Monitoring and Modeling Endosulfan in Chinese Surface Soil

Nov 17, 2010 - Financial support from Dalian Maritime University (Reward for .... 1997, 45, 2291– 2298 ...... poisoning cases and identify etiol. ag...
2 downloads 0 Views 2MB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 9279–9284

Monitoring and Modeling Endosulfan in Chinese Surface Soil H O N G L I A N G J I A , * ,† L I Y A N L I U , ‡ Y E Q I N G S U N , †,§ B I N G S U N , † DEGAO WANG,† YUSHAN SU,| K U R U N T H A C H A L A M K A N N A N , ‡,⊥ A N D Y I - F A N L I * ,†,‡,| International Joint Research Center for Persistent Toxic Substances (IJRC-PTS), Dalian Maritime University, Dalian, P. R. China, IJRC-PTS, State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, P. R. China, Environmental System Biology Institute, Dalian Maritime University, Dalian, P. R. China, Science and Technology Branch, Environment Canada, Toronto, Canada, and Wadsworth Center, New York State Department of Health, and Department of Environmental Health Sciences, School of Public Health, State University of New York at Albany Empire State Plaza, P.O. Box 509, Albany, New York 12201-0509, United States

Received August 14, 2010. Revised manuscript received October 10, 2010. Accepted November 4, 2010.

Endosulfan is a currently used organochlorine pesticide in China, with annual usage of 2300 t between 1994 and 2004. Concentrations of endosulfan (including R- and β-isomers and their metabolite endosulfan sulfate) were reported for surface soil collected in 2005 at 141 sites (6 background, 95 rural, and 40 urban) across China. The concentrations of total endosulfan (sum of R-endosulfan, β-endosulfan, and endosulfan sulfate) at all sites ranged from BDL (below detection limit) to 19000 pg/g dry weight (dw), with geometric mean (GM) 120 pg/g dw. Rural soils had the highest total endosulfan concentrations, with GM 160 pg/g dw, followed by urban soils (GM ) 83 pg/g dw) and background soils (GM ) 38 pg/g dw). The observed soil concentrations of R-endosulfan (GM ) 6.5 pg/g dw) were much lower than those of β-endosulfan (GM ) 49 pg/g dw) and endosulfan sulfate (GM ) 47 pg/g dw). The fractional abundance of R-endosulfan FR-endo [R-endosulfan/(R-endosulfan + β-endosulfan)] for all soils ranged from 0.00040 to 0.91, with GM 0.10, much lower than those in technical products (ranged from 0.67 to 0.7), which most likely reflects that R-endosulfan is more volatile and degrades faster than β-endosulfan in soil. Consequently, half-life of β-endosulfan in soil is expected longer than R-endosulfan. Significant correlation between endosulfan sulfate and its parent isomers suggested that the presence of endosulfan sulfate originated from its parent isomers. Based on multiple linear regression model, inventories of endosulfan sulfate in Chinese agricultural soil * Corresponding author phone: 86-411-8472-8489; fax: 86-4118472-8489; e-mail: [email protected] (H.J.); phone: (416)739-4892; fax: (416)739-4288; e-mail: [email protected] (Y.-F.L.). † International Joint Research Center for Persistent Toxic Substances (IJRC-PTS), Dalian Maritime University. ‡ Harbin Institute of Technology. § Environmental System Biology Institute, Dalian Maritime University. | Environment Canada. ⊥ State University of New York at Albany Empire State Plaza. 10.1021/es102791n

 2010 American Chemical Society

Published on Web 11/17/2010

in 2004 with a 1/4° longitude × 1/6° latitude resolution are established. Comparison between field measurements and modeling results showed significant correlations between the modeled and measured endosulfan concentrations, and 89%, 83%, and 70% of monitoring data fell between the lowest and the highest modeled concentrations for R- and β-endosulfan and endosulfan sulfate, respectively. The good agreement lends credibility to modeled soil concentrations of endosulfan. To our knowledge, this is the first soil concentration inventory for endosulfan sulfate, which paves the way for further study on its environmental behavior.

Introduction Cyclodiene insecticide endosulfan (6,7,8,9,10-hexachloro1,5,5a,6,9,9a-hexahydro-6,9-methano- 2,4,3-benzo[e]dioxathiepin-3-oxide) is applied as a technical mixture of R- and β-endosulfan. The two isomers have similar insecticidal (1, 2) but different physicochemical properties (3-6). Under environmental conditions, the cyclic sulfite group of endosulfan can be oxidized to corresponding sulfate (i.e., endosulfan sulfate) (7-9) or hydrolyzed to a less toxic diol (2). Endosulfan sulfate has been identified as the dominant metabolite of endosulfan degradation in soil (9-14). Sethunathan et al. (15) reported that endosulfan sulfate was the major metabolite product detected in nonflooded and flooded soil, whereas endosuflan diol was not detected in any soil samples. Study on endosulfan degradation in seven types of soil showed that endosulfan sulfate was the major transformation product under aerobic incubations (16). Endosulfan sulfate is less volatile than the R- and β-isomers and more persistent than its parents compounds in soil, sediment, and biota (17). Endosulfan sulfate and R- and β-endosulfan are generally considered to be equally toxic and classified by the US Environmental Protection Agency as priority pollutants (18). Endosulfan is listed as a candidate for new persistent organic pollutants (POPs) under the Stockholm Convention (19) and was recently banned for use in the USA (20). Endosulfan is currently extensively used in agriculture around the world including some food and nonfood crops. After application, endosulfan is released into atmosphere through spray drift, postapplication volatilization, and wind erosion. Some recent studies showed that endosulfan occurs at relatively high air concentrations across the globe compared to other organochlorine pesticides (OCPs) (21-23). Endosulfan has been applied in China to control pests in cotton since 1994 and in wheat, tea, tobacco, apple, and other fruits since 1998. The total usage of endosulfan in China was estimated to be approximately 25,700 t between 1994 and 2004 (24). The first Chinese gridded inventories of usage, emissions to air, and residues in soil inventories with a 1/4° longitude by 1/6° latitude resolution for R- and β-endosulfan were produced for 1994-2004 (24, 25). Air emissions were 7400 t for R-endosulfan and 3300 t for β-endosulfan between 1994 and 2004 (25). Soil residues in application areas in 2004 varied largely for R-endosulfan, ranging from 1.7 to 365 t but much less for β-endosulfan with a range from 119 to 263 t. Based on the emission and residue inventories, concentrations of R- and β-endosulfan in Chinese agricultural surface soil were further calculated for each grid cell (25). In order to evaluate the modeled concentration inventories, a survey of endosulfan in Chinese soil was carried out by the International Joint Research Center for Persistent Toxic Substances (IJRC-PTS) in 2005 as part of a Chinese persistent VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9279

TABLE 1. Summary of Endosulfan Concentrations and Statistical Resultsa GM (range) (pg/g dw) compounds

all sites (n ) 141)

background (n ) 6)

rural (n ) 95)

urban (n ) 40)

R-endosulfan β-endosulfan endosulfan sulfate total endosulfan FR-endo

6.6 (BDL-2300) 49 (BDL-4700) 47 (BDL-16000) 120 (BDL-19000) 0.10(0.0004-0.91)

4.9 (BDL-22) 9.1 (3.1-16) 16 (BDL-68) 38 (16-87) 0.28 (0.047-0.84)

6.5 (BDL-620) 65 (BDL-3400) 66 (BDL-1600) 160 (BDL-19000) 0.080 (0.0004-0.76)

7.2 (BDL-2300) 32 (BDL-4700) 25 (BDL-4100) 83 (BDL-9500) 0.15 (0.0053-0.91)

OF (%) kurtosis skewness 83 96 91 -----

91.0 31 70 48 1.8

9.0 5.2 7.7 6.3 1.5

a GM ) geometrical mean; OF ) occurrence frequency; BDL ) below detection limit; total endosulfan ) R-endosulfan + β-endosulfan + endosulfan sulfate; FR-endo ) R-endosulfan/(R-endosulfan + β-endosulfan).

toxic substances (PTSs) Soil and Air Monitoring Program, phase I (SAMP-I), in which concentrations of PTSs including endosulfan were monitored in Chinese air and soil. Some results of this program have been reported elsewhere, such as soil concentrations of polychlorinated biphenyls (PCBs) (26), and air concentrations of PCBs (27) and dechlorane plus (28) across China. The objective of this study is to report nationwide measurements of two endosulfan isomers and their metabolite endosulfan sulfate in Chinese soil. The soil measurements were further compared to the modeling results reported by Jia et al. (25) for evaluation purposes.

Methodology Field Sampling. Soil was sampled at 141 sites across China in 2005 including 6 background sites, 95 rural sites, and 40 urban sites; the details of the sampling sites, which covered site code, site type, and geographic coordinates, are given in the Supporting Information (SI). Details on soil sampling and analytical procedures were reported elsewhere (26), and a brief description is provided here. Surface soil samples (0-20 cm) were collected using clean stainless steel spades. Five portions of soil, taken in an area of several square meters, were bulked and mixed evenly to obtain one sample. Overlying vegetation was removed prior to sample collection. Samples were sealed in a solvent-rinsed glass bottle with Teflon cap and sent to the IJRC-PTS laboratory where they were stored frozen (-20 °C) until extraction. Analytical Procedures. Five grams of each soil sample was accurately measured into a precleaned extraction thimble and spiked with a labeled recovery standard (surrogate) containing 2,3,5,6-tetrachlorobiphenyl (CB-65), 2,2′,4,4′,6,6′hexachlorobiphenyl (CB-155), 1,3,5-tribromobenzene (TBB), and D8-p,p’-DDT (Accustandard, New Haven, CT, USA). Samples were Soxhlet extracted for 24 h with 200 mL of n-hexane/acetone (1:1 v/v). The extract was filtered through a funnel filled with anhydrous sodium sulfate and then rotaryevaporated to 1 mL. The extract was then passed through 10 g silica gel column and eluted with 80 mL of 50% hexane in dichloromethane (1:1, v/v). The elution was rotaryevaporated to 2 mL, solvent-exchanged to isooctane, and reduced to l mL under a gentle nitrogen stream. The internal standards, CB-30, CB-204, and octachloonaphthalene (OCN) were added to correct volume difference prior to GC-MS analysis. Endosulfan isomers and endosulfan sulfate were identified and quantified with GC-MS in NCI mode (Agilent 6890). A DB-5 MS capillary column was used with dimensions of 0.25 µm film thickness, 0.25 mm ID, and 30 m length. The column oven temperature was programmed at a rate of 10 °C/min from an initial temperature of 80 °C (2-min hold time) to 160 °C (1-min hold time) at a rate of 1.5 °C/min and then to 230 °C (15-min hold time) at a rate of 20 °C/min to 280 °C (10min hold time). Injector and transfer line temperatures were held at 250 and 280 °C, respectively. Quality Assurance/Quality Control. Values of 3 times the instrument detection limits (IDL) were used as the method 9280

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

detection limit (MDL), giving MDL values of 0.61 pg/g dry weight (dw) for R-endosulfan and 3.2 pg/g dw for both β-endosulfan and endosulfan sulfate. The values of 2/3 MDL were used to substitute the BDL (below detection limit) data for the statistical analysis. Procedural blanks were included once every 10 samples, and endosulfan concentrations in these blanks were all found below MDL. The recoveries were 105 ( 13% for CB-65, 111 ( 17% for CB-155, 103 ( 22% for TBB, and 105 ( 20% for D8-p,p’-DDT. The external spike samples for endosulfans were included at a rate of once every 10 soil samples by spiking clean sodium sulfate with the calibration standards and passing them through entire procedures. Average spike recoveries were 96 ( 10% for R-endosulfan, 60 ( 18% for β-endosulfan, and 61 ( 21% for endosulfan sulfate. The lower recovery for endosulfan sulfate was likely related to adsorption on active sites of glasswares during sample treatment and analysis because of its higher polarity (29, 30). Significant conversion of β-endosulfan to R-isomer was reported from a number of studies (6, 31, 32) but not vice versa. Schmidt et al. (6) reported that 10% of β-endosulfan was converted to R-isomer at a temperature of 160 °C. The conversion might also occur during sample preparation (e.g., Soxhlet extraction) and instrumental analysis (e.g., injection and column separation). To accommodate differences associated with experiment process and instrumental analysis, recovery corrections were performed for soil concentrations of β-endosulfan and endosulfan sulfate.

Results and Discussion Soil Concentrations and Spatial Distribution. Analytical results of each species for all soil samples are listed in Table SI-1. Endosulfan was found ubiquitous in Chinese surface soil. Occurrence frequencies (OFs) were high in the 141 soil samples and 83%, 96%, and 91% for R-endosulfan, β-endosulfan and endosulfan sulfate, respectively. Geometrical mean (GM) concentrations and their ranges are summarized in Table 1 for R-endosulfan, β-endosulfan, endosulfan sulfate, and total endosulfan (sum of R-endosulfan, β-endosulfan, and endosulfan sulfate). Concentrations of total endosulfan ranged from BDL to 19000 pg/g dw with GM ) 120 pg/g dw. Rural soils showed the highest total endosulfan concentrations (GM ) 160 pg/g dw), followed by urban soils (GM ) 83 pg/g dw) and background soils (GM ) 38 pg/g dw). Higher concentrations in rural soils are expected since endosulfan was widely applied in agriculture areas for pest control purpose. Soil measurements of endosulfan were reported in some other locations around the world, as shown in Table SI-2. Concentrations of endosulfan in this study were in line with previous measurements but close to the lower end of the concentration ranges. Spatial distribution of soil concentration for total endosulfan is shown in Figure 1. The highest total endosulfan concentration (19000 pg/g dw) was found at R51, a rural site in Yancheng, Jiangsu Province, which is located in an agricultural area where endosulfan was extensively used (24).

FIGURE 1. Distribution of total endosulfan (total endosulfan ) r-endosulfan + β-endosulfan + endosulfan sulfate) in Chinese surface soil from 141 sites, among which 6 are background sites (blue), 95 are rural sites (green), and 40 are urban sites (red). The numbers from 80 to 130 on top are longitude in °E, and the numbers from 20 to 50 at left are latitude in °N. High concentrations (>4000 pg/g dw) of total endosulfan were also found at some other rural sites like R12, R57, R69, and R73. Among urban sites, high concentrations of total endosulfans were found at U17 in Xizang Autonomous Region (Tibet) and U09 and U28 in Fujian Province. High concentrations of R-endosulfan were previously reported in urban air in southern China, for example, 348 pg/m3 in Guangzhou and 124 pg/m3 in Hong Kong (33). Local sources, such as possible use of this insecticide on vegetables and fruits, are a plausible explanation for high levels in these urban areas (25). High concentrations at the urban site U17 in Xizang were, however, unexpected. To our knowledge, no historical use of endosulfan was reported in this area, and no point sources (like pesticide factories) were found in the area. One possible explanation is that some amount of endosulfan was used locally, and the other is that high soil concentration at U17 came from long-range atmospheric transport (LRAT) inputs. India is one of major endosulfan producers, and its total production were 41033 t in 1995-2000 (34), a factor of 1.6 higher than total usage in China (25700 t) between 1994 and 2004 (24). Higher air concentrations of endosulfan (0.45-1120 pg/m3) were found in India than other regions around the world (34). Emissions of persistent organic pollutants (POPs) in India have a large influence on Southwestern China and particularly on Xizang (35). It was reported that the Indian source was the largest contributor (>50%) to γ-HCH in air in western China and was also responsible for approximately 55% of total deposition of γ-HCH in China in 2005, especially western China (35). In a recent study conducted in Xizang, much higher air concentrations of R-endosulfan were found at sites close to the China-India border than those at other sites (36). Higher soil concentrations at U17 were likely related to heavy usage of endosulfan in India and consequent LRAT input to Xizang. It is worthwhile to point out that atmospheric transport of endosulfan in a continental scale has also been reported by other scientists (37, 38). The use of chemicals can lead to high soil concentrations in source regions, forming primary sources, which are the sources where the chemicals are directly produced, stored, applied, or discharged. After release from a primary source, the chemicals will be dispersed through air and water current, and the secondary sources will be created almost immediately in the uncontaminated surrounding areas that do not have these chemicals before. At the initial stage of this process, a distribution pattern of chemicals in soil is called primary distribution. Continuous dispersal of chemicals through

runoff and repeated cycles of volatilization and deposition leads to that the chemicals in the primary sources decrease, while those in the secondary sources increase, gradually forming the secondary distribution (39). A primary distribution pattern, depending on the use pattern of the chemicals, is characterized by high concentration in the source regions and much lower in other regions, while a secondary distribution pattern has more uniform concentration pattern or the pattern dependent on the soil total organic carbon and temperature, not by the use pattern of the chemicals. Spatial distribution of soil concentrations of chemicals can be described through two statistical parameters, skewness and kurtosis. The term of skewness is a measure of “asymmetry”, and kurtosis is a measure of the “peakedness” for the probability distribution of a real-valued random variable. Higher kurtosis means more variance due to infrequent extreme deviations (primary distribution pattern), as opposed to frequent modestly sized deviations (secondary distribution pattern). When an OCP is banned for use, its air concentration distribution is expected to become more normally distributed spatially (i.e., lose its asymmetry and peakedness), indicating a secondary distribution pattern. One good example is atmospheric hexachlorobenzene in North America (23). As a current use pesticide, spatial distribution of endosulfan compounds displayed high skewness (>5) and high kurtosis (>30) (Table 1), an indication of “extreme” concentrations (Figure SI-1), a typical primary distribution pattern. Comparison between r- and β-Endosulfan and Endosulfan Sulfate in Soil. In the present study, the geometric mean soil concentration of β-endosulfan (49 pg/g dw) was much higher than that of R-endosulfan (6.5 pg/g dw). Observed FR-endo [R-endosulfan/(R-endosulfan + β-endosulfan)] in the 141 soil samples ranged from 0.00040 to 0.91, with GM ) 0.10, was much lower than that in technical products (0.67-0.7) (40, 41), which was also reported in soils measured in some other locations around the world (Table SI-2). Lower FR-endo values were likely related to the different degradation half-lives and physical-chemical properties for these two isomers (Table SI-3). As shown in Table SI-3, higher vapor pressure and Henry’s law constant for R-endosulfan than β-endosulfan suggest that R-endosulfan has a greater tendency to evaporate from surface media to air. It is apparent that half-life of β-endosulfan in soil is much longer than R-endosulfan, mainly because of its slower degradation (10, 42) and lower volatility (25, 43). This is also consistent with air measurements showing that R-endosulfan was the most abundant isomer in ambient air, accounting for more than 90% of total endosulfan (21-23). Soil concentration (GM) of endosulfan sulfate (47 pg/g dw) was comparable to that of β-endosulfan (49 pg/g dw) but much higher than that of R-endosulfan (6.5 pg/g dw). The high concentrations of endosulfan sulfate reflect historical use of technical endosulfan in China and its consequent degradation. A few degradation studies on degradation of Rand β-isomer to endosulfan sulfate have been reported. In aquatic systems, R-isomer is converted more readily to endosulfan sulfate than the β-isomer (44-46). Similar degradation studies in soil have been barely reported although Ghadiri and Rose (11) claimed slower degradation of β-endosulfan than R-endosulfan. In order to make further discussion of statistics for endosulfan residue data, log-transformation was performed for all the monitoring data. It appears that all compounds were log-normally distributed in Chinese surface soil (Figure SI-2). Pearson correlation of log-transformed concentrations between endosulfan sulfate and endosulfan isomers were found statistically significant at rural sites, urban sites, and all sites included (Table SI-4) but not at background sites. Elevated endosulfan sulfate was found at sites where concentrations of R- and β-endosulfan were high as well. It VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9281

suggests that sources of endosulfan sulfate are likely from historical application of endosulfan and consequent degradation. Better correlation for β-endosulfan than R-endosulfan implies longer half-life of β-endosulfan in soil (10, 25, 42) as the result of its slower degradation and lower volatility (25, 43, 47). Modeling Soil Concentrations of Endosulfan Sulfate. Endosulfan sulfate is as persistent and toxic as its parent compounds (19, 48). Its concentrations were found higher than those of R-endosulfan and comparable to those of β-endosulfan in Chinese surface soil. Therefore, study on soil residues of endosulfan sulfate is equally important as endosulfan isomers. The modeled lowest (preapplication) and highest (postapplication) annual concentrations for R- and β-endosulfan were reported in an earlier study for Chinese agricultural soil from 1995 to 2004 with 1/4° × 1/6° longitude and latitude resolution by using the Simplified Gridded Pesticide Emission and Residue Model (SGPERM) (25, 49). The model is an integrated modeling system combining mathematical model, database management system, and geographic information system. By using the gridded endosulfan usage inventories (24), gridded annual mean air and soil concentrations for Rand β-endosulfan in each cell were determined (The detailed description of the model is presented in the SI.). Soil concentrations for endosulfan sulfate were, however, not available because transformation mechanism from endosulfan isomers to endosulfan sulfate is not well understood. Significant and strong correlation between endosulfan sulfate and endosulfan isomers in Chinese soil provides an opportunity for us to compile soil concentration inventories for endosulfan sulfate based on the inventories of endosulfan isomers. Multiple linear regression was used in this study to calculate endosulfan sulfate soil concentrations using soil concentrations for R- and β-endosulfan. Measured soil logarithm concentrations of endosulfan sulfate at 141 sampling sites were treated as dependent variables, whereas those of R- and β-endosulfan were set as two independent variables. The multilinear regression was performed with SPSS V.10.0, and the coefficients of regression were found statistically significant (R ) 0.87, p < 0.0001) as shown in eq 1 log y ) 0.1338log x1 + 0.8895log x2 + 0.05688

(1)

where y, x1, and x2 are measured soil concentrations of endosulfan sulfate, R-endosulfan, and β-endosulfan, respectively. Gridded soil concentrations of endosulfan sulfate were calculated by using eq 1 using the modeled soil concentrations of R- and β-isomer (25). Figure 2 depicts the gridded lowest, highest, and annual mean concentrations of endosulfan sulfate in Chinese agricultural soil in 2004 with a 1/4° × 1/6° longitude and latitude resolution. The highest concentrations of endosulfan sulfate in Chinese agricultural soil were in Yunnan Province, where tobacco cultivation is common. Other regions with high soil concentrations included northern Gansu, southern Anhui, northern of Fujian, and parts of Xinjiang Autonomous Region. To our knowledge, this is the first soil residue and concentration inventories for endosulfan sulfate, which paves the way for further study on its environmental fate and pathways. Comparison between Monitoring and Modeling Results. The lowest (preapplication) and highest (postapplication) concentrations for R- and β-endosulfan were calculated in an earlier study in Chinese agricultural soil between 1994 and 2004 in 1/4° × 1/6° longitude and latitude resolution (25) and for endosulfan sulfate in this study. Our modeling work indicated that, before the pesticide application, the soils had the lowest residues due to endosulfan remaining in the soil from the previous years, and after application, the 9282

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

FIGURE 2. Distribution of (a) the lowest, (b) the highest, and (c) annual mean concentration (ng/g dw) of endosulfan sulfate in Chinese agricultural soil in 2004 with a 1/4° longitude by 1/6° latitude resolution. soils had the highest residues due to endosulfan remaining in the soil from the previous years plus the residues of current use. In the present study, it is hard to know the application time of endosulfan across the country. In order to further assess the differences between modeled and monitored data, annual average soil concentrations of endosulfan were also calculated for China in this study. Measured soil concentrations of R- and β-endosulfan and endosulfan sulfate at the 141 sites in 2005 in this study were compared to the modeled results in the grid cells containing these sites for 2004, and results are presented in Figure 3.

“outliers” appear both above the modeled highest and below the modeled lowest concentrations. The causes for these outliers could be due to several factors. First, the modeled gridded endosulfan usage inventories (24) based on which the soil concentrations of endosulfan were calculated have uncertainties. Second, the depositions of endosulfan to soil due to shortand long-range atmospheric transport were not considered in our model. A good example of the outlier is high concentration found at Site U17 in Xizang discussed in the previous section. Most importantly, the modeled gridded soil concentration is the mean concentration in agricultural soil in each cell (around 25 km ×25 km), while the measured concentration is just for the one sampling site in the grid cell, and thus, the existence of the outliers is not unexpected. Correlations between monitored and the annual mean modeled results (log-scale) in all soil samples were found significant for both isomers and endosulfan sulfate (P < 0.0001) with the correlation coefficient R ) 0.48 for R-endosulfan, R ) 0.80 for β-endosulfan, and R ) 0.67 for endosulfan sulfate (Figure SI-4). Better agreement for β-endosulfan and endosulfan sulfate confirms their longer halflife in soil (i.e., less differences between the modeled highest and lowest concentrations) for β-endosulfan and endosulfan sulfate than R-endosulfan in soil.

Acknowledgments We are grateful to many volunteers for helping nationwide soil sampling, especially students at the Harbin Institute of Technology, Dalian Maritime University, Chengdu University of Technology, and Northeast Forestry University and their families. Financial support from Dalian Maritime University (Reward for Candidates for Excellent Ph.D Dissertation) and the National Natural Science Foundation of China (No. 21077015 and 20807008) are highly appreciated. Thanks also go to the three anonymous reviewers for their valuable comments and suggestions.

Supporting Information Available Tables SI-1-SI-4, Figures SI-1-SI-4, and additional text. This material is available free of charge via the Internet at http:// pubs.acs.org. FIGURE 3. Comparison between monitoring soil concentrations in 2005 and modeling data in 2004 for (a) r-endosulfan, (b) β-endosulfan, and (c) endosulfan sulfate. Differences between the modeled lowest (preapplication) and highest (postapplication) concentrations were 3 orders of magnitudes for R-endosulfan but only 1 order of magnitude for β-endosulfan and endosulfan sulfate. The larger differences for R-endosulfan reflected its higher volatility, while the much smaller differences for β-endosulfan and endosulfan sulfate were due to their lower volatility and slower degradation in soil. GM concentration of R-endosulfan (6.6 pg/g dw, Table 1) lied well between the modeled lowest (0.42 pg/g dw) and highest (480 pg/g dw) concentrations. Overall, 89% of the 141 measured R-endosulfan concentrations fell within the modeled concentration range. Similarly, GM concentration of β-endosulfan (49 pg/g dw, Table 1) was in the range of the modeled lowest (27 pg/g dw) and highest (130 pg/g dw) concentrations, and GM concentration endosulfan sulfate (47 pg/g dw) was between the modeled lowest (19 pg/g dw) and highest (256 pg/g dw) concentrations. More importantly, 83% measured β-endosulfan and 70% measured endosulfan sulfate concentrations among the 141 samples were found within its modeled concentration range. Although differences between the modeled lowest and highest concentrations were only 1 order of magnitude for β-endosulfan and endosulfan sulfate, the modeling performance was fairly comparable to that for R-endosulfan with 3 order of magnitude differences. Figure 3 also shows that some

Literature Cited (1) Gupta, P. K.; Gupta, R. C. Pharmacology, toxicology and degradation of endosulfan A review. Toxicology 1979, 13, 115– 130. (2) Goebel, H.; Gorbach, S.; Knauf, W.; Rimpau, R. H.; Hutenbach, H. Properties, effects, residues, and analytics of the insecticide endosulfan. Residue Rev. 1982, 83, 1–174. (3) Singh, N. C.; Dasgupta, T. P.; Roberts, E. V.; Mansingh, A. Dynamics of pesticides in tropical conditions. 1. Kinetic studies of volatilization, hydrolysis, and photolysis of dieldrin and alphaand beta-endosulfan. J. Agric. Food Chem. 1991, 39, 575–579. (4) Peterson, S. M.; Batley, G. E. The fate of endosulfan in aquatic ecosystems. Environ. Pollut. 1993, 82, 143–152. (5) Rice, C. P.; Chernyak, S. M.; McConnell, L. L. Henry’s law constants for pesticides measured as a function of temperature and salinity. J. Agric. Food Chem. 1997, 45, 2291–2298. (6) Schmidt, W. F.; Bilboulian, S.; Rice, C. P.; Fettinger, J. C.; McConnell, L. L.; Hapernan, C. J. Thermodynamic, spectroscopic, and computational evidence for the irreversible conversion of beta- to alpha-endosulfan. J. Agric. Food Chem. 2001, 49, 5372–5376. (7) Chandler, G. T.; Scott, G. I. Effects of sediment-bound endosulfan on survival, reproduction and larval settlement of meiobenthic polychaetes and copepods. Environ. Toxicol. Chem. 1991, 10, 375–383. (8) Guerin, T. F.; Kennedy, I. R. Distribution and dissipation of endosulfan and related cyclodienes in sterile aqueous systems: implications for studies on biodegradation. J. Agric. Food Chem. 1992, 40, 2315–2323. (9) Kathpal, T. S.; Singh, A.; Dhankhar, J. S.; Singh, G. Fate of endosulfan in cotton soil under sub-tropical conditions of northern India. Pestic. Sci. 1997, 50, 21–27. VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9283

(10) Antonious, G. F.; Byers, M. E. Fate and movement of endosulfan under field conditions. Environ. Toxicol. Chem. 1997, 16, 644– 649. (11) Ghadiri, H.; Rose, C. W. Degradation of endosulfan in a clay soil from cotton farms of western Queensland. J. Environ. Manage. 2001, 62, 155–169. (12) Wan, M. T.; Kuo, J. N.; Buday, C.; Schroeder, G.; Aggelen, G.; Pasternak, J. Toxicity of R- and β-endosulfan and their formulated and degradation products to Daphnia magna, Hyalella azteca, Oncorhynchus mykiss, Oncorhynchus kisutch, and biological implications in streams. Environ. Toxicol. Chem. 2005, 24, 1146–1154. (13) Rao, D. M. R.; Murty, A. S. Persistence of endosulfan in soils. J. Agric. Food Chem. 1980, 28, 1099–1101. (14) Navarro, S.; Barba, A.; Segura, J. C.; Oliva, J. Disappearance of endosuflan residues from seawater and sediment under laboratory conditions. Pest. Manage. Sci. 2000, 56, 849–854. (15) Sethunathan, N.; Megharaj, M.; Chen, Z.; Singh, N.; Kookana, R. S.; Naidu, R. Persistence of endosulfan and endosulfan sulfate in soil as affected by moisture regime and organic matter. Bull. Environ. Contam. Toxicol. 2002, 68, 725–731. (16) Martens, R. Degradation of endosulfan-8, 9-14C in soil under different conditions. Bull. Environ. Contam. Toxicol. 1977, 17, 438–446. (17) Guerin, T. F. Abiological loss of endosulfan and related chlorinated organic compounds from aqueous systems in the presence and absence of oxygen. Environ. Pollut. 2001, 115, 219–230. (18) Keith, L. H.; Telliard, W. A. Priority pollutants: I-a perspective view. Environ. Sci. Technol. 1979, 13, 416–423. (19) UNEP (United Nations Environment Programme), 2007. Guidance on the Global Monitoring Plan for Persistent Organic Pollutants, Preliminary Version February 2007. Amended in May 2007. Stockholm Convention Secretariat, Geneva, p 147. (20) Lubick, N. Endosulfan’s exit: U. S. EPA pesticide review leads to ban. Science 2010, 328, 1466. (21) Pozo, K.; Harner, T.; Wania, F.; Muir, D. C. G.; Jones, K. C.; Barrie, L. A. Toward a global network for persistent organic pollutants in air: results from the GAPS study. Environ. Sci. Technol. 2006, 40, 4867–4873. (22) Pozo, K.; Harner, T.; Lee, S. C.; Wania, F.; Muir, D. C. G.; Jones, K. C. Seasonally resolved concentrations of persistent organic pollutants in the global atmosphere from the first year of the GAPS study. Environ. Sci. Technol. 2009, 43, 796–803. (23) Shen, L.; Wania, F.; Lei, Y. D.; Teixeira, C.; Muir, D. C. G.; Bidleman, T. F. Atmospheric distirution and long-range transport behavior of organochlorine pesticides in North America. Environ. Sci. Technol. 2005, 39, 409–420. (24) Jia, H.; Li, Y. F.; Wang, D.; Cai, D.; Yang, M.; Ma, J.; Hu, J. Endosulfan in China 1- gridded usage inventories. Environ. Sci. Pollut. Res. 2009, 16, 295–301. (25) Jia, H.; Sun, Y.; Li, Y. F.; Tian, C.; Wang, D.; Yang, M.; Ding, Y.; Ma, J. Endosulfan in China 2-emissions and residues. Environ. Sci. Pollut. Res. 2009, 16, 302–311. (26) Ren, N.; Que, M.; Li, Y. F.; Liu, Y.; Wan, X.; Xu, D.; Sverko, E.; Ma, J. Polychlorinated biphenyls in Chinese surface soils. Environ. Sci. Technol. 2007, 41, 3871–3876. (27) Zhang, Z.; Liu, L.; Li, Y. F.; Wang, D.; Jia, H.; Harner, T.; Sverko, E.; Wan, X.; Xu, D.; Ren, N.; Ma, J.; Pozo, K. Analysis of polychlorinated biphenyls in concurrently sampled Chinese air and surface soil. Environ. Sci. Technol. 2008, 42, 6514–6518. (28) Ren, N.; Sverko, E.; Li, Y. F.; Zhang, Z.; Harner, T.; Wang, D.; Wan, X.; McCarry, B. E. Levels and isomer profiles of dechlorane plus in Chinese air. Environ. Sci. Technol. 2008, 42, 6476–6480. (29) Barr, B. D. Biomonitoring of exposure to pesticides. J. Chem. Health Saf. 2008, 15, 20–29. (30) Sutherland, T. D.; Horne, I.; Lacey, M. J.; Harcourt, R. L.; Russell, R. J.; Oakeshott, J. G. Enrichment of an endosulfan-degrading

9284

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

(31)

(32) (33)

(34)

(35) (36)

(37)

(38) (39)

(40) (41)

(42) (43) (44) (45) (46)

(47) (48) (49)

mixed bacterial culture. Appl. Environ. Microb. 2000, 66, 2822– 2828. Schmidt, W. F.; Hapeman, C. J.; Fettinger, J. C.; Rice, C. P.; Bilboulian, S. Structure and asymmetry in the isomeric conversion of alpha to beta-endosulfan. J. Agric. Food Chem. 1997, 45, 1023–1025. Rice, C. P.; Chernyak, S. M.; Hapeman, C. J.; Biboulan, S. Airwater distribution of the endosulfan isomers. J. Environ. Qual. 1997, 26, 1101–1106. Li, J.; Zhang, G.; Guo, L.; Xu, W.; Li, X.; Lee, C. S. L.; Ding, A.; Wang, T. Organochlorine pesticides in the atmosphere of Guangzhou and Hong Kong: Regional sources and long-range atmospheric transport. Atmos. Environ. 2007, 41, 3889–3903. Zhang, G.; Chakraborty, P.; Li, J.; Sampathkumar, P.; Balasubramanian, T.; Kathiresan, K.; Takahashi, S.; Subramanian, A.; Tanabe, S.; Jones, K. C. Passive atemospheric sampling of organochlorine pesticides, polychlorinated biphenyls, and polybrominated diphenyl ethers in urban, rural, and wetland sites along the coastal length of India. Environ. Sci. Technol. 2008, 42, 8218–8223. Tian, C.; Ren, N.; Ma, J.; Li, Y. F. Impact of γ-HCH residue in the global soil on the Chinese environment. 2. Impact analysis. Chin. Environ. Sci. 2009, 29, 897–901, in Chinese. Wang, X.; Gong, P.; Yao, T.; Jones, K. C. Passive air sampling of organochlorine pesticides, polychlorinated biphenyls, and polybrominated diphenyl ethers across the Tibetan Plateau. Environ. Sci. Technol. 2010, 44, 2988–2993. Hoh, E.; Hites, R. Sources of Toxaphene and other organochlorine pesticides in North America as determined by air measurements and potential source contribution function analyses. Environ. Sci. Technol. 2004, 38, 4187–4194. Hafner, W.; Hites, R. Potential sources of pesticides, PCBs, and PAHs to the atmosphere of the Great Lakes. Environ. Sci. Technol. 2003, 37, 3764–3773. Li, Y. F.; Harner, T.; Liu, L.; Zhang, Z.; Ren, N. Q.; Jia, H.; Ma, J.; Sverko, E. Polychlorinated biphenyls in global air and surface soil: distributions, air-soil exchange, and fractionation effect. Environ. Sci. Technol. 2010, 44, 2784–2790. Herrmann, M. Preliminary risk profile of endosulfan; Umweltbundesamt: Berlin, Germany, 2002. Weber, J.; Halsall, C. J.; Muir, D.; Teixeira, C.; Small, J.; Solomon, K.; Hermanson, M.; Hung, H.; Bidleman, T. Endosulfan, a global pesticide: a review of its fate in the environment and occurrence in the Arctic. Sci. Total Evnviron. 2010, 408, 2966–2984. Beyers, R. A.; Woodham, D. W.; Bowman, M. C. Residues on coastal Bermuda grass, trash and soil treated with endosulfan. J. Econ. Entomol. 1965, 58, 160–161. Shen, L.; Wania, F. Compilation, evaluation, and selection of physical-chemical property data for organochlorine pesticides. J. Chem. Eng. Data 2005, 50, 742–768. Leonard, A. W.; Hyne, R. V.; Lim, R. P.; Leigh, K. A.; Le, J.; Beckett, R. Fate and toxicity of endosulfan in Namoi River water and bottom sediment. J. Environ. Qual. 2001, 30, 750–750. Mukherjee, I.; Gopal, M. Interconversion of stereoisomers of endosulfan on chickpea crop under field conditions. Pestic. Sci. 1994, 40, 103–106. Walse, S. S.; Scott, G. I.; Ferry, J. L. Stereoselective degradation of aqueous endosulfan in modular estuarine mesocosms: formation of endosulfan γ-hydroxycarboxylate. J. Environ. Monit. 2003, 5, 373–379. Wong, F.; Alegria, H. A.; Bidleman, T. F. Organochlorine pesticides in soils of Mexico and the potential for soil-air exchange. Environ. Pollut. 2010, 158, 749–755. Kullman, S. W.; Matsumura, F. Metabolic pathways utilized by phanerochaete chrysosporium for degradation of the cyclodiene pesticide endosulfan. Appl. Environ. Microb. 1996, 62, 593–600. Li, Y. F.; Venkatesh, S.; Li, D. Modeling global emissions and residues of pesticides. Environ. Model. Assess. 2004, 9, 237–243.

ES102791N