Environ. Sci. Technol. 2008, 42, 8395–8400
Organochlorine Pesticides Contaminated Surface Soil As Reemission Source in the Haihe Plain, China SHU TAO,* WENXIN LIU, YAO LI, YU YANG, QIAN ZUO, BENGANG LI, AND JUN CAO Laboratory for Earth Surface Processes, College of Urban and Environmental Sciences, Peking University, Beijing 100871, China
Received July 16, 2008
A large amount of organochlorine pesticides (OCPs) including hexachlorocyclohexane isomers (R-HCH, β-HCH, γ-HCH, andδ-HCH,ΣHCHasthetotal)anddichlorodiphenyltrichloroethane and metabolites (p,p′-DDT, p,p′-DDE, and p,p′-DDD, ΣDDT as the total) have been applied over the Haihe Plain (an area of 300 000 km2) in Northern China. Even though the agricultural application of these OCPs was terminated more than a decade ago, their residues remain in the environment and continue to represent significant public health concern. In this study, more than300 surface soil samples were collected from the Haihe Plain for measurement of these OCPs. The measured ΣHCH and ΣDDT residues were 3.9 ( 26 and 64 ( 260 ng/g as arithmetic means and standard deviations with median values of 0.38 and 6.5 ng/g, respectively. Although the levels were approximately 1 order of magnitude lower than those recorded in 1980s, it was estimated that there were 430 ( 110 tons of ΣHCH and 6100 ( 760 tons of ΣDDT in the surface soil of the area, respectively. The soils with high levels of OCP residuals were mostly distributed on the fringes of major cities, due to intensive farming and discharge from pesticide producers in the cities. The residuals of ΣHCHs and ΣDDTs were significantly correlated to soil organic matter content. Both R-HCH/β-HCH and p,p′DDT/p,p′-DDE ratios were log-normally distributed and negatively correlated to log(ΣHCH) and log(ΣDDT), respectively. Thus these ratios and correlations preclude certainty in distinguishing fresh application from historical usage. According to the total residuals and the distributions of R-HCH/β-HCH and p,p′DDT/p,p′-DDE ratios, it appears that significant recent input of either the OCPs is very unlikely. The calculated fugacities in soil and air provided quantitative evidence indicating a net and seasonally varied transport of ΣHCH (0.31 ton/year) and ΣDDT (1.9 ton/year) from soil to atmosphere in the study area. Such a surface-to-air transport suggested that after the ban, the surface soil had gradually converted from a major sink to an important emission source of OCPs and the reemission will continue for many years to come.
Introduction Hexachlorocyclohexane isomers (R-HCH, β-HCH, γ-HCH, δ-HCH, ΣHCH as the total) and dichlorodiphenyltrichloro* Corresponding author phone and fax: 0086-10-62751938; e-mail:
[email protected]. 10.1021/es8019676 CCC: $40.75
Published on Web 10/14/2008
2008 American Chemical Society
ethane and metabolites (p,p′-DDT, p,p′-DDE, and p,p′-DDD, ΣDDT as the total) were largely produced and extensively applied organochlorine pesticides (OCPs) in China in the past. It was estimated that the total amounts of ΣHCH and ΣDDT produced in China were more than 4 million tons and 0.46 million tons over the period from 1950 to 1983 with most of them being applied in agriculture (1, 2). As a result, high levels of these pesticides have been found in various environmental media, ecological compartments, and human tissues (3-5). With naturally occurring organic matter as an important component, soil is often the predominant sink of OCPs. Multimedia fate modeling has indicated that 75% of lindane stored in the surface soil in Tianjin at steady state before the ban (6). However, following the termination of anthropogenic application, OCP residues have declined much slower in soil than in atmosphere and water (7). Consequently, reequilibration among various bulk media has resulted in soil switching from being a major sink to a secondary emission source of OCPs. This source can act for decades because of large storage and high persistence. Therefore, the issues of adverse impacts on ecosystem and human health as well as the potential for long-range transport of OCPs released from contaminated soils are long-term public concerns (8, 9). The Haihe Plain, situated in northern China roughly between 114°E and 120°E and between 36°N and 42°N, is one of the most populated areas in China. It covers Beijing, Tianjin, Hebei and part of Shandong with a total population more than 130 million. The total size of the area is 300 000 km (2), which is approximately equal to the total size of Great Britain and Ireland. Many cities including Beijing, Tianjin, Tangshan, Shijiazhuang, Jinan, Zibo, and Baoding on the Haihe Plain are among the large industrial cities in China and numerous pesticide manufactures are located in these cities. On the other hand, the agricultural land stretches between the cities have been cultured for many decades and had received large amounts of ΣHCH and ΣDDT in the past. Because of both industrial discharge and agricultural application, the Haihe Plain is one of the most severely OCPcontaminated area in China. There have been several surveys on soil OCP contamination in the Haihe Plain; with studied areas ranged from a single orchard to an area around 20 000 km2 (3, 10-13). Despite agricultural application being banned years ago, high levels of OCP residuals in surface soils have often been reported. To understand the fate of these residuals in surface soil, we conducted an extensive investigation on OCPs in the entire Haihe Plain. In addition to assessing total quantities of ΣHCH and ΣDDT remaining in surface soil and the geographical distribution of these residues, the transport from soil to atmosphere was assessed using a fugacity approach.
Methodology Sampling. Three hundreds and two surface soil samples (1-10 cm) were collected based on an approximately equal longitude and latitude grid design from the Haihe Plain in 2004. The sampling locations are shown in the Supporting Information (Figure S1). At each location, five subsamples were collected from a 100 × 100 m2 plot (four in the corners and one in the center), and thoroughly mixed to form a composite sample. The samples were collected using stainless steel scoops and were air-dried, ground to pass through a 70-mesh sieve and maintained at 4 °C prior to analysis. Reagents. Solvents including n-hexane, acetone, and dichloromethane of analytical grade (Beijing Reagent Company, China) were purified by distillation. A mixture of VOL. 42, NO. 22, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8395
TABLE 1. Residuals of ΣHCH and ΣDDT in the Surface Soil Samples from the Haihe Plain (ng/g, n = 302) percentiles R-HCH β-HCH γ-HCH δ-HCH p,p′-DDT p,p′-DDE p,p′-DDD ΣHCH ΣDDT SOC a
mean ( SD
Gmean
min
max
P05
P10
P25
P50
P75
P90
P95
0.31 ( 2.01 2.25 ( 15.9 1.22 ( 18.4 0.11 ( 0.53 12.8 ( 56.5 48.7 ( 212 2.09 ( 7.60 3.90 ( 26.0 63.6 ( 256 0.76 ( 0.60
0.049 0.195 0.036 0.012 1.83 5.26 0.33 0.44 8.68 0.57
N.D. N.D. N.D. N.D. 0.06 0.07 N.D. 0.02 0.40 0.02
25.5 239 317 8.48 661 2140 95.9 349 2350 4.92
0.002 0.017 0.003 0.000 0.227 0.469 0.021 0.084 0.964 0.158
0.011 0.029 0.010 0.001 0.279 0.663 0.031 0.109 1.32 0.215
0.025 0.077 0.019 0.005 0.535 1.340 0.095 0.186 2.50 0.402
0.053 0.190 0.038 0.015 1.49 4.01 0.300 0.376 6.47 0.632
0.112 0.519 0.080 0.061 3.47 11.5 0.869 0.825 16.8 0.96
0.220 1.38 0.160 0.213 15.4 77.6 3.66 1.89 93.9 1.43
0.517 2.70 0.296 0.451 32.7 172 6.33 4.46 268 1.73
a
N.D.: not detectable.
organochlorine pesticide stock standard was prepared by diluting a commercial mixed standard (J&K chemical Ltd., USA) with n-hexane. The working standard solution was prepared by diluting the stock standard in n-hexane. 2,4,5,6tetrachloro-m-xylene (TCMX) and 4,4′-dichlorobiphenyl (J&K chemical Ltd., USA) were used as an a surrogate and internal standard. Silica gel (60-80 mesh, Beijing Chemical Reagent Co., China) was heated at 450 °C for 4 h, kept in a sealed desiccator, and reactivated at 130 °C for 16 h immediately prior to use. Granular anhydrous sodium sulfate (Beijing Chemical Reagent Co., China) was heated at 600 °C in a furnace for 6 h and stored in the sealed desiccator prior to use. All glassware was cleaned in an ultrasonic cleaner (KQ500B, Kunshan Ultrasonic Instrument, China) and heated at 400 °C for 6 h. Sample Extraction and Cleanup. Approximate 10 g of each dried soil sample was homogenized with 10 g of anhydrous sodium sulfate, and packed into a 34 mL stainless steel extraction cell. The sample was extracted with 20 mL mixture of n-hexane and acetone (1:1 v/v) using an accelerated solvent extractor (Dionex ASE300, USA). The extraction was carried out in one cycle with a 7-min heating followed by a 5-min static extraction at 140 °C and 105 kg/cm2, and eluted with another 20 mL of the mixed solvent. The extract was evaporated to near dryness under reduced pressure at 35 °C in a rotary evaporator. The concentrated extract was transferred with 2 mL of n-hexane onto the top of chromatography column (30 cm × 10 mm i.d.) filled with the silica gel and eluted with 20 mL of n-hexane (discarded) and a 50 mL mixture of dichloromethane and n-hexane (2:3 v/v) in sequence at a rate of 2 mL/min. The eluate was further concentrated on the rotary evaporator and the final volume was adjusted to ∼1 mL under a gentle stream of nitrogen. Sample Analysis. An Agilent Gas Chromatograph 6890 coupled with a HP-5 column (30 m × 0.32 mm i.d. × 0.25 µm film thickness) and a 63Ni-ECD detector was used for analysis. The samples were injected in splitless mode with a venting time of 0.75 min. The injector and detector temperatures were 220 and 280 °C, respectively. Oven temperature was held at 50 °C for 1 min, increased to 150 at 10 °C/min, then to 240 at 3 °C/min and maintained for 15 min. Nitrogen was used as both a carrier (1.0 mL/min) and a makeup (60 mL/min) gas. The target compounds were identified on the basis of the retention times (previously confirmed with GC-MS) and quantified by the internal standard. A mixed working standard was used for calibration. Soil organic carbon (SOC) contents were measured using a Shimadzu 5000A TOC analyzer (Japan). Quality Control. Two procedural blanks were determined simultaneously for each set of the sample analysis by going through the same extraction and cleanup procedures. The possible degradation of p,p′-DDT in the injector was routinely checked by running the standard solution. The measured 8396
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 22, 2008
procedure blanks were over 1 order of magnitude lower than most of the sample measurements. Two replicates were analyzed for randomly selected 106 samples to check for reproducibility and the mean value of the calculated standard deviations for the 106 pairs was 0.161 ng/g for ΣHCH. The recovery of the surrogate (TCMX) was 75 ( 10%. The average recoveries of the soil samples spiked with the standard solution (100 ng/g sample media) were 83, 76, 94, 105, 87, 77, and 116% for R-HCH, β-HCH, γ-HCH, δ-HCH, p,p′-DDT, p,p′-DDE, and p,p′-DDD, respectively, which were used for data correction. The detection limits based on 10 g soil sample were 0.05 ng/g for all HCH isomers and 0.3, 0.1, and 0.3 ng/g for p,p′-DDT, p,p′-DDE, and p,p′-DDD, respectively. Data Analysis. To address the potential of reemission of OCPs in soil to atmosphere, we calculated fugacities of the chemicals in soil and atmosphere after Harner (14). Similarly, equations developed by Mackay et al. were used to derive the soil-to-air transport flux (15). The detailed procedures for fugacity and flux calculation are provided in the Supporting Information. Statistica (v5.5, StatSoft) was used for performing all statistical analysis and a probability of p < 0.05 was considered as the significant level. The fugacities were calculated using Matlab (v.6.0, The MathWorks, Inc.), whereas Surfer (v.7.02, Surface Mapping System) was applied for mapping.
Results and Discussion Residuals of ΣHCH and ΣDDT in the Surface soil. On the basis of the calculated coefficients of skewness and kurtosis of the directly measured and the log-transformed data, the residuals appear to be log-normally distributed. The detailed results and the histograms are presented in the Supporting Information. Table 1 lists the descriptive statistics including arithmetic means and standard deviations (S.D.), geometric means (GMean), minimums and maximums (Min and Max), as well as various percentiles (from P05 to P95). On the basis of the measurements of soil samples collected from seven counties in this area in 2003, it was reported that the means and standard deviations of ΣHCH and ΣDDT were 4.01 ( 2.21 and 11.2 ( 17.3 ng/g, respectively (12). Another survey carried out in 2003 in Beijing found that the residuals of ΣHCH and ΣDDT in top soils were 1.47 ( 3.20 and 77.2 ( 516 ng/g, respectively (11). Although information on the levels of OCPs in 1980s, when extensive agricultural application was a common practice, is scarce, limited historical data available indicate that OCP residuals in the soil declined substantially. For example, residuals of ΣHCH and ΣDDT in the soils collected from 17 sites in an orchard in Beijing were as high as 1484 ( 2437 and 4280 ( 6054 ng/g, respectively, in 1993 and were reduced to 135 ( 222 and 951 ( 861 ng/g, respectively, at the same sites in 2003 (10). The order of magnitude decreases of lindane in various environmental media in Tianjin was also demonstrated by a fate model (7).
FIGURE 1. Spatial distributions of SOC, ΣHCH, and ΣDDT in the surface soil in the Haihe Plain in log-scale. Major cities in the area are marked on the maps. It is expected that without further input, the residuals will continue to decrease in the future. Geographical Distribution. On the basis of the measured concentrations of all the samples, the spatial distribution of SOC, ΣHCH, and ΣDDT are mapped in Figure 1. Although the detailed information on landuse, vegetation, landform, surface feature and irrigation were recorded for all the sampling sites and the data were carefully analyzed, no significant associations between OCP residuals and these physical conditions were identified. There was a general tendency that high levels of residuals were found on the fringes of major cities in the region. The contamination of the surface soil was caused predominately through two sources, i.e., the direct application in agriculture and discharge from pesticide manufactures. Like many other places in China, suburban areas of major cities are usually featured in intensive farming for vegetables and per unit area loading of pesticides was much higher than that in the field for grain production. In addition, all OCP manufactures were located in major cities including Tianjin, Beijing, Shijiazhuang and Handan. Consequently, the heavy contamination of suburban soils by ΣHCH and ΣDDT was expected. In addition, significant correlations were revealed between OCP residuals and SOC in the surface soil. With all the samples included, the Pearson correlation coefficients between logtransformed OCPs and SOC were statistically significant for both ΣHCH (p < 0.000) and ΣDDT (p ) 0.045). Detailed information on the correlations is presented in the Supporting Information (Figure S3). SOC is often found to be positively correlated to concentrations of persistent organic pollutants in soil, for example, it was found that ΣDDT and ΣHCH in surface soils from Teide Mountain depended significantly on SOC (16). In a multimedia modeling study, evidence collected suggested that the deposition rate and SOC were the two most important factors governing the spatial variation of phenanthrene in soil in Tianjin (17). In summary, the geographical distributions of ΣHCH and ΣDDT in the surface soil of the studied area were reflective of agricultural application, industrial discharge as well as soil property. Compositions of ΣHCH and ΣDDT. In China, technical mixture of ΣHCH (ca. R-HCH 71%, β-HCH 6%, γ-HCH 14% and δ-HCH 9%) was extensively applied in agriculture from the 1950s to the early 1980s while lindane (γ-HCH 99.9%) was used for another decade (18). The production of technical ΣHCH and lindane in Dagu Chemical located in Tianjin was not terminated until 2000 (19). Relative concentrations of R-HCH to β-HCH are often used to assess the history of application owing to their difference in persistency (20). Wang recently suggested that, on account of R-HCH/β-HCH ratios, illegal use of ΣHCH had continued at a few isolated locations
after the ban (13). However, the R-HCH/β-HCH ratios of the majority of the samples (293 out of 302) collected in this study were below 2.1, whereas the ratio in the technical mixture is 11.8. Among the ten samples with the ratio greater than 2.1, eight of them had ΣHCH residuals at least 1 order of magnitude below the all-sample average of 3.90 ng/g. The one sample (L11, Figure S1 in the Supporting Information) with both high ΣHCH residual (349 ng/g) and high R-HCH/ β-HCH ratio (3.0) was sampled near Dagu Chemical, where ΣHCH and lindane had been produced for several decades until production ceased in 2000. Our results suggest that continued use of ΣHCH across the Haihe Plain is unlikely to have occurred in recent years. The increase in γ-HCH fraction from 13.7% in the technical mixture to 31.3% in the surface soil can be explained in two ways: lindane was used for another decade after the technical mixture was banned in 1983 and degradation of γ-HCH is significantly slower than that of R-HCH (20). In fact, the residual levels of γ-HCH in the soil also decreased approximately an order of magnitude (7). Therefore, taking both ΣHCH composition and residual levels into consideration, fresh application of either ΣHCH or lindane on large scale in recent years in this area was very unlikely. It is interesting to note that there was a significantly negative correlation between log(R-HCH/β-HCH) (the ratio was also log-normally distributed) and log(ΣHCH) (Figure 2) with a Pearson correlation coefficient (r) of 0.366 (n ) 302, p < 0.001). With two outliers (M23 and L11) excluded, an r value of 0.512 (n ) 300, p < 0.001) was observed. Similar negative correlation was also revealed between log(γ-HCH/ ΣHCH) and log(ΣHCH) (r ) 0.285, n ) 302, p < 0.001). The correlations suggest that the ratios depend not only on how old the residuals were but also on how much OCP had been applied and the historic duration of the application. Such associations imply that a single value of R-HCH/β-HCH or similar ratios can not precisely distinguish between the historical remains and fresh applications. Further study may help to confirm such a statement and to develop a quantitative procedure for identifying fresh application of ΣHCH using information of both isomer ratios and residual levels. Since DDT was the dominant composition in the commercial product applied in agriculture (95.2%) and it gradually transforms to DDE or DDD in soil depending on redox conditionS, the relative concentrations of DDT and metabolites also provide information on the history of its application. In general, higher DDT/DDE or DDT/ΣDDT ratios indicate fresher usage and 0.5 was often applied as an arbitrary value of the ratio to distinguish the historical and recent applications. On the basis of these ratios, a few studies claimed that there were possible fresh applications of DDT in the studied area recently (10, 13). In contrast, the evidence VOL. 42, NO. 22, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8397
FIGURE 2. Observed frequency distribution of log(r-HCH/β-HCH) and the relationship between log(r-HCH/β-HCH) and log(ΣHCH). Two outliers are marked in gray.
FIGURE 3. Observed frequency distribution of log(p,p′-DDT/p,p′-DDE) and the relationship between log(p,p′-DDT/p,p′-DDE) and log(ΣDDT). Seven outliers are marked in gray. provided in this study do not support these observations. For the soil samples collected, the average percentages of p,p′-DDT, p,p′-DDE, and p,p′-DDD were 20.1, 76.6, and 3.3%, respectively, and p,p′-DDE was the dominant compound in most samples. Among the 302 soil samples collected, the only exceptions were nine samples with the ratio of p,p′DDT/p,p′-DDE greater than 2.0. However, seven of them had ΣDDT residuals (0.8- 4.1 ng/g) much lower than the mean value of 63.3 ng/g. The only two without reasonable explanation at this stage were J16 and D02 (see Figure S1 in the Supporting Information) with ΣDDT of 735 and 282 ng/g and p,p′-DDT/p,p′-DDE ratios of 3.0 and 2.3, respectively. Obviously, fresh application of DDT in this area is very unlikely except a few isolated sites, where detailed investigation is needed to clarify the situation. Like ΣHCH, the lognormally distributed p,p′-DDT/p,p′-DDE was also negatively correlated to total DDT residual (Figure 3). The Pearson correlation coefficient was 0.266 (p < 0.001) without removing any outlier (n ) 302) and became 0.399 (p < 0.001) with 7 outliers (marked in gray in Figure 3) excluded. Similar to ΣHCH, such correlation also suggests that time of application is not the only factor affecting the p,p′-DDT/p,p′-DDE ratio and the amount and duration of the historically application may also be important. Emission Potential of OCPs from the Contaminated Surface Soil. In a previous study, it was found that the OCP residues in the surface soil were uniformly distributed vertically in the plough horizon of 0 to 30 cm (Figure S4, the Supporting Information). Therefore, the total quantities (means and standard errors) of ΣHCH and ΣDDT remained in the surface (0-30 cm) soil of the studied area of 300,000 km2 in 2004 (excluding 5% of water bodies) were estimated to be 430 ( 110 and 6100 ( 760 tons, respectively (the detailed calculation is provided in the Supporting Information). In China, the total amounts of ΣHCH and ΣDDT produced over the period from 1950 to 1983 were more than 4.4 and 0.46 million tons, respectively, and most of them had been applied in agriculture (1, 2). Although detailed data on OCP applications in the studied area was not available, with the grain and vegetable productions in the studied area con8398
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 22, 2008
tributed to 8.6% and 18.5% of the national totals (21), it was estimated that the accumulative quantities applied in this area in the past were in the ranges of hundreds of thousands tons of ΣHCH and tens of thousands tons of ΣDDT, respectively. It appears that the persistence of ΣDDT in the soil was around 2 orders of magnitude longer than that of ΣHCH. Although the quantities of OCPs remained in the soil were only small fractions of the total inputs and the residuals of ΣHCH and ΣDDT were remarkably lower than what they were years ago, a large quantity of them will stay there for decades to come. After the termination of discharge and application, soil with large quantities of accumulated OCPs has converted gradually from a major sink to an important reemission source. The tendencies of soil-to-air emission can be described by calculating the fugacities of the chemicals in both soil and atmosphere.On the basis of the measured ΣHCH and ΣDDT in the surface soil (the current study) and in air (Supporting Information), the fugacities of individual HCH isomers and DDT and metabolites in the surface soil at individual sampling sites and in air as the averages of the entire area were derived (details provided in the Supporting Information). As two typical examples, the spatial distributions of the calculated fugacities (25 °C) of R-HCH and p,p′DDE are mapped in Figure 4. The distribution patterns of the fugacities were similar to but not the same as those of the concentrations, because the fugacities depend on both OCP residuals and SOC. In general, high tendencies of evaporation from soil to air occur at the sites with high OCP residuals and low SOC contents. For the same reason, high SOCs are favorable for accumulation of OCPs during application and unfavorable for escape of them after the application is terminated. For R-HCH, the annual mean fugacities in the surface soil of the studied area ranged from 9.30 × 10-12 to 9.94 × 107 Pa, with a median value of 4.17 × 10-9 Pa. In comparison, the mean R-HCH fugacity in air was 3.27 × 10-9 Pa, which was lower than the fugacities in the surface soil at 41% of the sampling sites. For p,p′-DDE, the annual mean fugacities in soil varied from 1.71 × 10-10 to 8.05 × 10-6 Pa with a median value of 9.45 × 10-9 Pa, and
FIGURE 4. Spatial distributions of annual mean fugacities of r-HCH and p,p′-DDE in the surface soils in the Haihe Plain in log-scale. The corresponding mean fugacities in atmosphere were 3.27 × 10-9 and 1.46 × 10-9 Pa for r-HCH and p,p′-DDE, respectively. The areas with net evaporation from soil to air and the areas with net deposition from air to soil are separated by the red contours, which are air fugacities.
FIGURE 5. Monthly soil-to-air fluxes of r-HCH and p,p′-DDE in the Haihe Plain based on the median residual values at all sampling sites. the soil fugacities at 85% of the sites were higher than air fugacity of 1.46 × 10-9 Pa. The red lines in Figure 4 mark contours where soil fugacity is equal to air fugacity, which separates the areas of net evaporation from soil to air (darker colored) and those of net deposition from air to soil (lighter colored). Since the fugacities are temperature-dependent, the seasonal variations in the fugacities as well as the transport flux across the air-soil interface were also modeled (see the Supporting Information for details). The median values of 302 samples were used for the calculation. As the typical examples, the calculated monthly soil-to-air transport flux densities for R-HCH and p,p′-DDE are plotted over time in Figure 5. In fact, all compounds of HCH isomers and DDT and metabolites had similar patterns of monthly variation following the change in temperature. The emissions from soil to air were significantly stronger in summer than those in the other seasons with peaks in July and August. The annual emissions of ΣHCH and ΣDDT from soil to air in the study area were estimated to be 0.31 tons and 1.9 tons, respectively, in 2004. Given the total residuals of 430 tons and 6100 tons in the area, only very small fractions of ΣHCH and ΣDDT evaporate from soil to air each year and this process is expected to last years. The emission of OCPs from the surface soil to atmosphere was also confirmed by the detectable levels of both ΣHCH and ΣDDT in air in the studied area many years after the ban of these pesticides (the Supporting Information). Like other semivolatile organic pollutants emitted from this area, ΣHCH and ΣDDT are also expected to be transported to remote areas by uplifting and westly transport. On the basis of a review on sources and pathways of selected organochlorine pesticides to the Arctic, it was suggested that R-HCH concentration in Arctic air
responded to the change in emission in China due to atmospheric transport (22). More importantly, ΣHCH and ΣDDT from the surface soil can have adverse impacts on local ecosystem and food production. As semivolatile hydrophobic pollutants, the dominant pathway from environmental media to grains and aerial parts of vegetables is the deposition from atmosphere (23). Such a diffusive transport is the primary reason leading to the contamination of locally grown grains and vegetables. In a survey on vegetable contamination in Tianjin, it was found that residuals of both ΣHCH and ΣDDT in aerial parts of eight vegetables growing at a heavily contaminated site were generally higher than those from a less contaminated site (24). The soil-to-air diffusion and foliar deposition are believed to be the major pathways of the contamination of the vegetables. Unfortunately, the heavily polluted sites are mostly located in the suburbs of major cities, where most vegetables are produced. Because the consumptions of vegetables and grains are two major pathways of human exposure to OCPs in the area (25), the contamination of ΣHCH and ΣDDT in edible parts of vegetables from this area is and will be a public concern. Particular attention should be given to the places where the organochlorine pesticides were produced or stored in the past. In the studied area, very high levels of ΣHCH (349 ng/g) and ΣDDT (2350 ng/g) in the surface soils were found at the sites near old production facilities. Risk caused by the dispersion of OCPs from these sites should be thoroughly assessed.
Acknowledgments The funding for this study was provided by the National Scientific Foundation of China (40730737, 40710019001), National Basic Research Program (2007CB407301), and the High Technology Research and Development Program of China (2007AA06Z408). The authors thank Dr. Brian Reid for his valuable comments on the manuscript.
Supporting Information Available Map of sampling locations; statistical distributions of ΣHCH and ΣDDT in the surface soil; correlations between OCPs and SOC; vertical distribution of OCPs in soil; calculation of total residuals of ΣHCH and ΣDDT in the surface soils; calculation of fugacities; measured concentration of OCPs in the surface soil and air; and calculation of soil-to-air flux (PDF). This material is available free of charge via the Internet at http://pubs.acs.org. VOL. 42, NO. 22, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8399
Literature Cited (1) Li, Y. F. Global technical hexachlorocylohexane usage and its contamination consequences in environment: from 1948 to 1997. Sci. Total Environ. 1999, 232, 123–60. (2) Li,Y. F.; Li,D. C. Global Emission Inventories for Selected Organochlorine Pesticides; Internal Report; Meteorological Service of Canada: Toronto, Canada, 2004. (3) Gong, Z. M.; Xu, F. L.; Dawson, R; Cao, J; Liu, W. X.; Li, B. G.; Shen, W. R.; Zhang, W. J.; Qin, B. P.; Sun, R; Tao, S. Residues of hexachlorocyclohexane isomers and their distribution characteristics in soils in the Tianjin area, China. Arch. Environ. Contam. Toxicol. 2004, 46, 432–437. (4) Sun, Y. Z.; Wang, X. T.; Li, X. H.; Xu, X. B. Distribution of persistent organochlorine pesticides in tissue/organ of silver carp (Hypophthalmichthys molitrix) from Guanting Reservoir, China. J. Environ. Sci. (China) 2005, 17, 722–726. (5) Wong, H. H.; Leugn, A. O. W.; Chang, J. K. Y.; Choi, M. P. K. A review on the usage of POP pesticides in China, with emphasis on DDT loadings in human milk. Chemosphere 2005, 60, 740– 752. (6) Cao, H. Y.; Tao, S.; Xu, F. L.; Coveney, R. M.; Cao, J.; Li, B. G.; Liu, W. X.; Wang, X. J.; Hu, J. Y.; Shen, W. R.; Qin, B. P.; Sun, R. Multimedia fate model for hexachlorocyclohexane in Tianjin, China. Environ. Sci. Technol. 2004, 38, 2126–2132. (7) Tao, S; Yang, Y; Cao, H. Y.; Liu, W. X.; Coveney, R. M.; Xu, F. L.; Cao, J; Li, B. G.; Wang, X. J.; Hu, J. Y.; Fang, J. Y. Modeling the dynamic changes in concentrations of gamma-hexachlorocyclohexane (gamma-HCH) in Tianjin region from 1953 to 2020. Environ. Pollut. 2006, 139, 183–193. (8) Soto, A. M.; Chung, K.; Sonnenshein, C. The pesticides endosulfan, toxaphene and dieldrin have estrogenic effects on human estrogen-sensitive cells. Environ. Health Perspect. 1994, 102, 380–383. (9) Abdullah, A. R.; Bajet, C. M.; Matin, M. A.; Khan, D. D.; Sulaiman, A. H. Ecotoxicology of pesticides in the tropical paddy field ecosystem. Environ. Toxicol. Chem. 1997, 16, 59–70. (10) Shi, Y. J.; Guo, F. F.; Meng, F. Q.; Lu, Y. L.; Wang, T. Y.; Zhang, H. Temporal trend of organic chlorinated pesticides residues in soils of orchard, Beijing. Acta Sci. Circum. 2005, 25, 313–318. (11) Zhang, H. Y.; Gao, R. T.; Jiang, S. R.; Huang, Y. F. Spatial variability of organochlorine pesticides (DDTs and HCHs) in surface soils of farmland in Beijing, China. Sci. Agric. Sin. 2006, 39, 1403– 1410. (12) Zhao, B. Z.; Zhang, J. B.; Zhu, A. N.; Xia, M.; Lu, X.; Jiang, Q. Residues of HCH and DDT in typical agricultural soils of Huang-
8400
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 22, 2008
Huai-Hai Plain, China. Acta Pedolog. Sin. 2005, 42, 916–922. (13) Wang, T. Y.; Lu, Y. L.; Shi, Y. J.; Giesy, J. P.; Luo, W. Organochlorine pesticides in soils around Guanting Reservoir, China. Environ. Geochem. Health 2007, 29, 491–501. (14) Harner, T.; Bidleman, T. F.; Jantunen, L. M. M.; Mackay, D. Soil-air exchange model of persistent pesticides in the U.S. Cotton Belt. Environ. Toxicol. Chem. 2001, 20, 1612–1621. (15) Mackay, D. Finding fugacity feasible. Environ. Sci. Technol. 1979, 13, 1218–1223. (16) Ribes, A.; Grimalt, J. O.; Garcia, C. J. T.; Cuevas, E. Temperature and Organic Matter Dependence of the Distribution of Organochlorine Compounds in Mountain Soils from the Subtropical Atlantic (Teide, Tenerife Island). Environ. Sci. Technol. 2002, 36, 1879–1885. (17) Tao, S.; Cao, H. Y.; Liu, W. X.; Li, B. G.; Cao, J.; Xu, F. L.; Wang, X. J.; Coveney, R. M.; Shen, W. R.; Qin, B. P.; Sun, R. Fate modeling of phenanthrene with regional variation in Tianjin, China. Environ. Sci. Technol. 2003, 37, 2453–2459. (18) Lin, Y. Can usage of lindane be extended ? Sci. Manage. Pestic. 1996, 59, 30. (19) On Implementation of the Notification from the Office of the State Department on Management of Pesticides and Animal Medicines; Report MCI [1992]77; Ministry of Chemical Industry: Beijing, 1992. (20) Chessells, M. J.; Hawker, D. W.; Connell, D. W.; Papajcsik, I. A. Factors influencing the distribution of lindane and isomers in soil of an agricultural environment. Chemosphere 1988, 17, 1741– 1749. (21) National Statistics Bureau. Chinese Statistics Year Book 2003; Beijing Statistics Press: Beijing, 2004. (22) Li, Y. F.; Macdonald, R. W. Sources and pathways of selected organochlorine pesticides to the Arctic and the effect of pathway divergence on HCH trends in biota: a review. Sci. Total Environ. 2005, 342, 87–106. (23) Tao, S; Jiao, X. C.; Chen, S. H.; Xu, F. L.; Li, Y. J.; Liu, F. Z. Uptake of vapor and particulate polycyclic aromatic hydrocarbons by cabbage. Environ. Pollut. 2006, 140, 13–15. (24) Tao, S.; Xu, F. L.; Wang, X. J.; Liu, W. X.; Gong, Z. M.; Fang, J. Y.; Zhu, L. Z.; Luo, Y. M. Organochlorine pesticides in agricultural soil and vegetables from Tianjin, China. Environ. Sci. Technol. 2005, 39, 2494–2499. (25) Guo, M.; Tao, S.; Yang, Y.; Li, B. G.; Cao, J.; Wang, X. J.; Xu, F. L.; Liu, W. X.; Wu, Y. N. Population exposure to HCH in Tianjin area. Environ. Sci. 2005, 26, 164–167.
ES8019676
