JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gaynor, J.D.
Right arrow Articles by Reynolds, W.D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gaynor, J.D.
Right arrow Articles by Reynolds, W.D.
Agricola
Right arrow Articles by Gaynor, J.D.
Right arrow Articles by Reynolds, W.D.
Related Collections
Right arrow Agricultural Pesticides
Right arrow Surface Water Quality
Right arrow Other Cropping Systems
Right arrow Other Environmental Contamination
Right arrow Water Pollution
Journal of Environmental Quality 31:300-308 (2002)
© 2002 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America

TECHNICAL REPORT
Surface Water Quality

Runoff and Drainage Losses of Atrazine, Metribuzin, and Metolachlor in Three Water Management Systems

J.D. Gaynor*,a, C.S. Tana, C.F. Drurya, T.W. Welackya, H.Y.F. Nga,b and W.D. Reynoldsa

a Crops Research Centre, Agriculture and Agri-Food Canada, 2585 County Road E, Harrow, ON, N0R 1G0 Canada
b National Water Research Institute, Environment Canada, Burlington, ON, L7R 4A6 Canada

* Corresponding author (gaynorj{at}em.agr.ca)

Received for publication January 22, 2001.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
Rainfall can transport herbicides from agricultural land to surface waters, where they become an environmental concern. Tile drainage can benefit crop production by removing excess soil water but tile drainage may also aggravate herbicide and nutrient movement into surface waters. Water management of tile drains after planting may reduce tile drainage and thereby reduce herbicide losses to surface water. To test this hypothesis we calculated the loss of three herbicides from a field with three water management systems: free drainage (D), controlled drainage (CD), and controlled drainage with subsurface irrigation (CDS). The effect of water management systems on the dissipation of atrazine (6-chloro-N2-ethyl-N4-isopropyl-1,3,5-triazine-2,4-diamine), metribuzin [4-amino-6-(1,1-dimethylethyl)-3-(methylthio)-1,2,4-triazine-5(4H)-one), and metolachlor [2-chloro-N-(2-ethyl-6-methylphenyl)-N-(2-methoxy-1-methylethyl)acetamide] in soil was also monitored. Less herbicide was lost by surface runoff from the D and CD treatments than from CDS. The CDS treatment increased surface runoff, which transported more herbicide than that from D or CD treatments. In one year, the time for metribuzin residue to dissipate to half its initial value was shorter for CDS (33 d) than for D (43 d) and CD (46 d). The half-life of atrazine and metolachlor were not affected by water management. Controlled drainage with subsurface irrigation may increase herbicide loss through increased surface runoff when excessive rain is received soon after herbicide application. However, increasing soil water content in CDS may decrease herbicide persistence, resulting in less residual herbicide available for aqueous transport.

Abbreviations: CD, controlled drainage • CDS, controlled drainage with subsurface irrigation • D, free drainage • DAA, days after application


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
HERBICIDES are most vulnerable for transport when excess rain is received soon after their application (Wauchope, 1978; Gaynor et al., 1995). Their presence in surface water can have a deleterious effect on aquatic organisms or contaminate drinking water supplies (Frank et al., 1990; Thurman et al., 1991). Measures to reduce herbicide losses from their site of application are needed.

Tile drainage can reduce herbicide transport by surface runoff. For example, tiling a silty clay loam reduced losses of metolachlor by 90% and trifluralin by 57% compared with no tile drains (Southwick et al., 1997). In an earlier study, Southwick et al. (1990) reported that tile drains reduced metolachlor and atrazine losses by 56% compared with no tile drainage because of a 38% reduction in surface runoff volume. Installation of tile drains reduces herbicide loss by channelling surface runoff to tile drainage, which increases herbicide adsorption to soil, thereby promoting degradation by soil microorganisms or chemical reactions. The effectiveness of tile drains to mitigate herbicide loss in runoff is a function of management practice, soil type, soil hydraulic properties, herbicide solubility, and environmental factors (Jury, 1986).

Water table control is an emerging technology to increase yield, improve nutrient usage by crops, and reduce off-site movement of herbicides (Tan et al., 1993; Drury et al., 1996; Jebellie et al., 1996). Some beneficial effects of water table control have been reported. On a loam soil, water table control at 15 cm depth reduced chemical loss by 45% compared with a water table at 60 cm depth because of reduced tile drainage (Sarwar and Kanwar, 1996). Tile depth and spacing affect drainage volumes with greater drainage from tile at narrow spacing, and from deep rather than shallow depth (Bolton and Hore, 1976; Kladivko et al., 1991, 1999). Drury et al. (1996) and Amatya et al. (1998) found a 43 to 47% reduction in nitrate or total Keldahl nitrogen (TKN) loss with water table control compared with no water management. Most of this reduction occurred in the noncrop period from October to April.

Essex county in southwestern Ontario has a land area of 130030 ha, which is primarily fine-textured soils with surface slopes < 1%. About 89% of the land area is systematically tile drained. Natural rain of 763 to 845 mm is equally distributed through the year; however, dry periods in May, June, and July frequently reduce crop yields (Drury and Tan, 1995). Thus, water table control has potential in Essex county to preserve soil moisture and thereby increase crop yield. Our objective was to monitor the effect of water management system on water quality as it relates to herbicide loss. Water quality from two water management treatments, controlled drainage (CD) and controlled drainage with subsurface irrigation (CDS), was compared with that from free drainage (D). We also monitored the effect of these treatments on persistence in the soil of the three herbicides, atrazine, metribuzin, and metolachlor as well as the metabolite des-ethyl atrazine [6-chloro-N-(1-methylethyl)-1,3,5-triazine-2,4-diamine].


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
The soybean [Glycine max (L.) Merr.]–corn (Zea mays L.) rotation study began in 1995 on a Brookston clay loam (fine, loamy, mixed, mesic, Typic Argiaquoll). Previous to this, the plots had been in continuous corn since 1991 and had received annual applications of atrazine (1.1 kg a.i. ha-1), metribuzin (0.5 kg a.i. ha-1), and metolachlor (1.68 kg a.i. ha-1) (Drury et al., 1996; Gaynor et al., 2000, 2001). The Ap horizon has an acidic pH (5.5) and contains 16 g kg-1 organic matter in the top 30 cm. The B horizon has a clay texture to 1.5 m, a pH of 6.5 and contains 2 g kg-1 organic matter.

The experimental site consists of fifteen 0.1-ha plots with two 104-mm-diameter tile drains per plot installed before 1965 at a depth of 0.6 m and spacing of 7.5 m. Each plot was isolated by (i) a berm surrounding all sides to contain surface runoff, (ii) double-layer 4-mil thick plastic barrier from the surface to a depth of 1.2 m to prevent leakage and subsurface interaction between adjacent treatments, and (iii) a 7.5-m-wide by 67-m-long buffer area with a single tile drain at 0.6 m to prevent cross contamination between plots. Treatments, arranged in a completely randomized design within the 15 plots, included (i) free drainage (D), (ii) controlled drainage (CD), and (iii) controlled drainage with subsurface irrigation (CDS). Each drainage treatment had two nitrogen (N) rates, N1 and N2. For soybean, N1 was 0 kg ha-1 and N2 50 kg ha-1 as ammonium nitrate. For corn, N1 was 150 and N2 200 kg ha-1 as urea at side dress at the six leaf stage. Corn received an initial side dress of 142 kg ha-1 18–46–0 (N–P2O5–K2O) at planting. There were three replicates for the N1 and two replicates for the N2 treatments, respectively, within the three water management treatments. Nitrogen rate had no effect on herbicide results, and therefore each water table treatment had five replicates (i.e., the results for the three replicates of N1 and two replicates of N2 treatments were combined).

Soybean, cultivar A2615, was planted at 97.4 kg ha-1 with a Great Plains (Great Plains Manufacturing, Assaria, KN) no-till drill in 7.5-cm-wide rows on 23 May 1995. Corn (Pioneer 3515) was seeded 28 May 1996 at a rate of 79700 seeds ha-1 in 75-cm-wide rows with a Kinze (Kinze Manufacturing Co., Williamsburg, IA) four row planter. No tillage, other than that associated with the planting and fertilizer operation, was employed.

Surface runoff, subirrigation, and tile drainage volumes were recorded continuously from January 1995 to May 1997 by Neptune T-10 water metres (Soultani et al., 1993). Water samples for herbicide analysis were collected automatically (Calypso 2000S; Buhler Gmbh and Co., Geneq Inc., Weston, ON) at preset volumes to provide representative changes in analyte concentration during each runoff event. For example, during the growing season, one sample was collected for each 500 L (0.05 mm) of surface runoff or 1500 L (0.15 mm) of tile drainage. Risers, installed on tile drain outlets in CD and CDS treatments, controlled the water table at a nominal depth of 30 cm. Risers were installed after planting, removed before harvest, and reinstalled after harvest to ensure that planting and harvesting operations would not be impeded by wet soil conditions. Subsurface irrigation on the CDS treatments was initiated when the water table receded to less than 30 cm at the riser, whereas no subsurface irrigation was added to the CD treatment. Excess rain was drained from the plots when the water table was greater than the preset riser height.

Weed control in soybean was provided by metribuzin (0.56 kg a.i. ha-1) and metolachlor (1.92 kg a.i. ha-1), which were applied, preemergence broadcast, on 26 May 1995. Corn was planted in 1996 and weed control was provided by atrazine (1.1 kg a.i. ha-1), metribuzin (0.5 kg ha-1) and metolachlor (1.68 kg ha-1), applied preemergence broadcast to all plots on 31 May. Herbicides were applied as a tank mix in 180 L water ha-1 at 210 kPa from saddle tanks mounted on a Ford 4600 tractor. The spray boom consisted of six 11004 VS Tee Jet XR nozzles (Sprayer Systems Co., Wheaton, IL) spaced at 51 cm and at a height of 30 cm from the ground.

After a runoff-producing rain event, water samples were collected and stored at 4°C until they were analyzed, typically within 30 d of collection. Previous studies found no change in herbicide concentration in samples stored up to six months. A suitable aliquot of runoff or tile drainage (100 to 500 mL) was filtered through a 0.45-µm filter (Gelman Cat GN-6; Gelman Sciences, Rexdale, ON). Herbicide was extracted from the water on a preconditioned cyclohexyl Sep Pak cartridge (Baker Cat. no. 7212-03; Mallinckrodt Baker, Phillipsburg, NJ). After loading, the cartridge was air-dried and then eluted with 1.5 mL methanol for analysis on a Varian 3400 gas chromatograph (Varian Canada, Mississauga, ON). Analytes were separated on a 15-m DB-5 capillary column (J&W Scientific, Folsom, CA) temperature programmed from 80 to 230°C at 20°C min-1 then held for 12.5 min. A thermionic sensitive detector at 300°C operated in N mode was used to detect the analytes. Herbicide mass in runoff or drainage was calculated as the sum of the herbicide concentration in the sample aliquot multiplied by the corresponding volume of runoff or drainage. Recovery of the analytes from fortified runoff or drainage (0.1 and 1 µg L-1, n = 2) was 80%. Results were not corrected for recovery. Our limit of detection in water varied with season and volume of water analyzed but was calculated (three times the standard deviation of samples) to be about 0.01 µg L-1 in spring samples. This was the minimum concentration used to calculate mass for events where concentrations were near the limit of detection.

Soil from each plot was sampled at regular intervals during the growing season with a 2.54-cm-diameter probe containing an acetate sleeve. Composite samples of 21 cores (14 adjacent to the tile and 7 between the tile) were taken from 0- to 10-, 10- to 15-, and 15- to 20-cm depths. The samples were stored at -10°C and analyzed within 30 d of collection. The samples from each depth were homogenized by hand and a representative aliquot of soil was extracted for 1 h on a New Brunswick Model V shaker (New Brunswick Scientific, New Brunswick, NJ) in 9:1 (v/v) methanol and water and then filtered through a Whatman (Maidstone, UK) No. 5 filter. An additional subsample was used for gravimetric water content determination. The extracts were reduced to about 10 mL on a Buchler evaporator (PTFE-1-6; Fisher Scientific, Nepean, ON), diluted to 100 mL with distilled water, extracted on a cyclohexyl Sep Pak cartridge, and analyzed as described for water analysis. Values are reported on a dry weight bases and were not corrected for extraction efficiency (80%) as calculated by extraction of fortified soil (60 to 500 µg kg-1, n = 3). Limit of detection was 0.01 µg kg-1.

Significance of water management treatments for the quantities measured were determined from Proc GLM in SAS (SAS Institute, 1989). The main effect of nitrogen rate for runoff, drainage, and herbicide loss was nonsignificant as was the N by water management interaction (p > 0.05), so the data were reanalyzed with five replicates for water management effect. When model main effects were significant (p < 0.05), differences among means for water management systems and herbicide residues in soil were identified by the least significant difference routine in SAS. Half-life of each herbicide in the top 10 cm of soil was calculated from the rate constant (k) derived using Proc REG from the log normal form of the first order rate equation: log C = kt + log Co, where C and Co = soil residue at time t (days after application) and to.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSION
 REFERENCES
 
Annual rain was 20% below the 10-yr average (763.1 mm) in 1995 and 6% below the average in 1996 (Table 1). To aid in presentation and discussion of the data, results are presented in terms of collection event intervals. These event intervals are characterized by the collection of samples following a period of sufficient rain to produce either surface runoff or tile drainage. Event intervals in relation to daily rainfall are presented in Fig. 1 with event intervals, accumulated rain, and number of samples collected from the 15 plots listed in Table 2. It is evident from Fig. 1 that several rainfall events may be included in an event interval, especially if they occurred within a few days of each other. The first event (Event 1, Table 2) following herbicide application had 30% less rain in 1995 and 13% less rain in 1996 than the average (76.7 mm, Table 2) rainfall for the period. These lower amounts of rain contributed to smaller annual losses (less than 0.8% of application) of the herbicides applied to the treatments compared with years of greater rain in the weeks following herbicide application (Gaynor et al., 1992, 1995, 2000). Herbicide loss is highly dependent upon several environmental factors including soil water content at the time of herbicide application, incidence of rain relative to time of herbicide application, and rain intensity (Gaynor et al., 1992, 1995, 2000; Zhang et al., 1997; Wauchope, 1978).


View this table:
[in this window]
[in a new window]
  		Table 1. Monthly rain and 10-yr average during the growing season at Woodslee, ON for 1995 and 1996.

 


View larger version (23K):
[in this window]
[in a new window]
  		Fig. 1. Daily rainfall at Woodslee, Ontario for 1995 and 1996. Numbers correspond to rain events in Table 2 and arrows indicate herbicide application.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Dates for collection of samples after rain events producing runoff or drainage, number of runoff and drainage samples collected during the event interval, days from herbicide application, accumulated rainfall within the event interval, and 10-yr average rainfall for the accompanying event interval. Event 1 is the first collection after herbicide application in 1995 and 1996.

 
Water Management
Subsurface irrigation for the CDS treatment occurred from 19 June to 7 Sept. 1995 and 8 July to 7 Sept. 1996. On average the CDS treatment received 109.0 and 101.6 mm of subsurface irrigation in 1995 and 1996, respectively. During these intervals, there were 196.0 and 71.25 mm of rainfall, respectively. Thus, the CDS treatments received 20 to 38% more water during the growing season (May to August) than the 10-yr average rainfall. In 1995, little runoff (<0.2 mm) occurred in the first event after herbicide application whereas >1.6 mm of runoff occurred in the corresponding event for 1996 (Fig. 2 and 3 , bottom graphs). Less tile drainage (69 to 155 mm) was recorded in 1995 than in 1996 (135 to 256 mm, Table 3). Subsurface irrigation contributed to greater surface runoff from the CDS treatment than from the controlled drainage (CD) and drained (D) treatments (Table 3). Tile drainage was the reverse of surface runoff in that CDS had the least drainage of the three water management treatments, although the effect was only significant in 1995.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 2. Cumulative herbicide loss in runoff and drainage (mg or g ha-1) and accompanying depth of runoff and drainage (mm) for each rain event from three water management treatments in 1995.

 


View larger version (37K):
[in this window]
[in a new window]
  		Fig. 3. Cumulative herbicide loss in runoff and drainage (g ha-1) and accompanying depth of runoff and drainage (mm) for each rain event from three water management treatments in 1996.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Depth of runoff and drainage and herbicide loss in surface runoff and tile drainage from three drainage control scenarios.

 
Herbicide Losses
Atrazine
No atrazine was applied to the treatments in 1995. However, small but significant amounts of residual atrazine and des-ethyl atrazine (0.03 and 0.01 g ha-1) from previous applications were found (Table 3). Atrazine and des-ethyl atrazine from previous applications have consistently been reported in tile drainage (Jayachandran et al., 1994) and piezometers (Kalita et al., 1997). Most of the atrazine and des-ethyl atrazine loss occurred in the third and fourth runoff or third to fifth drainage events in 1995 (Fig. 2). In 1995, there was significantly more des-ethyl atrazine in drainage from the D treatment than from the CDS and CD treatments (Table 3). This was probably related to greater herbicide residues in the D treatment relative to the CDS and CD treatments, although we were unable to detect differences (p > 0.05) in soil atrazine or des-ethyl atrazine residues among water management treatments for any of the sampling dates.

There was a significant decline in atrazine and des-ethyl atrazine residues among the different sampling dates (p < 0.001) in the 0- to 10-cm depth (Table 4). Residues of both atrazine and des-ethyl atrazine were detected to 20 cm with a gradual decrease in des-ethyl atrazine residues (p < 0.005) with time in the 10- to 15-cm depth. At this depth atrazine residues also decreased with time (p < 0.003) in the D and CD treatments but no differences among residues were detected for the CDS treatment. At the 15- to 20-cm depths, atrazine and, to a lesser extent, des-ethyl atrazine residues increased with time for the three water management treatments but the differences were only significant for the CD treatment (p = 0.03). The increase in residues may reflect leaching and slower degradation at the deeper depth. It is well known that the rate of herbicide degradation is slower in subsoil than topsoil (Walker and Welch, 1989). Jebellie et al. (1996) reported longer persistence of atrazine in soil at a soil water content of 20% than at 35 or 50%. The D treatment would be expected to have a lower overall soil water content than the CD or CDS treatments because tile drainage was unrestricted in the D treatment. Jayachandran et al. (1994) and others (Adams and Thurman, 1991) found an increase in the ratio of des-ethyl atrazine to atrazine with time from application, indicating the importance of this metabolite to the total triazine load.


View this table:
[in this window]
[in a new window]
  		Table 4. Atrazine and des-ethyl atrazine concentration in soil with three water management treatments, 1995 and 1996.

 
Eighty-nine percent of the total annual triazine loss in 1996 occurred in the first event after application (Fig. 3). The D and CD treatments transported similar proportions of atrazine in surface runoff (43%) and tile drainage (55%) (Fig. 3). A greater proportion of des-ethyl atrazine was transported in tile drainage (62%) than in surface runoff (35%) from these treatments. In the CDS treatment, a larger proportion of this herbicide and its metabolite were transported in surface runoff (59 to 67%) than in tile drainage (41 to 33%). Total triazine transport for the two years was significantly greater (p < 0.008) from the CDS treatment than from the D and CD treatments, mainly as a consequence of increased surface runoff (Table 3). Des-ethyl atrazine is the major degradation product of atrazine (Brouwer et al., 1990) and its greater occurrence in the CDS treatment compared with the D and CD treatments would be expected because of its higher soil water content from subirrigation. This metabolite has lower adsorption affinity for soil or sediments than atrazine (Adams and Thurman, 1991; Mersie and Seybold, 1996), and consequently it is readily found in rivers and ground water (Thurman et al., 1991).

Kalita et al. (1997) measured atrazine concentration in two soils with and without water management. They found lower atrazine concentrations in piezometers from fields where water tables were higher (0.25 to 0.30 m below surface) than from fields maintained at deeper water table depths (2.0 to 1.5 m below surface). In their study, atrazine concentrations decreased with subsurface irrigation, probably because of dilution and/or enhanced dissipation. Liaghat and Prasher (1996) investigated the effect of grass cover and water management on reduction of atrazine concentration in contaminated water. Atrazine concentration in drainage was lowest from grass cover with subsurface drainage than from bare soil with controlled drainage, presumably because of the greater water storage capacity available in the drained treatment. Jebellie et al. (1996) recovered a significantly larger atrazine concentration from soil maintained at 20% than 50% water content. Atrazine concentration in soil is significantly reduced at higher water contents (Walker, 1991). In this study, the half-life for atrazine in 1996, calculated from the first order regression coefficient, was 27 d with no differences among water management treatments.

Residues of atrazine and des-ethyl atrazine were largest in the 0- to 10-cm soil depth than at deeper depths in both 1995 and 1996 (Table 4). Atrazine residues decreased with time (p < 0.007) for all water table treatments and depths but leaching was evident in 1996 from an increase in concentration from 2.28 to 6.28 µg kg-1 between 59 and 110 days after application (DAA) in the 10- to 15-cm soil depth. An increase in residues of atrazine was also noted in the 15- to 20-cm depth at 26 and 110 DAA compared with the previous sampling values.

Metribuzin
Metribuzin was applied in both years. In 1995, more metribuzin was lost in tile drainage than surface runoff because little surface runoff occurred in the first event after application (Fig. 2). In absence of an early runoff event after herbicide application, water management had little effect on metribuzin loss (p > 0.05, Table 3), which would be expected because of its fast rate of dissipation and low affinity for adsorption by this soil (Gaynor et al., 2000). Ninety-six percent of total metribuzin loss (4.2 to 4.7 g ha-1) occurred in the first event after application.

Metribuzin rapidly dissipated to less than 30 µg kg-1 in the 0- to 10-cm soil depth within 41 DAA (Table 5). The half-life of metribuzin was 14 d in 1995 and soil residues were not affected by water management at any depth within each sampling time (p > 0.05) except for 13 DAA at 10 to 15 cm (p = 0.04). At this sampling date, more metribuzin (2.63 µg kg-1) was recoved from D than CD or CDS (1.15 µg kg-1). Metribuzin was found to 20 cm but residues decreased with time since application (p < 0.05).


View this table:
[in this window]
[in a new window]
  		Table 5. Metribuzin concentration in soil with three water management treatments, 1995 and 1996.

 
In 1996, a greater proportion of metribuzin was lost in surface runoff than tile drainage from the CDS treatment (Fig. 3), while D and CD treatments transported similar proportions of metribuzin in runoff and drainage. The total loss of metribuzin from the CDS (5.9 g ha-1) was greater than from the D or CD treatments (2.5 and 3.3 g ha-1, respectively; Table 3) because of surface runoff. Most metribuzin loss occurred in the first event after application.

Metribuzin residues in the 0- to 10-cm depth decreased with time (p < 0.001) in 1996 for all water table treatments (Table 5). Although a high level of significance was calculated for the DAA main effect from the analysis of variance (ANOVA) for metribuzin residues with time at this depth, mean separation using the LSD test did not detect differences in metribuzin residues from 11 to 356 DAA. Metribuzin residues differed among sampling dates in the 10- to 15- and 15- to 20-cm depths (p < 0.01) but higher residues were found at 110 and 356 DAA compared with previous residues because of leaching (Table 5). Residues were found to 20 cm soil depth with a slight increase at 110 DAA after an accumulated rainfall of 144 mm. Jebellie and Prasher (1998) recovered more metribuzin from soil maintained under free drainage than from treatments where water table was controlled at 0.4 or 0.8 m from the soil surface. In another study, metribuzin concentrations were significantly higher (p < 0.009) in soil incubated at 20% than 50% water content (Jebellie et al., 1996).

The half-life of metribuzin at the 0- to 10-cm depth in 1996 was significantly longer (p = 0.05) in the D and CD treatments (45 d) than in CDS (33 d). Metribuzin appeared to be more persistent (half-life = 41 ± 7.4 d) than atrazine (25 ± 1.6 d) or metolachlor (36 ± 5.9 d), contrary to what was reported in previous studies at this field site. For example, Gaynor et al. (2000) found that the rate of metribuzin dissipation in this soil was somewhat faster than that of atrazine (29 vs. 45 d) or metolachlor (59 d) based on data collected from 1992 to 1994. This was attributed to the fact that metribuzin has a lower adsorption affinity than atrazine or metolachlor for this soil and is readily translocated to deeper depths with rain (Gaynor et al., 2000).

Metolachlor
As with metribuzin, metolachlor loss in 1995 occurred primarily (91% of total loss) through tile drainage (Table 3). Water management had no effect on metolachlor transport, most of which (92%) occurred in the first event after herbicide application (Fig. 2). Total loss averaged 10.2 g ha-1, representing less than 0.5% of applied. On average, less metolachlor was lost in 1996 (8.5 g ha-1) than in 1995. In 1996, a greater proportion of the total mass of metolachlor (p < 0.007) was transported in surface runoff from the CDS (69%) than the D (46%) or CD (54%) treatments compared with the proportion transported in drainage (Fig. 3). This resulted in more total metolachlor transported from this treatment relative to the other water management treatments (Table 3). The first event after herbicide application in 1996 transported 67 to 77% of the total herbicide loss from all water management treatments (Fig. 3).

Soil metolachlor residues in 1995 and 1996 gradually declined with time (p < 0.001) in the 0- to 10-cm depth for all water table treatments. Metolachlor residues at the 10- to 15- and 15- to 20-cm depths in 1995 were not significantly different with time except in the D treatment (p < 0.03). In 1996, metolachlor residues differed among sampling times (p < 0.007) for all water management treatments except CD in the 15- to 20-cm depth (p = 0.06). There was an increase in metolachlor concentration (p < 0.001) at 110 DAA in 1996 in the 10- to 15-cm depth (Table 6) as a result of leaching after a high rainfall event but no increase was noted in the 15- to 20-cm depth at this date. Metolachlor residues decreased with depth as the herbicide moved to the tile drains. Water management had no effect on metolachlor dissipation at any depth within each of the sampling times (p > 0.05), and half-lives of 18 and 36 d were calculated for 1995 and 1996, respectively, in the 0- to 10-cm depth. This is within the range that has been reported for previous years at this site (Gaynor et al., 2000). Jebellie et al. (1996) also failed to detect a difference in metolachlor persistence with soil water content. Metolachlor concentration was similar in soil incubated at 20, 35, and 50% water content. In contrast, Walker (1991) found less persistence of metolachlor with increasing soil water content.


View this table:
[in this window]
[in a new window]
  		Table 6. Metolachlor concentration in soil with three water management treatments, 1995 and 1996.

 

   CONCLUSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
REFERENCES
 
Water management had no consistent effect on total herbicide loss in surface runoff and tile drainage. Controlled drainage with subsurface irrigation altered the source of herbicide transport through increased surface runoff and herbicide loss in surface runoff but reduced drainage and herbicide loss in tile drainage. The net result was no significant change in total herbicide loss with the exception of an increase in atrazine and des-ethyl atrazine loss. We theorized that the CDS treatment increased the water content of the subsoil, thereby reducing infiltration of rain and increasing surface runoff and herbicide transport. In this study, drain tile was installed at 0.6 m with water table control at 0.3 m. Tile at 0.9 m depth with water table control at 0.6 m would provide greater water storage capacity in the soil profile and possibly reduce surface runoff and herbicide loss compared with no drainage control (Kalita et al., 1997). Herbicide loss in runoff, drainage, and the sum of runoff and drainage was similar between D and CD. Natural rainfall was insufficient to alter the soil moisture content in the CD treatment to cause a significant effect on herbicide loss between the D and CD treatments. It was hypothesized that the greater soil water content in the CDS treatment would enhance herbicide degradation as noted by a reduction in half-life. However, only the half-life of metribuzin was reduced by the CDS treatment compared with the D and CD treatments and this only occurred in 1996. The failure to detect a change in half-life with water management treatment may have been a consequence of our sampling protocol, where one-third of our samples were collected between the tile drains and two thirds near the drains. The low hydraulic conductivity of this soil (<0.1 mm d-1) may have increased spatial variability of soil water content by subirrigation, reducing the ability to detect a change in half-life of the herbicides.


   ACKNOWLEDGMENTS
 
Appreciation is expressed to the following individuals: M. Soultani, V. Bernyk, D. MacTavish, D. Pohlman, G. Stasko, T. Oloya, K. Rinas, J. St. Denis, S. Mannell, W. McLean, A. Szabo, M. Bissonnette, and S. Duransky, who contributed through technical, secretarial, and field support. Appreciation is further extended to Big "O" Inc., which donated some of the nonperforated pipes used in this project, and to Novartis Crop Protection Canada Inc. for analytical standards of the herbicides.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
M. Larose, G. C. Heathman, L. D. Norton, and B. Engel
Hydrologic and Atrazine Simulation of the Cedar Creek Watershed Using the SWAT Model
J. Environ. Qual., March 1, 2007; 36(2): 521 - 531.
[Abstract] [Full Text] [PDF]

Home page
 Vadose Zone JHome page
K. Noborio, R. G. Kachanoski, and C. S. Tan
Solute Transport Measurement Under Transient Field Conditions Using Time Domain Reflectometry
Vadose Zone J., March 8, 2006; 5(1): 412 - 418.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gaynor, J.D.
Right arrow Articles by Reynolds, W.D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gaynor, J.D.
Right arrow Articles by Reynolds, W.D.
Agricola
Right arrow Articles by Gaynor, J.D.
Right arrow Articles by Reynolds, W.D.
Related Collections
Right arrow Agricultural Pesticides
Right arrow Surface Water Quality
Right arrow Other Cropping Systems
Right arrow Other Environmental Contamination
Right arrow Water Pollution

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome