Environ Monit Assess (2015) 187:94 DOI 10.1007/s10661-015-4329-5
Spatial distributions and seasonal variations of organochlorine pesticides in water and soil samples in Bolu, Turkey Hatice Karadeniz & Serpil Yenisoy-Karakaş
Received: 8 August 2014 / Accepted: 25 January 2015 # Springer International Publishing Switzerland 2015
Abstract In this study, a total of 75 water samples (38 groundwater and 37 surface water samples) and 54 surface soil samples were collected from the five districts of Bolu, which is located in the Western Black Sea Region of Turkey in the summer season of 2009. In the autumn season, 17 water samples (surface water and groundwater samples) and 17 soil samples were collected within the city center to observe the seasonal changes of organochlorine pesticides (OCPs). Groundwater and surface water samples were extracted using solid phase extraction. Soil samples were extracted ultrasonically. Sixteen OCP compounds in the standard solution were detected by a gas chromatography-electron capture detector (GC-ECD). Therefore, the method validation was performed for those 16 OCP compounds. However, 13 OCP compounds could be observed in the samples. The concentrations of most OCPs were higher in samples collected in the summer than those in the autumn. The most frequently observed pesticides were endosulfan sulfate and 4,4′-dichlorodiphenyltrichloroethane (DDT) in groundwater samples, α-HCH in surface water samples, and endosulfan sulfate in soil samples. The average concentration of endosulfan sulfate was the highest in water and soil samples. Compared to the literature values, the average concentrations in this study Electronic supplementary material The online version of this article (doi:10.1007/s10661-015-4329-5) contains supplementary material, which is available to authorized users. H. Karadeniz : S. Yenisoy-Karakaş (*) Faculty of Science and Art, Department of Chemistry, Abant Izzet Baysal University, 14280 Bolu, Turkey e-mail: [email protected]
were lower values. Spatial distribution of OCPs was evaluated with the aid of contour maps for the five districts of Bolu. Generally, agricultural processes affected the water and soil quality in the region. However, non-agricultural areas were also affected by pesticides. The concentrations of pesticides were below the legal limits of European directives for each pesticide. Keywords Organochlorine pesticides . Groundwater . Surface water . Soil . Seasonal changes . Spatial distribution
Introduction Pesticides are any substance or mixture of substances intended for preventing, destroying, or controlling any pest, and they are divided into many classes according to target organism, mode of action, and chemical structure (Arias-Estevez et al. 2008). Although many of these chemicals are useful for pest and disease control, crop production, and industry, they have had unexpected effects on human health and the environment. In the Stockholm Convention, 90 countries agreed to reduce or eliminate the production, use, and/or release of 12 key POPs. This dirty dozen contains eight pesticides (aldrin, chlordane, endrin, dieldrin, dichlorodiphenyltrichloroethane (DDT), hexachlorobenzene, mirex, toxaphene, and heptachlor) (EPA 2013). Organochlorine pesticides are the most persistent organic micropollutants present in water (Dong et al. 2005). They can contaminate surface water directly as
94
Environ Monit Assess (2015) 187:94
Page 2 of 12
spray drift and/or run-off and drainage from the treated soil. Their accumulative property helps to transport these compounds in many parts of the hydrosphere (Vega et al. 2005). For this reason, pesticide usage is a risk for water quality since they can contaminate groundwater and surface water in agricultural areas. Waters are contaminated through direct discharges by industry, sewage, leaking underground storage tanks, and landfills. However, a major source of pollution comes from nonpoint sources (Calhoun 2005). Therefore, another source is the pollution from the atmospheric deposition. Pesticides can be distributed in the environment by physical processes such as sedimentation, adsorption, and volatilization. After these processes, they can be degraded by chemical and/or biological processes. Chemical processes generally occur in water or in the atmosphere. After application, pesticides may be lost from the soil, water, and air system to the atmosphere in two stages. Pesticide residues in the soil or on the crop dissipate into the atmosphere by diffusion and turbulent mixing (Taylor 1995). These losses also depend on the properties of pesticides. The remaining part in the soil is partitioned between the soil and soil water. In this study, the concentrations of organochlorine pesticides were determined in groundwater and surface water samples and in soil samples collected from five districts of Bolu (the city center, Mudurnu, Göynük, Seben, and Kıbrıscık). Using the levels of OCPs in the region, pollution maps were prepared. Seasonal trends were evaluated.
Experimental Sampling sites Bolu is a city in the Western Black Sea Region of Turkey (Fig. 1). It has nine districts and 511 villages. It is located between 30° 32′ and 32° 36′ east longitudes and 40° 06′ and 41° 01′ north latitudes. The average altitude of the city and the city center are 1000 and 725 m, respectively. Approximately 55 % of the total area is covered with forests and mountains. An area of 17.7 % is used for agriculture. Despite the small agricultural area usage, this city is important in agriculture for Turkey (Erşahin and Şerifeken 2002). The agricultural activities are generally performed around the city center and in the districts such as Göynük, Mudurnu, and Seben. The climate of the city is similar to the Black Sea region, which is generally
warm in summer, cold in winter, and rainy in almost all seasons. More than 100 lakes are present in Bolu. In addition, the total surface area of lakes and streams are 2000 and 260 ha, respectively. The most important agricultural products are wheat, barley, corn, oats, rice as cereals, potatoes, sugar beets, and sugar beet seed as industrial plants. The sunflower plants are produced in some part of the city. There is little fruit and vegetable production. The most common fruit is an apple which is grown in the Seben district. Related to these products is endosulfan, which is one of the most extensively used pesticides in the study area. Endosulfan is applied to the water and soil by mixing before planting. The pesticides in the study area are applied in spring and autumn seasons. Sampling Before sampling, 1-L amber glass bottles were washed with commercial detergent in hot water and rinsed with tap water, deionized water, acetone, and several times with hexane. Then they were dried in an oven at 100 °C and stored until usage. The aluminum foil used for the soil samples was washed several times with acetone and hexane and then dried in an oven. A 2.5 % solution of NaN3 was prepared for water samples to protect the water samples from microbial growth. Water samples were collected into pre-cleaned 1L amber glass bottles. One milliliter of 2.5 % NaN3 was added to each bottle and closed with stopper. Surface soil samples were also collected around the water sampling points with a steel spoon (0.5 cm below the surface). The coordinates of the sampling points were determined by Geographical Positioning System (GPS) (Garmin eTrex Vista® HCx). All soil samples were packaged with aluminum foil and then dried in an oven at 40 °C for 1 day. All the samples were stored at 4 °C until the analysis. A total of 75 water samples (37 groundwater and 38 surface water samples) and 54 surface soil samples were collected from the five districts of Bolu in summer season of 2009 (June). In the autumn season (October), 17 water samples and 17 soil samples were collected only within the city center to observe the seasonal changes of organochlorine pesticides. The sampling points were numbered between 1 and 75. Maps that show groundwater and surface water samples and surface soil samples sampling points are shown in Fig. 1.
Environ Monit Assess (2015) 187:94
Page 3 of 12 94
Fig. 1 Groundwater, surface water, and soil sampling points
Reagents and materials Acetone, acetonitrile, dichloromethane, ethyl acetate, and hexane (GC grade with 99.99 % purity) and petroleum benzene (40–60 °C) with 90 % purity were used. All were purchased from Merck (Germany). The SPE (C18) cartridges were purchased from the company, Biotech (USA). Anhydrous sodium sulfate (99 % purity), florisil (0.150– 0.250 mm), and neutral aluminum oxide (0.063–0.2 mm) were purchased from Merck (USA). Glass microfiber filters (47 mm) were from Whatman (USA). Calibration standard solutions (Mix163) were purchased from Dr. Ehrenstorfer (Germany). Intermediate stock solutions were prepared from the main stock standards with appropriate dilutions with hexane. Gas tight glass syringes (Hamilton, USA) were used in the preparation of standard solutions. Extraction of samples To extract the water samples, a solid phase extraction technique was used. One liter of water sample was filtered from the 47-mm glass fiber filters to separate the particles. Then 500-mL water samples were
measured with volumetric flask. Conditioning of the cartridge was done with 10 mL of 1:1 acetonitrile (ACN) and dichloromethane (DCM) mixture, 5 mL methanol, and 5 mL of deionized water, sequentially. After the conditioning, the sample was passed through a cartridge. The analytes were trapped on the cartridge. To remove the water from the cartridge, it was dried for half an hour. Pesticides were eluted with 9 mL of ACN:DCM mixture and 3 mL hexane from the cartridge. The mixture was evaporated by gentle stream of nitrogen near dryness and the volume was completed to 0.5 mL with hexane. Soil samples were sieved from mesh number 30, with pore width 0.595 mm to remove stones and organic fossils. Then, two 30-g subsamples were weighed. All soil samples were held in a 30-mL mixture of dichloromethane (DCM) and petroleum ether (PE) (1:4) for 24 h. Then, 30 min ultrasonic extraction was applied. Finally, sodium sulfate was added into the filter paper to remove traces of water, and the samples were filtered and washed several times with DCM and PE (1:4) mixture. The washings and the collected filtrate were concentrated under the nitrogen flow and transferred to the 10-mL vials. The samples were stored at 4 °C in the refrigerator
94
Page 4 of 12
until a cleanup procedure. For the cleanup, a column was prepared using 1 g of aluminum oxide, 1 g of florisil, and 1 g of sodium sulfate (top), consecutively. This column was activated with 10 mL of hexane, and the sample was poured into the column. Analytes were eluted with 40 mL of 1:1 ethyl acetate:hexane mixture and collected in a beaker. The analyte mixture was preconcentrated by rotary evaporator until about 1 mL remained. This volume was further evaporated by the gentle stream of nitrogen near dryness and the volume was completed to 0.5 mL with hexane. Analysis For halogenated organic compounds, sensitive and comprehensive instrumental techniques are required. Therefore, the researchers selected an instrument, a Hewlett Packard (HP) 6890 N series gas chromatography equipped with an electron capture detector (Agilent, USA). A 30-m, 0.25-mm id., 0.25-μm film thickness, cross-linked 5 % phenyl methyl siloxane, (HP-5MS), capillary column, and Agilent 7683B series automatic injector (Agilent, USA) were used. The operating parameters of the GC-ECD instrument are given in a supplementary document (Table A.1). Method and validation studies Linearity, method detection limit (MDL), and limit of quantification (LOQ) Pesticide working solutions were prepared from the stock solution in the range of 0.05 ng mL−1 and 50 ng L−1. A sample chromatogram of 25 ng L−1 solution is given in a supplementary document (Fig. A.1). Calibration curves were obtained from the eight calibration solutions. Linearity was evaluated by using r2 values. The method detection limit (MDL) and limit of quantification (LOQ) were calculated for GC-ECD by analyzing 2.5 ng L−1 standards ten times. MDL values were three times standard deviation of ten replicates, and LOQ values were ten times standard deviation of ten replicates. The results were summarized in a supplementary document (Table A.2). Recovery and repeatability of method The SPE procedure used in this work was taken from the study of Binici et al. (2014). This procedure was first
Environ Monit Assess (2015) 187:94
checked to obtain reasonable recoveries for deionized water, groundwater, and surface water samples. For that purpose, 250 mL deionized water was poured into a 500-mL volumetric flask. Then the 500 μL of 15 ng mL−1 pesticide mixture, which was approximately the concentration in the middle part of the linear range of the calibration curve, was added and diluted to 500 mL with deionized water, surface water, and groundwater. The six prepared parallel water samples were extracted on the same day. These procedures were repeated with three types of water for 5 days. The recoveries of each pesticide in each water sample were evaluated, and the results were given in a supplementary document (Fig. A.2). The reasonable recovery values (greater than 70 %) were obtained for all of the compounds except for aldrin and dieldrin. The recovery values of aldrin and dieldrin were greater than 50 %. The ultrasonic extraction procedure used in this study was based on a previous study (Öz 2009). Six 30-g soil samples were weighed and placed in six different beakers, and 500 μL of 15 ng mL−1 pesticide mixture was added to each sample. Then, all prepared soil samples were held in a 30-mL mixture of dichloromethane (DCM) and petroleum ether (PE) (1:4) for 24 h. Then, 30 min ultrasonic extraction was applied. Finally, sodium sulfate was added into the filter paper to remove traces of water, and the samples were filtered and washed several times with DCM and PE (1:4) mixture. The washings and the collected filtrate were concentrated under the nitrogen flow and transferred to the 10-mL vials. The spiking was performed at the middle fortification level of the calibration plot. The recoveries of each pesticide in each soil sample were evaluated. The recoveries of heptachlor epoxide, aldrin, 4,4′ DDE, and endosulfan II ranged from 40 to 50 %. Due to the low recoveries, the correction was applied for the results. The related data were given in a supplementary document (Table A.3).
Results and discussion Statistical treatment of data The statistical evaluations of OCP concentrations in summer season in two types of water and soil samples were completed, and the results are shown in Table 1. Endosulfan is widely used in the region and has very low water solubility. It is applied to the soil before or after watering the plants. This could explain the higher
4
3
7
4
6
DDE
Endrin
Endosulfan II
DDD 1.22 0.68
Endosulfan Sulfate 33 1.76±1.87
DDT
0.74
1.42
0.29
0.20
0.81
0.27
0.25
0.72
0.58
0.44
– –
– – – –
0.3–2.34
28 1.1±0.48
1.07
1.11
0.79
–
– – 5 1.34±1.12
0.68 –
– –
0.53
0.62
21 0.69±0.05
16 0.54±0.03
32 0.96±1.05
0.84–11.15 29 1.4±0.62
0.16–0.77
0.19–0.24
0.35–9.10
0.24–0.29
0.23–0.27
0.54–1.27
0.52–1.02
0.36–1.09
1.01
1.30
–
–
1.00
–
–
0.69
0.53
0.73
–
21 0.98±3.9
16 0.21±0.39 0.47–2.52 18 0.29±0.78
0.9–3.29
–
17 0.34±0.95
–
0.051
0.07
0.062
0.1
–
–
0.19
0.057±0.0074 0.06 –
–
0.029 0.025±0.0054 0.023
0.041±0.032
18 0.62±1.31
4
9
0.074
0.11
0.07
0.093
–
–
0.12
0.34
0.024
0.033
0.015–3.21
0.025–18
0.0022–1.59
0.015–4
–
–
0.0044–5.54
0.049–0.63
0.02–0.037
0.016–0.078
Art. mean±SD Median Geo. mean Range
0.39–3.03 –
–
–
0.620.83
0.5–0.61
0.36–5.77 3
N
Soil sample (Ns =54)
Ns number of samples, N number of samples observed, Art. mean arithmetic mean, SD standard deviation, Geo. mean geometric mean
33 0.88±0.55
0.23
0.20
0.50
0.29
0.15±0.24
0.21±0.02
0.88±3.23
0.14±0.03
0.24
0.69
22 0.73±0.15
δ-HCH
Dieldrin
0.15±0.01
0.53
12 0.59±0.14
0.39
19 0.47±0.19
α-HCH
γ-HCH
Art. mean±SD Median Geo. mean Range
N
N
Art. mean± Median Geo. Mean Range SD
Surface water sample (Ns =37)
Groundwater sample (Ns =38)
Table 1 Statistical evaluation of surface water, groundwater, and soil samples in the summer season (ng L−1 for water samples, ng g−1 for soil samples)
Environ Monit Assess (2015) 187:94 Page 5 of 12 94
94
Page 6 of 12
concentrations of endosulfan and its derivatives in groundwater samples than in surface water samples. Endosulfan sulfate had both the highest concentration and was the most frequently observed OCP in the soil samples. Moreover, the percentage observation of αHCH in surface water was higher than that in groundwater. 4,4′-DDT, and its derivatives were also observed in surface water, groundwater, and soil samples. Heptachlor, heptachlor epoxide, and methoxychlor were not observed in any of the samples.
Seasonal changes of the OCPs Figure 2a shows the seasonal difference for each observed pesticide in water samples (combination of surface water and groundwater samples) collected only in the city center. Average value (at least three values) of the concentrations of each pesticide (without using the substitution for the not detected concentrations) was used. Fig. 2 a Seasonal differences in average concentrations for each observed pesticide in water samples taken from the city center (N=17). b Seasonal differences in average concentrations for each observed pesticide in soil samples taken from the city center (N=17)
Environ Monit Assess (2015) 187:94
In the summer sampling, 10 pesticides were observed in the surface water and groundwater samples. However, in the autumn season, only four pesticides were observed in surface water and groundwater samples. Endosulfan sulfate had the highest arithmetic mean in both seasons because of the extensive usage of endosulfan in the past, since over time; the endosulfan isomer mixture is converted into endosulfan sulfate (Walse et al. 2003). Seasonal changes were also observed for the soil samples. For the soil samples, the concentrations in summer samples were higher than the autumn season, except for 4,4′-DDT (Fig. 2b). The concentrations were higher in the agricultural areas in both seasons for all samples.
Spatial distributions In the five districts of Bolu, the spatial distributions of the each target analytes were presented by the contour maps by using summer season data. The missing data
Environ Monit Assess (2015) 187:94
was filled with half of the detection limit values for hexachlorocyclohexane isomers, aldrin, dieldrin, and endrin to draw the maps. For other pesticides, the detection limit values were used to fill the missing values. With the aid of these maps, the relationship between the observed concentrations and the characteristics of sampling points were evaluated and the degradation processes were discussed. Endosulfan isomers and their derivatives Endosulfan has still been used in agricultural activities in Turkey, even though the use of endosulfan is under discussion since 2007 in Turkey. Commercial endosulfan contains endosulfan I and endosulfan II isomers in ratios from 2:1 to 7:3, depending on the technical mixture (Herrmann 2002). It is applied directly to the soil or plant. For this reason, endosulfan II concentrations were higher than the other sampling points, especially in the agricultural areas of Mudurnu (Fig. 3a). Moreover, the two isomers can be converted to endosulfan sulfate by oxidation in biotic and abiotic degradation (Walse et al. 2003). Both endosulfan II and endosulfan sulfate are scavenged from the atmosphere easily by precipitation, due to their higher solubility than endosulfan I, and as such, they can be carried to the surface water by the atmospheric longrange transport (Chan et al. 1994, 2003). Figure 3b, c shows the spatial distributions of endosulfan sulfate concentrations in surface water and groundwater samples. When the two maps were compared, it was seen that concentrations in surface waters might have been affected by atmospheric long-range transport of endosulfan sulfate because the concentrations of surface water samples were higher than the concentrations in groundwater samples. As a result, the concentrations of the surface water and groundwater and soil compartments of Bolu were affected by the current and previous use of commercial endosulfan mixture. Aldrin and dieldrin Under most environmental conditions, aldrin is largely converted to dieldrin and dieldrin is significantly more persistent. Most environmental releases of aldrin and dieldrin are directly to the soil. They have low water solubility and a tendency to bind tightly to the soils. Therefore, both compounds migrate downward very slowly through soils or into surface water or groundwater. Most surface water aldrin/dieldrin has been
Page 7 of 12 94
attributed to particulates by surface runoff. For this reason, it is possible that significant volatilization of aldrin/dieldrin might occur, with subsequent atmospheric photo degradation or rainfall (USEPA 2003). According to the local authorities, aldrin has not been in use after the ban in the 1980s in Turkey. Because of the past usage of aldrin, it was observed only in one sample. However, dieldrin was observed in agricultural areas. Therefore, the observed presence of dieldrin can be explained by the past usage of aldrin.
Hexachlorobenzene isomers (HCHs) HCHs were one of the most widely used organochlorine pesticides. They have a relatively high volatility, and they can be found in many environmental samples, including pristine locations such as the Arctics. They have been used as an insecticide in wheat production (Willet et al. 1998). In technical grade HCH, the α- to γ-isomer ratio is between 3 and 7. Global sources of HCHs and the transfer of HCHs away from the source can be evaluated using the α- to γ-isomer ratios. For example, the current usage of lindane (γ-HCH) shows this ratio near or less than unity (Hargrave et al. 1988). Therefore, higher α to γ ratios in water samples can be explained by differential gas exchange across air-water interfaces, washout by precipitation or volatilization-adsorption variations between the isomers (Iwata et al. 1993). The α- to γ-isomer ratios in surface water and groundwater samples are presented (Fig. A.3 and Fig A.4) in the supplementary document. The α- to γ-isomer ratios in Bolu ranged between 0.65 and 5.45 in surface water samples. Besides the high wheat production areas, high α- to γ-isomer ratios can also be seen in non-agricultural regions, such as Abant Lake (32) and Seben Lake (52). In particular, Seben Lake is a new artificial lake, and this lake was filled with snow water in 2008. Long-range transport can be effective in this lake (Fig. 4a). For groundwater samples, α- to γ-isomer ratios ranged between 0.44 and 0.77. In some sampling points, all isomers were observed. The major isomer of surface water was α-HCH. The higher existence of α-HCH in surface water indicated the photo degradation of lindane (Willet et al. 1998; Rissato et al. 2006). In groundwater samples, the major isomer was δ-HCH. The high δ-HCH content in groundwaters in this study indicated that the transformation of lindane was dominant (Buser and Muller 1995).
94
Page 8 of 12
Environ Monit Assess (2015) 187:94
Environ Monit Assess (2015) 187:94
Page 9 of 12 94
Fig. 4 a The spatial distribution of α-HCH in surface water samples. b The spatial distribution of α-HCH in groundwater samples
Endrin Endrin is nearly insoluble in water. The detection of endrin in groundwater and surface water is rare. 3 a The spatial distribution of endosulfan II in soil samples. b Fig. The spatial distribution of endosulfan sulfate in surface water samples. c The spatial distribution of endosulfan sulfate in groundwater samples
Although endrin has low volatility, it was found in higher concentrations in the air during its manufacture. It was never found in urban areas, where it was rarely used (ATSDR 1996). Endrin was used in the production of sugar cane, grain crops, sugar beets, and rice as an insecticide. These plants are produced in the sampling region of this study.
94
Page 10 of 12
Endrin was found in some of the water samples in Bolu. The previous use of endrin could be the cause of the presence in water samples. Endrin concentrations in surface water were high in the agricultural region in Mudurnu (sampling points 22, 24, and 25). It was also observed in non-agricultural regions, such as sampling point 53. Endrin can also be seen in groundwater samples. The Seben region is a grain-producing region. The groundwater sample
Environ Monit Assess (2015) 187:94
taken from sampling point 59 in an agricultural area (Seben) had the highest endrin concentration in this study. Endrin degradation in soil did not significantly appear. Approximately 41 % of the applied endrin can remain in a field up to 14 years after application (Nash and Woolson 1967). Endrin was not detected in any soil samples. It was thought that it was carried from the atmosphere to bodies of water (Fig. 5a, b).
Fig. 5 a The spatial distribution of endrin in surface water samples. b The spatial distribution of endrin in groundwater samples
Environ Monit Assess (2015) 187:94
Page 11 of 12 94
Fig. 6 The spatial distribution of 4,4′-DDT in surface water samples
4,4′-DDT, 4,4′-DDE, 4,4′-DDD 4,4′-DDT can contaminate to surface waters, primarily by runoff, atmospheric transport, and drift or by direct application. 4,4′-DDT is persistent in soil and does not move easily to groundwater (EPA 2000). The ratios of 4,4′-DDT/ (4,4′-DDD+4,4′-DDE) and 4,4′-DDE/4,4′-DDD have been used to indicate the degree of DDT composition and to find a sign for fresh input of DDTs. If the ratio is higher than unity, the fresh usage of these pesticides can be possible (Lee et al. 2001). The 4,4′-DDT/ (4,4′-DDD+ 4,4′-DDE) ratios ranged from 3.9 to 61.1 in surface water samples and from 0.9 to 40.6 in groundwater samples in the summer season in this study. In addition to the persistency of 4,4′-DDT, it was used in very large amounts in this area; the illegal use of 4,4′-DDT is also possible in Turkey (Acara 2004). The results of this study can reveal possible fresh input of DDTs in this region, as was seen in the surface water samples (Fig. 6). In the current study, 4,4′-DDE was not observed in soil samples, although 4,4′DDT and 4,4′-DDD were seen in some soil samples.
Conclusion In the current study, the concentrations of OCPs in surface water, groundwater, and soil samples were
determined. The sampling was performed in the city center, Seben, Kıbrıscık, Göynük, and Mudurnu districts of Bolu. To observe the seasonal changes, the sampling was repeated only in the city center of Bolu. Solid phase extraction and ultrasonic extraction techniques were used for the extraction of compounds from water and soil samples, respectively. The samples were analyzed by GC-ECD. The concentrations of 13 OCPs in water samples and 11 OCPs in soil samples were determined. The spiked samples were used for the recovery tests. The recoveries ranged from 72 to 108 % for water samples, except for aldrin and dieldrin, and from 40 to 131 % in soil samples. They were generally within the limit of acceptance (70–130 %) according to the EPA, except for aldrin, heptachlor epoxide, and 4, 4′ DDE in the soil samples. Seasonal variations can be observed in the city center. The concentrations of endosulfan sulfate in water samples and 4, 4′-DDE in soil samples were higher in summer than in autumn. The contour maps showed that endosulfan was the currently used pesticide in Bolu and it was the only pesticide determined in the five districts. Although the use of 4, 4′-DDT was banned in 1985 in Turkey, it was also generally observed in Bolu. This demonstrated the possibility of current illegal use of this pesticide. Generally, the districts of Bolu were affected by pesticides used for agricultural applications. The
94
Page 12 of 12
concentrations were lower than 100 ng L−1, which is the limit of the European Union for each pesticide in water resources. The atmospheric long-range transport of pesticides, precipitation, and transformation by photo degradation is the effective mean to contaminate surface waters, particularly in Bolu. Acknowledgments The authors would like to thank Prof. Dr. Okan Külköylüoğlu and Prof. Dr. Muzaffer Dügel for their technical support and Elif Özlü and İlker Köprü for their assistance with the sampling. The authors are grateful to Uğur Saklangiç, Emre Aslan, and Onur Çar for their assistance with the sample preparation and analysis.
References Acara, A. (2004). Türkiye’de Kalıcı Organik Kirleticiler (POP’s) Envanter Özet Raporu. Resource Document. http://www. ttgv.org.tr/content/docs/envanter-ozet-raporu.doc. Accessed 4 Oct 2013. Arias-Estevez, M., Lopez-Periago, E., Martinez-Carballo, E., Simal-Gandara, J., Mejuto, J. C., & Garcia-Rio, L. (2008). The mobility and degradation of pesticides in soils and the pollution of groundwater resources. Agriculture, Ecosystem & Environment, 123, 247–260. ATSDR (Agency for Toxic Substances and Disease Registry). (1996). Toxicological Profile for endrin and endrin aldehyde (update). Atlanta: Research Triangle Institute. Binici, B., Yenisoy-Karakaş, S., Bilsel, M., & Durmaz-Hilmioğlu, N. (2014). Sources of polycyclic hydrocarbons and pesticides in soluble fraction of deposition samples in Kocaeli, Turkey. Environmental Science and Pollution Research, 21, 2907– 2917. Buser, H., & Muller, M. D. (1995). Isomer and enantioselective degradation of hexachlorocyclohexane isomers in sewage sludge under anaerobic conditions. Environmental Science and Technology, 29, 664–672. Calhoun, Y. (2005). Water pollution. New York: Chelsea House Publisher. Chan, C., Bruce, G., & Harrison, B. (1994). Wet deposition of organochlorine pesticides and polychlorinated biphenyls to the Great Lakes. Journal of Great Lakes Research, 20, 546–560. Chan, C., Williams, D., Neilson, M., Harrison, B., & Archer, M. (2003). Spatial and temporal trends in the concentrations of selected organochlorine pesticides (OCs) and polynuclear aromatic hydrocarbons (PAHs) in Great Lakes Basin precipitation, 1986 to 1999. Atmospheric Environment, 40, 1563–1578.
Environ Monit Assess (2015) 187:94 Dong, C., Zeng, Z., & Yang, M. (2005). Determination of organochlorine pesticides and theirderivations in water after HSSPME using polymethylphenylvinylsiloxane-coated fiber by GC-ECD. Water Research, 39, 4204–4210. Erşahin, G., Şerifeken, İ. (2002). Bolu ili Raporu, Devlet Planlama Teşkilatı Müsteşarlığı. DPT:2651, Ankara, Turkey. EPA (2000) Revised Draft Lake Michigan Lakewide Management Plan for Toxic Pollutants. Clean water act. Canada. EPA (2013). www.epa.gov/oia/toxics/pop.html. Accessed 4 Oct 2013. Hargrave, B. T., Vass, W. P., Erickson, P. E., Fowler, B. R. (1988). Atmospheric transport of organochlorines to the Arctic Ocean. Tellus, 480–493. Herrmann M. (2002). Preliminary risk profile of endosulfan. Berlin Germany. Iwata, H., Tanabe, S., Sakal, N., & Tatsukawa, R. (1993). Distribution of persistent organochlorines in Oceanic air and surface seawater and the role of ocean on their global transport and fate. Environmental Science and Technology, 27, 1080–1098. Lee, K. T., Tanabe, S., & Koh, C. H. (2001). Distribution of organochlorine pesticides in sediments from Kyeonggi Bay and nearby areas, Korea. Environmental Pollution, 114, 207– 213. Nash, R. G., & Woolson, E. A. (1967). Distribution of chlorinated insecticides in cultivated soil. Soil Science Society of America Journal, 32, 525–527. Öz, M. (2009). Characterization of PCBs, PAHs, and Organochlorine Pesticides in the Atmosphere of Bolu and Source Apportionment: Gas to Particle, Soil to Particle, Soil to Gas Exchanges. Bolu, Turkey. Rissato, S. R., Galhiane, M. S., Ximenes, V. F., de Andrade, R. M., Talamoni, J. L., Libanio, M., de Almeida, M. V., Apon, B. M., & Cavalari, A. A. (2006). Organochlorine pesticides and polychlorinated biphenyls in soil and water samples in the Northeastern part of Sao Paulo State, Brazil. Chemosphere, 65, 1949–1958. Taylor, A. W. (1995). Environmental behaviour of agrochemicals. Chichester: John Wiley & Sons. USEPA (2003). Health Effect support Document for Aldrin/ Dieldrin. Washington, D.C. Vega, A. B., Frenich, A. G., & Vidal, J. L. M. (2005). Monitoring of pesticides in agricultural water and soil samples from Andalusia by liquid chromatography coupled to mass spectrometry. Analytica Chimica Acta, 538, 117–127. Walse, S. S., Scott, G. I., & Ferry, J. L. (2003). Stereoselective degradation of aqueous endosulfan in modular estuarine mesocosms: formation of endosulfan γ-hydroxycarboxylate. Journal of Environmental Monitoring, 5, 373–379. Willet, K. L., Ulrich, E. M., & Hites, R. A. (1998). Differential toxicity and environmental fates of hexachlorocyclohexane isomers. Environmental Science and Technology, 32, 2197– 2207.
Spatial distributions and seasonal variations of organochlorine pesticides in water and soil samples in Bolu, Turkey Hatice Karadeniz & Serpil Yenisoy-Karakaş
Received: 8 August 2014 / Accepted: 25 January 2015 # Springer International Publishing Switzerland 2015
Abstract In this study, a total of 75 water samples (38 groundwater and 37 surface water samples) and 54 surface soil samples were collected from the five districts of Bolu, which is located in the Western Black Sea Region of Turkey in the summer season of 2009. In the autumn season, 17 water samples (surface water and groundwater samples) and 17 soil samples were collected within the city center to observe the seasonal changes of organochlorine pesticides (OCPs). Groundwater and surface water samples were extracted using solid phase extraction. Soil samples were extracted ultrasonically. Sixteen OCP compounds in the standard solution were detected by a gas chromatography-electron capture detector (GC-ECD). Therefore, the method validation was performed for those 16 OCP compounds. However, 13 OCP compounds could be observed in the samples. The concentrations of most OCPs were higher in samples collected in the summer than those in the autumn. The most frequently observed pesticides were endosulfan sulfate and 4,4′-dichlorodiphenyltrichloroethane (DDT) in groundwater samples, α-HCH in surface water samples, and endosulfan sulfate in soil samples. The average concentration of endosulfan sulfate was the highest in water and soil samples. Compared to the literature values, the average concentrations in this study Electronic supplementary material The online version of this article (doi:10.1007/s10661-015-4329-5) contains supplementary material, which is available to authorized users. H. Karadeniz : S. Yenisoy-Karakaş (*) Faculty of Science and Art, Department of Chemistry, Abant Izzet Baysal University, 14280 Bolu, Turkey e-mail: [email protected]
were lower values. Spatial distribution of OCPs was evaluated with the aid of contour maps for the five districts of Bolu. Generally, agricultural processes affected the water and soil quality in the region. However, non-agricultural areas were also affected by pesticides. The concentrations of pesticides were below the legal limits of European directives for each pesticide. Keywords Organochlorine pesticides . Groundwater . Surface water . Soil . Seasonal changes . Spatial distribution
Introduction Pesticides are any substance or mixture of substances intended for preventing, destroying, or controlling any pest, and they are divided into many classes according to target organism, mode of action, and chemical structure (Arias-Estevez et al. 2008). Although many of these chemicals are useful for pest and disease control, crop production, and industry, they have had unexpected effects on human health and the environment. In the Stockholm Convention, 90 countries agreed to reduce or eliminate the production, use, and/or release of 12 key POPs. This dirty dozen contains eight pesticides (aldrin, chlordane, endrin, dieldrin, dichlorodiphenyltrichloroethane (DDT), hexachlorobenzene, mirex, toxaphene, and heptachlor) (EPA 2013). Organochlorine pesticides are the most persistent organic micropollutants present in water (Dong et al. 2005). They can contaminate surface water directly as
94
Environ Monit Assess (2015) 187:94
Page 2 of 12
spray drift and/or run-off and drainage from the treated soil. Their accumulative property helps to transport these compounds in many parts of the hydrosphere (Vega et al. 2005). For this reason, pesticide usage is a risk for water quality since they can contaminate groundwater and surface water in agricultural areas. Waters are contaminated through direct discharges by industry, sewage, leaking underground storage tanks, and landfills. However, a major source of pollution comes from nonpoint sources (Calhoun 2005). Therefore, another source is the pollution from the atmospheric deposition. Pesticides can be distributed in the environment by physical processes such as sedimentation, adsorption, and volatilization. After these processes, they can be degraded by chemical and/or biological processes. Chemical processes generally occur in water or in the atmosphere. After application, pesticides may be lost from the soil, water, and air system to the atmosphere in two stages. Pesticide residues in the soil or on the crop dissipate into the atmosphere by diffusion and turbulent mixing (Taylor 1995). These losses also depend on the properties of pesticides. The remaining part in the soil is partitioned between the soil and soil water. In this study, the concentrations of organochlorine pesticides were determined in groundwater and surface water samples and in soil samples collected from five districts of Bolu (the city center, Mudurnu, Göynük, Seben, and Kıbrıscık). Using the levels of OCPs in the region, pollution maps were prepared. Seasonal trends were evaluated.
Experimental Sampling sites Bolu is a city in the Western Black Sea Region of Turkey (Fig. 1). It has nine districts and 511 villages. It is located between 30° 32′ and 32° 36′ east longitudes and 40° 06′ and 41° 01′ north latitudes. The average altitude of the city and the city center are 1000 and 725 m, respectively. Approximately 55 % of the total area is covered with forests and mountains. An area of 17.7 % is used for agriculture. Despite the small agricultural area usage, this city is important in agriculture for Turkey (Erşahin and Şerifeken 2002). The agricultural activities are generally performed around the city center and in the districts such as Göynük, Mudurnu, and Seben. The climate of the city is similar to the Black Sea region, which is generally
warm in summer, cold in winter, and rainy in almost all seasons. More than 100 lakes are present in Bolu. In addition, the total surface area of lakes and streams are 2000 and 260 ha, respectively. The most important agricultural products are wheat, barley, corn, oats, rice as cereals, potatoes, sugar beets, and sugar beet seed as industrial plants. The sunflower plants are produced in some part of the city. There is little fruit and vegetable production. The most common fruit is an apple which is grown in the Seben district. Related to these products is endosulfan, which is one of the most extensively used pesticides in the study area. Endosulfan is applied to the water and soil by mixing before planting. The pesticides in the study area are applied in spring and autumn seasons. Sampling Before sampling, 1-L amber glass bottles were washed with commercial detergent in hot water and rinsed with tap water, deionized water, acetone, and several times with hexane. Then they were dried in an oven at 100 °C and stored until usage. The aluminum foil used for the soil samples was washed several times with acetone and hexane and then dried in an oven. A 2.5 % solution of NaN3 was prepared for water samples to protect the water samples from microbial growth. Water samples were collected into pre-cleaned 1L amber glass bottles. One milliliter of 2.5 % NaN3 was added to each bottle and closed with stopper. Surface soil samples were also collected around the water sampling points with a steel spoon (0.5 cm below the surface). The coordinates of the sampling points were determined by Geographical Positioning System (GPS) (Garmin eTrex Vista® HCx). All soil samples were packaged with aluminum foil and then dried in an oven at 40 °C for 1 day. All the samples were stored at 4 °C until the analysis. A total of 75 water samples (37 groundwater and 38 surface water samples) and 54 surface soil samples were collected from the five districts of Bolu in summer season of 2009 (June). In the autumn season (October), 17 water samples and 17 soil samples were collected only within the city center to observe the seasonal changes of organochlorine pesticides. The sampling points were numbered between 1 and 75. Maps that show groundwater and surface water samples and surface soil samples sampling points are shown in Fig. 1.
Environ Monit Assess (2015) 187:94
Page 3 of 12 94
Fig. 1 Groundwater, surface water, and soil sampling points
Reagents and materials Acetone, acetonitrile, dichloromethane, ethyl acetate, and hexane (GC grade with 99.99 % purity) and petroleum benzene (40–60 °C) with 90 % purity were used. All were purchased from Merck (Germany). The SPE (C18) cartridges were purchased from the company, Biotech (USA). Anhydrous sodium sulfate (99 % purity), florisil (0.150– 0.250 mm), and neutral aluminum oxide (0.063–0.2 mm) were purchased from Merck (USA). Glass microfiber filters (47 mm) were from Whatman (USA). Calibration standard solutions (Mix163) were purchased from Dr. Ehrenstorfer (Germany). Intermediate stock solutions were prepared from the main stock standards with appropriate dilutions with hexane. Gas tight glass syringes (Hamilton, USA) were used in the preparation of standard solutions. Extraction of samples To extract the water samples, a solid phase extraction technique was used. One liter of water sample was filtered from the 47-mm glass fiber filters to separate the particles. Then 500-mL water samples were
measured with volumetric flask. Conditioning of the cartridge was done with 10 mL of 1:1 acetonitrile (ACN) and dichloromethane (DCM) mixture, 5 mL methanol, and 5 mL of deionized water, sequentially. After the conditioning, the sample was passed through a cartridge. The analytes were trapped on the cartridge. To remove the water from the cartridge, it was dried for half an hour. Pesticides were eluted with 9 mL of ACN:DCM mixture and 3 mL hexane from the cartridge. The mixture was evaporated by gentle stream of nitrogen near dryness and the volume was completed to 0.5 mL with hexane. Soil samples were sieved from mesh number 30, with pore width 0.595 mm to remove stones and organic fossils. Then, two 30-g subsamples were weighed. All soil samples were held in a 30-mL mixture of dichloromethane (DCM) and petroleum ether (PE) (1:4) for 24 h. Then, 30 min ultrasonic extraction was applied. Finally, sodium sulfate was added into the filter paper to remove traces of water, and the samples were filtered and washed several times with DCM and PE (1:4) mixture. The washings and the collected filtrate were concentrated under the nitrogen flow and transferred to the 10-mL vials. The samples were stored at 4 °C in the refrigerator
94
Page 4 of 12
until a cleanup procedure. For the cleanup, a column was prepared using 1 g of aluminum oxide, 1 g of florisil, and 1 g of sodium sulfate (top), consecutively. This column was activated with 10 mL of hexane, and the sample was poured into the column. Analytes were eluted with 40 mL of 1:1 ethyl acetate:hexane mixture and collected in a beaker. The analyte mixture was preconcentrated by rotary evaporator until about 1 mL remained. This volume was further evaporated by the gentle stream of nitrogen near dryness and the volume was completed to 0.5 mL with hexane. Analysis For halogenated organic compounds, sensitive and comprehensive instrumental techniques are required. Therefore, the researchers selected an instrument, a Hewlett Packard (HP) 6890 N series gas chromatography equipped with an electron capture detector (Agilent, USA). A 30-m, 0.25-mm id., 0.25-μm film thickness, cross-linked 5 % phenyl methyl siloxane, (HP-5MS), capillary column, and Agilent 7683B series automatic injector (Agilent, USA) were used. The operating parameters of the GC-ECD instrument are given in a supplementary document (Table A.1). Method and validation studies Linearity, method detection limit (MDL), and limit of quantification (LOQ) Pesticide working solutions were prepared from the stock solution in the range of 0.05 ng mL−1 and 50 ng L−1. A sample chromatogram of 25 ng L−1 solution is given in a supplementary document (Fig. A.1). Calibration curves were obtained from the eight calibration solutions. Linearity was evaluated by using r2 values. The method detection limit (MDL) and limit of quantification (LOQ) were calculated for GC-ECD by analyzing 2.5 ng L−1 standards ten times. MDL values were three times standard deviation of ten replicates, and LOQ values were ten times standard deviation of ten replicates. The results were summarized in a supplementary document (Table A.2). Recovery and repeatability of method The SPE procedure used in this work was taken from the study of Binici et al. (2014). This procedure was first
Environ Monit Assess (2015) 187:94
checked to obtain reasonable recoveries for deionized water, groundwater, and surface water samples. For that purpose, 250 mL deionized water was poured into a 500-mL volumetric flask. Then the 500 μL of 15 ng mL−1 pesticide mixture, which was approximately the concentration in the middle part of the linear range of the calibration curve, was added and diluted to 500 mL with deionized water, surface water, and groundwater. The six prepared parallel water samples were extracted on the same day. These procedures were repeated with three types of water for 5 days. The recoveries of each pesticide in each water sample were evaluated, and the results were given in a supplementary document (Fig. A.2). The reasonable recovery values (greater than 70 %) were obtained for all of the compounds except for aldrin and dieldrin. The recovery values of aldrin and dieldrin were greater than 50 %. The ultrasonic extraction procedure used in this study was based on a previous study (Öz 2009). Six 30-g soil samples were weighed and placed in six different beakers, and 500 μL of 15 ng mL−1 pesticide mixture was added to each sample. Then, all prepared soil samples were held in a 30-mL mixture of dichloromethane (DCM) and petroleum ether (PE) (1:4) for 24 h. Then, 30 min ultrasonic extraction was applied. Finally, sodium sulfate was added into the filter paper to remove traces of water, and the samples were filtered and washed several times with DCM and PE (1:4) mixture. The washings and the collected filtrate were concentrated under the nitrogen flow and transferred to the 10-mL vials. The spiking was performed at the middle fortification level of the calibration plot. The recoveries of each pesticide in each soil sample were evaluated. The recoveries of heptachlor epoxide, aldrin, 4,4′ DDE, and endosulfan II ranged from 40 to 50 %. Due to the low recoveries, the correction was applied for the results. The related data were given in a supplementary document (Table A.3).
Results and discussion Statistical treatment of data The statistical evaluations of OCP concentrations in summer season in two types of water and soil samples were completed, and the results are shown in Table 1. Endosulfan is widely used in the region and has very low water solubility. It is applied to the soil before or after watering the plants. This could explain the higher
4
3
7
4
6
DDE
Endrin
Endosulfan II
DDD 1.22 0.68
Endosulfan Sulfate 33 1.76±1.87
DDT
0.74
1.42
0.29
0.20
0.81
0.27
0.25
0.72
0.58
0.44
– –
– – – –
0.3–2.34
28 1.1±0.48
1.07
1.11
0.79
–
– – 5 1.34±1.12
0.68 –
– –
0.53
0.62
21 0.69±0.05
16 0.54±0.03
32 0.96±1.05
0.84–11.15 29 1.4±0.62
0.16–0.77
0.19–0.24
0.35–9.10
0.24–0.29
0.23–0.27
0.54–1.27
0.52–1.02
0.36–1.09
1.01
1.30
–
–
1.00
–
–
0.69
0.53
0.73
–
21 0.98±3.9
16 0.21±0.39 0.47–2.52 18 0.29±0.78
0.9–3.29
–
17 0.34±0.95
–
0.051
0.07
0.062
0.1
–
–
0.19
0.057±0.0074 0.06 –
–
0.029 0.025±0.0054 0.023
0.041±0.032
18 0.62±1.31
4
9
0.074
0.11
0.07
0.093
–
–
0.12
0.34
0.024
0.033
0.015–3.21
0.025–18
0.0022–1.59
0.015–4
–
–
0.0044–5.54
0.049–0.63
0.02–0.037
0.016–0.078
Art. mean±SD Median Geo. mean Range
0.39–3.03 –
–
–
0.620.83
0.5–0.61
0.36–5.77 3
N
Soil sample (Ns =54)
Ns number of samples, N number of samples observed, Art. mean arithmetic mean, SD standard deviation, Geo. mean geometric mean
33 0.88±0.55
0.23
0.20
0.50
0.29
0.15±0.24
0.21±0.02
0.88±3.23
0.14±0.03
0.24
0.69
22 0.73±0.15
δ-HCH
Dieldrin
0.15±0.01
0.53
12 0.59±0.14
0.39
19 0.47±0.19
α-HCH
γ-HCH
Art. mean±SD Median Geo. mean Range
N
N
Art. mean± Median Geo. Mean Range SD
Surface water sample (Ns =37)
Groundwater sample (Ns =38)
Table 1 Statistical evaluation of surface water, groundwater, and soil samples in the summer season (ng L−1 for water samples, ng g−1 for soil samples)
Environ Monit Assess (2015) 187:94 Page 5 of 12 94
94
Page 6 of 12
concentrations of endosulfan and its derivatives in groundwater samples than in surface water samples. Endosulfan sulfate had both the highest concentration and was the most frequently observed OCP in the soil samples. Moreover, the percentage observation of αHCH in surface water was higher than that in groundwater. 4,4′-DDT, and its derivatives were also observed in surface water, groundwater, and soil samples. Heptachlor, heptachlor epoxide, and methoxychlor were not observed in any of the samples.
Seasonal changes of the OCPs Figure 2a shows the seasonal difference for each observed pesticide in water samples (combination of surface water and groundwater samples) collected only in the city center. Average value (at least three values) of the concentrations of each pesticide (without using the substitution for the not detected concentrations) was used. Fig. 2 a Seasonal differences in average concentrations for each observed pesticide in water samples taken from the city center (N=17). b Seasonal differences in average concentrations for each observed pesticide in soil samples taken from the city center (N=17)
Environ Monit Assess (2015) 187:94
In the summer sampling, 10 pesticides were observed in the surface water and groundwater samples. However, in the autumn season, only four pesticides were observed in surface water and groundwater samples. Endosulfan sulfate had the highest arithmetic mean in both seasons because of the extensive usage of endosulfan in the past, since over time; the endosulfan isomer mixture is converted into endosulfan sulfate (Walse et al. 2003). Seasonal changes were also observed for the soil samples. For the soil samples, the concentrations in summer samples were higher than the autumn season, except for 4,4′-DDT (Fig. 2b). The concentrations were higher in the agricultural areas in both seasons for all samples.
Spatial distributions In the five districts of Bolu, the spatial distributions of the each target analytes were presented by the contour maps by using summer season data. The missing data
Environ Monit Assess (2015) 187:94
was filled with half of the detection limit values for hexachlorocyclohexane isomers, aldrin, dieldrin, and endrin to draw the maps. For other pesticides, the detection limit values were used to fill the missing values. With the aid of these maps, the relationship between the observed concentrations and the characteristics of sampling points were evaluated and the degradation processes were discussed. Endosulfan isomers and their derivatives Endosulfan has still been used in agricultural activities in Turkey, even though the use of endosulfan is under discussion since 2007 in Turkey. Commercial endosulfan contains endosulfan I and endosulfan II isomers in ratios from 2:1 to 7:3, depending on the technical mixture (Herrmann 2002). It is applied directly to the soil or plant. For this reason, endosulfan II concentrations were higher than the other sampling points, especially in the agricultural areas of Mudurnu (Fig. 3a). Moreover, the two isomers can be converted to endosulfan sulfate by oxidation in biotic and abiotic degradation (Walse et al. 2003). Both endosulfan II and endosulfan sulfate are scavenged from the atmosphere easily by precipitation, due to their higher solubility than endosulfan I, and as such, they can be carried to the surface water by the atmospheric longrange transport (Chan et al. 1994, 2003). Figure 3b, c shows the spatial distributions of endosulfan sulfate concentrations in surface water and groundwater samples. When the two maps were compared, it was seen that concentrations in surface waters might have been affected by atmospheric long-range transport of endosulfan sulfate because the concentrations of surface water samples were higher than the concentrations in groundwater samples. As a result, the concentrations of the surface water and groundwater and soil compartments of Bolu were affected by the current and previous use of commercial endosulfan mixture. Aldrin and dieldrin Under most environmental conditions, aldrin is largely converted to dieldrin and dieldrin is significantly more persistent. Most environmental releases of aldrin and dieldrin are directly to the soil. They have low water solubility and a tendency to bind tightly to the soils. Therefore, both compounds migrate downward very slowly through soils or into surface water or groundwater. Most surface water aldrin/dieldrin has been
Page 7 of 12 94
attributed to particulates by surface runoff. For this reason, it is possible that significant volatilization of aldrin/dieldrin might occur, with subsequent atmospheric photo degradation or rainfall (USEPA 2003). According to the local authorities, aldrin has not been in use after the ban in the 1980s in Turkey. Because of the past usage of aldrin, it was observed only in one sample. However, dieldrin was observed in agricultural areas. Therefore, the observed presence of dieldrin can be explained by the past usage of aldrin.
Hexachlorobenzene isomers (HCHs) HCHs were one of the most widely used organochlorine pesticides. They have a relatively high volatility, and they can be found in many environmental samples, including pristine locations such as the Arctics. They have been used as an insecticide in wheat production (Willet et al. 1998). In technical grade HCH, the α- to γ-isomer ratio is between 3 and 7. Global sources of HCHs and the transfer of HCHs away from the source can be evaluated using the α- to γ-isomer ratios. For example, the current usage of lindane (γ-HCH) shows this ratio near or less than unity (Hargrave et al. 1988). Therefore, higher α to γ ratios in water samples can be explained by differential gas exchange across air-water interfaces, washout by precipitation or volatilization-adsorption variations between the isomers (Iwata et al. 1993). The α- to γ-isomer ratios in surface water and groundwater samples are presented (Fig. A.3 and Fig A.4) in the supplementary document. The α- to γ-isomer ratios in Bolu ranged between 0.65 and 5.45 in surface water samples. Besides the high wheat production areas, high α- to γ-isomer ratios can also be seen in non-agricultural regions, such as Abant Lake (32) and Seben Lake (52). In particular, Seben Lake is a new artificial lake, and this lake was filled with snow water in 2008. Long-range transport can be effective in this lake (Fig. 4a). For groundwater samples, α- to γ-isomer ratios ranged between 0.44 and 0.77. In some sampling points, all isomers were observed. The major isomer of surface water was α-HCH. The higher existence of α-HCH in surface water indicated the photo degradation of lindane (Willet et al. 1998; Rissato et al. 2006). In groundwater samples, the major isomer was δ-HCH. The high δ-HCH content in groundwaters in this study indicated that the transformation of lindane was dominant (Buser and Muller 1995).
94
Page 8 of 12
Environ Monit Assess (2015) 187:94
Environ Monit Assess (2015) 187:94
Page 9 of 12 94
Fig. 4 a The spatial distribution of α-HCH in surface water samples. b The spatial distribution of α-HCH in groundwater samples
Endrin Endrin is nearly insoluble in water. The detection of endrin in groundwater and surface water is rare. 3 a The spatial distribution of endosulfan II in soil samples. b Fig. The spatial distribution of endosulfan sulfate in surface water samples. c The spatial distribution of endosulfan sulfate in groundwater samples
Although endrin has low volatility, it was found in higher concentrations in the air during its manufacture. It was never found in urban areas, where it was rarely used (ATSDR 1996). Endrin was used in the production of sugar cane, grain crops, sugar beets, and rice as an insecticide. These plants are produced in the sampling region of this study.
94
Page 10 of 12
Endrin was found in some of the water samples in Bolu. The previous use of endrin could be the cause of the presence in water samples. Endrin concentrations in surface water were high in the agricultural region in Mudurnu (sampling points 22, 24, and 25). It was also observed in non-agricultural regions, such as sampling point 53. Endrin can also be seen in groundwater samples. The Seben region is a grain-producing region. The groundwater sample
Environ Monit Assess (2015) 187:94
taken from sampling point 59 in an agricultural area (Seben) had the highest endrin concentration in this study. Endrin degradation in soil did not significantly appear. Approximately 41 % of the applied endrin can remain in a field up to 14 years after application (Nash and Woolson 1967). Endrin was not detected in any soil samples. It was thought that it was carried from the atmosphere to bodies of water (Fig. 5a, b).
Fig. 5 a The spatial distribution of endrin in surface water samples. b The spatial distribution of endrin in groundwater samples
Environ Monit Assess (2015) 187:94
Page 11 of 12 94
Fig. 6 The spatial distribution of 4,4′-DDT in surface water samples
4,4′-DDT, 4,4′-DDE, 4,4′-DDD 4,4′-DDT can contaminate to surface waters, primarily by runoff, atmospheric transport, and drift or by direct application. 4,4′-DDT is persistent in soil and does not move easily to groundwater (EPA 2000). The ratios of 4,4′-DDT/ (4,4′-DDD+4,4′-DDE) and 4,4′-DDE/4,4′-DDD have been used to indicate the degree of DDT composition and to find a sign for fresh input of DDTs. If the ratio is higher than unity, the fresh usage of these pesticides can be possible (Lee et al. 2001). The 4,4′-DDT/ (4,4′-DDD+ 4,4′-DDE) ratios ranged from 3.9 to 61.1 in surface water samples and from 0.9 to 40.6 in groundwater samples in the summer season in this study. In addition to the persistency of 4,4′-DDT, it was used in very large amounts in this area; the illegal use of 4,4′-DDT is also possible in Turkey (Acara 2004). The results of this study can reveal possible fresh input of DDTs in this region, as was seen in the surface water samples (Fig. 6). In the current study, 4,4′-DDE was not observed in soil samples, although 4,4′DDT and 4,4′-DDD were seen in some soil samples.
Conclusion In the current study, the concentrations of OCPs in surface water, groundwater, and soil samples were
determined. The sampling was performed in the city center, Seben, Kıbrıscık, Göynük, and Mudurnu districts of Bolu. To observe the seasonal changes, the sampling was repeated only in the city center of Bolu. Solid phase extraction and ultrasonic extraction techniques were used for the extraction of compounds from water and soil samples, respectively. The samples were analyzed by GC-ECD. The concentrations of 13 OCPs in water samples and 11 OCPs in soil samples were determined. The spiked samples were used for the recovery tests. The recoveries ranged from 72 to 108 % for water samples, except for aldrin and dieldrin, and from 40 to 131 % in soil samples. They were generally within the limit of acceptance (70–130 %) according to the EPA, except for aldrin, heptachlor epoxide, and 4, 4′ DDE in the soil samples. Seasonal variations can be observed in the city center. The concentrations of endosulfan sulfate in water samples and 4, 4′-DDE in soil samples were higher in summer than in autumn. The contour maps showed that endosulfan was the currently used pesticide in Bolu and it was the only pesticide determined in the five districts. Although the use of 4, 4′-DDT was banned in 1985 in Turkey, it was also generally observed in Bolu. This demonstrated the possibility of current illegal use of this pesticide. Generally, the districts of Bolu were affected by pesticides used for agricultural applications. The
94
Page 12 of 12
concentrations were lower than 100 ng L−1, which is the limit of the European Union for each pesticide in water resources. The atmospheric long-range transport of pesticides, precipitation, and transformation by photo degradation is the effective mean to contaminate surface waters, particularly in Bolu. Acknowledgments The authors would like to thank Prof. Dr. Okan Külköylüoğlu and Prof. Dr. Muzaffer Dügel for their technical support and Elif Özlü and İlker Köprü for their assistance with the sampling. The authors are grateful to Uğur Saklangiç, Emre Aslan, and Onur Çar for their assistance with the sample preparation and analysis.
References Acara, A. (2004). Türkiye’de Kalıcı Organik Kirleticiler (POP’s) Envanter Özet Raporu. Resource Document. http://www. ttgv.org.tr/content/docs/envanter-ozet-raporu.doc. Accessed 4 Oct 2013. Arias-Estevez, M., Lopez-Periago, E., Martinez-Carballo, E., Simal-Gandara, J., Mejuto, J. C., & Garcia-Rio, L. (2008). The mobility and degradation of pesticides in soils and the pollution of groundwater resources. Agriculture, Ecosystem & Environment, 123, 247–260. ATSDR (Agency for Toxic Substances and Disease Registry). (1996). Toxicological Profile for endrin and endrin aldehyde (update). Atlanta: Research Triangle Institute. Binici, B., Yenisoy-Karakaş, S., Bilsel, M., & Durmaz-Hilmioğlu, N. (2014). Sources of polycyclic hydrocarbons and pesticides in soluble fraction of deposition samples in Kocaeli, Turkey. Environmental Science and Pollution Research, 21, 2907– 2917. Buser, H., & Muller, M. D. (1995). Isomer and enantioselective degradation of hexachlorocyclohexane isomers in sewage sludge under anaerobic conditions. Environmental Science and Technology, 29, 664–672. Calhoun, Y. (2005). Water pollution. New York: Chelsea House Publisher. Chan, C., Bruce, G., & Harrison, B. (1994). Wet deposition of organochlorine pesticides and polychlorinated biphenyls to the Great Lakes. Journal of Great Lakes Research, 20, 546–560. Chan, C., Williams, D., Neilson, M., Harrison, B., & Archer, M. (2003). Spatial and temporal trends in the concentrations of selected organochlorine pesticides (OCs) and polynuclear aromatic hydrocarbons (PAHs) in Great Lakes Basin precipitation, 1986 to 1999. Atmospheric Environment, 40, 1563–1578.
Environ Monit Assess (2015) 187:94 Dong, C., Zeng, Z., & Yang, M. (2005). Determination of organochlorine pesticides and theirderivations in water after HSSPME using polymethylphenylvinylsiloxane-coated fiber by GC-ECD. Water Research, 39, 4204–4210. Erşahin, G., Şerifeken, İ. (2002). Bolu ili Raporu, Devlet Planlama Teşkilatı Müsteşarlığı. DPT:2651, Ankara, Turkey. EPA (2000) Revised Draft Lake Michigan Lakewide Management Plan for Toxic Pollutants. Clean water act. Canada. EPA (2013). www.epa.gov/oia/toxics/pop.html. Accessed 4 Oct 2013. Hargrave, B. T., Vass, W. P., Erickson, P. E., Fowler, B. R. (1988). Atmospheric transport of organochlorines to the Arctic Ocean. Tellus, 480–493. Herrmann M. (2002). Preliminary risk profile of endosulfan. Berlin Germany. Iwata, H., Tanabe, S., Sakal, N., & Tatsukawa, R. (1993). Distribution of persistent organochlorines in Oceanic air and surface seawater and the role of ocean on their global transport and fate. Environmental Science and Technology, 27, 1080–1098. Lee, K. T., Tanabe, S., & Koh, C. H. (2001). Distribution of organochlorine pesticides in sediments from Kyeonggi Bay and nearby areas, Korea. Environmental Pollution, 114, 207– 213. Nash, R. G., & Woolson, E. A. (1967). Distribution of chlorinated insecticides in cultivated soil. Soil Science Society of America Journal, 32, 525–527. Öz, M. (2009). Characterization of PCBs, PAHs, and Organochlorine Pesticides in the Atmosphere of Bolu and Source Apportionment: Gas to Particle, Soil to Particle, Soil to Gas Exchanges. Bolu, Turkey. Rissato, S. R., Galhiane, M. S., Ximenes, V. F., de Andrade, R. M., Talamoni, J. L., Libanio, M., de Almeida, M. V., Apon, B. M., & Cavalari, A. A. (2006). Organochlorine pesticides and polychlorinated biphenyls in soil and water samples in the Northeastern part of Sao Paulo State, Brazil. Chemosphere, 65, 1949–1958. Taylor, A. W. (1995). Environmental behaviour of agrochemicals. Chichester: John Wiley & Sons. USEPA (2003). Health Effect support Document for Aldrin/ Dieldrin. Washington, D.C. Vega, A. B., Frenich, A. G., & Vidal, J. L. M. (2005). Monitoring of pesticides in agricultural water and soil samples from Andalusia by liquid chromatography coupled to mass spectrometry. Analytica Chimica Acta, 538, 117–127. Walse, S. S., Scott, G. I., & Ferry, J. L. (2003). Stereoselective degradation of aqueous endosulfan in modular estuarine mesocosms: formation of endosulfan γ-hydroxycarboxylate. Journal of Environmental Monitoring, 5, 373–379. Willet, K. L., Ulrich, E. M., & Hites, R. A. (1998). Differential toxicity and environmental fates of hexachlorocyclohexane isomers. Environmental Science and Technology, 32, 2197– 2207.
