Spatial Heterogeneity of Heavy Metals in Arid Oasis Soils and Its Irrigation Input–Soil Nutrient Coupling Mechanism

Abstract

Soil environmental quality in arid oases is crucial for regional ecological security but faces multi-source heavy metal (HM) contamination risks. This study aimed to (1) characterize the spatial distribution of soil HMs (As, Cd, Cr, Cu, Hg, and Zn) in the Ka Shi gar oasis, Xinjiang, (2) quantify the driving effect of irrigation water, and (3) elucidate interactions between HMs, soil properties, and land use types. Using 591 soil and 12 irrigation water samples, spatial patterns were mapped via inverse distance weighting interpolation, with drivers and interactions analyzed through correlation and land use comparisons. Results revealed significant spatial heterogeneity in HMs with no consistent regional trend: As peaked in arable land (5.27–40.20 μg/g) influenced by parent material and agriculture, Cd posed high ecological risk in gardens (max 0.29 μg/g), and Zn reached exceptional levels (412.00 μg/g) in gardens linked to industry/fertilizers. Irrigation water impacts were HM-specific: water contributed to soil As enrichment, whereas high water Cr did not elevate soil Cr (indicating industrial dominance), and Cd/Cu showed no significant link. Interactions with soil properties were regulated by land use: in arable land, As correlated positively with EC/TN and negatively with pH; in gardens, HMs generally decreased with pH, enhancing mobility risk; in forests, SOM adsorption immobilized HMs; in construction land, Hg correlated with SOM/TP, suggesting industrial-organic synergy. This study advances understanding by demonstrating that HM enrichment arises from natural and anthropogenic factors, with the spatial heterogeneity of irrigation water’s driving effect critically regulated by land use type, providing a spatially explicit basis for targeted pollution control and sustainable oasis management.

1. Introduction

As a key unit for maintaining the stability of desert ecosystems, the quality of the soil environment in arid zone oases is directly related to regional ecological security and sustainable development [1,2]. Heavy metal contamination of soils poses a major threat to crop safety, groundwater quality, and human health, especially in arid zones, where pollutants are prone to cumulative effects through continuous irrigation [3,4]. It has been shown that heavy metal enrichment of soils in arid oases may originate from multiple sources, including irrigation water, agrochemicals and atmospheric deposition, and that heavy metal contamination of soils, driven by irrigation activities, industrial discharges and historical land use, poses a significant threat to the health of soils, the safety of crops and the quality of groundwater in the region [5].

The ecological vulnerability of the Ka Shi gar region in southern Xinjiang, as a typical agro-industrial complex oasis, which has long been subjected to multi-source pollution coercion from metal smelting, fertilizer application, and natural weathering, may exacerbate the complexity of pollutants’ environmental behaviour [6]. In recent years, geostatistical methods (e.g., Inverse Distance Weight Interpolation, IDW) have provided an important tool for modelling the spatial distribution of soil contaminants and have shown particular strengths in revealing anthropogenically driven spatial heterogeneity of contamination [7]. Although land use types (e.g., agricultural area and industrial land) have been demonstrated to have a significant role in regulating the distribution of soil heavy metals (e.g., enrichment of Cd and Pb from agricultural sources), there is still a significant knowledge gap on the coupling mechanism between heavy metals and soil physicochemical properties (organic matter, pH, and nutrients) under different land use modes (arable land, parkland, forest land, construction land, and grassland) in the arid oasis area [8,9].

For example, arable land may lead to heavy metal–nutrient co-accumulation due to frequent irrigation and fertilization, whereas intensely anthropogenically disturbed built-up land may weaken the fixation capacity of heavy metals by destroying the soil structure [10]. However, the quantitative resolution of such land-use-specific geochemical processes in arid zones has not been systematically carried out. In addition, although irrigation water is widely recognized as a key mediator of heavy metal transport, existing studies are mostly limited to simple comparisons of water and soil concentrations, failing to quantify the gradient effects of irrigation water points on the spatial patterns of heavy metals in surrounding soils through spatial interpolation techniques (e.g., IDW), and in particular the heterogeneity of such spatial associations in the context of different land use types (e.g., cropland vs. grassland) remains unrevealed.

Current studies have paid insufficient attention to the following key issues: heavy metal–physical and chemical property interaction networks regulated by land use types (the generally high pH of arid zone soils may alter the bioavailability of metals such as Cu and Cr by regulating heavy metal adsorption-desorption equilibrium in a synergistic manner with land use practices, e.g., localized reduction of pH due to cropland fertilizer applications) [11,12]; spatial pollution pathway resolution driven by irrigation water (traditional statistical methods are difficult to capture the spatial non-homogeneity of the impacts of irrigation water, and need to be combined with geostatistical tools to identify the concentration decay pattern of heavy metals in soils around the irrigation water points in the GIS platform).

In view of this, this study focuses on five major land use types—arable land, orchards, forest land, construction land, and grassland—in a typical arid oasis region. It combines agricultural irrigation dependency with a comprehensive dataset covering soil heavy metals (Cu, Cr, Ni, Pb, Cd, etc.), soil nutrients (total nitrogen, total phosphorus, total potassium, and organic matter), and irrigation water quality. We hypothesize the following: (1) the spatial variation of soil heavy metals is driven by the synergistic effects of irrigation water input and land use-regulated soil properties (pH, organic matter, total nitrogen, and phosphorus); (2) the gradient attenuation effect of irrigation water on surrounding soil heavy metals exhibits land use type dependency.

To validate these hypotheses, based on 591 soil samples (covering heavy metals and physical and chemical indicators) and 12 irrigation water samples, we used IDW spatial interpolation to analyze the characteristics of soil heavy metal content and its spatial heterogeneity. Combining irrigation water point data, we identified the potential driving effects of irrigation water heavy metals on surrounding soil pollution and revealed the interactive mechanisms between soil heavy metal accumulation and key environmental factors (total nitrogen/phosphorus/potassium, organic matter, pH, etc.). The results of this study provide a spatially explicit assessment framework for pollution risks associated with irrigation in arid oases and, by elucidating the synergistic effects of soil nutrient status and irrigation management on heavy metal fate, provide a theoretical basis for the sustainable management of arid agricultural systems.

2. Study Area and Method

2.1. Study Area

The study area is located in the western edge of the Taklamakan Desert in the Kashgar Delta and the front edge of the Gezi River alluvial fan, in the oasis area of the middle reaches of the Gezi River (39°14′13″–39°18′08″ N, 75°51′25″–76°06′58″ E), and belongs to the shallow buried area of groundwater in the middle reaches of the Gezi River in the front edge of the alluvial plain and alluvial fan (Figure 1). It belongs to the shallow buried area of the alluvial plain and the front edge of the alluvial fan in the middle reaches of the Gezi River, with a relatively flat terrain and an elevation of 1272–1313 m. The average annual temperature of this area is 12.8 °C, the annual precipitation is 71.6 mm, the evaporation is 2242.8 mm, the relative humidity is 49.9%, the annual absolute humidity is 6.8 mm, the dryness is 13.84, the average frost-free period is 215.3 days, the average annual sunshine is 2822 h, and the sunshine rate is 65.5%. The average frost-free period is 215.3 days, the average annual sunshine hours is 2822 h, and the sunshine rate is 65.5%. The current water area is 14.2 km², accounting for 24.8% of the town planning area. The main types are ditch and river water surface, and the main soil types are lime meadow soil, meadow salt soil, irrigation and silt tidal soil, salinized tidal soil, salinized meadow soil, and so on.

2.2. Sample Collection

2.2.1. Surface Soil Sampling

In the experimental area, 591 sampling points were laid out to collect topsoil samples from 0 to 20 cm, weeds, grass roots, gravel, bricks, fertilizer clumps, etc., were removed, and samples were taken using the five-point sampling method within the range of 50–100 m around the GPS fixed point locations. After the samples were collected, they were promptly put into cloth sample bags and covered with special polyethylene plastic bags to avoid mutual contamination between samples.

2.2.2. Irrigation Water Sampling

Twelve groups of samples of irrigation water sources in the test area were collected. Before collection, the container was washed with the water samples to be taken, and irrigation water was collected at the irrigation water outlet during the irrigation period of the farmland in the state of natural water flow, without disturbing the water flow and the bottom sediment to ensure the representativeness of the samples. During sampling, the mouth of the sampler faced the direction of water flow, and the screw-necked stopper was tightened in time after sampling was completed and sealed with paraffin.

2.3. Measurement Methods

Soil samples covered 10 indicators, including 8 heavy metal elements (As, Cu, Cr, Cd, Hg, Ni, Zn, and Pb) and soil physicochemical indicators (total nitrogen, total phosphorus, total potassium, organic matter, effective phosphorus, fast-acting potassium, pH, and cation exchange). The indicators of water samples involved (B, Cd, Cr, Cu, Pb, pH, Se, Hg, and As), the processing and determination process in strict accordance with the GB/T 17141-1997 method [13] as a normative standard, the analytical method for each element is shown in

Table 1. Sample analytical methods.

Index	Assay Method	Index	Assay Method
Cu	ICP-MS	CEC	Ammonium acetate exchange method (pH 7.0)
Cr	ICP-OES	pH	Glass electrode method (soil/water = 1:2.5)
Ni	ICP-MS	B	ICP-MS
Zn	ICP-OES	Total N	Kjeldahl method
Pb	ICP-MS	Total P	Molybdenum antimony anti-spectrophotometry
Cd	ICP-MS	Total K	HF-HClO4 digestion-Flame photometry
Se	ICP-MS	Organic matter	Potassium dichromate oxidation-external heating method
As	ICP-MS	Available P	Olsen method (sodium bicarbonate extraction)
Hg	CVAAS	Available K	Ammonium acetate extraction-Flame photometry

.

2.4. Data Processing

Soil heavy metal spatial distribution maps were generated based on the inverse distance weighting method, soil heavy metal and nutrient-related data were analyzed using EXCEL 2021, SPSS 27, R 4.1.2, and mapping was performed using OriginPro 2024, R.

3. Results and Analyses

3.1. Characterization of the Content of Irrigation Water by Indicator

Statistical analysis of multiple heavy metals and physicochemical indicators in 12 irrigation water samples is shown in Figure 2. The content range of each index of irrigation water samples) revealed that the concentration range and degree of dispersion of each heavy metal showed differences. The concentration range of boron (B) is 0.01–0.60 mg/L, with an average of 0.07 mg/L (standard deviation 0.17 mg/L). Sample 9 (0.60 mg/L) was significantly higher than the other samples (all 0.01 mg/L), and this sample was collected from the northern construction site of the Gazi River (Figure 1), indicating that construction sites pollute water resources, leading to an increase in boron concentration. The concentration range of hexavalent chromium (Cr) was relatively wide (0.001–0.015 mg/L), with high values concentrated in samples 3 (0.015 mg/L), 5 (0.009 mg/L), 6 (0.007 mg/L), and 7 (0.005 mg/L). The high-value samples were collected from farmland, construction land, and orchards, respectively, and their heterogeneous distribution reflects intermittent pollution inputs. The total arsenic concentration ranged from 0.0005–0.006 mg/L (mean 0.0026 mg/L, standard deviation 0.0016 mg/L). Sample 12 (0.006 mg/L) and samples 1, 3, 4, 5, and 8 (0.003 mg/L) showed higher values, with these samples distributed across forest land, farmland, industrial land, and other land types, necessitating attention to potential geological or anthropogenic pollution sources. Selenium (Se) concentrations showed minimal variation (0.0002–0.0007 mg/L), with only Sample 1 (construction land) and Sample 6 (garden land) showing slightly higher values (0.0007 mg/L). It is noteworthy that cadmium (Cd), copper (Cu), lead (Pb), and total mercury concentrations were identical across all samples (0.05, 0.05, 0.01, and 0.0004 mg/L, respectively), with zero standard deviations reflecting a high degree of homogeneity in the background values of the test environment. In terms of physicochemical parameters, pH values were neutral to alkaline (7.36–7.98), with sample 9 having the lowest pH (7.36), which is potentially correlated with abnormally high B concentrations.

3.2. Characterization of Soil Heavy Metal Content

The distribution range of heavy metal content in different land use types was derived by analyzing soil samples of different land use types in the study area to test their heavy metal content. As shown in Figure 3, the content of heavy metals in different soil types, the range of As content varied among different land use types, with the following ranges: arable land (5.27–40.20 μg/g) > forest land (6.26–23.10 μg/g) > garden land (6.04–17.90 μg/g) > grassland (7.88–15.70 μg/g) > construction land (7.32–14.00 μg/g), and the maximum value of cultivated land at the anomaly (40.20 μg/g) was significantly higher than that of the other types. The range of cadmium (Cd) content was 0.10–0.29 μg/g in garden land, 0.05–0.22 μg/g in cultivated land, and 0.05–0.22 μg/g in forest land.(0.05–0.22 μg/g) > grassland (0.10–0.19 μg/g) > woodland (0.09–0.18 μg/g) > construction land (0.11–0.17 μg/g), and the anomalies were located in the garden (0.29 μg/g). The value of 0.29 μg/g was the highest, and in terms of consistency of mean values, the mean values of all types were close to each other (0.14–0.16 μg/g), and only slightly higher in gardens, which should be a concern regarding the ecological risk. The range of chromium (Cr) content was cropland (23.60–87.00 μg/g) > woodland (28.50–74.30 μg/g), and forested land (28.50–74.30 μg/g). Cr (23.60–87.00 μg/g) > forest land (28.50–74.30 μg/g) > garden land (27.00–72.30 μg/g) > grassland (30.40–58.30 μg/g) > construction land (34.40–43.60 μg/g); Cu (Cu) content range: cultivated land (9.32–44.70 μg/g) > forest land (11.10–42.10 μg/g) > garden land (11.20–39.60 μg/g) > grassland (12.10–28.30 μg/g) > construction land (13.40–21.20 μg/g). High values in agricultural land: the mean values in cultivated land and garden land (22.21 and 24.78 μg/g) were significantly higher than those in non-agricultural land; the differences in the mercury (Hg) content range were small (0.01–0.04 μg/g), and the maximum values in cultivated land and garden land were slightly higher (0.04 μg/g). The lead (Pb) content range was garden land (16.20–26.40 μg/g) > cultivated land (12.40–26.30 μg/g) > forest land (15.50–25.20 μg/g) > grassland (14.50−25.20 μg/g). The range of lead (Pb) content was garden land (16.20–26.40 μg/g) > cropland (12.40–26.30 μg/g) > forest land (15.50–25.20 μg/g) > grassland (14.00–23.30 μg/g) > building land (17.30–20.80 μg/g) > construction land (17.30–20.80 μg/g). The content of nickel (Ni) ranged from arable land (10.40–43.40 μg/g) to forest land (12.70–38.80 μg/g) to garden land (12.50–37.30 μg/g) to grassland (14.50–29.30 μg/g). Grassland (14.50–29.20 μg/g) > constructed land (15.80–22.10 μg/g). Zn content ranged from 38.00 to 412.00 μg/g in garden land to 40.30 to 1111.00 μg/g in constructed land (−111.00 μg/g) to cropland (32.50 to 102.00 μg/g) to forest land (39.40 to 95.80 μg/g) to grassland (38.80 to 76.50 μg/g). The maximum value of 412.00 μg/g in the garden was exceptionally prominent, and the mean values were higher in comparison with those of construction land (66.20 μg/g) than those of cultivated land (65.77 μg/g).

3.3. Matching Analysis of Spatial Interpolation Results of Soil Heavy Metals with Irrigation Water Sampling Points

As can be seen by the inverse distance weighting method in Figure 4, as a whole, the distribution of each heavy metal content did not show an obvious regular trend of gradually increasing or decreasing from one direction to another, but rather showed a complex, patchy, and fragmented distribution. The As content is differentiated by different colors, with the light purple area (5.6–10) more widely distributed, covering most of the region; the dark purple area (10.1–15), yellowish-green area (15.1–20), orange area (20.1–30), and red area (30.1–40) are scattered in patches; the light grey area with a lower Cd content (0.052–0.5) is more widely distributed than the dark purple area (10.1–15, 0.131–0.15) occupy a larger area. As the content increased, the different color regions (0.131–0.15, 0.151–0.16, 0.161–0.2, 0.201–0.3) were distributed in scattered small patches. The light green areas (25–40) of Cr content are widely distributed, but there are distinct areas of high red values (61–90) concentrated in the northern and right-central parts of the region. Areas with light green Cu content (10–20) are more predominant, and areas with higher content (25–28, 28–30, 30–45) are scattered patches. The light purple area (0.006–0.015) of Hg content is widely distributed, while the high value area (0.031–0.04) is sporadic and scattered, mainly concentrated in the central and southern parts of the region. The light gray area (11–20) of Ni content occupies a large area, and the high value area (31–45) is concentrated in the eastern and northern edges of the region. The light blue areas (12–19) of Pb content are widely distributed, and the high value areas (24–27) are scattered in small patches, mainly concentrated in the central and southern parts of the region. The light purple area (34–60) of Zn content is widely distributed, and the high values (251–450) are concentrated in the right-central and southern parts of the region. Overlaying the land use types (Figure 1) reveals that the high value areas of As are located in cropland and grassland, and the medium and low value areas are distributed in various land use types without significant differences. While the high value areas of Cd and Cr are located in the garden land, which indicates that the garden land has a significant effect on the accumulation of Cd and Cr elements, the high value areas of Cu, Hg, and Ni are located in the forest land, garden land and construction land, which indicates that the forest land, garden land, and construction land are important for Cu, Hg, and Ni, which are significantly higher than those in cropland and grassland; Pb is mainly enriched in cropland, and the rest of the land types are scattered with low values.

By analyzing the spatial distribution characteristics of soil heavy metals around the irrigation water sample points, it was found (Figure 1 and Figure 4) that the soil heavy metals around the water sample point (No. 3) with extremely high values of As showed a high value (20.1–30) distribution area, and the rest of the water sample points around the water samples did not have obvious characteristics due to the water samples of the As content, and the corresponding soil heavy metal content did not have obvious distribution characteristics. The spatial distribution of soil heavy metals around the water sample points with high Cr content (3, 5) is in the low value area. On the contrary, the distribution of soil heavy metals around the rest of the water sample points with Cr content (6, 7, 8, and 9) is aggregated; there is no change in the overall water sample points of Cd, Cu, Hg, and Pb (0.05 mg/L) and the distribution of soil heavy metals has different characteristics, which means that the distribution of soil heavy metals is not correlated with the irrigation of the water samples. This indicates that the distribution characteristics of soil heavy metals are not correlated with the irrigation of water samples, but with the land use type and other factors.

3.4. Correlation Analysis Between Soil Heavy Metals and Soil Nutrients

The distribution range of nutrient content of different land use types was derived by analyzing and testing the nutrient content of soil samples of different land use types in the study area. As shown in Figure 5, the range of pH content varies among different land use types, with the content ranging as follows: grassland > cropland > building land > woodland > garden land, and the CEC content ranging as follows: woodland > garden land > cropland > grassland > building land. Organic matter (SOM) content ranges from forest land > garden land > arable land > grassland > building land; quick-acting potassium (AK) content ranges from grassland > arable land > building land > forest land > garden land. Quick-acting phosphorus (AP) content ranges as forest land > garden land > construction land > cropland > grassland. Total Nitrogen (TN) content ranges from forest land > garden land > cropland > grassland > building land; total phosphorus (TP) content ranges from forest land > garden land > cropland > building land > grassland; total potassium (TK) content ranges from forest land > garden land > cropland > grassland > building land.

The study of the relationship between soil nutrients and heavy metals through correlation analysis revealed (Figure 6) that As was significantly and positively correlated with CEC, Cd was significantly and positively correlated with AP, Hg was significantly and positively correlated with pH, Cr, Cu, and Pb were significantly and positively correlated with TK, and Ni and Zn were significantly and positively correlated with TP. Cropland correlation analysis revealed that As was significantly and positively correlated with CEC and TN, and significantly and negatively correlated with AK, TK, and pH. Cd was significantly and positively correlated with AP, CEC, SOC, TN, and TP, and significantly and negatively correlated with AK and pH. Cr, Cu, Hg, Zn, and Ni showed significant positive correlation with AP, CEC, SOC, TK, TN, and TP, and significant negative correlation with pH. Pb showed significant positive correlation with TK and TN and significant negative correlation with AK, pH, and TP. Construction land correlation analysis found that Hg was significantly and positively correlated with SOC and TP. Forest land correlation analysis revealed that As and Cd were significantly and positively correlated with CEC, SOC, and TN, while Cr, Cu, Hg, Zn, Hg, and Ni were significantly and positively correlated with CEC, SOC, TK, and TN. Garden correlation analysis revealed that As was significantly and positively correlated with AK, CEC, SOC, TK, TP, and TN, and significantly and negatively correlated with pH. Cd, Cr, Cu, Hg, Zn, Hg, and Ni showed significant positive correlation with AK, AP, CEC, SOC, TK, TP, and TN and significant negative correlation with pH. Pb showed significant positive correlation with AP, CEC, SOC, TK, TP, and TN, and significant negative correlation with pH. Zn showed significant positive correlation with CEC, SOC, TK, TP, and TN, and significant negative correlation with pH.

4. Discussion

4.1. Characterization of Irrigation Water and Soil Physicochemical Property Contents

The distribution range of heavy metal content in different land use types was derived by analyzing soil samples of different land use types in the study area to test their heavy metal content. In this study, we found that the range of As content varied among different land use types, with the content ranging from cropland > forest land > garden land > grassland > construction land, this is consistent with background concentration studies in Region As of Florida, and the maximum value of cropland in the anomalies was significantly higher than that of other types, which related to the background of high arsenic in the matrices of the soil-forming materials or to the agricultural activities (e.g., arsenic-containing pesticides) [11]. Comparison of the mean values shows that cropland > garden land > forest land, reflecting that agricultural land is more significantly affected by human interference, indicating that there is no obvious single increasing or decreasing trend of As content among different land types, and the fluctuation ranges of As content in each land type overlap. Cadmium (Cd) content ranges from garden > arable land > grassland > forest land > construction land. The anomaly is located in the garden with the highest maximum value. This is consistent with the findings of Khatun et al. Khatun et al., who pointed out that long-term application of cadmium-containing phosphate fertilizers or acidic soils enhances cadmium activity. In terms of mean consistency, the means of all types are close, with only the garden plot being slightly higher, and attention should be paid to its ecological risks [12]. Chromium (Cr) content ranges from cropland > forest land > garden land > grassland > construction land. Chrysochoou et al. found that high Cr values in farmland, forest land, and orchards are associated with parent material (such as weathered bedrock) or industrial pollution diffusion (such as irrigation water sources), while low values in construction land are due to human disturbance of topsoil, which interrupts the natural accumulation of Cr [14,15]. Copper (Cu) content ranged from cropland > forest land > garden land > grassland > construction land, with high values for agricultural land; the mean values for cropland and garden land were significantly higher than those for non-agricultural land. McLaughlin et al. pointed out that high heavy metal levels in agricultural land are directly related to the long-term use of copper-based fungicides and organic fertilizers (animal manure containing Cu). Du et al. pointed out that low heavy metal levels in construction land are related to soil compaction or backfill soil diluting heavy metal content [16,17]. Mercury (Hg) content ranges varied little by type, with slightly higher maxima in cropland and gardens. Li et al. pointed out in Scientific Reports that vegetable Hg contamination is related to atmospheric deposition (such as coal combustion and industrial exhaust) and pesticide residues, but the overall risk level in this study is low [18]. Lead (Pb) content ranges garden land > cultivated land > forest land > grassland > construction land. Based on the terrain location shown in Figure 1 and the results of Chaney et al.’s study, the factors contributing to heavy metal exposure in urban gardens were identified. High values in garden plots were attributed to the dispersion of traffic exhaust (near roads) or the accumulation of industrial waste, while construction sites exhibited more uniform Pb distribution due to soil disturbance [19]. Nickel (Ni) content ranges from cultivated land > forest land > garden land > grassland > construction land, and the natural source dominant Ni content has a high correlation with soil-forming matrices (e.g., basaltic rocks). The influence of agricultural activities on Ni content was lower than that of Cu and Cd. The range of zinc (Zn) content was garden land > construction land > cropland > forest land > grassland, with an unusually high maximum in garden land, which related to historical industrial pollution (e.g., near zinc mines) and irrational application of Zn-containing fertilizers (e.g., excessive sludge), and a comparison of the mean values with construction land > cropland. Sharifi et al. pointed out that heavy metal ions around lead–zinc mining and processing plants have a significant impact on soil. The high values are related to the release of Zn from building materials (such as corroded galvanized components) [20]. In the comprehensive analysis, the parkland should focus on Zn and Cd, the cultivated land should focus on As and Cu, and the construction land should focus on the abnormal accumulation of Zn. The dominant factors of man-made influences are as follows: agricultural activities (fertilizer and pesticide application) are the main reasons for the accumulation of Cu, Cd, and As; industrial pollution and traffic sources are the important sources of Zn and Pb.

Differences in the physical and chemical properties of soils in different land-use types are mainly influenced by the combined effects of the intensity of human activities, the type of vegetation, management measures, and natural soil-forming factors. The pH value of grasslands is relatively high, and Chen et al. noted that this is associated with the accumulation of base ions and reduced secretion of organic acids. In contrast, the pH values of forest and garden soils are relatively low. Jiang et al. noted that soil organic matter plays a significant role in acid buffering and reducing aluminum leaching in acidic forest soils. The organic matter in forest and garden soils is primarily derived from the decomposition of litter, which produces organic acids and results in strong leaching, leading to lower pH levels [21,22]. CEC values follow the order forest land > orchard land > farmland > grassland > construction land. Carmo et al. [23] noted that soil fertility and electrical conductivity are influenced by organic matter content and nutrient input. Forest and orchard lands have higher values due to the release of soluble base ions from litter decomposition, while construction land has the lowest values due to reduced ion leaching caused by surface hardening [23]. Organic matter (SOM) content was highest in forested land, benefiting from large amounts of litter input and slow decomposition, while construction land had the lowest organic matter accumulation due to surface soil destruction. This is consistent with the findings of Henderson et al. and Solomon et al. [24,25]. Quick-acting potassium (AK) was highest in grassland, associated with root activity and return of grazing manure, while woodland and garden land had lower levels due to vegetative uptake and leaching [26]. Available phosphorus (AP) is higher in forest and garden soils, consistent with the findings of Sorkau et al. [27], who noted that land use and plant diversity significantly enhance microbial phosphorus in forest and grassland soils. This phenomenon is associated with the release of phosphorus through litter decomposition and the promotion of phosphorus solubility by acidic environments. However, grasslands exhibit lower phosphorus availability due to their tendency to bind with calcium [27]. The whole amount of nutrients (TN, TP, and TK) all showed forest land > garden land > cropland > grassland > construction land, reflecting the closed nature of the biological cycle under the natural vegetation cover and the disturbance of nutrient cycling by human activities. Overall, these differences reflect the interaction between the efficiency of ecosystem material cycling and the intensity of human activities, providing an important basis for land management and ecological restoration.

4.2. Analysis of Spatial Heterogeneity of Heavy Metals in Soil Around Irrigation Water

In the present study, 12 samples of irrigation water were analyzed for heavy metals, and the results showed significant differences in the range of concentrations and degree of dispersion of different heavy metals. Boron (B) concentrations were significantly higher in Sample 9 (0.60 mg/L) than in the other samples (0.01 mg/L), which is located in a construction site in the northern part of the Gezi River, suggesting that construction activities may result in the introduction of B-containing pollutants (e.g., construction materials, industrial wastewater) into the water body. The high values of irrigation water in B are associated with construction land, similar to the pollution pattern of “Cd and Zn input from construction activities” in the Tai’an study [28]. The concentration distribution of chromium (Cr) showed significant heterogeneity (0.001–0.015 mg/L), with high values concentrated in samples 3, 5, 6, and 7, in the Tai’an study, the spatial variation coefficient of Cr reached 18.3%, and the semivariogram model showed that 72% of its distribution was controlled by structural factors (parent material) [25], while intermittent industrial emissions caused fluctuations in Cr(VI) [29].

Cr (VI) is highly toxic, and its fluctuating distribution in water bodies suggests the need for increased monitoring, especially of irrigation water sources around industrial areas. The high value of total arsenic (As) (0.006 mg/L) appeared in Sample 12, and the rest of the higher values (0.003 mg/L) were distributed in different areas, such as forested land, cultivated land, and industrial land, suggesting that the source of As a combination of both geologic backgrounds (e.g., weathering of arsenic-containing minerals) and anthropogenic activities (e.g., pesticides and smelting wastewater). Cadmium (Cd), copper (Cu), lead (Pb), and total mercury (Hg) concentrations were consistent across all samples (0.05, 0.05, 0.01, and 0.0004 mg/L), consistent with regional background values. This conclusion is consistent with the conclusion in the oilseed system study that “heavy metals in non-agricultural areas mainly originate from parent material [29]”. In addition, Sample 9 had the lowest pH (7.36), which is associated with unusually high B concentrations and may reflect the effect of acidic wastewater inputs on water quality.

It was found that the content of As in soil varied greatly in different locations without a clear trend of concentration, which was influenced by factors such as soil parent material and regional geochemical background. The distribution of Cd in soil is relatively discrete, possibly due to localized human activities (e.g., industrial emissions, agricultural fertilization, etc.) or differences in soil properties that lead to different degrees of enrichment at different locations. Cr is significantly enriched in some areas, which are related to the presence of chromium-containing industrial activities, waste chromium slag stockpiles, etc., in the vicinity of these areas, while most of the areas have relatively low levels. The distribution of Cu in the soil is relatively decentralized, and the areas of high content are related to copper mining, industrial wastewater discharge containing copper, etc. The adsorption and accumulation capacity of Cu in soil varies in different areas. Hg is mainly concentrated in the central and southern parts of the region, indicating that the content of Hg is low in most of the soils, but there is anomalous enrichment in some localized areas, which is due to the use of Hg-containing pesticides and the deposition of industrial exhaust gases. Ni is not uniformly distributed in the soil, and parts of the margins may have elevated levels due to Ni-related geological sources or anthropogenic inputs. Pb is mainly concentrated in the central and southern parts of the region, suggesting that the distribution of Pb in the soil is locally enriched, which is related to activities such as transportation emissions (e.g., residuals from past use of leaded gasoline), lead mining, and smelting. The high Zn content areas are concentrated in the right central and southern parts of the region, indicating that the Zn content is relatively low in most of the region, but there is a significant enrichment in specific areas, which related to factors such as Zn mining, industrial Zn-containing wastewater discharge, and the use of Zn-containing fertilizers in agriculture. Overall, heavy metal pollution of irrigation water presents a distribution pattern combining point sources (e.g., construction land and industrial areas) and surface sources (e.g., agricultural areas), which needs to be combined with the type of land use for targeted control.

4.3. Correlation Analysis Between Heavy Metals and Physical and Chemical Properties of Soils of Different Land Use Types

The interaction between soil nutrients and heavy metals is an important part of soil environmental chemistry research, which directly affects soil quality, crop safety, and ecological health. In this study, the correlation characteristics between soil nutrients (e.g., organic carbon, nitrogen, phosphorus, potassium, etc.) and heavy metals (As, Cd, Cr, Cu, Hg, Pb, Ni, Zn, etc.) under different land-use types (arable land, built-up land, forest land, garden land, and grassland) were revealed through correlation analyses, which provided a scientific basis for the risk control and nutrient management of soil pollution. From the overall results of the analysis, there was a significant correlation between soil nutrients and heavy metals, and different heavy metals showed a preference for specific nutrients. As is significantly positively correlated with CEC, which is consistent with the results of studies in the western part of the Hetao Basin in Inner Mongolia, indicating that the migration of As is related to soil salinity. Higher CEC values typically indicate higher levels of soluble salts in the soil, which can promote the dissolution and diffusion of As [30,31]. The positive correlation between Cd and AP is related to the application of phosphate fertilizers. Suciu et al. pointed out that phosphate fertilizers often contain Cd impurities, and long-term application can lead to the accumulation of Cd in the soil [32]. The positive correlation between Hg and pH is somewhat unusual. Farella et al. pointed out that the release of mercury in deforested soils is caused by the accumulation of alkaline cations. However, in this study, Hg accumulated in soils with higher pH, which may be related to its binding with organic matter or the influence of specific land use practices (such as construction land) [29]. The positive correlation between Cr, Cu, Pb, and TK is attributed to the application of potassium fertilizer or the symbiotic relationship between heavy metals and potassium minerals in the parent material of the soil [33,34]. This conclusion supports the potassium ore coexistence hypothesis [29]. The positive correlation between Ni and Zn and TP is consistent with the findings of Jalali et al., reflecting the contribution of phosphorus fertilizer or organic fertilizer application to the input of Zn and Ni [35]. These results suggest that the dynamics of soil nutrients may directly affect the fate and bioavailability of heavy metals, especially in agricultural activities, and that fertilizer management needs to take into account its potential impact on heavy metal accumulation.

Nutrient–heavy metal relationships in cropland are significantly influenced by anthropogenic activities (e.g., fertilizer application and irrigation). As was positively correlated with total nitrogen (TN) and negatively correlated with immediate potassium (AK) and pH, suggesting that nitrogen fertilizer application may promote the release of As, while acidic soils may enhance As activity [36]. Cd was positively correlated with AP, organic carbon (SOC), and TN, and it is possible that agricultural inputs (e.g., phosphorus fertilizers, organic fertilizers) are important sources of Cd. The positive correlation of heavy metals such as Cr, Cu, Ni, and Zn with SOC, TN, and TP further suggests that organic matter and nutrient enrichment promote the immobilization or co-precipitation of heavy metals but may also increase their bioavailability through chelation [37].

The positive correlation between Hg, SOC, and TP in construction land is attributed to industrial pollution or the input of organic waste. Hg readily binds with organic matter to form stable complexes [38]. This result suggests the risk of industrial organic waste input, similar to the Taian study on “Hg in urban soil combined with organic matter [28]”. The correlation patterns were more similar for forest and garden land, and heavy metals such as As, Cd, and Cr were all significantly and positively correlated with SOC and TN, suggesting that the accumulation of organic matter plays a dominant role in the adsorption of heavy metals under natural vegetation cover [39]. It is worth noting that almost all heavy metals in orchard soils are negatively correlated with pH, suggesting that acidic conditions may enhance the mobility of heavy metals, especially in orchard management, where one needs to be wary of the risk of contamination due to acidification [40]. Pb was positively correlated with AP and SOC in the garden and related to the historical use of Pb-containing pesticides. The correlation between soil nutrients and heavy metals was significantly influenced by land use type and anthropogenic activities. The contribution of nutrient management to heavy metal accumulation was particularly strong in cropland and garden land, whereas the pattern of association in woodland and built-up land was more dependent on natural processes.

Future research should focus on investigating the interaction mechanisms between soil heavy metals and their physical and chemical properties under different land use types, particularly the impact of agricultural activities on heavy metal accumulation and the role of natural processes in heavy metal distribution. Advanced analytical techniques, such as isotope tracing and nanoscale characterization, should be employed to elucidate the sources of heavy metals, their migration and transformation pathways, and changes in their bioavailability. Additionally, monitoring and research on irrigation water quality and its contribution to soil heavy metal pollution in surrounding areas should be strengthened to accurately assess the potential risks of irrigation water to agricultural ecosystems. Furthermore, conducting long-term field experiments to simulate the effects of different management measures (such as fertilizer types, irrigation regimes, crop rotation, etc.) on soil heavy metal dynamics will provide a scientific basis for developing effective soil pollution prevention strategies and nutrient management plans. Finally, by integrating geographic information systems (GISs) and big data analysis technologies, a soil heavy metal pollution risk warning system should be established to achieve precise management and ecological restoration, ensuring soil health and sustainable agricultural development.

5. Conclusions

5.1. Characterization of Soil Heavy Metal Content and Its Spatial Heterogeneity

Soil heavy metal content in the arid oasis area showed land-use-dependent differences, and soil heavy metal content varied among different land-use types: As was found in the widest range (5.27–40.20 μg/g) in arable land, with maximum values significantly higher than in other types, and was influenced by both soil-forming parent material and agricultural activities; Cd had the highest maximum value (0.29 μg/g) in the garden, and the mean value was slightly higher than the other types, which need to be alerted to ecological risks; Cr and Cu were higher in cropland, woodland, and garden land, and were associated with soil-forming matrices or agricultural inputs; Zn showed abnormally high values (412.00 μg/g) in the garden, which was related to industrial pollution or fertilizer application. In spatial distribution, heavy metals were scattered in patches with no regular trend, in which high Cr values were concentrated in cultivated land and garden land in the north and the right of the center. The high values of Zn are distributed in the right center and in the southern part of the parkland and construction land, indicating that human activities (industry, agriculture) and natural factors drive the spatial heterogeneity.

5.2. Potential Driving Effects of Heavy Metals in Irrigation Water on Surrounding Soil Contamination

Differences in the match between heavy metal concentrations in irrigation water and neighboring soils are as follows: The soil around the water sample site (No. 3) with high As values shows a corresponding high value area, indicating that irrigation water is a possible source of As input. Soils around the high Cr water samples (#3 and #5) showed low values, suggesting that soil Cr enrichment was dominated by industrial sources rather than irrigation water. The concentrations of Cd, Cu, Hg, and Pb were stable in water samples, with no significant correlation with soil distribution, and soil enrichment was mainly affected by the type of land use (e.g., fertilizer application in gardens, disturbance of construction land). Overall, heavy metal pollution in irrigation water is a “combination of point source (e.g., industrial wastewater) and surface source (e.g., agricultural irrigation)” pattern, which needs to be combined with the type of land use (e.g., gardening, cropland) for targeted control.

5.3. Interaction Mechanisms Between Soil Heavy Metals and Soil Physicochemical Properties

Soil nutrient and heavy metal correlations were significantly regulated by land use type: In arable land, As was positively correlated with CEC and total nitrogen (TN), and negatively correlated with pH, reflecting the enhancement of its activity by N fertilizer application and acidic conditions. Cd was positively correlated with quick-acting phosphorus (AP) and organic matter (SOC), and directly correlated with phosphorus and organic fertilizer inputs. In the garden, almost all heavy metals were negatively correlated with pH, and the acidic environment exacerbated the risk of heavy metal migration. Pb was positively correlated with AP and SOC, which were related to the historical use of lead-containing pesticides. In forest land, heavy metals were significantly positively correlated with SOC and TN, and organic matter adsorption dominated heavy metal fixation. In construction land, Hg was positively correlated with SOC and total phosphorus (TP), suggesting that industrial pollution and organic matter complexation need to be emphasized. In summary, soil nutrient cycling (e.g., fertilization, organic matter decomposition) affects heavy metal storage patterns and bioefficacy, forming a “land use-nutrient-heavy metal” interactions network, which needs to be combined with the development of differentiated pollution prevention and control strategies for different functional areas.