Pollution status, distribution, source analysis, and risk assessment of OCPs in soil from the Hexi Corridor in Northwest China

Study area and sample collection

The study region (92°13′-108°46′ E and 32°31′-42°57′ N) was represented by five main cities. Fifty soil samples were collected in November 2018 from Jiuquan (A1-A10), Jiayuguan (B1-B10), Zhangye (C1-C10), Jinchan (D1-D10), and Wuwei (E1-E10) [Figure 1]. Collect soil samples using the uniform distribution method, refer to GB/T 36197-2018 for specific operations. Five subsamples were gathered from each sampling site, taken from the same 100 m² area with a depth range of 0-20 cm, which were then combined to create a composite sample. After eliminating stones and residual roots, the samples were air-dried at room temperature and filtered through a 1-mm mesh screen. These samples were subsequently placed in glass bottles and kept at 20 °C below zero until they were analyzed. Soil pH and organic carbon content were evaluated using previously described techniques^[24].

Figure 1. Sampling locations in the study area.

Chemicals and reagents

OCPs mixed standards were purchased from AccuStandard and included 24 individual compounds: α-HCH, β-HCH, δ-HCH, γ-HCH, p,p'-DDD, p,p'-DDT, p,p'-DDE, o,p'-DDT, dicofol (DIC), HCB, heptachlor (HE), heptachlor epoxide (HPE), cis-chlordane (cis-CD), trans-chlordane (trans-CD), α-endosulfan (α-ED), β-endosulfan (β-ED), endosulfan sulfate (EDS), aldrin (AD), dieldrin endrin (DE), endrin aldehyde (EDA), endrin ketone (EDK), mirex (ME), and methoxychlor (MEC). The 100-200 mesh silica gel was purchased from Qingdao Haiyang Chemical Co. (Shandong, China). It was activated by heating it at 130 °C for 16 h, then deactivated by adding purified water (3.3%), and stored in a desiccator. To use, anhydrous Na₂SO₄ was baked at 600 °C for 6 hours. All solvents used were of HPLC grade (Fisher Chemical Company, USA) and all glassware was thoroughly cleaned with detergent, K₂Cr₂O₇-H₂SO₄ solution, running water, deionized water, and acetone. It was then baked at 400 °C for 3.5 h before reusing.

Sample pretreatment and instrumental analysis

Our team followed the procedure detailed in our previous report to evaluate the target OCPs^[25]. First, 10 g soil samples were weighed for each group and mixed with 10 g of anhydrous Na₂SO₄. Next, we spiked this mixture with a surrogate (4-Chloro-m-xylene with a concentration of 2.5 × 10^-4μg/g) and used a low-temperature ultrasonic extractor to extract it with a 200-mL combination of hexane/acetone (1:1 v/v). The resulting extracts were concentrated, and their solvent was exchanged to hexane. We utilized a silica gel column of dimensions 25 cm × 1 cm I.D. containing 6 g of deactivated silica gel topped with 2 cm of anhydrous sodium sulfate. To prepare the column, we first eluted it with 40 mL of hexane before injecting the extract. Then, we proceeded by eluting the column with 80 mL of n-hexane, followed by 35 mL of n-hexane/DCM (2:3 v/v). The process used to determine OCPs using gas chromatography with a micro-electron capture detector (GC-µECD). The detector was equipped with a fused silica capillary column and helium was used as a carrier gas. The temperature program involved maintaining an initial temperature of 60 °C for 2 min, followed by an increase to 190 °C at a rate of 5 °C/min, and then an increase to 280 °C at a rate of 10 °C/min. Quantification of OCPs was performed using a five-point calibration method, and the results were confirmed with an Agilent 6890GC-5975MSD system, which utilized electron impact ionization mode and selected ion monitoring (SIM) mode.

Quality assurance

Table 1 presents the limits of detection (LOD), analysts' recoveries, and relative standard deviations (RSD) obtained by testing natural soil spiked with individual OCP standards (n = 5) to assess the recovery efficiency of the procedure. To ensure proper procedural performance and monitor matrix effects, different blank and triplicate samples were evaluated for each of the five samples, with additional surrogates added to each sample. Soil samples spiked with surrogates showed recoveries ranging from 63.4% to 92.0%, while the OCP variabilities in triplicate samples were no more than 10%. The results were presented on a dry weight basis, and were not adjusted for recoveries.

Pollution characteristics and residual levels of OCPs

In this paper, we collected 50 soil samples from five cities located in the Hexi Corridor and determined the concentrations for 24 individual OCPs. The statistical results of OCPs in the soils are shown in Table 1, and the residual contents of OCPs in the soil of five cities are shown in [Supplementary Table 1]. From Table 1, except for a few sampling locations for individual OCPs, such as EDS, EDA, EDK, ME, and MEC, other individual OCPs were detected in the soil at all sampling locations, which indicated that the OCP contamination in the Hexi Corridor was diverse and ubiquitous. Based on historical reports, the usage of DDTs and HCHs was 15,589 tons and 64,991 tons, respectively, from the 1950s to the 1980s in Gansu Province^[26]. As may be expected, the entire study location, as a main grain and vegetable production location, has relatively widespread OCP residues, mostly likely due to historical usage as well as recent input in the study areas. The contents of ∑₂₄OCPs ranged from 23.1-393 ng/g (average: 94.7 ng/g), the ∑DDTs contents ranged from 4.96-167 ng/g, and the ∑HCHs contents were in a range of 3.40-97.5 ng/g. The sum of AD, DE, EDA, and EDK (∑Drins) was in a range of 4.04-64.3 ng/g, accounting for 12.6% of ∑₂₄OCPs. The total concentration of cis-CD and trans-CD (ΣCHLs) ranged from 0.740-24.1 ng/g. The order of the dominant individual OCPs in the Hexi Corridor was as follows: DDTs (average: 35.8 ng/g) > HCHs (average: 16.0 ng/g) > ∑Drins (average: 11.6 ng/g) > Nonachlor (average: 6.27 ng/g) > Methoxychlor (average: 4.86 ng/g) > CHLs (average: 4.32 ng/g) > HCB (average: 4.28 ng/g) > Endrin aldehyde (average: 4.18 ng/g). The average concentration of heptachlor epoxide (3.30 ng/g) was higher than that of heptachlor (1.75 ng/g), indicating an existing aerobic degradation environment in the areas sampled. Among 24 individual OCPs, the contents of ΣDDTs and ΣHCHs were much higher than that of the other OCPs, accounting for 38.7% and 16.1% of OCPs, respectively. The OCPs in soils from the Hexi Corridor were mainly HCHs and DDTs, and the concentration of DDTs was higher than that of HCHs, which may be attributed to (1) larger amounts of DDTs used in agricultural activities compared to HCHs, and (2) the higher resistance to bio-degradation of DDTs than HCHs^[27]. Among individual OCPs, EDS was relatively low (average: 0.34 ng/g), but it was detected in 98% of the soil samples.

There are many studies on OCP contamination in soil from China and worldwide [Table 2]. The ∑OCPs levels in urban soils vary considerably, ranging from 7.6 to 37,331 ng/g with a median value of 53.7 ng/g. The average content of OCPs is 2,861 ng/g, which is several orders of magnitude higher than the values (58.9 ng/g) reported for agricultural soils in China^[12]. Compared with other regions and countries in the world, the content of ∑HCHs in this study was lower than that of Kalashah Kaku, Pakistan^[28], Spain^[29], and Tianjin^[30] in China. ∑HCHs were comparable to the results from Urumqi^[31], Taiyuan^[32], and Fujian Province^[8] in China. But ∑HCHs were higher than those reported for Hong Kong^[33], Beijing^[34], Guangzhou^[22], and Shanghai^[25]; seven cities in India^[35]; Punjab Province of Pakistan^[36]; Canada; and the UK^[37]. The content of ∑DDTs in soil was lower than that in Romanian soil^[38,39], Germany^[40], Beijing^[34] and Guangzhou^[22] in China; but higher than that reported for the Pearl River Delta^[41], Urumqi^[31] and Hong Kong^[33] in China as well as that reported for India^[42], Nepal^[43], Italian^[13,14], and Poland^[44]. In general, compared with other areas in China and worldwide, the contents of HCHs and DDTs in the soil in the Hexi Corridor fell in the upper middle level. The study area is located in the temperate continental climate, which is characterized by arid and semiarid areas with less wind and rain, and fewer microorganisms in soil, resulting in a slower degradation of HCHs and DDTs.

Spatial distribution of OCPs

Figure 2 and [Supplementary Table 1] show the distribution characteristics and spatial distribution of OCPs in the soils from the study area. OCPs are normally associated with agricultural application contamination^[25], which may explain the differences in OCP levels obtained among the five cities [Supplementary Table 1]. From [Supplementary Table 1], Jiuquan is known to have a long and intensive agricultural and pesticide production history, and therefore the highest average level with a mean of 179 ng/g (ranging from 49.2-393 ng/g) is not surprising. The other three cities, despite some extreme values, had relatively low contents, with Jiayuguan showing a mean of 75.3 ng/g (ranging from 45.8-153 ng/g), very similar to Wuwei with a mean of 74.9 ng/g (ranging from 23.1-129 ng/g) and Jinchang with a mean of 84.1 ng/g (ranging from 56.9-125 ng/g). very similar to Wuwei with a mean of 74.9 ng/g (ranging from 23.1-129 ng/g) and Jinchang with a mean of 84.1 ng/g (ranging from 56.9-125 ng/g). Zhangye showed the lowest mean value (60.4 ng/g and ranging from 38.8-96.7 ng/g). A little surprising, as the major agricultural production area, this low level might be attributed to the rise of organic agriculture and facilities farming, resulting in a high degradation of OCPs. However, this hypothesis needs further investigation. For individual sampling locations, it can be seen from [Supplementary Figure 1] that the highest value (393 ng/g) of OCPs is located in A3 (Jiuquan), and the lowest value (23.1 ng/g) is located in E2 (Wuwei). The highest HCHs value (29.2 ng/g) was in A7 (Jiuquan), and the lowest HCHs value (14.7 ng/g) was in E1 (Wuwei). The highest value of DDTs (42.9 ng/g) was in A4 (Jiuquan), and the lowest value (6.93 ng/g) was in C3 (Zhangye).

Figure 2. Spatial distribution of (A) for ∑₂₄OCPs; (B) for ∑HCHs; and (C) for ∑DDTs in soil from Hexi Corridor in Northwest China.

The contents of p,p'- DDT and p,p'-DDD were the highest among detected individual OCPs, which were 1.35-93.6 ng/g and 1.54-40.7 ng/g with a mean of 18.3 ng/g and 8.33 ng/g, respectively. As can be seen from Figure 2, the highest contents of DDTs appear at sampling locations A3, A4, and A7 (Jiuquan City). The order of content of ∑₂₄OCPs in the soils of five cities from highest to lowest in Hexi was Jiuquan > Jinchang > Jiayuguan > Wuwei > Zhangye. The order of ∑HCHs residual content was Jiuquan > Wuwei > Jinchang > Jiayuguan > Zhangye. The order of ∑DDTs residual content was Jiuquan > Jinchang > Jiayuguan > Wuwei > Zhangye. Most of the sampling locations in Jiuquan were near cultivated land and farmland. Previous research data indicates that before the prohibition of OCPs, the proportion of HCHs and DDTs in the region accounted for more than 50% of the total pesticide use^[23]. Therefore, there were high residual amounts of HCHs and DDTs. In addition, Jiuquan is a city where industry and agriculture jointly develop, while Zhangye is an agricultural city. Therefore, the sources of OCPs in Jiuquan soil are more extensive than those in Zhangye. Thus, the residual OCPs in the soil of Jiuquan are higher compared to other cities.

The composition of OCPs

Composition characteristics of HCHs

Figure 3A displays the HCH composition present in the soil of the Hexi Corridor. All four HCH isomers (α, β, γ, and δ) were detected with a 100% detection rate, and the total residual concentration of HCHs ranged from 3.40 ng/g to 97.5 ng/g, with an average of 16.0 ng/g. The residual concentration of individual isomers was as follows from highest to lowest: δ-HCH (6.20 ng/g) > β-HCH (4.33 ng/g) > γ-HCH (3.79 ng/g) > α-HCH (1.71 ng/g). The percentage of HCH isomers found in soil samples was as follows: δ-HCH (30.1%) > β-HCH (29.9%) > γ-HCH (27.0%) > α-HCH (9.03%). The percentage of δ-HCH found in Jiuquan soil (30.1%) was the highest among all sampling sites, indicating the possibility of historical residues in the study areas in the Hexi Corridor. Data show that before the prohibition of OCPs, the proportion of HCHs and DDTs in the region accounted for more than 50% of the total pesticide use^[23]. Moreover, δ-HCH, as a difficult-to-biodegradable isomer of HCHs, has a high residual amount in soil.

Figure 3. Percentage of the individual isomers of HCH (A) and DDTs (B) in soils. DDTs: Dichlorodiphenyltrichloroethanes; HCH: Hexachlorocyclohexane.

The relationship between total organic carbon and OCPs

TOC in soil has the ability to adsorb OCPs due to its hydrophobicity properties. Furthermore, TOC provides a sufficient amount of carbon to reduce the levels of OCPs through bio-degradation^[46,47]. Accordingly, TOC can affect the behavior and risk of OCPs in soil. In the present study, TOC in soil was in a range of 4.43-28.4 g/kg with a mean of 11.9 g/kg. To get insights into the role of TOC, the contents of Σ₂₄OCPs, ΣHCHs, and ΣDDTs were plotted with TOC concentration using a scatter plot [Supplementary Figure 2]. The results showed that the OCPs were positively correlated with TOC concentrations in soil (Σ₂₄OCPs with R² = 0.609, ΣDDTs with R² = 0.671, and ΣHCHs with R² = 0.409 at P < 0.05 [Supplementary Figure 2]. This suggests TOC in soil plays a crucial role in adsorbing OCPs^[25,48]. In an equilibrated soil-air system, there is a proportional relationship between the residual OCP concentrations and the soil organic matter (SOM)^[49]. However, if the system is imbalanced or if other complex soil conditions exist, deficient correlations may result^[50]. In a previous report, we also noticed a weak correlation between TOC and OCPs^[25]. Nevertheless, our current study has found a stable linear correlation between OCPs and TOC, which agrees with prior investigations in Central Europe^[17] and the Central Tibetan Plateau^[51].

Source apportionment of OCPs in soil

The proportion of parent compound to its metabolites can offer valuable insights into the origin of contamination^[52]. The technical HCH compound typically contains around 60%-70% α-HCH, 5%-12% β-HCH, 10%-12% γ-HCH, and 6%-10% δ-HCH (UNEP). In addition, lindane (> 99% β-HCH) was also used, despite the ban on technical HCH^[53]. β-HCH is the most easily absorbed by SOM and less likely to evaporate compared to other isomers. Meanwhile, α-HCH and γ-HCH can transform into β-HCH in soil, and the molecules' chlorine atoms in β-HCH resist degradation from microorganisms in soil, causing its accumulation. This study employed the ratios of α-HCH/γ-HCH and β-HCH/(α-HCH+γ-HCH) to trace the source of HCHs. Ratios of α-HCH/γ-HCH greater than 3 but less than 7 indicate technical HCH contamination, whereas a ratio less than 3 suggests new lindane input^[54]. For the β-HCH/(α-HCH + γ-HCH) ratio, values over 0.5 denote historical contamination, while lower values indicate pollution from recent pesticide use and atmospheric deposition. Results reveal the presence of HCH contamination in all sampling sites, with low ratios of α-HCH/γ-HCH, indicating early-stage lindane input. The β-HCH/(α-HCH+γ-HCH) ratios are in a range of 0.12-3.57, with 70% of sampling locations > 0.5 [Figure 4A], signifying historical application, mainly in the Zhangye region, indicating contamination of the Hexi corridor with HCHs from past sources and possible new contamination.

Figure 4. Ratios of the individual isomers of HCH(A); DDT(B); CDs, and EDs (C) in soils

Studies indicate that DDT is a commonly used insecticide, with p,p'-DDE and p,p'-DDD being the primary products resulting from the dechlorination of p,p'-DDT^[55]. The ratio of p,p'-DDT to (p,p'-DDE + p,p'-DDD) is utilized to determine whether the input of DDT in the environment is increasing or decreasing. A low ratio of p,p'-DDT/(p,p'-DDE + p,p'-DDD) suggests historical DDT use, while a value > 2.0 denotes recent application^[56]. As observed in Figure 4B, the majority of values of p,p'-DDT/(p,p'-DDE + p,p'-DDD) were < 2.0 (ranging from 0.204 to 3.70), indicating that DDT in the soil of the sample points primarily originated from the past and had been bio-converted to DDD and DDE. Additionally, a DDD/DDE ratio of less than 1 was observed in 24% of sampling sites, suggesting that DDT had undergone aerobic bio-degradation, influenced by the environment's oxygen content and soil porosity. Notably, the technical endosulfan accounted for 70% and 30% of α and β-ED, correspondingly, with a ratio of almost 2.33^[25]. A ratio of α-ED/β-ED < 2.33 indicates aged soil residues as the α-isomer decomposes more effortlessly in comparison to the β-isomer in soil. The ratio of cis-CD/trans-CD in technical mixtures is reported to be about 0.77^[57]. Previous reports showed that trans-CD is easier to degrade than cis-CD in the environment^[58], and the ratio of cis-CD/trans-CD > 1 is generally indicative of aged chlordane^[59]. The concentration range of cis-CD was 0.113-4.92 ng/g, with an average of 1.05 ng/g, while trans-CD ranged from 0.292 to 21.1 ng/g, with an average of 3.30 ng/g. The cis-CD/trans-CD ratio range was 0.66-18.8, with a mean value of 5.16 [Figure 4C], indicating that CD residues mainly resulted from past use. Additionally, the level of HE was commonly lower than that of HPE, indicating no recent input in the study areas in Northwest China's Hexi Corridor. Nonetheless, a high level of HE was found in soils of numerous locations, suggesting HE is still being introduced into the study area's soil.

Pollution and risk assessment

Health risk assessment

Exposure risk assessment of OCPs in soils was conducted by the USEPA method^[61]. The non-carcinogenic and carcinogenic risks for children and adults under three exposure pathways (ingestion, dermal contact, and inhalation) were estimated and established upon Equations (1)-(5). The applied equations of OCP exposure in different exposure pathways were presented as follows:

where C is the concentration of OCPs, CDI_Ing is ingestion exposure, OSIR is daily soil intake, ED is exposure years, EF is the rate of exposure, CF is conversion factor, BW is average weight, AT is average contact time, SAE is skin area touching the soil, SSAR is soil-skin adhesion coefficient, ABS is absorption factor, DAIR is daily air volume, PEF is formation coefficient of soil dust, CDI_Dermal refers to dermal contact exposure, and CDI_Inh is inhalation exposure. The exposure risk is divided into the carcinogenic risk (CR) and the non-carcinogenic effect (HI).

where CR is the composite index of carcinogenic risk, CR_i represents the three exposure routes, HI is the composite non-carcinogenic risk index, HQ_i expresses the non-carcinogenic risk of an exposure path, CDI_i is the soil exposure amount under a certain exposure path, R_fD_i is the reference dose of a non-carcinogen for an exposure route. The related parameter used in the Eq(1)-Eq(5) is presented in Supplementary Table 2 and Supplementary Table 3.

The CR (HCHs, DDTs, HCB, aldrin, and chlordane) and the HI (HCHs, DDTs, HCB, aldrin, CDs, and EDs) were evaluated in the soil of the Hexi Corridor [Supplementary Table 4]. Generally, CR < 1.0 × 10^-6 is an acceptable risk level, while CR ranging from 1.0 × 10^-6 to 1.0 × 10^-4 indicates a medium cancer risk to humans. CR above 1.0 × 10^-4 is considered to be a relatively high cancer risk^[14]. Meanwhile, HI > 1 indicates a higher health damage risk, while < 1 indicates a low health risk to humans. Based on the calculated results, the total CR fluctuated in a range of 1.86 × 10^-7 to 5.62 × 10⁻⁶ for children and 1.21 × 10^-7 to 1.91 × 10^-6 for adults. Moreover, for the three exposure pathways, the CR through ingestion was higher or equal to the risk through dermal contact, but two or three orders of magnitude higher than the risk through inhalation. The order of the means of carcinogenic risk of OCPs in soil was Jiuquan (1.80 × 10^-6 for children, 6.12 × 10^-7 for adults) > Jinchang (7.60 × 10^-7 for children, 2.58 × 10^-7 for adults) > Jiayuguan (7.26 × 10^-7 for children, 2.46 × 10^-7 for adults) > Wuwei (7.31 × 10^-7 for children, 2.48×10⁻⁷ for adults) > Zhangye (5.60 × 10^-7 for children, 1.90 × 10^-7 for adults) [Supplementary Table 4]. According to the CR index [Supplementary Table 4], the CR index of Jiuquan is higher than that for other regions, and the carcinogenic risk of OCPs in the soil was higher for children than for adults. HI of OCPs for children and adults ranged from 0.032 to 0.932 (average: 0.458) and 0.093 to 0.281 (average: 0.155), respectively, and all of these were < 1, indicating a low health risk. For HI, HCHs contributed the most, accounting for more than 50% on average, followed by DDT, and CHL [Figure 5]. In summary, fortunately, the levels of OCPs in very few of the collected soil samples have certain carcinogenic and other health risks to humans; however, these should receive further research and management attention. According to the risk index of different exposure routes, the main exposure pathway was ingestion, followed by dermal contact and inhalation. These findings align with prior research indicating that children are a demographic that is more susceptible to the adverse health effects of these pollutants^[8,19].

Figure 5. Carcinogenic and non-carcinogenic risk of OCPs in soils. (A): Carcinogenic risk of OCPs in soils; (B): non-carcinogenic risk of OCPs in soils. CR: the composite index of carcinogenic risk; HI: the composite non-carcinogenic risk index.