Migration, accumulation, and risk assessment of potentially toxic elements in soil-plant (shrub and herbage) systems at typical polymetallic mines in Northwest China

In grassland systems of the semi-arid mining area, the migration, accumulation, and bioavailability of potentially toxic elements (PTEs) are important ecological and health risk issues. Thirty-eight pairs of topsoil (0–20 cm) and plant samples were collected around Baiyin City and in Dongdagou stream valley to investigate the migration of PTEs in soils, transfer of PTEs in soil-plant (shrub and herbage) systems, and assess the risk in soils and plants. The total concentrations of PTE (Hg, As, Cu, Zn, Cd, and Pb) were analyzed following digestion in mixture acid solution, and bioavailable PTE was extracted with a strong chelating agent (DTPA-TEA-CaCl2). The transfer factor (TF) and bioaccumulation factor (BCF) were calculated to examine the migration of PTEs in soil-plant. Hazard quotient (HQ) and total hazard index (THI) were calculated to assess the risk and migration of PTEs in soils. The results showed that PTEs in soils and plants of study area exceeded the soil background value and Hygienic Standard for Feeds. Correlation among the total Hg, As, Cu, Zn, Cd, and Pb in soils of Dongdagou stream valley was significant at p < 0.01. A good correlation was exhibited between PTEs in root/aboveground parts of plants and DTPA-soil extractable. Difference of TF and BCF was existed between Dongdagou stream valley and around Baiyin City. Hg, Cu, Zn, Cd, and Pb were mainly accumulated in soils near the mining area. The calculated THI exceeded 1, and As and Pb were the major risk factors. The ability to absorb and transfer Hg, As, Cu, and Pb of plants was lower in more serious polluted area. As had a stronger migration capacity in study area. PTEs in soils had an adverse health effect for residents, and PTEs in plants may cause toxicity to cattle and sheep.


Introduction
Potentially toxic element (PTE) pollution of soil caused by mining activities was a serious problem worldwide (Ding et al. 2017). During the weathering process of mining waste, heavy metals will diffuse into soil and sediments (Souza et al. 2015;Li et al. 2017a). Waste water from mining activities discharged into ditches posed serious risks to the ecosystem in the basins. The normal function of soil ecosystems and human health faces a risk, due to the long-term enrichment of PTEs in soil (Odukudu et al. 2014;Duan et al. 2015;Wu et al. 2016a). Many studies have shown that soils near the industrial area have been contaminated by PTEs, which lead to the enrichment of PTEs in plants (Xue et al. 2014;Wang et al. 2016;Li et al. 2017b). The mobility of PTEs also attracts great attention. Santos-Francés et al. (2011) reported that distribution of total Hg was related to atmospheric input. In chemical processes such as ion exchange and precipitation, adsorption-desorption has been found to be the most relevant process affecting the transport, toxicity, and bioavailability of heavy metals on soil or sediment colloidal surfaces (Holm et al. 1996;Singh and McLaughlin 1999). The immobilization and mobility of PTEs are dependent on such factors as pH, temperature, ionic strength, cation exchange capacity (CEC), surface area, and clay particle size (Gray et al. 1999;Galunin et al. 2014). In general, for most metals, lower pH results in an increase in the solubility of many forms of the metal, with the exception of metals in the form of oxyanions or amphoteric species (Borma et al. 2003). Kraus and Wiegand (2006) revealed that heavy weathering accelerated PTE migration and contamination of the surrounding soil, and Cd and Zn were highly mobile and showed strong displacements in acidic surroundings, Pb was the immobile element, while Cu and As showed a variable distribution in soils and sediments.
The transfer of PTEs from soil to plants is one of the major routes of human exposure to contaminated soil (Khan et al. 2007;Liu et al. 2013). Some elements (e.g., Cu, Cr, Ni, and Zn) are essential for the healthy functioning and reproduction of plants and animals. However, these essential elements may cause adverse effects at high concentrations (Adamo et al. 2014). For some non-essential elements (e.g., As, Pb, and Hg), even low contents can result in toxicity to plants or animals (Nagajyoti et al. 2010). Chemical forms, mobility, and bioavailability of PTEs determined their potential toxicity in soil (Sanjay et al. 2011;Huang et al. 2012). Some studies reported that the uptake, accumulation, and redistribution of metals by plants depend on the total and bioavailable concentrations of PTEs in soils (Liu et al. 2015;Zeng et al. 2011). The availability of PTEs in soil is often affected by soil properties, including pH, total organic matter (TOM), and soil redox potential (Dai et al. 2004;McBride 2002;Zeng et al. 2011).
Baiyin City is an important copper mining in northwest of China, and sierozem is the main soil type (Zang et al. 2017). Several researchers found that the agricultural soil of Baiyin, located at the south of the mining area, was seriously contaminated by PTEs (Nan and Zhao 2000;Zang et al. 2017;Zhang 2018;Li et al. 2019). In the semi-arid area like Baiyin City, grassland is an important ecosystem, and how the mobility of PTEs in soil-plant systems and space is an attractive hot spot. Migration of PTEs, especially Hg and As, in soil and soil-plant systems in semi-arid industrial and mining oasis region of northwestern China has been scarcely studied so far. Therefore, this study takes Baiyin, a typical industrial and mining city in northwest China, as a case study to complement the research in this area. The findings obtained in this study can provide reference for similar research fields and other fields.
In this study, we measured PTEs (Hg, As, Cu, Zn, Cd, and Pb) of grassland soils and plants in Dongdagou stream valley, which is a stream that flows through the mining area, and around Baiyin City. The aims of this study were to (1) examine the accumulation of PTEs in grassland soils and plants, (2) investigate the migration of PTEs in soils and soil-plant systems of study area, and (3) assess the ecological risk of PTEs in soils and plants.

Study area and sample collection
The study area is located in Baiyin City, Gansu Province, northwest of China, as shown in Fig. 1. Baiyin City is a major non-ferrous metal mining, especially copper resources, in China. The average temperature of the study area is 9.2 °C, and the annual precipitation is 244 mm, which belongs to the temperate semi-arid climate. Dongdagou stream is one of the main rivers in Baiyin City, which received treated and untreated domestic and industrial wastewater. Dongdagou flows through the eastern part of Baiyin City and finally into the Yellow River, and almost all nonferrous metal mining areas in Baiyin City are located in the upper reach of Dongdagou.
Thirty-eight pairs of grassland topsoil and plant (shrub and herbage) samples were collected in Dongdagou stream valley and around Baiyin City in November 2018, as shown in Table S1. A 1-m 2 area was set as the sampling area in each sampling point, and three topsoil (0-20 cm) samples were collected and mixed thoroughly. The whole plants with roots were collected in the sampling area and then separated the roots and aboveground parts in the laboratory. The collected plant species include Cynodon dactylon (L.) Persoon, Halogeton glomeratus (Bieb.) C. A. Mey., Suaeda glauca (Bunge) Bunge, Achnatherum splendens (Trin.) Nevski, Lolium perenne L., Setaria viridis (L.) P. Beauv., Nitraria tangutorum Bobrov, and Reaumuria soongarica (Pall.) Maxim. Collected samples were transported to the laboratory for processing within 2 days.

Chemical analysis of soils and plants
The soil samples were air-dried at room temperature. The plant samples were thoroughly cleaned with deionized water, and then oven dried at 105 °C for 2 h and oven dried at 75 °C to constant weight. The soil pH and EC were determined by a pH meter (PHS-3C, REX, Shanghai, China) and an EC meter (DDS-307, REX, Shanghai, China) in a soil/ water suspension (1:2.5). OM was determined by potassium dichromate oxidation method (Lu 2000). The CaCO 3 content of soil was titrated with HCl. Microwave digestion system (Anton Paar, Multiwave PRO 3000) was used to digested soil and plant samples. Hg and As in samples were extracted by a digestion solution of HCl and HNO 3 (Zhang et al. 2018), while Cu, Zn, Cd, and Pb in soils were extracted by a digestion solution of HNO 3 , HCl, HF, and H 2 O 2 , and in plants were extracted by a digestion solution of HNO 3 and H 2 O 2 . The DTPA extractable PTEs in soils were extracted by DTPA-TEA-CaCl 2 . Hg and As in extraction solution were measured by atomic fluorescence spectrophotometry (AFS-8220, Beijing Jitian, China), and other metals were measured by atomic absorption spectrophotometry (Thermo Fisher, SOLAAR M6).

Quality control and assurance
Total contents of PTEs in soils and plants were determined in triplicate. Standard samples GBW-07408 (GSS-8), GBW10046 (GSB-24), and blanks were used to ensure the accuracy of metal element content, and the recoveries were within 100 ± 10%. All glass and plastic containers were soaked in 10% (v/v) HNO 3 solution for more than 48 h and washed with deionized water before utilization. The chemical reagents used were guarantee reagent.

Statistical analysis
Data normality was determined using the Kolmogorov-Smirnov (k-s) normality test method before considering the statistical test. After testing the data for normality, the homogeneity of variances was tested using one-way analysis of variance ( Fig. 3 represented the distance between sampling point and the edge of the mining area (green line in Fig. 1), which was measured by ArcGIS 10.5, and the average value was taken when distance between sampling points is less than 50 m.

Health risk assessment, TF, BCF, and geochemical background
Human beings may be exposed to soil PTEs mainly through three ways: direct ingestion, inhalation, and dermal absorption. In this study, the health risk assessment procedure provided by the Ministry of Ecology and Environment of China (MEEPRC 2019) was used to calculate the exposure risk of residents. The exposure from direct ingestion, dermal absorption, and inhalation was calculated as follows: The values and descriptions of the factors used to estimate risk are shown in Table S2. HQ or THI < 1 indicates that health risk from soil PTEs is at acceptable level, while HQ or THI > 1 indicates adverse health is likely to occur on the human body.
The TF was used to determine the translocation of PTEs from root to the aboveground parts of plants (Gupta et al. 2008), and the TFs were calculated as follows: where ρ and C i represent the PTE concentrations in the aboveground parts and root (mg kg −1 ), respectively.
The BCF referred to the ratio of PTE concentrations in aboveground parts of plants to that in soil (Yao et al. 2019), and the BCFs were calculated as follows: where σ is the contents of PTE in the aboveground parts of plants (mg kg −1 ), and C j was the contents of PTEs in soils (mg kg −1 ).
The calculated distribution function (CDF) method suggested by Matschullat et al. (2000) was used to calculate the geochemical background (GB). The method considers that the lower values should be free from anthropogenic influences, so that a new dataset (reduced data set) can be constructed with minimum to median values. Under this premise, a distribution function can be constructed. The new calculated values may HQ i = HQ ois + HQ dcs + HQ pis depict the natural range of the distribution of elements in the environmental medium. The new values were calculated as follows (Pujiwati et al. 2022): where C Nn is the nth new calculated values, median is the median concentration of the original dataset, and C On is the nth concentration value below the median in the original dataset. The GB range was calculated as follows: The potential non-carcinogenic risks of PTEs to human body were calculated as follows: where range is the GB range, mean is the arithmetic mean, and θ is the standard deviation for the calculated dataset.

Soil properties and contents of PTE in soil
As shown in Table 1, the pH of soils in Dongdagou stream valley and around Baiyin City were ranged from 6.47 to 7.91 and 6.45 to 9.4, respectively. The mean pH in soils of Dongdagou stream valley was obviously lower than the background value, which indicated that soils in Dongdagou stream valley were acidified due to mining activities Ma et al. 2021). Soil in the study area was seriously salinized, which could be found from the EC value of soils. CaCO 3 content in soils of Dongdagou stream valley was significantly lower than the background value for sierozem (118 g kg −1 ) (CNEMC 1990), and the loss of CaCO 3 was probably caused by soil acidification in the study area.
The concentrations of PTE in almost all soil samples of the study area exceeded the background values. And the GB results calculated by CDF suggested that the study area had a high GB value of Cu, Zn, Cd, and Pb, which was due to the existence of particular mineral deposits in the study area. PTEs of most soil samples exceeded the GB upper values, which indicated some contamination and enrichments of PTEs in the study area. It was worth noting that PTEs in 13 soil samples of Dongdagou stream valley even exceeded the risk intervention values. The maximum concentrations of As, Cd, and Pb in Dongdagou stream valley were reached 222.69 mg kg −1 , 148.13 mg kg −1 , and 5612.30 mg kg −1 , respectively, which indicated that PTEs in soils may pose risks to human health (MEEPRC 2018). Compared with the grassland around Baiyin City, the pollution of PTEs in soil of Dongdagou stream valley was more serious.
The CV of PTEs in soil from Dongdagou valley decreased in the order Pb (164%) > Hg (152%) > Cd (145%) > Cu (129%) > Zn (67%) > As (56%). The large CV for Pb, Hg, Cd, and Cu indicated that PTE contents varied greatly at different sites (Xiao et al. 2015). The biggest CV of PTEs in soils around Baiyin City was Cd (179%). The contents of soil PTE in the east of Baiyin City were much higher than that in other directions of Baiyin City, because the eastern region of Baiyin was more easily affected by mining activities.

Correlation analysis
As shown in Table 2, the correlation among the total Hg, Cu, Zn, Cd, and Pb in soils of Dongdagou was significant at p < 0.01, but in the soils around Baiyin City, this correlation was not so obvious. PTEs except As exhibited significant negative correlation with pH in soils of Dongdagou. These results implied that there was a high degree of homology among Hg, Cu, Zn, Cd, and Pb in soils of Dongdagou stream valley (Manta et al. 2002), and Hg, Cu, Zn, Cd, and Pb concentrations decreased with increasing pH in Dongdagou. However, this result was not found in the grassland soil around Baiyin City, which may be because soil pH around Baiyin City was not significantly affected by mining activities, making this correlation weaker than that of Dongdagou stream valley.
Although the As concentration in soils of Dongdagou stream valley was very high, there was no good correlation between As concentration and physicochemical properties (except CaCO 3 contents) and other PTEs. Different from other metal ions, oxidation conditions and alkaline conditions promote the mobility of As (Williams et al. 2005;Biswas et al. 2014;Wu et al. 2016b;Torres et al. 2017), which may be the reason for the poor correlation between As and other metals.

Migration of PTEs in soil-plant systems
The TF and BCF in plants from Dongdagou stream valley and around Baiyin City are presented in Table 3. The mean TF and BCF from Dongdagou valley decreased in the order Pb (1.12) = Zn (1.12) > Hg (0.8) > Cd (0.62) > Cu (0.36) > As (0.31) and Cu (0.46) > Hg (0.32) = Pb (0.32) > Cd (0.29) > Zn (0.21) > As (0.12), while the mean TF and BCF around Baiyin City decreased in the order Hg (2.34) > Pb (1.59) > As (1.09) > Zn (0.96) > Cu (0.8) > Cd (0.41) and Cu (0.67) > Pb (0.47) > Hg (0.39) > Zn (0.34) > As (0.28) > Cd (0.13). Both TF and BCF of Hg, As, Cu, and Pb in Dongdagou stream valley were lower than that around Baiyin City, which may be because the absorption level of PTEs by plants in different regions had changed under metal stress (Bhat et al. 2019;Li et al. 2021). Dian and Giok (2017) reported that the efficiency of metal translocation may be affected by the systems responsible for capillary action in plants. Due to the adaptive responses of plant to pollutants, many plant species had become metal tolerant, as these plants were growing in the contaminated area from a long period (Sainger et al. 2011). The results also demonstrated that efficiency of Pb and Zn transported from roots to aboveground parts in Dongdagou stream valley was higher than that of other PTEs which was consistent with Singh et al. (2010) and Loris et al. (2022). Migration of Cu, Hg, and Pb from soil to plant was higher than that of As, Cd, and Zn in this study, which indicated that Cu, Hg, and Pb in plants of study area should be given attention more than other metals. In other studies, the transfer coefficient value for Cd was higher than other metals (Bergmann 1992;Coumar et al. 2015). This may be because plants in the study area have only so much ability to accumulate heavy Table 1 Grassland soil properties and PTE contents in Dongdagou stream valley and around Baiyin City   Chang et al. (1987) and Barbarick et al. (1995).
As shown in Fig. 2, a good correlation was exhibited between PTEs in root/aboveground parts of plants and DTPA extractable PTEs in Dongdagou stream valley (except Hg and Zn) and around Baiyin City (except Hg and As). The results indicated that DTPA extractable PTEs could well reflect the absorption efficiency of plants for metals (Soriano-Disla et al. 2010;Wan et al. 2020;Xing et al. 2020). The DTPA extraction of Cu, Zn, and Pb had been satisfactorily predicted in some studies (Hooda et al. 1997;Brun et al. 2001;Meers et al. 2007;Soriano-Disla et al. 2010). But extraction with DTPA was not a good method to assess phytoavailability of Hg, which was consistent with Luis Rodríguez et al. (2017). Higueras et al. (2003) reported that most Hg was in cinnabar form or bound to organic matter near mining area, and both forms seemed not to be phytoavailable.
It could be seen that there was a significant positive correlation between contents of PTEs in the aboveground parts and roots in Dongdagou stream valley, but this result was not found in Hg and As around Baiyin City. It showed that in the study area, PTEs could still be transported from the root to the aboveground part even if plants were stressed by complex PTEs. At the same time, it showed that the absorption capacity of roots to PTEs was one of the main factors affecting the migration of PTEs from soil to aboveground parts of plants. In fact, migration of PTEs from soil to plant depended on the uptake and transport capacity of plant roots and cells for PTEs (Clemens et al. 2020;Yang et al. 2005).

Distribution of physicochemical properties and PTEs in Dongdagou stream valley soil
The bioavailability of PTEs was correlated with soil pH, organic matter content, and texture (Pietrzykowski et al. 2014). Therefore, it was necessary to understand the distribution of soil physiochemical properties in the study area. As shown in Fig. 3, the soil pH was weakly acidic in the upper reaches of Dongdagou, but with the increase of distance, the soil pH increased rapidly and become alkaline. This was because the acid waste water or wastes produced by mining activities had dramatically reduced the soil pH, while the soil in the river valley had a strong buffer effect  on the pH, making the soil pH return to normal soon. EC in soils first decreased and then increased, which was mainly affected by the agricultural activities in the middle reaches of Dongdagou. Salinization was one of the main impacts of agriculture on soil quality (Zalidis et al. 2002), and human activities may lead to soil salinity accumulation in multiple direct and indirect ways (Misopolinos 1990). However, the content of OM in Dongdagou stream valley has always shown a significant downward trend. The migration of PTEs in topsoil along Dongdagou valley is shown in Fig. 4. The contents of Hg, Zn, Cd, and Pb in soils increased first and then decreased. It was worth

Distance(m)
noting that the content of Cd and Pb decreased to a very low level at the end of Dongdagou stream valley. The concentration of Cu in soils was almost always decreasing, but the concentration of As had no obvious trend. These results indicated that Hg, Zn, Cd, and Pb had a similar migration ability, while As had a stronger migration ability than other PTEs in study area. The mobility and bioavailability of Hg in soils was low (Santos-Francés et al. 2011), because Hg 0 , which had a low reactivity, accounted for a large proportion in soils near the smelting plant and mining area (Kocman et al. 2004). Hg 0 was important in terms of environmental risk because it allowed Hg to accumulate near pollution sources without moving further with soil or water. Some previous studies had shown that the retention and mobility of Cd were dependent on such factors as pH, temperature, CEC, and ionic strength (Gray et al. 1999;Lair 2006). Galunin et al. (2014) found that the acidity of environmental samples resulted in more intense competition between Cd 2+ and protons, resulting in weakened cadmium adsorption, which was consistent with the change of Cd in Dongdagou stream valley soil. Mobility of PTEs was also affected by the high affinity in the TOM in topsoil. PTE concentrations in soil profiles decreased with depth, confirming the high accumulation of these elements in organic and organo-mineral topsoil (Godbold and Hüttermann 1985). This was due to the high adsorption and accumulation capacity of PTEs by organic matter and mineralhumus soil complexes (Kabata-Pendias and Pendias 1992). There was also easier migration of PTEs into the soil in sandy soils (Pietrzykowski et al. 2014). The migration of As was possible under extremely high pH conditions (pH > 8) because high pH promoted the dissolution of As forms under reducing conditions, and other anions promoted the desorption of As(V) from the soil solid phase, especially under alkaline conditions (Krysiak and Karczewska 2017). The release of As under alkaline conditions could be explained by its anionic character and the balance between adsorption and desorption processes (Bowell 1994;Liu et al. 2001). Biswas et al. (2014) and Torres et al. (2017) found that a high As(V) adsorption was existed at a lower pH, whereas As(V) tended to be less adsorbed on hydrous oxides at elevated pH levels. Krysiak and Karczewska (2017)

Ecological risk assessment of PTEs in plants
Cattle and sheep were the main herbivorous livestock in the study area. Therefore, the effects of plant PTEs on cattle and sheep were analyzed (Table 4). The maximum tolerable toxic metal level, defined as the dietary level, did not impair accepted indicators of animal health or performance when fed for a defined period of time. It could be seen from It was well established that toxic metals had potential adverse effects in livestock, and naturally occurring episodes of acute and toxic intoxication had been much described in the literature (López-Alonso 2012). Large amounts of PTEs may be transferred from contaminated soils to plants, leading to accumulation of these PTEs in cattle and sheep (López-Alonso et al. 2003;Miranda et al. 2005). Accumulation of PTEs could cause toxic effects in herbivores, also in humans who consume meat and milk contaminated with toxic metals (González-Weller et al. 2006;Vromman et al. 2008;Cai et al. 2009). Therefore, the study area should continue to have PTE pollution of plant safety as a target.

Non-carcinogenic risk assessment of PTEs in soil
The spatial distribution maps of HQ and THI in soils from study area are presented in Figs. 5 and 6. The HQ or THI values were indicated by different colors in the map, and red indicated non-carcinogenic risk beyond the acceptable risk level. The distribution of HQ and THI in study area showed that the non-carcinogenic risk of Hg, Cu, Zn, and Cd in soils of the study area to human body was at an acceptable level, and the high-risk area was distributed in the Dongdagou stream valley, especially near the mining area. The HQ values of Pb in soils near the mining area, which was the upper of Dongdagou stream, exceeded acceptable level, while the risk level of As in Dongdagou stream valley from upstream to downstream was beyond the acceptable level. These results were consistent with the migration of PTEs along Dongdagou stream valley. In terms of the total noncarcinogenic risk of PTEs in soils, Dongdagou stream valley was the risk area, while the health risk of grassland soils around Baiyin City far away from the mining area was at an acceptable level. This result revealed that Dongdagou stream was the most important way of pollutant diffusion in the study area.
Comparing HQ of As and THI distribution maps, it could be found that As contributed the most to health risk in the study area. Cao et al. (2014) found that the health risks associated with arsenic exposure in people should be taken seriously, with ingestion appearing to be the main route of exposure to arsenic and other PTEs, followed by inhalation exposure. Zang et al. (2017) revealed that that Cu was slightly polluted, Zn was moderately polluted, and Pb was practically unpolluted in the middle and end of Dongdagou stream valley. The non-carcinogenic risk assessment results in this study were consistent with the research results of Zang et al. (2017). Compared with previous studies, this study found that the higher risk caused by PTEs was located at the upstream of Dongdagou near the mining area. From the upstream to the downstream of Dongdagou, the risk caused by PTEs, except As, decreased rapidly. The ecological risks caused by Hg and As in Dongdagou stream valley were not mentioned in previous studies. This study revealed that the ecological risks caused by Hg did not spread in a large scale, mainly concentrated in the upstream of Dongdagou, but the high ecological risks caused by As spread throughout the Dongdagou stream valley. Due to the high mobility and toxicity of As, the risks caused by As should be paid more attention. In addition to the strong mobility of As in the study area, the application of various fertilizers and pesticides

Fig. 5
Spatial distribution maps of soil HQ in study area (Lines in the figure are isolines. The darker the blue color means the higher the HQ value, i.e., the higher the health risk, and the red color means the HQ > 1, i.e., adverse health is likely to occur on the human body) in agricultural activities in Dongdagou area may also be another contributor to the serious pollution of As in soil (Luo et al. 2009;Zhou et al. 2018).

Conclusion
The results from this study revealed that soils and plants in grassland of Dongdagou stream valley were contaminated by Hg, As, Cu, Zn, Cd, and Pb due to the influence of mining activities in the past. Ecological risk assessment and non-carcinogenic risk assessment showed that long-term exposure to polluted soil particles may have adverse effects on human health, and plants were not suitable for livestock feed in the study area. Correlation analysis showed that high homology between Hg, Cu, Zn, Cd, and Pb in soils of Dongdagou stream valley was existed. Migration analysis of PTEs indicated that the mobility of Hg, Cu, and Pb from soils to plants was higher than that of As, Cd, and Zn, and DTPA extraction well reflected the phytoavailability of PTEs (except Hg) in soils of the study area. Plants had a lower ability to absorb and transfer Hg, As, Cu, and Pb in more serious polluted area. According to the results of PTEs migration and health risk assessment in study area, As had a stronger migration capacity than other metals and contributed the most to health risk. Pollution of As and plants safety should be paid attention to, and the migration mechanism of As should be further studied.
Authors' contributions Material preparation, data collection, and analysis were performed by Qianfang Yang. Shengli Wang and Zhongren Nan provided the experimental conditions and funds, directed the experiment, and designed the framework of the paper. The first draft of the manuscript was written by Qianfang Yang, and all the authors commented on previous versions of the manuscript. All the authors read and approved the final manuscript.
Funding The work was supported by the National Key Research and Development Program of China (2018YFC1802905).

Data availability
The datasets used or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication Not applicable.
Competing interests The authors declare no competing interests.  Fig. 6 Spatial distribution maps of soil THI in study area (Lines in the figure are isolines. The darker the blue color means the higher the THI value, i.e., the higher the total health risk, and the red color means the THI > 1, i.e., adverse health is likely to occur on the human body)