Science of the Total Environment 478 (2014) 90–97

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Organochlorine pesticides and polychlorinated biphenyls in surface soils from Ruoergai high altitude prairie, east edge of Qinghai-Tibet Plateau Nan Gai a, Jing Pan a,⁎, Hua Tang b, Shu Chen a, Dazhou Chen b, Xiaohua Zhu a, Guohui Lu a, Yongliang Yang a,⁎⁎ a b

National Research Center for Geoanalysis, Beijing 100037, China Division of Metrology in Chemistry, National Institute of Metrology, Beijing 100013, China

H I G H L I G H T S • • • • •

OCPs and PCBs in soils along a transect from source area to Ruoergai highland were measured OCP levels in wetland soils were higher than in grassland soils. Good correlation was observed between TOC and PCBs in the soils. The compounds with higher concentrations were α-HCH, β-HCH, HCB, and PCB 28. The POPs' behaviors in high plateau areas are similar to Polar Regions.

a r t i c l e

i n f o

Article history: Received 16 September 2013 Received in revised form 31 December 2013 Accepted 1 January 2014 Available online 12 February 2014 Keywords: Persistent organic pollutants High plateau Ruoergai Surface soil Elevation gradient

a b s t r a c t Organochlorine pesticides (OCPs) and polychlorinated biphenyls (PCBs) in surface soils along a transect from source areas (a petro-chemical industrial city, Lanzhou and its adjacent agricultural areas) to Ruoergai highland prairie (3552 m above sea level (a.s.l.)), where livestock farming was the only human economic activity, were studied. OCPs in Ruoergai soils were dominated by HCHs. The land types, organic carbon contents and pH affected the POP preservation in soil. OCPs and PCBs in surface soils in Ruoergai wetland and grassland showed different contamination patterns; OCP levels in wetland soils were higher than those in grassland. Significant correlations were observed between total organic carbon (TOC) contents and PCB concentrations in the soils. The land type determines TOC content in soils, which in turn was a major factor on soil concentrations of POPs. The transect was divided into two sections: The first section (Gradient I) is from Lanzhou (1740 m a.s.l.) to Luqu (2400 m a.s.l.) with decreasing agricultural activities, and the second section (Gradient II) is from Luqu to Ruoergai (3500 m a.s.l.) with grassland as the main land type. Soils of Ruoergai area were dominated by α-HCH, β-HCH, HCB, and PCB28, suggesting that the behaviors of POPs in the high plateau region were different from high mountain cold-trapping effect, and that the POPs' behaviors in high plateau region were similar to Polar Regions. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Some of the semi-volatile persistent organic pollutants (POPs) can undergo long-range transport (LRT) in the form of vapor or adsorbed onto atmospheric aerosols to remote clean areas via atmospheric circulation at regional or even global scales (Wania and Mackay, 1993). These POPs may deposit onto the surface media through dry and/or wet precipitations or air–soil exchange, posing potential threat to human health and ecosystem in remote areas (Hansen 2000; UNEP 2001). More volatile components of the mixtures, such as the lower

⁎ Correspondence to: J. Pan, National Research Center for Geoanalysis, Xicheng District, 26 Baiwanzhuang Road, Beijing 100037, China. Tel.: +86 10 6899 9562. ⁎⁎ Correspondence to: Y. Yang, National Research Center for Geoanalysis, Xicheng District, 26 Baiwanzhuang Road, Beijing 100037, China. Tel.: +86 10 5790 9211. E-mail addresses: [email protected] (J. Pan), [email protected] (Y. Yang). 0048-9697/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2014.01.002

chlorinated PCBs, hop poleward more efficiently than the higher chlorinated ones, leading to a compositional shift to more volatile constituents with increasing Northern latitude (Ockenden et al., 1998). Previous studies have shown that high mountains and cold regions become the sink and important reservoirs of POPs (Blais et al., 1998; Vandrooge et al., 2004). POPs can deposit via wet precipitation and they can be trapped by ice, snow, soils, and vegetations (Grollert et al., 1997; Weiss et al., 2000). Changes in POP composition in soil with altitude of the Peruvian Andes and the Italian Alps have been noted (Tremolada et al., 2008; Wania and Westgate, 2008). Wania and Westgate (2008) discussed the differences in the fractionation pattern along latitudinal and elevation gradients and proposed that the precipitation scavenging efficiency of organic chemicals was temperature dependent. According to the “Mountain-POP” model, POPs moving towards higher altitude along with decreasing temperature, are fractionated due to different chemico-physical properties. Several studies on POPs in top-soils from the Tibetan Plateau have been reported

N. Gai et al. / Science of the Total Environment 478 (2014) 90–97

(Wang et al., 2007, 2012;P. Wang et al., 2009). However, the climate and topography of high plateaus are different from high mountains. The behaviors of POPs in high plateaus are expected to be different from ordinary high mountains, and therefore, these are ideal places to study the long-range transport of POPs via atmosphere and its relationship with seasonal change. Up to date, little research on precipitation of POPs in highland pasture areas is available. Ruoergai (Zoige) highland prairie, located in the eastern edge of the Qinghai-Tibetan Plateau, was chosen as our study area. The altitude of Ruoergai highland is 3500 m a.s.l. on average, and livestock grazing is the major human activity. To the north (approximately 600 km) of Ruoergai area is Lanzhou city (1900 m a.s.l.), Gansu Province, representing agricultural and industrial regions. Northwesterly is prevailing in Ruoergai area in both summer (July, 2011) and winter (November, 2011). The present work investigated and quantified the levels of organochlorine pesticides (OCPs) and polychlorinated biphenyls (PCBs) in top-soils from Ruoergai highland prairie and along a transect from Lanzhou to Ruoergai high plateau. Ten OCPs and six indicator PCB congeners were analyzed by highresolution gaschromatographhigh resolution mass spectrometry (HRGC-HRMS). The aims of this paper are to further understand the behaviors of POPs in remote areas and provide information on fractionation of POPs from source areas to a high plateau prairie. 2. Methods 2.1. Research areas and the geographic setting Ruoergai highland prairie of the Aba Tibetan and Qiang Autonomous Prefecture, is located at the eastern edge of the Qinghai-Tibetan Plateau at altitudes of 3200–3600 m a.s.l. , north of Sichuan Province, and the convergence zone of East Asia monsoon and the Qinghai-Tibetan Plateau climate system. The eco-environment is pasture and alpine wetlands. There is 808,000 hectare natural grassland in Ruoergai pasture, one of the largest three grasslands in China. The economics is mainly livestock farming (yak, sheep, and horse). Most of population is Tibetan. The prairie has an area of 53,000 km2, is the largest alpine wetland and a special geographic area in the World. The climate of Ruoergai highland grassland is characterized by cold weather, with annual average temperature of 2 °C, annual average precipitation of 600 mm, mostly during period between June and August. Summer is short (June and July); the average temperature in summer is 10.8 °C whereas the average temperature in winter is − 5 °C. The prevailing wind is mainly westerly and northwesterly. In recent decades, the climate has changed dramatically, showing deceasing precipitation and increasing evaporation trend, especially since the 1990s. This trend has been leading to shrinking of the wetland area, water resources drastically reduced, grassland degradation, and accelerated soil desertification. The soils in the grassland are soaked in water for long time with low temperature, mainly as marsh soils, characterized by high gleization and rich organic matter and peat. Due to the cold and moist climate, low evaporation and poor drainage, the surface soils are often in a wet state, favorable to swamp development. Lanzhou, the capital of Gansu Province, is located on the northwest of the Loess Plateau, at altitudes of 1500–2000 m a.s.l.. The average temperatures in summer and winter in Lanzhou are 21 °C and −3.9 °C, respectively. Lanzhou is a petrochemical industry city and the surrounding areas at altitude below 2500 m a.s.l. are agricultural in nature. Therefore, Lanzhou becomes a pesticide and PCB emission source area for Ruoergai highland prairie with the northwesterly wind. 2.2. Sample collection The sampling sites (n = 36) in Ruoergai area are shown in Fig. 1(b). In order to distinguish the relationship between POPs in surface soils of

91

Ruoergai high Plateau and the surrounding industrial and agricultural areas, the sampling strategy of soils was designed to be along the transect from Lanzhou to Ruoergai area at increasing altitude (Fig. 1a). The details of the samples and sampling locations are provided in Table 1. Surface soils (0–5 cm) were collected in summer (July) and winter (November) of 2011, using a small stainless shovel. The samples were wrapped in aluminum foil and stored frozen until analysis. After air-dried at room temperature, all samples were ground, sieved through a 60 mesh size, and stored in refrigerator until analysis.

2.3. Chemical analysis of OCPs and PCBs HPLC grade Acetone, n-hexane, and dichloromethane were obtained from Tianjin Kermel Chemical Reagent Co. Inc., China; analytical grade methanol was from Jinan Chemical Reagent Co. Inc., China; guaranteed reagent grade sulfuric acid was from Qingdao Chemical Reagent Co. Inc., China; copper chips and analytical grade granular anhydrous sodium sulfate were from Tianjin Huazhen Chemical Reagent Co. Inc., China. Two surrogate standards (PCB209 and 2,4,5,6-tetrachloro-mxylene (TMX)) and Certified Reference Materials including PCB28, PCB52, PCB101, PCB118, PCB138, PCB153, PCB180, α-, β-, γ- and δ-HCH, p,p′-DDD, p,p′-DDE, p,p′-DDT, o,p′-DDT, α- and β-endosulfans were purchased from the National Research Center for Certified Reference Materials. The quantitative standards of PCBs were from the Wellington Laboratories (Guelph, ON, Canada). The internal standards 13 C mass-labelled PCBs (13C-PCB28, 13C-PCB52, 13C-PCB101, 13C-PCB118, 13 C-PCB153, 13C-PCB138, 13C-PCB180 were from Accustandard USA. Granular anhydrous sodium sulfate was activated at 650 °C in a furnace for 6 h and then kept in sealed desiccators. Florisil was activated at 130 °C for 16 h before use. All glassware was cleaned in an ultrasonic cleaner and heated at 350 °C for 12 h. The extraction and cleanup procedures for OCPs and PCBs were carried out as follows: After adding surrogate standards and mass-labelled standards to the soil samples, they were Soxhleted with 150 mL DCM for 48 h. After the extraction, 2 g copper chips were added to remove sulfur. The extract was reduced to 5–10 mL using a rotary evaporator, and then further reduced to 1 mL after 10–15 mL hexane was added. This concentration step was repeated three times and then the sample solution was transferred to a 5 mL cell, and they were passed through a silica–alumina column (7 mm i.d.; from bottom to top, 10 g 3% activated silica, 10 g 3% de-activated alumina, and 1 g dehydrated sodium sulfate). The column was eluted with 35 mL hexane/DCM (v/v ratio 1:1) solution. The eluent was reduced to 0.5 mL using a rotary evaporator, then further concentrated to 0.2 mL under a gentle high purity nitrogen gas. The sample was subject to HRGC-HRMS analysis after adding injection internal standards.

2.4. HRGC-HRMS measurement The OCPs and PCBs in the samples were analyzed by an HRGC-HRMS system (HP-6890 high resolution gas chromatograph coupled to Finngan MAT 900XL HRMS operated in an electron impact (EI) and selective ion monitoring (SIM) mode. The GC column was a HP-5MS (30 m × 0.32 mm i.d., with film thickness of 0.25 μm). The GC conditions are as follows: the temperature settings for the injector and detector were 280 °C and 320 °C, respectively; the initial column temperature of 50 °C was held for 2 min, then increased at a rate of 10 °C min− 1 to 180 °C, held for 2 min, and then increased to 220 °C at 2 °C min−1, and to 290 °C at 10 °C min− 1, held for 15 min. Oxygen-free nitrogen (99.999% purity) was used as the carrier gas at a constant flow rate of 1.5 mL min − 1 . The sample (1.0 μL) was injected with splitless injection mode. The make-up gas was high pure nitrogen.

92

N. Gai et al. / Science of the Total Environment 478 (2014) 90–97

Fig. 1. Sampling locations for surface soils from (a) the transect from Lanzhou to Ruoergai high plateau and (b) Ruoergai County.

2.5. Quality control Field blank, reagent blank, and procedure blank were conducted. One blank sample was carried out during each batch (ten samples) of

experiment to evaluate any contamination during the analysis. No detectable concentrations were found in the blank samples. Samples were analyzed in duplicate. Identification was carried out using retention times and mass ratios; the retention time of a component peak

Table 1 The details of sample information and sampling locations along the transect. Sampling site

Field no.

Distance from Lanzhou

Location

Altitude

Land type

Ruoergai County Huahu Lake North of Rierlang Mt. Rierlang Mt. Gongba Village Hezuo Quao Village Maji Village Linxia Lanzhou

S-32 S-30 110701 111101 111102 110703 110704 110705 111104 111105

(km) 460 420 410 400 370 320 245 230 200 0

103°10.863′; 33°26.427′ 102°49.137′; 33°54.977′ 102°43′713″; 34°3′562″ 102°36′892″; 34°45′796″ 102°52′244″; 34°56′476″ 102°53′795″; 35°5′603″ 102°47′245″; 35°17′68″ 102°56′379″; 35°14′402″ 103°09′205″; 35°32′295″ 103°11′667″; 35°43′148″

(m) 3552 3473 3116 3103 2813 2414 2201 2109 1985 1740

grassland and wetland grassland and wetland grassland grassland grassland corn field corn field corn field corn field loess

N. Gai et al. / Science of the Total Environment 478 (2014) 90–97

should be exactly same with that of corresponding 13C mass-labelled standard and the mass ratio of the two single ions of a compound within 0.8–1.0. Internal calibration using corresponding 13C mass-labelled standards was used to quantify the concentrations. The recoveries of OCPs and PCBs were in the range of 69.5%–105% and 75.6%–121.8%, respectively. All the data were corrected for recovery. The detection limits were 0.04–0.1 pg g−1 for OCPs and 0.36–0.71 pg g−1 for PCBs. 3. Results and discussions 3.1. Concentrations in soils of Ruoergai area Measureable concentrations of HCHs, HCB, DDTs, endosulfans (alpha and beta) and PCBs were detected in all 36 surface soil samples in Ruoergai area during summer and winter of 2011 (Table 2,) and concentrations of TOC, OCPs and PCBs in top-soils at each sampling site of Ruoergai high plateau are provided in Supplementary material (SM) Table SI-1. The ∑HCHs accounted for 44.24% of the total OCPs. DDTs and HCHs were the predominant OCPs. The detection frequencies for α-HCHs, β-HCH, γ-HCH, HCB, p,p′-DDE were 100%, followed by p,p′DDD (92.6%), p,p′-DDT (92.6%), p,p′-DDT (70.4%), α-endosulfan (14.8%), and β-endosulfan (66.7%). The comparisons of OCP and PCB concentrations in Ruoergai soils between summer and winter 2011 are given in Fig. 2. The average sum of α-HCHs, β-HCH, γ-HCH (∑HCHs) was higher in summer than those in winter, while ∑DDTs in summer and winter were similar (Table 3). The average ∑ HCHs of the two seasons were 1.86 ± 1.49 ng g− 1 d.w. (range: 0.43–6.72 ng g − 1 dry weight (d.w.), while the average sum of DDTs (∑ DDTs) was 1.63 ± 1.20 ng g− 1 d.w. (range: 0.29–4.34 ng g− 1 d.w.). Endosulfans in soil were detected in some sampling sites, accounting for 6.67% of the total OCPs. The average concentrations of α- and βendosulfan were 0.05 and 0.28 ng g−1 d.w., respectively. The endosulfan concentrations in Ruoergai soils were lower than those observed in soils from the northeast China (3.34 ng g−1 d.w.) (X. Wang et al., 2009) and southern China (1.31 ng g−1 d.w.) (Huang et al., 2010). The mean HCB concentrations were 0.93 ±0.59 ng g−1 d.w. (range: 0.23 to 2.60 ng g−1 d.w.) accounting for 18.4% of the total POPs. HCB

93

concentrations in Ruoergai soils were higher than those in some background areas in the world, such as Antarctic area (0.034–0.17 ng g−1 dw) and European high mountain areas (0.24 ± 0.24 ng g− 1 dw) (Borghini et al., 2005; Tremolada et al., 2008). PCB28 and PCB52 were detected in all surface soils in Ruoergai area. PCB101 were detected in 44.3% and 33.3% of the samples in summer and in winter, respectively. PCB153 was detected in 11.1% in the samples from summer (mainly from the north of the Huahu Lake), but not detected in winter. PCB153, 138, and 180 were not detected in any of the samples. The mean ∑6 PCBs were 0.88 ng g− 1 d.w. in summer (range: 0.34–2.31 ng g− 1 d.w.) and 1.13 ng g− 1 d.w. in winter (range: 0.22–1.96 ng g− 1 d.w.). ∑ HCHs, HCB, and PCBs in Ruoergai soils were higher than those from the Tibetan Plateau (0.064–0.847, 0.024–0.564 and 0.075–1.021 ng g − 1 d.w., respectively); ∑ DDTs was lower than those from the Tibetan Plateau (0.013–7.7 ng g − 1 d.w.)(Wang et al., 2012). The topography and the source areas (such as India, Yunnan and Sichuan Provinces, China) in Tibetan Plateau could be more diversified than that of Ruoergai high plateau and there are only agricultural activities in Tibet. 3.2. Isomer concentrations and composition changes from Lanzhou to Ruoergai The concentrations of OCPs and PCBs in soils from Lanzhou collected in 2011 summer are shown in Fig. 2(c) and SM SI-2. The ∑ HCHs (2.53 ng g−1 d.w.) and ∑DDTs (3.35 ng g−1 d.w.) were approximately two-fold higher than those of Ruoergai area. p,p′-DDE was the predominant isomer in Lanzhou soils. The ratio of o,p′-DDT/p,p′-DDT was 0.52, suggesting that the DDTs was residue from historical usage. Endosulfans were not analyzed for soil samples along the transect from Lanzhou to Ruoergai. The concentrations of OCPs in surface soils along the transect from Lanzhou to Ruoergai area are shown in Fig. 3. The transect was divided into two sections according to the changes in POP concentrations with altitude. The first section (denoted as Gradient I in Fig. 3) is from Lanzhou (1740 m a.s.l.) to Hezuo (2400 m a.s.l.) and the second section

Table 2 Concentrations of TOC (% w/w), OCPs and PCBs (ng g−1 d.w.) in top soils from Ruoergai high plateau.

a

Summer (n = 27)

Min.

Max.

Ave.

Median

TOC α-HCH β-HCH γ-HCH α/γ-HCHb ΣHCHs HCB p,p′-DDE p,p′-DDD o,p′-DDT p,p′-DDT o,p′/p,p′-DDTc ΣDDTs β-endosulfan α-endosulfan ΣOCPs

1.52 0.11 0.18 0.10 0.68 0.43 0.23 0.16 nda nd nd 0.31 nd nd 0.97

31.7 2.69 6.52 1.39 2.60 10.6 2.6 3.26 2.08 1.06 0.83 3.00 5.72 0.92 0.48 17.6

8.0 0.62 1.46 0.38 1.60 2.46 0.89 0.83 0.51 0.27 0.34 0.86 1.95 0.32 0.05 5.67

5.2 0.44 0.59 0.34 1.59 1.49 0.86 0.72 0.41 0.24 0.35 0.66 1.69 0.40 0.00 4.63

PCB28 PCB52 PCB101 PCB153 PCB138 PCB180 ΣPCBs

0.17 0.03 nd nd nd nd 0.22

1.58 0.30 0.10 0.09 0.08 nd 1.96

0.69 0.14 0.04 0.01 nd nd 0.88

0.65 0.13 0.05 nd nd nd 0.87

Detection frequency (%)

100 100 100

100 100 92.6 70.4 92.6

66.7 14.8

100 100 59.3 11.1 7.4

Winter (n = 9)

Min.

Max.

Ave.

Median

TOC α-HCH β-HCH γ-HCH α/γ-HCH ΣHCHs HCB p,p′-DDE p,p′-DDD o,p′-DDT p,p′-DDT o,p′/p,p′-DDT ΣDDTs β-endosulfan α-endosulfan ΣOCPs

1..6 0.2 0.43 0.17 1.15 0.80 0.38 0.1 nd nd nd 0.29 nd nd 1.58

24.5. 0.67 0.99 0.44 2.61 1.98 1.92 0.92 0.6 0.81 1.85 1.31 3.84 0.48 0.34 7.72

8.9 0.41 0.66 0.32 2.12 1.40 0.97 0.49 0.30 0.30 0.48 0.73 1.56 0.17 0.05 4.16

5.5 0.41 0.62 0.33 2.18 1.36 0.92 0.50 0.25 0.26 0.39 0.73 1.46 0.18 0.00 4.12

PCB28 PCB52 PCB101 PCB153 PCB138 PCB180 ΣPCBs

0.28 0.06 nd nd nd nd 0.34

2.01 0.24 0.11 nd nd nd 2.31

0.97 0.13 0.03 nd nd nd 1.13

0.92 0.12 nd nd nd nd 1.02

nd: not detected. bα/γ-HCH = concentration ratio of α−HCH/γ-HCH. c o,p′/p,p′-DDT = concentration ratio of o,p′-DDT/p,p′-DDT.

Detection frequency (%)

100 100 100

100 100 88.9 66.7 88.9

55.6 22.2

100 100 33.3 0 0 0

94

N. Gai et al. / Science of the Total Environment 478 (2014) 90–97

Fig. 3. Altitudinal gradients of OCP and PCB concentrations along the transect from Lanzhou to Ruoergai highland area in summer, 2011.

DDTs have lower vapor pressure (PL at 20 °C) (0.00097–0.0033 Pa) (Shen and Wania, 2005) than those of HCHs (0.013–3.31 Pa)(Xiao et al., 2004) and thus they cannot be transported further than HCHs in atmosphere. 3.3. Factors affecting the distribution of OCPs and PCBs in surface soils in Ruoergai area

Fig. 2. Concentrations of ∑OCPs and ∑PCBs in top-soils from (a) Ruoergai (summer), (b) Ruoergai (winter) (only the data with the same sampling sites for the two season samplings are used in the plots), and (c) Lanzhou (summer) in 2011.

(Gradient II) is from Hezuo to Ruoergai (3500 m a.s.l.). Gradient I is related to the distance from agricultural and industrial areas (in Lanzhou and the adjacent areas); the general trend for HCHs and DDTs in surface soils in Gradient I decreased with distance from Lanzhou, but PCB28 and PCB52 showed an increasing trend with distance from Lanzhou. Soils in Gradient II are mostly grassland and wetland, and have never been used for agriculture due to the high altitude (N 2800 m a.s.l.). The trends for HCHs, HCB, PCB28, and PCB52 were increasing in Gradient II, whereas all DDT isomers showed a decreasing trend toward Ruoergai Plateau. This phenomenon can be explained by cold-trapping effect proposed by Wania and Mackay (1996); the compounds with relatively higher vapor pressure can undergo atmospheric transport to remote areas and are trapped in cold areas such as Polar Regions or high mountains.

3.3.1. Land type and TOC Land types may affect the levels and compositions of OCPs and PCBs in surface soils. The land in Ruoergai area is classified into wetland and grassland. The soils in wetland are enriched water and organic matter; the soils of latter are mainly grassland for livestock grazing. The total organic carbon content (TOC) in soil is an important factor affecting POPs' soil–air exchange and long-range atmospheric transport (Walker et al., 1999). The TOC concentrations in typical wetland were in the range of 5.0%–34.7%, higher than the TOC (1.2%–4.8%) in grassland. In Ruoergai wetland, generally most of POP concentrations were positively correlated with TOC (Fig. 4). In contrast, no correlations were found between concentrations of POPs and TOC in soils of neighboring low altitude agricultural areas (the southern Lanzhou city, Gansu province), especially γ-HCH, probably due to the use of lindane in those areas. The average concentrations and compositions of POPs for the two different types of land are shown in Fig. 5 (a). β-HCH accounts for 63.10% and 43.24%, α-HCH accounts for 23.10% and 33.33%, and γ-HCH accounts for 13.52% and 24.32% of the total HCHs in typical wetland and grassland, respectively. Higher β-HCH in wetland soils can be explained by its coplanar structure with all chlorine atoms in a plane, which makes β-HCH more stable, less volatile, more water soluble, and difficult to be biodegraded (Wu et al., 1997). In general, the concentrations of all compounds measured in wetland soils were higher than those in grassland soils. For example, the mean HCH concentrations in wetland soils (3.55 ng g−1 d.w.) were higher than those in grassland soils (1.11 ng g −1 d.w.); and the mean DDT concentrations in the typical wetland and grassland were 2.27 and 1.52 ng g −1 d.w., respectively. 3.3.2. pH The pH values of the soils in the study area were in the range of 4.5–8.5. Here we define that soils with pH 4.5–6.5 as acid soil, and 6.5–7.5 and 7.5–8.5 are neutral and alkaline soils, respectively. The

Table 3 Precipitation (mm) and air temperature (°C) in Ruoergai area during the sampling periods. Year

2011

Month Precipitation (mm) Temperature °C

May 79.1 6.7

2012 June 121.5 10.2

July 200.1 11.3

Aug. 25.1 11.1

Sept. 122.3 8.3

Oct. 36.6 2.5

Nov. 3.6 −5.6

Dec. 3.9 −6.5

Jan. 5.7 −8.0

Feb. 10.5 −5.6

Mar. 16.9 −3.0

Apr. 44.0 2.2

N. Gai et al. / Science of the Total Environment 478 (2014) 90–97

OCPs (ng g-1 dw)

20.0

95

(a)

15.0

R = 0.363 10.0 5.0 0.0 0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

TOC (%)

2.0

(b)

R = 0.699

(b) pH

1.5 1.0 0.5 0.0 0.0

α-HCH β-HCH γ-HCH p,p’-DDE p,p’-DDD o,p’-DDT p,p’-DDT HCB PCB28 PCB52

2.0

Conc. (ng g-1 dw)

PCBs (ng g-1 dw)

2.5

5.0

10.0

15.0

20.0

25.0

30.0

1.6

β-HCH R2 = 0.9686 α -HCH R2 = 0.4511

1.2

γ -HCH R2 = 0.6000

0.8

PCB28 R2 = 0.2185

0.4 PCB52 R2 = 0.4545

35.0 0.0

TOC (%)

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

pH Fig. 4. Correlations (p b 0.05) between TOC and (a) OCPs, (b) PCBs in summer top-soils from Ruoergai high plateau.

analysis of linear correlation indicated that soil pH influenced the residues of α-HCH and β-HCH, respectively (Fig. 5b). HCH compounds showed more negative correlations with pH (α-HCH: R = − 0.672, β-HCH: R = −0.984, γ-HCH: R = − 0.775, p b 0.05), while DDT isomers had no correlations with pH. PCB28 (R = − 0.467) and PCB52 (R = −0.674, p b 0.05) showed weak correlation with pH. In general, the concentrations of OCPs and PCBs in acid soils were higher than those in neutral and alkaline soils. A previous study has shown that pH affects the concentrations of OCPs and PCBs via changing the structure of humus matter in soil (Wu et al., 1997). Transformations of HCHs in soil have also been reported by pH (Yu et al., 2013). Significant negative correlations among α-HCH, γ-HCH, δ-HCH and p,p′-DDE with pH were observed in soils along the north coastal areas of the Bohai Sea (Hu et al., 2010). Wenzel et al. (2002) have suggested that pH may influence the microbiological activity in the soil that affects the biotransformations of these chemicals (Wenzel et al., 2002). The effects of pH on conservation of POPs in soils are shown in Fig. 5(c). Concentrations of HCHs, DDTs in acid soil were generally higher than those of the alkaline soils. Acidity may restrain the microbiological activity in the soil and reduce biological degradation of these compounds. HCB and PCB28 concentrations in alkaline soils were as high as those in acid soils. The behaviors of these compounds in soils with different acidity require further investigation.

Fig. 5. Concentrations of OCPs and PCBs as functions of (a) land type, (b) soil pH (p b 0.05), and (c) soil acidity.

was a stable correlation between α-HCH and γ-HCH (Fig. 6), suggesting a common source of HCHs. β-HCH was the isomer with highest concentration among HCH isomers in the soils possibly due to its resistance to degradation and a relatively higher water solubility (SWL: 0.58 mol m−3 at 25 °C; Xiao et al., 2004) compared to those of α-HCH (SWL: 0.325 mol m− 3 at 25 °C) and γ-HCH (SWL: 0.269 mol m−3 at 25 °C). It was reported that rain scavenging is much more efficient for β- than for α-HCH (UNEP, 2007). Increasing wet precipitation in summer may directly lead to an increase in concentrations of HCHs especially β-HCH in the surface soils. In addition, laboratory- and field-based data including a long-

3.4. Isomer composition and source identification for OCPs and PCBs in Ruoergai area

γ-HCH (ng g-1 dw)

3.4.1. HCHs In the Ruoergai soils, the HCH isomer levels were in the order of β-HCH N α-HCH N γ-HCH. β-HCH (mean: 1.26 ng g−1 d.w.) was predominant HCH isomer, followed by α-HCH (0.57 ng g−1 d.w.) and γ-HCH (0.36 ng g−1 d.w.). In technical products of HCHs, the isomer composition is α-HCH: 60%–70%, β-HCH: 5%–12%, γ-HCH: 10%–12% and δ-ΗCΗ: 6%–10% (Walker et al., 1999). The α-/γ-HCH ratio less than 3 may be caused by lindane usage in the neighboring areas and the ratio between 3 and 7 may suggest unchanged HCHs technical products being used (Walker et al., 1999). In the Ruoergai soils, α-/γ-HCH ratios were mostly in the range of 1–2, suggesting that HCHs were from the areas where lindane were being used. There

1.6

R = 0.960 1.2 0.8 0.4 0.0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

α-HCH (ng g-1 dw) Fig. 6. Correlations between α-HCH and γ-HCH in top-soils of Ruoergai high plateau (p b 0.05).

96

N. Gai et al. / Science of the Total Environment 478 (2014) 90–97

term soil study suggest that β-HCH is persistent in soil, especially under low temperatures (UNEP, 2007).

3.4.2. DDTs ∑DDTs accounted for 37.4% of the total OCPs. The mean concentrations of p,p′-DDE were the highest among the four isomers, followed by p,p′-DDD, p,p′-DDT, o,p′-DDT. Recent studies have shown that after the ban of production and use of DDTs, dicofol became a new source of DDTs to the environment. Dicofol contains DDTs as by-products, in which the ratio of o,p′-DDT/p,p′-DDT is between 1.3 and 9.3 (Qiu et al., 2005). In the surface soils of Ruoergai area, the mean ratios were 0.83 in summer and 0.76 in winter, suggesting that the use of dicofol might be excluded in the neighboring regions. In all of the wetland and grassland soils the sum of DDD + DDE N DDT, showing that most of the DDTs had been degraded into DDE and DDD and there was no new DDT usage in the neighboring regions. The DDTs found in the soils were mainly coming from the areas where DDTs had been weathered for a long time and transported via atmospheric transport and precipitation. DDT concentrations in soils of densely populated regions of China were generally higher than those of HCHs, such as in Lanzhou. In contrast, OCPs in Ruoergai soils were dominated by HCHs (relatively more volatile than DDTs), indicating that the OCPs in the soils of Ruoergai area were from LRT via atmosphere.

3.4.3. Endosulfans Endosulfans were mainly distributed in the north and south parts of the Ruoergai area. This pattern may be related to the influence from the agricultural regions in southern Gansu Province and the northwestern China. Technical grade endosulfan is a mixture of two isomers, alpha endosulfan (64%–67%) and beta endosulfan (29%–32%) (GFEA-U, 2007). Compared to other OCPs, endosulfans are less persistent chemicals (Gupta and Gupta, 1979). It was reported that the mean concentrations of endosulfans in agricultural soils in northern India were 0.95 ± 0.53 ng g−1 d.w. (range: 0.01–22.9 ng g−1 d.w.) (Kumar et al., 2011), but endosulfans were observed at low levels (not detected) in soils around Mt. Qomolangma (Wang et al., 2007). Endosulfans are widely used in China as pesticides for cotton (mostly in Xinjiang Uygur Autonomous Region in the northwestern China), fruits, and tobacco plantation (Jia et al., 2012), and are manufactured and used mostly in the eastern part of China. Endosulfans can undergo atmospheric long-range transport (Schenker et al., 2007; Becker et al., 2011). The northwesterly in winter and southeasterly in summer may carry endosulfans from Xinjiang Uygur Autonomous Region to Ruoergai area.

3.4.4. HCB and PCBs Due to their relatively higher volatility, HCB, PCB28, and PCB52 are more apt to LRT than other OCPs and higher chlorinated PCBs. The theoretical long-range transport potential of HCB is 1.1 × 105 km, while that of p,p′-DDT is much shorter (800 km) (Beyer et al., 2000). It was reported that HCB in the Tibetan surface soils showed a weak correlation with temperature but showed a significant correlation with soil organic carbon (Wang et al., 2012). The models predicted that PCB congeners of medium hydrophobicity (5–6 chlorine atoms) are the most important for re-volatilization globally from surface soil, and the delivery of contaminants to high latitudes is more efficient than previously suggested (Lammel and Stemmler, 2012). The detection frequencies of PCB28 and PCB52 in the soils from Ruoergai high plateau were both 100% in summer and winter; while those for PCB101 were 59.3% and 33.3%, respectively in summer and winter; and the higher chlorinated congeners were not detected. As discussed in the following section, Ruoergai high plateau was similar to high latitudes with respect to the behaviors of PCBs.

3.5. Implication for the behaviors of POPs in high plateau areas Whether a semi-volatile compound can be detected in surface soils of remote high altitude areas depends not only on its physical–chemical properties, i.e., the LRT potential (Wania, 2003), but also on the distance from the sources, climate conditions, precipitation intensity, and the temperature gradient along the mountain slope (Daly and Wania, 2005). In some high mountain areas, the low-volatile compounds such as higher chlorinated PCBs are able to be enriched in soils or vegetations along the altitude gradient, because of the gradients of the favorable wet precipitation and temperature along the mountain slope. Examples are Wolong area of Sichuan Province, China where temperature decreases with an increasing altitude in a short range of 50 km (Pan, et al., 2013). On the other hand, high plateau areas are far from the pollution sources, with low altitude gradient and relatively flat topography, and lack of significant temperature gradient in a vast area. Therefore, the cold temperature and wet precipitation are in favor of transport of relatively higher volatile compounds such as PCB28 and PCB52 to deposit in these areas. It has been recognized that POP composition along the altitude gradient showed a quite different patterns in Alps and Andes (Tremolada et al., 2008). Wania and Westgate (2008) compared the difference between polar cold-trapping effect and mountain cold-trapping effect on POP compounds with plots of log KOA versus log KWA. The POP chemicals which show polar cold-trapping and high mountain cold trapping effects lie in the central part of the area of the plots between log KOA 8–10 and log KWA 2–4.5. The chemicals enriched at high mountains are less volatile than those enriched in polar areas, with the difference as large as two orders of magnitude. The polar model predicts that lighter PCBs can be relatively enriched in polar areas, in contrast to some high mountain areas, where higher chlorinated PCBs can be relatively enriched (Daly and Wania, 2005; Tremolada et al., 2008). The air masses in high plateau areas are often from high altitude atmosphere in nature. The altitude gradients in terrains of high plateau regions are smaller than mountains. For example, the transition zone between low altitude areas and the Qinghai-Tibetan Plateau can cross a distance of hundred or even thousand kilometers. The altitudes of high plateau regions are often greater than 3500 m, subject to special climate's influence, and with longer snow precipitation period. Von Waldow et al. (2010) through model experiment indicated that no tendency of increased fractionation of PCBs due to decreasing temperatures, and fractionation occurs as a function of distance from the emission source regardless of the temperature gradient. It has been observed that HCB, low molecular weight PCBs and HCHs are able to readily exchange between the Tibetan surface and the gas phase with net movement of these POPs to soils of higher TOC (Wang et al., 2012). However, the topography in Tibetan is more diversified than that of Ruoergai high plateau. Ruoergai high plateau can be served as a natural laboratory to study the behaviors of POPs due to the flat and long altitude gradient and stable cold climate pattern. Distance from source areas is the main factor for fractionation of POP isomers with different volatility. Our results show that high plateau areas (e.g., Ruoergai area) are more like the polar areas, and thus the semi-volatile compounds α-HCH, β-HCH, HCB, and PCB 28 in high plateau soils were detected. Therefore the polar cold-trapping model (Wania and Mackay, 1996) can be applied to high plateau areas. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.01.002. Acknowledgements We thank Dr. Leo W.Y. Yeung (University of Toronto) for suggestions and corrections on the manuscript. This study was supported by the Natural Science Foundation of China (Project Nos. 41003044, 41073011) and the China Geological Survey (Project No. 1212011220058).

N. Gai et al. / Science of the Total Environment 478 (2014) 90–97

References Becker L, Scheringer M, Schenker U, Hungerbühler K. Assessment of the environmental persistence and long-range transport of endosulfan. Environ Pollut 2011;159: 1737–43. Beyer A, Mackay D, Matthies M, Wania F, Webster E. Assessing long-range transport potential of persistent organic pollutants. Environ Sci Technol 2000;34:699–703. Blais JM, Schindler DW, Muir DCG, Kimpe LE, Donald DB, Rosenberg B. Accumulation of persistent organochlorine compounds in mountains of western Canada. Nature 1998;395:585–8. Borghini F, Grimalt JO, Sanchez-Hernandez JC, Bargagli R. Organochlorine pollutants in soils and mosses from Victoria Land (Antarctica). Chemosphere 2005;58:271–8. Daly GL, Wania F. Organic contaminants in mountains. Environ Sci Technol 2005;39: 385–98. GFEA-U. Endosulfan. Draft Dossier prepared in support of a proposal of endosulfan to be considered as a candidate for inclusion in the CLRTAP protocol on persistent organic pollutants. Umweltbundesamt, Berlin: German Federal Environment Agency; 2007. Grollert C, Kasper A, Puxbaum H. Organic compounds in high Alpine snow. Int J Environ Anal Chem 1997;67:213–22. Gupta P, Gupta R. Pharmacology, toxicology and degradation of endosulfan. A Rev Toxicol 1979;13:115–30. Hansen JC. Environmental contaminants and human health in the Arctic. Toxicol Lett 2000;112/113:119–25. Hu WY, Wang TY, Khim JS, Luo W, Jiao WT, Lu YL, et al. Organochlorine pesticides (HCHs and DDTs) in soils along the north coastal areas of the Bohai Sea, China. Chem Ecol 2010;26:339–52. Huang DY, Peng PA, Xu YG, Sun CX, Deng HM, Deng YY. Distribution, regional sources and deposition fluxes of organochlorine pesticides in precipitation in Guangzhou, South China. Atmos Res 2010;97:115–23. Jia HL, Liu LY, Sun YQ, Cai DJ, Hu JX, Ren NQ, et al. Endosulfan in the Chinese environment: monitoring and modeling. Frontiers Environ Sci Eng 2012;6:32–44. Kumar B, Kumar S, Gaur R, Goel G, Mishra M, Singh SK, et al. Persistent organochlorine pesticides and polychlorinated biphenyls in intensive agricultural soils from North India. Soil Water Res 2011;6:190–7. Lammel G, Stemmler I. Fractionation and current time trends of PCB congeners: evolvement of distributions 1950–2010 studied using a global atmosphere–ocean general circulation model. Atmos Chem Phys Discuss 2012;12:11699–731. Ockenden WA, Sweetman AJ, Prest HF, Steinnes E, Jones KC. Towards an understanding of the global atmospheric distribution of persistent organic pollutants: the use of semipermeable membrane devices as time-integrated passive samplers. Environ Sci Technol 1998;32:2795–803. Pan J, Yang YL, Taniyasu S, Yeung LWY, Falandysz J, Yamashita N. Altitudinal distributions of PCDD/Fs, dioxin-like PCBs and PCNs in soil and yak samples from Wolong high mountain area, eastern Tibet-Qinghai Plateau, China. Sci Total Environ 2013;444: 102–9. Qiu XH, Zhu T, Yao B, Hu JX, Hu SW. Contribution of dicofol to the current DDT pollution in China. Environ Sci Technol 2005;39:4385–90. Schenker U, Scheringer M, Hungerbühler K. Including degradation products of persistent organic pollutants in a global multi-media box model. Environ Sci Pollut Res 2007;14: 145–52. Shen L, Wania F. Compilation, evaluation and selection of physical-chemical property data for organochlorine pesticides. J Chem Eng Data 2005;50:742–68.

97

Tremolada P, Villa S, Bazzarin P, Bizzoto E, Comolli R, Vighi M. POPs in mountain soils from the Alps and Andes: suggestions for a ‘precipitation effect’ on altitudinal gradients. Water Air Soil Pollut 2008;188:93–109. UNEP. Final act of the plenipotentiaries on the Stockholm Convention on persistent organic pollutants. United Nations environment program chemicalsGeneva; 2001445. UNEP. Report of the Persistent Organic Pollutants Review Committee on the work of its third meeting-Risk profile on beta hexachlorocyclohexane. UNEP/POPS/POPRC.3/ 20/Add.9; 2007. p. 1–25. [Geneva, 19–23 November]. Vandrooge BL, Grimalt JO, Camarero L, Catalan J, Stuchlik E, Garclia CT. Atmospheric semivolatile organochlorine compounds in European high-mountain areas (Central Pyrenees and High Tatras). Environ Sci Technol 2004;38:3525–33. Von Waldow H, Macleod M, Jones K, Scheringer M, Hungerbuhler K. Remoteness from emission sources explains the fractionation pattern of polychlorinated biphenyls in the Northern Hemisphere. Environ Sci Technol 2010;44:6183–8. Walker K, Vallero DA, Lewis RG. Factors influencing the distribution of lindane and other hexachlorocyclohexanes in the environment. Environ Sci Technol 1999;33:4373–8. Wang P, Zhang QH, Wang YW, Wang T, Li XM, Li M, et al. Altitude dependence of polychlorinated biphenyls (PCBs) and polybrominated diphenyl ethers (PBDEs) in surface soil from Tibetan Plateau, China. Chemosphere 2009a;76:1498–504. Wang X, Ren NQ, Qi H, Ma WL, Li YF. Levels, distributions, and source identification of organochlorine pesticides in the topsoils in Northeastern China. J Environ Sci 2009b;21: 1386–92. Wang XP, Yao TD, Cong ZY, Yan XL, Kang SC, Zhang Y. Distribution of persistent organic pollutants in soil and grasses around Mt. Qomolangma, China. Arch Environ Contam Toxicol 2007;52:153–62. Wang XP, Sheng JJ, Gong P, Xue YG, Yao TD, Jones KC. Persistent organic pollutants in the Tibetan surface soil: spatial distribution, air-soil exchange and implications for global cycling. Environ Pollut 2012;170:145–51. Wania F. Assessing the potential of persistent organic chemicals for long-range transport and accumulation in Polar Regions. Environ Sci Technol 2003;37:1344–51. Wania F, Mackay D. Tracking the distribution of persistent organic pollutants. Environ Sci Technol 1996;30:390A–6A. Wania F, Mackay D. Global fractionation and cold condensation of low volatility organochlorine compounds in polar regions. Ambio 1993;22:10–8. Wania F, Westgate JN. On the mechanism of mountain cold-trapping of organic chemicals. Environ Sci Technol 2008;42:9092–8. Weiss P, Lorbeer G, Scharf S. Regional aspects and statistical characterizations of the load with semivolatile organic compounds at remote Austrian forest sites. Chemosphere 2000;40:1159–71. Wenzel KD, Manz M, Hubert A, Schuurmann G. Fate of POPs (DDX, HCHs, PCBs) in upper soil layers of pine forests. Sci Total Environ 2002;286:143–54. Wu WZ, Xu Y, Schramm KW, Kettrup A. Study of sorption, biodegradation and isomerization of HCH in stimulated sediment/water system. Chemosphere 1997;35:1887–9. Xiao H, Li NQ, Wania F. Compilation, evaluation, and selection of physical-chemical property data for α-, β-, and γ-hexachlorocyclohexane. J Chem Eng Data 2004;49:173–85. Yu HY, Li FB, Yu WM, Li YT, Yang GY, Zhou SG, et al. Assessment of organochlorine pesticide contamination in relation to soil properties in the Pearl River Delta, China. Sci Total Environ 2013;447:160–8.