Environ. Sci. Technol. 2005, 39, 8374-8381
Leaching of Metribuzin Metabolites and the Associated Contamination of a Sandy Danish Aquifer J E A N N E K J Æ R , * ,† P R E B E N O L S E N , ‡ TRINE HENRIKSEN,† AND MARLENE ULLUM† Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen, Denmark, and Danish Institute of Agricultural Sciences, Research Centre Foulum, DK-8830 Tjele, Denmark
As degradation products of metribuzin have received little attention as potential groundwater contaminants, we evaluated leaching of metribuzin and its primary metabolites desaminometribuzin (DA), desaminodiketometribuzin (DADK), and diketometribuzin (DK) at a sandy test site in Denmark. Soil water and groundwater were sampled monthly over a four-year period. Leaching of metribuzin and DA was negligible. DK and DADK leached from the root zone (1 meter below ground surface (mbgs)) in average concentrations considerably exceeding the EU limit value for drinking water (0.1 µg/L). Both metabolites appear to be relatively stable and persisted in soil water and groundwater several years after application. Past application of metribuzin at the site had contaminated the groundwater with both DK and DADK, which were detected in 99% and 48%, respectively, of the groundwater samples analyzed. Except for three of the groundwater samples, the DADK concentration never exceeded the EU limit value. In contrast, the annual concentration of DK exceeded 0.1 µg/L at 90% of the screens analyzed. The present findings suggest that as the degradation products of metribuzin can leach through sandy soil in high concentrations, they could potentially contaminate the groundwater. In view of this risk DK and DADK should both be included in monitoring programs and their ecotoxicological effects should be further investigated.
Introduction Degradation products of agrochemicals used in modern agricultural production are frequently detected in the groundwater, often more frequently than the parent substances (1-4). Over the past decades, degradation products have thus become increasingly more important when evaluating the risk of groundwater contamination. The degradation products of the herbicide metribuzin have received very little attention as potential groundwater contaminants. In the early 1990s, metribuzin was widely used for pre- and post-emergent control of weeds in a variety of vegetable crops, as well as in soybean and wheat. Although usage in soybean decreased markedly in the United States during the 1990s, metribuzin has remained the herbicide * Corresponding author phone: +45 3814 2333; fax: +45 3814 2050; e-mail:
[email protected]. † Geological Survey of Denmark and Greenland. ‡ Danish Institute of Agricultural Sciences. 8374
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 21, 2005
most frequently applied to potato and tomato crops (64% and 38% of the 2001 crop area, respectively) (5). Available data from eight European countries shows that metribuzin also remains the most frequently applied herbicide in potato crops in Denmark (6) as well as Sweden, Ireland, and Finland (7). Dissipation of metribuzin is usually fast with DT50 (dissipation time for 50% of applied pesticide) values ranging from 11 to 46 days in both laboratory and field studies (812). Usually less than 10% of the applied metribuzin is mineralized to CO2, however, with the highest reported mineralization being 20% within 90 days (13). Despite the fast dissipation, the majority of the applied metribuzin will consequently remain in the soil either adsorbed, as bound residues, or as degradation products. The primary degradation products of metribuzin are desaminometribuzin (DA), desaminodiketometribuzin (DADK), and diketometribuzin (DK) (Figure 1), although there are reports of unidentified metabolites (13, 15, 16). Information on the fate of metribuzin metabolites is sparse. Henriksen et al. (17) reported rapid initial dissipation of both metribuzin and its metabolites in a sandy topsoil, with 50% disappearance within 30-40 days. Minor amounts of DA, DK, and DADK corresponding to 10% of the applied metribuzin were still detected one year after application, however. Minor amounts of DK, DADK, and DA were also detected in the soil several months after metribuzin application in laboratory and field dissipation studies (13, 16). Low sorption capacity of both metribuzin and its metabolites has been reported in batch experiments using Danish sandy topsoils (17) and in lysimeter studies (8, 15). One of these lysimeter studies (8) concluded that metribuzin and its metabolites are considerably more mobile than atrazine. In both studies, Kf for metribuzin, DA, DK, and DADK were found to be less than 1 (8, 15). In contrast, a field dissipation study concluded that neither metribuzin nor its metabolites moved in appreciable amounts (16). Most reports on metribuzin metabolites focus on the quantification of sorption and degradation processes in the soil, while quantification of leaching to the aquatic environment has been accorded little attention. What little information is available on leaching processes derives from laboratory and lysimeter studies (8, 15). To our knowledge, controlled experiments quantifying the leaching of metribuzin, and in particular its metabolites, under actual field conditions have not previously been reported. The present study thus examined the transport of metribuzin and its primary metabolites under field conditions in order to assess the risk that these compounds might leach to the groundwater in Denmark.
Experimental Section Site Description. The test site is located at Tylstrup in the North of Denmark. The site covers 1.1 ha (70 × 166 m) and is practically flat with windbreaks along the eastern and western borders. The soil is classified as a Humic Psammentic Dystrudept (18). The topsoil is characterized as loamy sand with approximately 6% clay and approximately 1.6% total organic carbon (Table 1). The aquifer comprises an approximately 20-m layer of marine sand sediment deposited in the Yoldia Sea. The southern part is rather homogeneous, consisting entirely of fine-grained sand, whereas the northern part is more heterogeneous due to the intrusion of several silt and clay lenses (19). During the monitoring period the groundwater table was located 3-4.5 meters below ground surface (mbgs). The overall direction of groundwater flow was toward the west (Figure 2). 10.1021/es0506758 CCC: $30.25
2005 American Chemical Society Published on Web 09/29/2005
from two of the downstream monitoring wells (M4 and M5). In addition, more intensive monitoring encompassing all suction cups and all monitoring wells except M7 and M2 was performed every four months. Data were collected for four years. Soil Samples. Topsoil (0-20 cm) was sampled on three occasions: October 22, 2002 (>40 months since metribuzin was last applied), January 28, 2003, and April 23, 2003. A total of 25 soil samples were collected diagonally across the field, homogenized, mixed, and sieved through a 2-mm mesh in situ. A 0.5-kg portion of the sieved soil was frozen at -20 °C until analysis.
FIGURE 1. Degradation pathways for metribuzin (14). Agricultural Management. Cultivation of the site was in line with conventional agricultural practice in the region. Metribuzin was applied in the maximum dose (0.25 kg metribuzin/ha) permitted under current regulations. During the potato-growing season of 1999 (planted May 5, harvested October 20) the field was sprayed with metribuzin twice: 0.2 kg/ha of Sencor WG being applied on May 25, 1999 and 0.15 kg/ha on June 7, 1999. Bromide tracer (30 kg/ha KBr) was applied on May 27, 1999 to confirm that the water samples actually originated from the test site. Due to the wet growing season, irrigation was performed only once during cultivation of the crop (33 mm on September 12). Monitoring. To avoid unintended leaching of pesticide due to the installation and presence of the sampling equipment in the ground, all installations were restricted to a buffer zone surrounding the treated area, as was all soil sampling deeper than 20 cm (Figure 2). Suction cups and monitoring wells were installed in March 1999 prior to initiation of monitoring in May 1999. A tipping bucket rain gauge system was used for local measurement of precipitation. Soil Water Sampling. Soil water was sampled monthly at locations S1 and S2 (Figure 2) using 16 Teflon suction cups (Prenart, DK) clustered 4 by 4 at depths of 1 and 2 m. Each cluster covered a horizontal distance of 2 m. The suction cups were installed from two excavation pits at the edge the test field via holes drilled obliquely to the desired depth. This procedure ensured that the soil directly above the suction cups remained undisturbed. Each suction cup was connected via a single length of PTFE tubing to a glass bottle refrigerated at 5 °C. Extraction of water was performed over a seven-day period using a continuous vacuum of approximately 80 KPa. The chemical analyses were performed on pooled samples from each cluster. Groundwater Sampling. Groundwater was sampled monthly from a number of the vertical monitoring wells (Figure 2) each consisting of four 1-m screens covering the upper approximately 4 m of the saturated zone. In August 2001, three additional 1-m screens were installed 6-9 mbgs near wells M4 and M5. To facilitate water sampling, each screen was permanently connected to a whale pump. Prior to water sampling the screens were purged by removing a volume of water equivalent to 3× the volume of water present in the screen. Bromide analysis was performed monthly on water samples from all the monitoring wells and from all four groups of suction cups. Pesticide analysis was performed monthly on water samples from the suction cups located 1 mbgs and
Analysis Methods. The analyses of metribuzin, DA, DK, and DADK in the water samples were all performed in commercial laboratories accredited for pesticide analyses by the Danish EPA. Metribuzin was analyzed by a LC/MS method already developed for detection of multiple pesticides while the degradation products (DA, DK, and DADK) were analyzed separately, using GC-MS. Metribuzin. The water samples (200 mL) were first adjusted to pH 2 using HCl. An internal standard (2,4-D-d5) was then added and the metribuzin concentration was determined by on-line solid-phase extraction (Prospekt, Spark Holland) coupled to an LC/MS system (1100 from Agilent). The analytes were separated in a C18 BDS hypersil column (Thermo, 2.1 × 250 mm, 5 µm particles) with a gradient program running from 70% of eluent A (10 mM ammoniumacetate) and 30% eluent B (acetonitrile), to 2% A within 25 min. DK and DADK. The water samples (1000 mL) were first adjusted to pH 3.3 using HCl, after which 2,4-dichlorobenzamide was added as internal standard. The samples were then extracted with 275 mL of dichloromethane for 20 min, and finally the extract was filtered with water-free sodium sulfate and evaporated to virtual dryness. After being redissolved in ethyl acetate the samples were measured by GC-MS (GCQ, Finnegan MAT) using a CP-SIL 8-CB/MS GCcolumn (length 50 m, 0.32 mm i.d./0.25 µm). The temperature program was initiated at 60 °C for 2 min, then raised by 13 °C/min to 220 °C and kept for 2 min, and finally increased to 310 °C for 2 min. DADK was identified using m/z 127 and 154 by scanning the ion-trap MS from m/z 90-350. DK was identified using m/z 168, which was the only specific ion obtainable with a sufficient response. Detection quality for DK was instead ensured by the use of internal standard and by comparing DK detection in internal control samples with that in the real samples. DA. DA was measured by the same GC-MS method applied for DK and DADK (see above), except that the dichloromethane extract was redissolved in 10% N-methyltrimethylsilyltrifluoroacetamide/ethyl acetate, since a derivatization step was required. The DA was identified using m/z 229 and 256. With all four compounds, calibration was performed on standards taken through the whole procedure, using internal standard calculations. As part of the internal quality control, two replicates of control samples (concentration level 0.03 µg/L) were included in each series of samples. The limit of detection (LOD) was 0.01 µg/L for metribuzin and 0.02 µg/L for DADK, DA, and DK. The coefficient of variation (CV) was less than 10% for all compounds, measured for concentrations about 5× the LOD. Blank samples consisting of HPLCgrade water (Rathburn Chemicals Ltd, Walkerburn, Scotland) were sent to the laboratory each month together with the groundwater samples. Blank samples were labeled with coded reference numbers so that the laboratories were unaware of which samples were blanks and which were groundwater samples. No pesticides were detected in any of the blank VOL. 39, NO. 21, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8375
TABLE 1. Physical and Chemical Properties of the Soil at the Tylstrup Test Site horizon
depth (cm)
claya (%)
silta (%)
sanda (%)
OMb (%)
C/N
CEC (meq/100 g)
pHCaCl2
Fec (mg/kg)
Alc (mg/kg)
Ap Ap2 Bv BC C
0-32 32-40 40-60 60-105 105-160
6.4 5.8 5.3 3.6 2.5
3.6 4.2 3.7 0.9 0.5
87.3 87.5 89.0 95.2 96.9
2.7 2.5 2.0 0.3 0.1
13 15 13 6 2
7.6 8.4 8.4 3.2 5.1
4.5 4.6 4.5 4.7 4.8
1660 2272 2432 1516 340
1034 2160 3660 1520 1008
a Clay 30 mm/ day) occurred during the second month after the pesticide was last applied (Figure 4). This precipitation patternsin terms of daily or monthly precipitationsis not unusual for the Tylstrup region, as similar patterns have occurred at other times during the period 1990-2004 (24). Over the four-year monitoring period, the amount of DADK and DK leached equaled 2.2-3.8% and 0.96-1.5%, respectively, of the applied metribuzin (Table 2). The amount leached is obviously highly site-specific, depending on the soil properties and hydrological conditions following application of the pesticide. As far as we are aware, quantitative field data on leaching of metribuzin metabolites have not been previously reported. Nevertheless, the potential leaching risk of metribuzin metabolites has also been suggested by Bowman (8) based on transport studies in 70-cm lysimeters packed with sandy soil; metribuzin, and in particular DADK and DK, leached through lysimeters irrigated with 1038 mm of water within 21 weeks. In similar lysimeters irrigated with only 227 mm, leaching was negligible. In laboratory experiments using undisturbed 10-cm microlysimeters with a loamy soil irrigated with 20 mm of water on a weekly basis over a 6-week period, 14C-metribuzin leached as metribuzin (12%,), DADK (4.5%), DK (4.2%), and DA (1.7%) (15). In contrast, Conn et al. (16) concluded from a field dissipation study on silty loam that neither metribuzin nor its metabolites moved in appreciable amounts, as less than 1% of the applied metribuzin leached from the top soil. Information is not available on precipitation at their site following metribuzin application, however, thus making it difficult to relate their findings to our results. Compared to other routes of dissipation, 500 >500
Kd (L/kg)b 0.94 1.83 0.72 0.71 ∼0 1.01 ∼0 0.54
a Apart from DADK following first-order kinetics in the topsoil, degradation rate decreased over time following a two-compartment model. b Determined using initial aqueous concentrations at 250 µg/L. c DT50 for DK in the topsoil could not be calculated.
With DK, the concentration level in the unsaturated zone (1-2 mbgs) was somewhat similar to that of the uppermost groundwater (3-5 mbgs). Thereafter the concentration level increased with increasing depth, and all samples from the deepest screen located 8-9 mbgs contained more than 0.1 µg/L DK. The concentration pattern exhibited by DADK was the opposite. In the root zone (1 mbgs), leaching of DADK was higher than that of DK. Unlike with DK, however, the concentration decreased down through the unsaturated zone and was markedly lower in the groundwater. That root zone leaching (1 mbgs) of DADK exceeded that of DK is in accordance with Bowman (8) performing transport studies in lysimeters packed with sandy soils. Henriksen et al. (17) evaluated the sorption and degradation of metribuzin and its metabolites in Tylstrup soil by means of batch experiments (Table 4). In their study sorption and degradation characteristic of DADK and DK were similar in the top soil, and they ascribed a higher leaching risk of DADK (1 mbgs) to the fact that more DADK than DK is produced as it can be formed via the DA degradation pathway and the degradation of DK (Figure 1). The supposition that more DADK than DK is produced is supported by the findings of Sharom and Stephenson (31) that, following six months of incubation of 14C-metribuzin in a silt loam, the radioactivity is distributed 10% in metribuzin, 20% in DK, 20% in DA, and 50% in DADK. That the concentration of DADK decreased down through the unsaturated zone while that of DK remained more constant could be ascribed to the different long-term leaching pattern of the two compounds (Figure 3). By far the majority of the DADK leached to 1 mbgs within the first year following application. The concentration decreased with time and was negligible at the very end of the monitoring period (Figure 3A and C). With DK, in contrast, less leached, but did so over a much longer period of time with concentrations reaching 0.1 µg/L, being detected as long as four years after application (Figure 3A and C and Table 2). While DADK appears to be formed and released shortly after application, thus yielding a pulse that moves downward through the unsaturated zone, DK seems to be released over a much longer period of time resulting in a much more constant concentration level at both 1 and 2 mbgs (Figure 3). As published studies on the fate of metribuzin focus on processes occurring within the first year after application, the literature does not contain experimental data that might explain the long-term leaching behavior observed in the present study. A likely explanation for the observed leaching pattern, though, is that the degradation pathway taken by metribuzin shifts during the course of time. While the DA degradation pathwaysmainly induced by photodegradationsprevails in the summer period after application, the DK degradation pathway might occur 8380
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 21, 2005
during a much longer period. Moreover, the slightly higher sorption capacity of DADK in the subsoil could also affect the downward transport at Tylstrup. Henriksen et al. (17) found that sorption of DADK was minor in subsoil, while that of DK was negligible (Table 4). Thus, once DADK and DK have reached the subsoil, some of the DADK could be retained, allowing DK to reach the groundwater to a higher extent than DADK (17). As a result of the different leaching patterns described above, the concentration of DK in the uppermost groundwater was somewhat similar to that in the unsaturated zone; in contrast, the concentration of DADK was markedly lower (Figures 5 and 6). No experimental data are available that might explain the increasing concentration of DK observed in the deepest screens. When evaluating Figure 6, though, it should be kept in mind that the data for 1 and 2 mbgs derive from suction cups located in the unsaturated zone, whereas the data for 3-9 mbgs derive from the monitoring screens located in the saturated zone. While the water sampled 1-2 mbgs represents water that has infiltrated at the test site, the water sampled 3-9 mbgs represents a mixture of water that has infiltrated at both the test site and at fields situated upstream of the test site. Water sampled in the deepest screens thus tends to represent water that has infiltrated in more distant areas. Metribuzin is widely used and has frequently been applied to the neighboring fields located upstream of the test site (28). The marked contamination with DK observed in the deepest screens is thus presumably due to previous applications of metribuzin at more distant locations upstream of the test site. On those occasions, a higher dose of metribuzin may have been used since the maximum permitted dose was reduced from 0.49 to 0.25 kg/ha in Denmark in 1994. DA was not detected in any of the water samples. According to Henriksen et al. (17), DA is presumably retained in the topsoil, where sorption and degradation to DADK prevents it from leaching.
Acknowledgments This study was funded by the Danish Pesticide Leaching Assessment Programme. We thank the many people who have contributed to this work over the years, including Bo Lindhardt, Christian Abildtrup, Henrik Vosgerau, Henning Hougaard, and Finn Plauborg (establishment of field sites and monitoring design) and Carsten Kamper, David Croft, Søren H. Jepsen, Birgit Sørensen, Carl H. Hansen, Søren Nielsen, and Lasse Gudmundsson (ongoing field monitoring and data preparation).
Literature Cited (1) Kolpin, D. W.; Thurman, E. M.; Linhart, S. M. Finding minimal herbicide concentrations in groundwater? Try looking for their degradates. Sci. Total Envir. 2000, 248, 115-222. (2) Kolpin, D. K.; Schnoebelen, D. J.; Thurman E. M. Degradates provide insight to spatial and temporal trends of herbicides in groundwater. Groundwater 2004, 42, 601-608. (3) Jørgensen, L. F. Groundwater monitoring 2003. Geological Survey of Denmark: Copenhagen, 2003. (4) Minnesota Department of Agriculture. Pesticide monitoring in water resources: Annual data report; Saint Paul, MN, 2005. http://www.mda.state.mn.us/appd/ace/reports/2005annual. pdf. (5) National Agricultural Statistical Services. Agricultural Chemical Use Databases; U.S. Department of Agriculture: Washington, DC, 2005. http://www.pestmanagement.info/nass/app_usage. cfm. (6) Danish Environmental Protection Agency. Bekæmpelsesmiddelstatestikken 2003, Orientering fra Miljøstyrelse nr. 9, 2004 (in Danish); Danish EPA: Copenhagen, 2004. http:// www.mst.dk/udgiv/publikationer/2004/87-7614-316-3/html/ bil01.htm.
(7) Eurostat. The use of plant protection products in the European Union. Data 1992-1999; ISBN 92-894-4818-0; European Communities: Luxembourg, 2002. http://europa.eu.int/estatref/ info/sdds/fr/milieu/use_of_plant_protection_products.pdf (visited August 23, 2005) (See metadata description on http://europa.eu.int/estatref/info/sdds/fr/milieu/agri_pest_sm. htm#compilation). (8) Bowman, B. T. Mobility and dissipation studies of metribuzin, atrazine and their metabolites in plainfield sand using field lysimeters. Environ. Toxicol. Chem. 1991, 10, 573-579. (9) Di, H. J.; Aylmore, L. A. G.; Kookana, R. S. Degradation rates of eight pesticides in surface and subsurface soils under laboratory and field conditions. Soil Sci. 1998, 163 (5), 404-411. (10) Gallaher, K.; Mueller, T. C. Effect of crop presence on persistence of atrazine, metribuzin, and clomazone in surface soil. Weed Sci. 1996, 44, 698-703. (11) Hyzak, D. L.; Zimdahl, R. L. Rate of metribuzin degradation and two analogs in soil. Weed Sci. 1974, 22, 75-79. (12) Lechon, Y.; Garcia-Valcarcel, A. I.; Matienzo, T.; SanchezBrunete, C.; Tadeo, J. L. Laboratory and field studies on metribuzin persistence in soil and its prediction by simulation models. Toxicol. Environ. Chem. 1997, 63, 47-61. (13) Moorman, T. B.; Harper, S. S. Transformation and mineralization of metribuzin in surface and subsurface horizons of a Mississippi delta soil. J. Environ. Qual. 1989, 18, 302-306. (14) Prestel, D.; Weisgerber, I.; Klein, W.; Korte, F. Bilanz der Verteilung und Umwaldlung von Metribuzin-14C (Sencor) in Kartoffeln, Mo¨hren und Boden unter Freilandbedingungen. Chemosphere 1976, 2, 137-144. (15) Locke, M. A.; Harper, S. S.; Gaston, L. A. Metribuzin mobility and degradation in undisturbed soil columns. Soil Sci. 1994, 157 (5), 279-288. (16) Conn, J. S.; Kosknen, W. C.; Werdin, N. R.; Graham, J. S. Persistence of metribuzin and metabolites in two Arctic soils. J. Environ. Qual. 1996, 25, 1048-1053. (17) Henriksen, T.; Svensmark, B.; Juhler, R. K. Degradation and sorption of metribuzin and primary metabolites in a sandy soil. J. Environ. Qual. 2004, 33, 619-627. (18) Soil Survey Staff Soil Taxonomy. A Basic System for Soil Classification for Making and Interpreting Soil Surveys, 2nd ed.; Agricultural Handbook Number 436; U. S. Department of Agriculture: New York, 1999. (19) Lindhardt. B.; Abildtrup, C.; Vosgerau, H.; Olsen, P.; Torp. S.; Iversen, B. V.; Jørgensen, J. O.; Plauborg, F.; Rasmussen, P.; Graversen, P. The Danish Pesticide Leaching Assessment Programme: Site Characterization and Monitoring Design; Geological Survey of Denmark and Greenland: Copenhagen, 2001. http://www.pesticidvarsling.dk. (20) Vognsen, L. HPLC metode til bestemmelse af anioner i vand. Husdyrforskning 1996, 4 (2), 9-10.
(21) Henriksen, T.; Svensmark, B.; Juhler, R. K. Analysis of metribuzin and transformation products in soil by pressurized liquid extraction and liquid chromatographic-tandem mass spectrometry. J. Chromatogr. 2002, 957, 79-87. (22) Larsbo, M.; Jarvis, N. MACRO 5.0. A model of water flow and solute transport in macroporous soil. Technical description; Department of Soil Sciences, Swedish University of Agricultural Sciences: Uppsala, Sweden, 2003. http://www.mv.slu.se/bgf/ Macrohtm/Macro5_0/Technical%20report%20macro%205. pdf. (23) Ullum, M.; Kjær, J.; Plauborg, F. Calibration and validation of Macro 5.1: Water flow and bromide transport on two sandy field sites. to be submitted to J. Hydrol. . (24) Kjær, J.; Ullum, M.; Olsen, P.; Sjelborg, P.; Helweg, A.; Mogensen, B. B.; Plauborg F.; Grant, R.; Fomsgaard, I. S.; Bru ¨ sch, W. The Danish Pesticide Leaching Assessment Programme: Monitoring Results, May 1999-July 2002; Geological Survey of Denmark and Greenland: Copenhagen, 2003. http://www.pesticidvarsling.dk. (25) Boesten, J. J. T. I. From laboratory to field: uses and limitations of pesticides behavior models for the soil/plant system. Weed Res. 2000, 40, 123-138. (26) Webster, G. R. B.; Reimer, G. J. Field degradation of the herbicide metribuzin and its degradation products in a Manitoba sandy loam soil. Weed Res. 1976, 16, 191-196. (27) Burgard, D. J.; Dowdy, R. H.; Koskinen, W. C.; Cheng, H. H. Movement of metribuzin in a loamy sand soil under irrigated potato production. Weed Sci. 1994, 42, 446-452. (28) Kjær, J.; Ullum, M.; Olsen, P.; Sjelborg, P.; Helweg, A.; Mogensen, B.; Plauborg F.; Jørgensen, J. O.; Iversen, B. V.; Fomsgaard, I.; Lindhardt, B. The Danish Pesticide Leaching Assessment Programme: Monitoring results, May 1999-July 2001; Geological Survey of Denmark and Greenland: Copenhagen, 2002. http:// www.pesticidvarsling.dk. (29) Lawrence, J. R.; Eldan, M.; Sonzogni, W. C. Metribuzin and metabolites in Wisconsin (U.S.A.) well water. Wat. Res. 1993, 27, 1263-1268. (30) Standers, T. G.; Ward, R. G.; Loftis, J. C.; Steele, T. D.; Adrian, D. D.; Yevjevich, V. Design of Networks for Monitoring Water Quality; Water Resources Publications: Littleton, CO, 1994. (31) Sharom, M. S.; Stephenson, G. R. Behavior and fate of metribuzin in eight Ontario soils. Weed Sci. 1976, 24, 153-160. (32) McKeague, J. A.; Day, J. H. Dithionite- and oxalate-extractable Fe and Al as aids in differentiating various classes of soils. Can. J. Soil Sci. 1966, 46, 13-22.
Received for review April 8, 2005. Revised manuscript received August 26, 2005. Accepted August 29, 2005. ES0506758
VOL. 39, NO. 21, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8381
