Effect of Turfgrass Cover and Irrigation on Soil ... - Semantic Scholar

Effect of Turfgrass Cover and Irrigation on Soil Mobility and Dissipation of Mefenoxam and Propiconazole D. S. Gardner and B. E. Branham* ABSTRACT

Some pesticides dissipate more quickly in thatch compared with soil (Hurto et al., 1979; Gold et al., 1988; Gardner et al., 2000). Horst et al. (1996) found that the half-lives of metalaxyl [N-(2,6-dimethylphenyl)-N(methoxyacetyl)-DL-alanine methyl ester], pendimethalin [N-(1-ethylpropyl)-3,4-dimethyl-2,6-dinitrobenzenamine], chlorpyrifos [O,O-diethyl-O-(3,5,6-trichloro-2-pyridinyl) phosphorothioate], and isazofos (O-5-chloro-1-isopropyl1H-1,2,4-triazol-3-yl O,O-diethylphosphorothioate) applied to turfgrass were 16, 12, 10, and 7 d, respectively. This compares with previously published soil t1/2 data of 70, 34, 30, and 90 d, respectively (Balogh and Anderson, 1992). Mefenoxam is the resolved, biologically active stereoisomer of metalaxyl. Mefenoxam is a fungicide for the control of certain diseases in turfgrass and other crops. A considerable amount of work has been published concerning metalaxyl (Horst et al., 1996; Starrett et al., 1996; Sukul and Spiteller, 2000). Metalaxyl has a water solubility of 8.4 g L⫺1 and a Koc of 29 to 287, indicating the potential for considerable leaching through the turfgrass–soil profile. Metalaxyl has a variable t1/2, ranging from 7 to 160 d (Balogh and Anderson, 1992). To date, no work has been published on the environmental fate of mefenoxam; however, it would be expected to be very similar to metalaxyl. Yet, mefenoxam has a higher water solubility, 26 g L⫺1, than metalaxyl. Propiconazole is a triazole fungicide used to control several pathogens. Propiconazole has a water solubility of 110 mg L⫺1 and a Koc of 387 to 1147, indicating the potential for some leaching through the turfgrass–soil profile. Propiconazole is persistent, with a t1/2 of 109 to 120 d (Wauchope et al., 1991). Our objective was to investigate the effect of creeping bentgrass cover and irrigation on the mobility and persistence of mefenoxam and propiconazole. We also wished to determine whether the attenuating effect of thatch on pesticide mobility and persistence was uniform across pesticides of different chemical properties.

Irrigation effects on pesticide mobility have been studied, but few direct comparisons of pesticide mobility or persistence have been conducted on turfgrass versus bare soil. The interaction of irrigation practices and the presence of turfgrass on the mobility and dissipation of mefenoxam [N-(2,6-dimethylphenyl)-N-(methoxyacetyl)-D-alanine methyl ester] and propiconazole (1-[[2-(2,4-dichlorophenyl)-4-propyl-1,3-dioxolan-2-yl]methyl]-1H-1,2,4-triazole) was studied. Sampling cylinders (20-cm diam.) were placed in either creeping bentgrass [Agrostis stolonifera L. var. palustris (Huds.) Farw.] or bare soil. Mefenoxam was applied at 770 g a.i. ha⫺1 and propiconazole was applied at 1540 g a.i. ha⫺1 on 14 June 1999. Sampling cylinders were removed 2 h after treatment and 4, 8, 16, 32, and 64 days after treatment (DAT) and the cores were sectioned by depth. Dissipation of mefenoxam was rapid, regardless of the amount of surface organic matter or irrigation. The half-life (t1/2 ) of mefenoxam was 5 to 6 d in turf and 7 to 8 d in bare soil. Most mefenoxam residues found in soil under turfgrass were in the 0- to 1-cm section at 0, 4, and 8 DAT. Residues were found in the 15- to 30-cm section at 4, 8, 16, 32, and 64 DAT, regardless of turf cover or irrigation. The t1/2 of propiconazole was 12 to 15 d in turfgrass and 29 d in bare soil. Little movement of propiconazole was observed in either bare soil or turf.

P

ublic awareness of possible environmental contamination resulting from the use of pesticides in turfgrass systems has increased in recent years (Balogh and Anderson, 1992). However, turfgrass is a unique system in that, except at establishment, pesticides are applied directly to the plant material and thatch, a layer of dead and living stems and roots between the green vegetation and soil surface. Movement of turf-applied pesticides into the soil is attenuated by the high organic carbon content of turfgrass thatch (Branham and Wehner, 1985; Branham, 1994; Horst et al., 1996). Laboratory studies show that turfgrass leaves and thatch strongly sorb organic compounds and thus should have a significant effect on the fate of pesticides applied to turfgrass (Niemczyk et al., 1988; Dell et al., 1994; Lickfeldt and Branham, 1995). Sears and Chapman (1979) showed that even a thin layer of thatch could significantly retard pesticide movement on sandy soils. Retention of pesticides by thatch may result in reduced mobility of pesticides applied to turfgrass (Stahnke et al., 1991; Smith and Bridges, 1996).

MATERIALS AND METHODS Field Procedures Field experiments were conducted in ‘Penneagle’ and ‘Seaside II’ creeping bentgrass turf in 1999 at the University of Illinois Landscape Horticulture Research Center in Urbana, Illinois. The soil was a Flanagan silt loam (fine, smectitic, mesic Aquic Argiudoll; 52 g kg⫺1 organic matter, 1.38 g cm⫺3 bulk density, and pH 6.5). Very few macropores were detected throughout the study, based on visual examination of soil cores. Creeping bentgrass thatch was 10 to 14 mm thick and had a bulk density of 0.53 g cm⫺3, 241 g kg⫺1 organic matter, and pH 6.7. Bare soil plots were prepared by stripping the sod from the plot with a sod-cutter.

D.S. Gardner, Dep. of Horticultural and Crop Science, 2021 Coffey Rd., The Ohio State University, Columbus, OH 43210-1086. B.E. Branham, Dep. of Natural Resources and Environmental Sciences, 1102 S. Goodwin Ave., Univ. of Illinois, Urbana, IL 61801-4798. Part of a thesis by the senior author in partial fulfillment of the requirements for the Ph.D. degree at the University of Illinois. Mention of product and equipment trade names is for identification purposes only and does not imply a warranty or endorsement to the exclusion of other products that may be similar. Received 14 July 2000. *Corresponding author ([email protected]). Published in J. Environ. Qual. 30:1612–1618 (2001).

Abbreviations: DAT, days after treatment; t1/2, half-life.

1612

1613

GARDNER & BRANHAM: MOBILITY OF MEFENOXAM AND PROPICONAZOLE

Sampling cylinders were constructed of 20-cm-diam. Schedule 40 polyvinylchloride (PVC) pipe cut into 30-cm lengths and beveled at one end to ease insertion into the soil. Sampling cylinders were inserted into each plot on 11 June 1999 using a hydraulic press (Alden Enterprises, Okemos, MI) attached to a tractor. Mefenoxam (Subdue Maxx; Sygenta, Greensboro, NC) or propiconazole (Banner Maxx; Sygenta) was applied on 14 June 1999 at 770 g a.i. ha⫺1 and 1540 g a.i. ha⫺1, respectively. The pesticides were applied with a backpack sprayer equipped with a TEEJET 8006E (Spraying Systems Co., Wheaton, IL) nozzle at a height of 36 cm, with an effective spray width of 30 cm. The pesticides were applied in 1120 L water ha⫺1 at 276 kPa. Irrigation (4 mm) was applied to the plots immediately after treatment. The experimental area was mowed three times per week at 1.8 cm with a reel mower and clippings were removed. Half of the plots were irrigated with 10 mm of water five times per week (high irrigation schedule). The other plots (low irrigation schedule) were watered as necessary to replace 80% of estimated evapotranspiration (Table 1).

Sampling and Analysis of Pesticides Sampling cylinders were removed from three replicate blocks of each level of turf cover and irrigation 2 h after treatment and 4, 8, 16, 32, and 64 days after treatment (DAT). Sampling dates were 14 June, 18 June, 22 June, 30 June, 16 July, and 17 August. Verdure and thatch were separated from the cores that had turfgrass. The soil cores were partitioned into 0- to 1-, 1- to 3-, 3- to 5-, 5- to 15-, and 15- to 30-cm soil depth sections. Samples or subsamples were weighed, placed in glass mason jars with aluminum foil-capped lids, and stored at ⫺20⬚C until residue analysis. Pesticides were extracted from thawed samples by placing a representative 20-g sample (3–10 g for verdure) in a 500mL Erlenmeyer flask. Pesticides were extracted from the samples by shaking with ethyl acetate (100 mL) for 4 h (3 h for soil samples) on a platform shaker at 200 rpm. The extracts were vacuum-filtered by passing through a Whatman (Maidstone, UK) G6 glass fiber filter. Ethyl acetate was removed by rotary evaporation at 40⬚C. The evaporatory flask was rinsed three times with 5 mL methylene chloride and the rinsate transferred. The methylene chloride was evaporated to 2 mL using a reacti-vap (Pierce, Inc., Rockford, IL). The extract was then passed through a 0.45-␮m nylon membrane filter (Gelman Sciences, Ann Arbor, MI) and transferred to an autosampler vial for analysis on a high pressure liquid chromatograph (Beckman Coulter, Fullerton, CA). Mefenoxam and propiconazole were separated on a 15-cm, 4.6-mm-i.d. column with a bonded 5-␮m C-18 phase (Beckman Coulter). Mefenoxam was separated from verdure, thatch, and soil coextractives by injecting 40 ␮L into a mobile phase of 30:70 (acetonitrile to water). After 8 min the mobile phase was increased to 100% acetonitrile over a 10-min period. Mefenoxam was detected with a UV-Vis detector (Beckman Coulter) at 210 nm with a retention time of 14.3 min. Propiconazole was separated from verdure, thatch, and soil coextractives by injecting 40 ␮L into a mobile phase of 55:45 (acetonitrile to water). After 5 min the mobile phase was increased to 100% acetonitrile over a 6-min period. Propiconazole was detected with a UV-Vis detector at 210 nm with a retention time of 9.4 min. Mefenoxam residues were quantified by peak area measurements in comparison with a 10 ␮g mL⫺1 external standard. Propiconazole residues were quantified by peak height mea-

Table 1. Rainfall and irrigation from date of pesticide application to last sample collection date. Pesticides application date was 14 June 1999. Days after treatment

Irrigation regime Low

High

Days after treatment

Irrigation regime Low

mm 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

mm

2.5

2.5

5.1

12.7

7.6

12.7 7.6 15.2

15.2 7.6 12.7

15.2 20.3

12.7 2.5

12.7 2.5 5.1 7.6

7.6

15.2

12.7

7.6 10.2

10.2 10.2

High

12.7 12.7 12.7 10.2 10.2 10.2 10.2 10.2 10.2 10.2

33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64

10.2 10.2 27.9 5.1

10.2 10.2 27.9 5.1

7.6

17.8

10.2 20.3 27.9

15.2 25.4 27.9

10.2

10.2 10.2 10.2 20.3

96.5

10.2 10.2 20.3 10.2 10.2 10.2 96.5

surements in comparison with a 10 ␮g mL⫺1 external standard. The limit of detection for both pesticides was 10 ␮g kg⫺1. Calibration standards were included after every sixth sample. A control sample fortified at 1 mg kg⫺1 and a method blank were included with each batch of 22 samples. Mefenoxam recovery from soil, verdure, and thatch samples averaged 91% with a coefficient of variation of 6%. Propiconazole recovery from soil, verdure, and thatch samples averaged 90% with a coefficient of variation of 5%. The mass of mefenoxam or propiconazole remaining in each soil profile was estimated from the concentration of pesticide present in a soil core section and the mass of the core section. The experiment was designed as a split-split-plot with irriga-

Fig. 1. Total mefenoxam residue in verdure, thatch, and soil as a function of sampling time.

1614

J. ENVIRON. QUAL., VOL. 30, SEPTEMBER–OCTOBER 2001

Fig. 2. Distribution of mefanxoam residues among verdure, thatch, and different soil depths over time in 1999. Horizontal T lines represent standard error of the means.

tion as the whole plot, bare soil vs. turf cover as the sub-plot, and soil depth section as the sub-sub-plot. On each sampling date, data were subjected to analysis of variance (SAS Institute, 1990). Half-life values were estimated by regressing the log of pesticide residues remaining versus time.

RESULTS Dissipation of mefenoxam was rapid, regardless of the presence of turfgrass cover or amount of irrigation applied (Fig. 1). The calculated t1/2 in turfgrass was 6 ⫾

1615


Table 2. Analysis of variance (ANOVA) on total mefenoxam residues in verdure, thatch, and soil profile components on different sampling dates during 1999. Probability of greater F ratio (P ⬎ F ) for surface organic matter content and soil profile components. Source†

df Day 0 Day 4 Day 8 Day 16 Day 32 Day 64

(R)eplication 2 (I)rrigation 1 R ⫻ I (Error A) 2 (T)urf cover 1 I⫻T 1 R⫻I⫻T (Error B) 4 (S)oil depth section 6 T⫻S 4 I⫻S 6 I⫻T⫻S 4 R⫻I⫻T⫻S (Error C) 40

NS‡ NS

NS NS

NS NS

NS NS

NS NS

NS NS

** NS

** NS

** NS

NS NS

* NS

NS NS

*** *** NS NS

*** *** NS NS

*** *** NS NS

*** *** NS NS

NS NS NS NS

*** NS NS *

* Significant at the 0.05 probability level. ** Significant at the 0.01 probability level. *** Significant at the 0.001 probability level. † I ⫽ high or low irrigation, T ⫽ bare soil or creeping bentgrass turf, S ⫽ soil depth section (verdure, thatch, and 0 to 1, 1 to 3, 3 to 5, 5 to 15, and 15 to 30 cm). ‡ Not significant.

1 d (R2 ⫽ 0.99, P ⫽ 0.001) under high irrigation and 5 ⫾ 1 d (R2 ⫽ 0.99, P ⫽ 0.002) under low irrigation. The calculated t1/2 in bare soil was 8 ⫾ 4 d (R2 ⫽ 0.95, P ⫽ 0.022) under high irrigation and 7 ⫾ 2 d (R2 ⫽ 0.99, P ⫽ 0.006) under low irrigation. There was rapid vertical movement of mefenoxam through the soil profile, regardless of turfgrass cover or irrigation regime. Irrigation regime did not affect the distribution of mefenoxam in the soil on any sampling date. Differences in the distribution of mefenoxam residues in the soil layers due to turfgrass cover were observed at 2 h after treatment and 4, 8, and 16 DAT, but not at 32 or 64 DAT (Table 2). Most of the mefenoxam was found in thatch on plots containing turfgrass or in the 0- to 1-cm soil layer on bare soil plots at 0 and 4 DAT (Fig. 2). By 8 DAT, however, the majority of mefenoxam applied to turf had moved into the soil. Residues were found in the 15- to 30-cm soil depth at 4, 8, 16, 32, and 64 DAT, regardless of turfgrass cover or irrigation regime. However, due to rapid dissipation, the total amount of mefenoxam recovered on any date in the 1- to 3-, 3- to 5-, 5- to 15-, and 15- to 30-cm soil sections did not exceed 15% of the total amount applied. The t1/2 of propiconazole varied due to turfgrass cover (Fig. 3). The calculated t1/2 in turfgrass was 12 ⫾ 1 d (R2 ⫽ 0.99, P ⫽ 0.001) under high irrigation and 15 ⫾ 2 d (R2 ⫽ 0.99, P ⫽ 0.001) under low irrigation. The calculated t1/2 in bare soil was 29 ⫾ 12 d (R2 ⫽ 0.90, P ⫽ 0.003) under high irrigation and 29 ⫾ 13 d (R2 ⫽ 0.86, P ⫽ 0.007) under low irrigation. There were differences in the distribution of propiconazole in the soil layers due to turfgrass cover on all sampling dates (Table 3). Little vertical movement of propiconazole in the soil was observed (Fig. 4). When propiconazole was applied to turf, at least 95% of the residues were recovered from the verdure and thatch on all sampling dates. At least 87% of propiconazole residues applied to bare soil were recovered from the

Fig. 3. Total propiconazole residue in verdure, thatch, and soil as a function of sampling time in 1999.

0- to 1-cm soil layer. Small amounts of propiconazole were found in the 1- to 3-, 3- to 5-, and 5- to 15-cm soil layers when applied to bare soil. Propiconazole applied to turfgrass was detected in trace amounts in the 1- to 3- and 3- to 5-cm soil layers 16 DAT. No propiconazole applied to turfgrass was found below the 0- to 1-cm soil layer on any other sampling date.

DISCUSSION The total amount of irrigation and rainfall recorded on the plots throughout the study differed by nearly 2:1 (Table 1). However, only minor differences in soil mobility or dissipation due to precipitation were observed with mefenoxam. The only difference due to precipitation observed with propiconazole was a slightly shorter t1/2 in turfgrass under high irrigation compared with low irrigation. Total irrigation received was similar between the high and low irrigation regimes for the first 3 d of the study and this may have affected the results. Previous work has shown that soil mobility of certain Table 3. Analysis of variance (ANOVA) on total propiconazole residues in verdure, thatch, and soil profile components on different sampling dates during 1999. Probability of greater F ratio (P ⬎ F ) for surface organic matter content and soil profile components. Source†

df Day 0 Day 4 Day 8 Day 16 Day 32 Day 64

(R)eplication 2 (I)rrigation 1 R ⫻ I (Error A) 2 (T)urf cover 1 I⫻T 1 R⫻I⫻T (Error B) 4 (S)oil depth section 6 T⫻S 4 I⫻S 6 I⫻T⫻S 4 R⫻I⫻T⫻S (Error C) 40

NS‡ NS

NS NS

NS NS

NS NS

* NS

NS NS

** NS

** NS

** NS

** NS

** NS

** NS

*** *** NS NS

*** *** NS NS

*** *** NS NS

*** *** NS NS

*** *** NS NS

*** *** NS NS

* Significant at the 0.05 probability level. ** Significant at the 0.01 probability level. *** Significant at the 0.001 probability level. † I ⫽ high or low irrigation, T ⫽ bare soil or creeping bentgrass turf, S ⫽ soil depth section (verdure, thatch, and 0 to 1, 1 to 3, 3 to 5, 5 to 15, and 15 to 30 cm). ‡ Not significant.

1616


Fig. 4. Distribution of propiconazole residues among verdure, thatch, and different soil depths over time in 1999. Horizontal T lines represent standard error of the means.

pesticides applied to turfgrass is not affected by irrigation or precipitation events (Niemczyk and Krueger, 1987; Cisar and Snyder, 1996). However, Niemczyk and Krause (1994) found that the mobility of some preemergence herbicides was correlated with major rainfall events.

Because of its high water solubility and low soil Koc, irrigation and rainfall are expected to have a significant effect on leaching of mefenoxam. Metalaxyl leached through 40-cm-deep lysimeters that received 100 mm of precipitation (Odanaka et al., 1994). Horst et al. (1996) found at least 28% of applied metalaxyl in soil under Kentucky


bluegrass (Poa pratensis L.) moving through the entire 60-cm soil column. Metalaxyl movement through 25cm-deep soil columns was 0, 9, 73, and 83% of applied material after being subjected to 50, 100, 150, and 200 mm, respectively, of simulated rainfall (Sharom and Edgington, 1986). Starrett et al. (1996) found that depth and frequency of application of irrigation water can affect the movement of mobile pesticides such as mefenoxam. The authors found 7.7% of the applied metalaxyl in the leachate from soil columns containing Kentucky bluegrass that received heavy irrigation (four, 25.4-mm applications). In contrast, 0.2% was measured in leachate from the soil columns receiving light irrigation (16, 6.4-mm applications). In our study, irrigation regime was varied by frequency of application, not amount at application. It is likely that an irrigation regime consisting of fewer, larger amounts would have resulted in increased leaching of mefenoxam, as shown by Starrett et al. (1996). Turfgrass cover did not affect mobility of mefenoxam. In a previous study, metalaxyl recovered in leachate from plots with varying amounts of turfgrass cover was 36.4, 26.7, 14.1, and 16.5% on plots with turfgrass densities of 0, 33, 66, or 100%, respectively (Petrovic et al., 1996). However, based on our soil distribution data, the amount of mefenoxam leaching below the deepest sampling depth, if measured, would have been similar on plots with turfgrass or bare soil. The main reason for the high rate of leaching in the study by Petrovic et al. was the unrealistically high irrigation rate of 19.5 mm d⫺1. The t1/2 for mefenoxam was 5 to 8 d, which is in close agreement with previous studies in turf (Horst et al., 1996). Mefenoxam residues in turf were significantly less than in bare soil at both 4 and 8 DAT (Table 2), indicating more rapid microbial metabolism of mefenoxam in verdure and thatch than in soil. Sharom and Edgington (1982) reported that metalaxyl was stable in sterilized soils but degraded in unsterilized soils, indicating that degradation by microorganisms is an important factor governing the fate of mefenoxam. Since mefenoxam is not strongly bound by thatch, the higher microbial activity in thatch may be circumvented by more frequent irrigation. The t1/2 was 1 d longer in soil and turf under more frequent irrigation; however, this difference was not statistically significant. The t1/2 of propiconazole in this study was between 10 and 30 d, which is much lower than what was previously reported in Wauchope et al. (1991). Bai and Liu (1987) observed a t1/2 of about 5 d when propiconazole was applied to wheat (Triticum aestivum L.). Similar results were found by Garland et al. (1999), who observed halflives of 8 to 10 d when propiconazole was applied to leaves of a peppermint (Mentha ⫻ piperita L. nothosubsp. piperita) crop. Limited downward movement of propiconazole was observed, which agrees with previously published studies (Kim and Suh, 1998; Liu and Weber, 1986). Kim and Suh (1998) found 4.4% of applied propiconazole in leachate from a sandy loam soil after 16 wk. The majority (97%) of the material remained in the upper 20 cm of the profile. Similar results were reported by

1617

Liu and Weber (1986). When applied to wheat, more residues were found on straw and leaves than were found in soil (Bai and Liu, 1987). Other avenues of pesticide fate include runoff, volatilization, and photodegradation (Frederick et al., 1996). Our study was conducted on plots with little slope, which minimized the possibility for runoff losses. Volatilization can be a significant mode of loss of mefenoxam (Petrovic et al., 1996). However, volatilization of propiconazole is of minor significance (Balogh and Anderson, 1992). A study conducted by Petrovic et al. (1996) measured pesticide residues in leachate on plots with various amounts of turfgrass cover. The present study is one of the first to directly examine the mobility and dissipation of pesticides in turfgrass compared with bare soil, with climate and soil type held constant. Post-treatment irrigation practices may not be as important in determining the soil mobility of pesticides as is soil moisture at application time or large precipitation events within 14 d of application. The results of this study indicate that the effect of turfgrass cover or irrigation practices may be more important in determining leaching and dissipation of moderately mobile pesticides such as propiconazole. However, pesticides with high water solubilities and low soil Koc values, such as mefenoxam, are prone to leaching regardless of turf cover or precipitation. ACKNOWLEDGMENTS Financial support for this research was provided by the United States Golf Association, Far Hills, NJ.

REFERENCES Bai, Q., and C. Liu. 1987. Dissipation of propiconazole residues in and on wheat (Triticum aestivum L.). Pestic. Sci. 19:229–234. Balogh, J.C., and J.L. Anderson. 1992. Environmental impacts of turfgrass pesticides. In J.C. Balogh and W.J. Walker (ed.) Golf course management and construction: Environmental issues. Lewis Publ., Boca Raton, FL. Branham, B.E. 1994. Herbicide fate in turf. p. 109–151. In A.J. Turgeon (ed.) Turf weeds and their control. ASA and CSSA, Madison, WI. Branham, B.E., and D.J. Wehner. 1985. The fate of diazinon applied to thatched turf. Agron. J. 77:101–104. Cisar, J.L., and G.H. Snyder. 1996. Mobility and persistence of pesticides applied to a USGA green. III: Organophosphate recovery in clippings, thatch, soil, and percolate. Crop Sci. 36:1433–1438. Dell, C.J., C.S. Throssell, M. Bischoff, and R.F. Turco. 1994. Estimation of sorption coefficients for fungicides in soil and turfgrass thatch. J. Environ. Qual. 23:92–96. Frederick, E.K., C.S. Throssell, M. Bischoff, and R.F. Turco. 1996. Fate of vinclozolin in creeping bentgrass turf under two application frequencies. Bull. Environ. Contam. Toxicol. 57:391–397. Gardner, D.S., B.E. Branham, and D.W. Lickfeldt. 2000. Effect of turfgrass on soil mobility and dissipation of cyproconazole. Crop Sci. 40:1333–1339. Garland, S.M., R.C. Menary, and N.W. Davies. 1999. Dissipation of propiconazole and tebuconazole in peppermint crops [Mentha piperita (Labiatae)] and their residues in distilled oils. J. Agric. Food Chem. 47:294–298. Gold, A.J., T.G. Morton, W.M. Sullivan, and J. McClory. 1988. Leaching of 2,4-D and dicamba from home lawns. Water Air Soil Pollut. 37:121–129. Horst, G.L., P.J. Shea, N. Christians, D.R. Miller, C. Stuefer-Powell, and S.K. Starrett. 1996. Pesticide dissipation under golf course fairway conditions. Crop Sci. 36:362–370. Hurto, K.A., A.J. Turgeon, and M.A. Cole. 1979. Degradation of

1618


Benefin and DCPA in thatch and soil from a Kentucky bluegrass (Poa pratensis ) turf. Weed Sci. 27:154–157. Kim, I.S., and Y.T. Suh. 1998. Behaviour of fungicide 14C-propiconazole in a lysimeter of sandy loam. Han’guk Nonghwa Hakhoechi 41:253–257. Lickfeldt, D.W., and B.E. Branham. 1995. Sorption of nonionic organic compounds by Kentucky bluegrass leaves and thatch. J. Environ. Qual. 24:980–985. Liu, S.L., and J.B. Weber. 1986. Adsorption/desorption, and mobility of propiconazole and prometryn in soils. Proc. South. Weed Sci. Soc. 39:460–467. Niemczyk, H.D., Z. Filary, and H. Krueger. 1988. Movement of insecticides residues in turfgrass thatch and soil. Golf Course Management. February. p. 22–23. Niemczyk, H.D., and A. Krause. 1994. Behaviour and mobility of preemergence herbicides in turfgrass: A field study. J. Environ. Sci. Health. B29(3):507–539. Niemczyk, H.D., and H.R. Krueger. 1987. Persistence and mobility of isazofos in turfgrass thatch and soil. J. Econ. Entomol. 80:950–952. Odanaka, Y., T. Taniguchi, Y. Shimamura, K. Iijima, Y. Koma, T. Takechi, and O. Matano. 1994. Runoff and leaching of pesticides in golf course. Nippon Noyaki Gakkaishi 19:1–10. Petrovic, A.M., W.C. Barrett, I. Larsson-Kovach, C.M. Reid, and D.J. Lisk. 1996. The influence of a peat amendment and turf density on downward migration of metalaxyl fungicide in creeping bentgrass san lysimeters. Chemosphere 33:2335–2340.

SAS Institute. 1990. SAS/STAT user’s guide. Vol. 2. 4th ed. SAS Inst., Cary, NC. Sears, M.K., and R.A. Chapman. 1979. Persistence and movement of four insecticide residues applied to turfgrass. J. Econ. Entomol. 71:272–274. Sharom, M.S., and L.V. Edgington. 1982. The adsorption and persistence of metalaxyl in soil and aqueous systems. Can. J. Plant Pathol. 4:334–340. Sharom, M.S., and L.V. Edgington. 1986. Mobility and dissipation of metalaxyl in tobacco soils. Can. J. Plant Sci. 66:761–771. Smith, A.E., and D.C. Bridges. 1996. Movement of certain herbicides following application to simulated golf course greens and fairways. Crop Sci. 36:1439–1445. Stahnke, G.K., P.J. Shea, D.R. Tupy, and R.N. Stougaard. 1991. Pendimethalin dissipation in Kentucky bluegrass turf. Weed Sci. 39: 97–103. Starrett, S.K., N.E. Christians, and T.A. Austin. 1996. Movement of pesticides under two irrigation regimes applied to turfgrass. J. Environ. Qual. 25:566–571. Sukul, P., and M. Spiteller. 2000. Metalaxyl: Persistence, degradation, metabolism, and analytical methods. Rev. Environ. Contam. Toxicol. 164:1–26. Wauchope, R.D., T.M. Butler, A.G. Hornsby, P.W.M. AugustijnBeckers, and J.P. Burt. 1991. The SCS/ARS/CES pesticide database for environmental decision-making. Rev. Environ. Contam. Toxicol. 123:1–155.

Prediction of Soil Organic Partition Coefficients by a Soil Leaching Column Chromatographic Method Feng Xu, Xinmiao Liang,* Bingcheng Lin, Fan Su, Karl-Werner Schramm, and Antonius Kettrup ABSTRACT The soil organic partition coefficient (Koc ) is one of the most important parameters to depict the transfer and fate of a chemical in the soil–water system. Predicting Koc by using a chromatographic technique has been developing into a convenient and low-cost method. In this paper, a soil leaching column chromatograpy (SLCC) method employing the soil column packed with reference soil GSE 17201 (obtained from Bayer Landwirtschaftszentrum, Monheim, Germany) and methanol–water eluents was developed to predict the Koc of hydrophobic organic chemicals (HOCs), over a log Koc range of 4.8 orders of magnitude, from their capacity factors. The capacity factor with water as an eluent (kⴕw) could be obtained by linearly extrapolating capacity factors in methanol–water eluents (kⴕ) with various volume fractions of methanol (␸). The important effects of solute activity coefficients in water on kⴕw and Koc were illustrated. Hence, the correlation between log Koc and log kⴕw (and log kⴕ) exists in the soil. The correlation coefficient (r ) of the log Koc vs. log kⴕw correlation for 58 apolar and polar compounds could reach 0.987, while the correlation coefficients of the log Koc–log kⴕ correlations were no less than 0.968, with ␸ ranging from 0 to 0.50. The smaller the ␸, the higher the r. Therefore, it is recommended that the eluent of smaller ␸, such as water, be used for accurately estimating Koc. Correspondingly, the r value of the log Koc–log kⴕw correlation on a reversed-phase Hypersil ODS (Thermo Hypersil, Kleinostheim, Germany) column was less than 0.940 for the same solutes. The SLCC method could provide a more reliable route to predict Koc indirectly from a correlation with kⴕw than the reversed-phase liquid chromatographic (RPLC) one. F. Xu, X. Liang, B. Lin, and F. Su, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 161 Zhongshan Road, Dalian 116011, P.R. China. K.-W. Schramm and A. Kettrup, GSF-National Research Center for Environment and Health, Institute of Ecological Chemistry, Ingolsta¨dter Landstrasse 1, D-85764 Neuherberg, Germany. Received 22 Nov. 2000. *Corresponding author (liangxm@ mail.dlptt.ln.cn). Published in J. Environ. Qual. 30:1618–1623 (2001).

T

he soil organic partition coefficient in soil–water systems (Koc ) plays a significant role in modeling the sorption behavior of a pollutant in soil and eventually leaching to ground water (OECD, 1996). The term Koc is defined as the concentration ratio of a solute sorbed onto the soil phase and in the aqueous phase at equilibrium state. The batch equilibrium method was a conventional method for Koc measurement, but it would generally take 4 to 96 h or longer to reach equilibrium. So the method is not suitable for rapid screening of lots of compounds at one time. As an alternative, many indirect RPLC methods have been proposed in which liquid chromatographic packing materials, such as octadecyl-, octyl-, cyanopropyl-, and phenyl-silica (Ko¨rdel et al., 1995; Kaune et al., 1998) were adopted to determine the capacity factor (k⬘) through a chromatogram and then to estimate Koc from log Koc–log k⬘ correlations. These methods are based on the fact that the compound with a higher Koc value elutes later on some stationary phases than the compound with a lower Koc value. Because of the large discrepancy existing between RPLC packing phases and soil (in the batch method), the Koc estimation is not completely satisfactory, especially for compounds of different classes. Therefore, other packing materials such as immobilized humic acid, which to some degree imitate soil, were proposed (Pussemier et al., 1994). A column packed with a naturally occurring soil should be the most appropriate material in the log Koc– Abbreviations: HOC, hydrophobic organic chemical; RPLC, reversedphase liquid chromatography; SLCC, soil leaching column chromatography.