Table 11, the overall carrying status of water and soil resources in the study area from 2002 to 2018 characterizes the evolution of a severe bearing area and a slight bearing area decreasing, and the area of critical bearing area, safe bearing area and good bearing area were gradually increasing, with the area of good bearing area increasing by 8.38% and the area of severe bearing area decreasing by 6.4%. With the continuous improvement of irrigation and water distribution facilities in the irrigation area, the area of arable land in the area has been increasing and the area of sand and grass has been decreasing, and as a result of this feedback, the vegetation coverage in the irrigation area has increased, the light and heat conditions have improved and the coordination of water and soil has tended to be suitable. With the combined effects of unreasonable exploitation of water and soil resources in the irrigation area by human production activities and geological structural conditions, surface ecological degradation, mainly soil salinization and secondary salinization, has occurred in the eastern part of the irrigation area in Luyang, Caowotan and Wufo. Gobi and sandy land are the main areas for use in Zhitan, Xijing, southern Manshuitan and northern Shangshawo, the water and soil environment has been greatly improved with water irrigation, but it is still a serious carrying area for water and soil resources due to the limitation of its environmental background. At the same time, the increase in construction and transportation land and the intensification of land contamination load bring new challenges to the water and soil resources carrying status of the irrigation area, showing a spatial state of decreasing from the urban area to the surrounding townships in an arc.

The spatial change mapping of water and soil resources carrying state is calculated using Equation (4), and the spatial flow direction of water and soil resources carrying state in the irrigation area from 2002 to 2018 is analyzed. The results are listed in Table 12.

As obtained from Table 12, the overall intensity of the evolution pattern of water and soil resources carrying state in the study area from 2002 to 2018 is: continuous changing type > pre-changing type > post-changing type > continuous stable type > recurrent changing type. In the continuous changing type, the water and soil resource carrying state of the irrigation area is in a benign evolutionary trend, indicating that the regional water and soil coordination capacity is in an improved state; in the recurrent changing type, the contradiction between human production activities and water and soil coordination in the local area is intensified and begins to become a constraint factor for water and soil resource carrying. In the pre-changing type, the regional water and soil coordination ability improved continuously with water lifting and irrigation at the early stage, but at the later stage, it was mainly restricted by natural geographic features and human production activities, and the state of water and soil resources was stabilized at the ‘ safe bearing’ level; the post-changing type was concentrated in Phase II, and after the completion of water-lifting project in the middle and later stage, the water and soil resources carrying state only tends to evolve healthily. Due to the constraints of the original environmental background in the study area, the continuous stable type is mainly distributed in areas with very poor coordination of water and soil resources such as Gobi, sandy land and uncultivated land.

Using the spatial transfer matrix statistics to calculate the spatial transfer process of different water and soil resources carrying states, the transfer area matrix of water and soil resources carrying states in different periods in the irrigation area is listed in Tables 13 and 14, where the rows represent the initial period and the columns represent the final period.

According to the above spatial transfer area matrix, the largest transfer area of irrigation area from 2002 to 2010 is the transfer of ‘critical bearing–safe bearing’, followed by the transfer of ‘slight bearing–critical bearing’ ; from 2010 to 2018, the transfer area of ‘safe bearing–good bearing’ is the largest, followed by ‘critical bearing–safe bearing’. Through comprehensive analysis of the overall transfer process, it is found that the carrying capacity of water and soil resources in the irrigation area from 2002 to 2018 presents a transition trend of ‘severe bearing–slight bearing–critical bearing–safe bearing–good bearing’, and the spatial transfer speed is gradually weakening. The reason is that the evolution rate of the environmental restoration process is slowing due to the vulnerability of the environment, and the predatory exploitation of water and soil resources by human production activities further slows down this healthy evolution.

• (1)

  The comprehensive evaluation cloud model parameters of the study area in 2002, 2010 and 2018 are Z₂₀₀₂ = (0.5034, 0.0236, 0.0071), Z₂₀₁₀ = (0.5586, 0.0218, 0.0062), Z₂₀₁₈ = (0.5989, 0.0249, 0.0061), respectively. The water and soil resources carrying status of each study node are ‘critical bearing’, ‘critical bearing–good bearing’, ‘critical bearing–good bearing’, and the overall in a state of continuous improvement, but the improvement rate is gradually decreasing.

• (2)

  The degree of influence of the driving factors on the evolution of the carrying state of water and soil resources is: water-soil coordination > groundwater depth > mineralization of groundwater > soil salinity > surface irrigation volume > surface temperature > land contamination load > average annual evaporation > average annual precipitation > vegetation coverage > soil electrical conductivity > elevation > slope, and the key driving factors are water-soil coordination, groundwater depth, mineralization of groundwater and soil salinity.

• (3)

  The trend of changes in the carrying status of water and soil resources in the study area from 2002 to 2018 showed a decrease in the area of severe bearing and slight bearing areas, and an increase in the area of critical bearing, safe bearing and good bearing areas. The evolution pattern of water and soil resources in the irrigation area is: continuous changing type > pre-changing type > post-changing type > continuous stable type > recurrent changing type, and its spatial and temporal evolution trend is ‘severe bearing–slight bearing–critical bearing–safe bearing–good bearing’.

• (4)

  During 2002–2018, the northwestern and south-central areas of the irrigation area showed a good evolution of water and soil resources carrying status, and the water and soil resources carrying capacity level has been improved to a certain extent. The serious bearing area is mainly distributed in the areas of Zhitan, Xijing, southern Manshuitan, northern Shang Shawo, etc. The bearing state of water and soil resources presents a spatial state of decreasing from the town to the surrounding areas by arc shooting type.

This research was funded by the Support Program for Leading Talents in Science and Technology Innovation in Central Plains, grant number 204200510048; the Zhejiang Province Basic Public Welfare Research Program Project, grant number LZJWD22E090001; Zhejiang Province Key R&D Program Project, grant number 2021C03019; the Henan Province Science and Technology Tackling Project, grant number 212102310273; and the Henan Province Key Scientific Research Projects Program for Higher Education Institutions, grant number 20A570006.

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

The authors declare there is no conflict.