| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
BRGM, Water Division, Avenue C Guillemin, BP 6009, F-45060 Orléans Cedex, France
* Corresponding author (n.baran{at}brgm.fr).
Received for publication December 17, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ESA, ethanesulfonic acid • OA, oxanilic acid
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The agrochemicals most often detected in ground water are atrazine (2-chloro-4-ethylamino-6-isopropylamino-1,3,5-triazine) and deethylatrazine (2-amino-4-isopropylamino-6-chloro-s-triazine) (Leistra and Boesten, 1989; Barbash et al., 2001). The use of atrazine is, therefore, strictly controlled in some countries (Denmark) and completely banned in others (e.g., Germany, Italy, Austria, Sweden, Norway) (Blanchoud et al., 2002). In France, as a result of national pressure, atrazine was restricted to agricultural uses in February 1997, with dose limitations of 1000 g ha–1. Despite this restriction, atrazine continues to be detected in ground water. As a result, authorities in some regions have decided to reduce the use of atrazine, and have set up substitution programs in the protection zones around some drinking water wells. Sales of atrazine became forbidden in France starting in July 2003.
Acetochlor, an herbicide used for maize, has been on the United States market since 1994, following approval by the USEPA. This approval will be renewed, however, only if the total quantity of other herbicides used on this crop, including atrazine, decreases (Kolpin et al., 1996). Acetochlor was approved in France in 2000 and is now used in substitution programs. Although the sorption, leaching, and degradation of herbicides in soils have been studied extensively over the last decade, few data are available for chloroacetanilides in general, very few for acetochlor (Zheng and Ye, 2002), and even fewer for the acetochlor metabolites. Some laboratory studies have been performed, but field studies are rare despite the fact that this is the only way to take into account all the processes that do or could control the fate of agrochemicals in the environment.
Weber and Peter (1982), studying adsorption on seven different soils using radiolabeled molecules, found that acetochlor was more sorbed than metolachlor [2-chloro-6'-ethyl-N-(2-methoxy-1-methylethyl)-o-acetoluidide] or alachlor [2-chloro-2',6'-diethyl-N-(methoxymethyl) acetanilide], and concluded that acetochlor is less mobile than the other two chloroacetanilides. Balinova (1997), however, using percolation experiments on soil columns, showed that acetochlor, like the other chloroacetanilides (metolachlor and alachlor), is a potential ground water contaminant. Zheng and Ye (2002) also concluded, based on adsorption and thin-layer chromatography experiments, that acetochlor presents a risk of ground water contamination, particularly in sandy soil or if aquifers are shallow. Barbash et al. (2001) sampled ground water in 20 major hydrologic basins in the United States and detected acetochlor in some wells tapping shallow aquifers just one year after the first applications, thus confirming the potential mobility of the molecule. Furthermore, Kalkhoff et al. (1998) and Kolpin et al. (1998) detected acetochlor and its major degradation products (acetochlor OA and acetochlor ESA) in ground water in the USA, with degradation products being detected in larger concentrations and more frequently than the parent product. For acetochlor, and chloroacetanilides in general, ESA is detected in ground waters more frequently and at higher concentrations than OA (Kalkhoff et al., 1998; Kolpin et al., 1998). The preponderance of the ESA degradate in ground waters has not yet been fully explained. The mobility and production of ESA might be greater than those of OA, and/or OA might be less stable. Ground waters could, therefore, be contaminated by leaching not only of the parent molecule but also its degradation products.
To assess the risk of ground water contamination by acetochlor and its degradates, the fate of the herbicide in the soil must first be studied. In soils, very few studies addressed the parent molecule and the two degradation products simultaneously. Phillips et al. (1999) refer to Lebaron et al. (1988) who suggest that only the OA degradate of metolachlor is present in soils. However, ESA was in fact not determined by Lebaron et al. (1998). The purpose of the present study was therefore to determine, under field conditions, the fate of acetochlor and its two major metabolites in two different agricultural soils under the same weather conditions. We attempted to determine whether degradate production, notably the difference between their relative proportions, was affected by soil type, and to characterize leaching to deeper soil horizons of the various degradates, in relation, also, to the parent molecule.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The two very different soils selected for the study are a deep silty soil (luvisol; United Nations Food and Agriculture Organization [FAO] classification) and a shallower, more pebbly, more calcareous soil (calcisol; FAO classification) that is also slightly more clayey in the surface layers (Table 1).
|
Treatment of the Plots
The experimental fieldwork was done on two 100- x 70-m maize fields, one in each soil type. A commercial liquid pesticide (Trophee; Zeneca Sopra, Velizy Villacoublay, France) was sprayed directly onto the soil after planting and before emergence on 19 Apr. 2000 using the farmer's customary equipment. This was the first time that acetochlor had been applied to these plots. A solution of KBr had been applied a few hours before the pesticide so that bromide, due to its low background concentrations in our soils and its low biological and chemical reactivity, could be used to trace water movement. There was almost no wind when the product was sprayed. The average temperature that day was 12.5°C with a minimum of 8°C and a maximum of 17°C.
To determine the heterogeneity of the treatment, 40 aluminum sampling cups (external diameter: 11 cm) were placed in each plot one hour before the bromide was applied. Within half an hour after the application of acetochlor the cups were placed in dark glass flasks. Back in the laboratory, analytical-grade methanol was poured into the flasks to cover the cups and these were stirred for one hour. After this extraction, the acetochlor and the bromide collected in each cup were measured by high performance liquid chromatography (HPLC)/UV and ion chromatography, respectively. The average application rates were (i) 1530 g ha–1 (±387 g) of acetochlor and 19967 g ha–1 (±7638 g) of bromide for the calcisol plot and (ii) 2021 g ha–1 (±504 g) of acetochlor and 32589 g ha–1 (±13730 g) of bromide for the luvisol plot. The two plots belong to two different farmers. They were treated with the same sprayer for bromide, but acetochlor was applied using two different sprayers. The intra-plot application heterogeneity is hard to explain since the climate conditions were excellent and the sprayers were checked before use. In addition, a detailed study of the quantites applied as a function of the location of the sampling cups (data not presented) did not reveal any systematic malfunctioning of the sprayer.
Weather Data
A national meteorological network rain gauge 3.5 km from the site records the daily rainfall. Due to the proximity (850 m) of the two plots, rainfall was considered to be the same for both (Tables 2 and 3). The first rain (10 mm) fell on 20 April (one day after treatment).
|
|
Soil Sampling
Soil cores were collected before the herbicide was applied and 7, 21 (calcisol), 27 (luvisol), 56, 160, 230, and 344 d after the application. Each plot was divided into four equal-area subplots and four samples were taken from each subplot during each sampling, thus giving a total of 16 cores per plot per sampling.
A 10-cm-diameter percussion corer was used for sampling, and the maximum sampled depth was 1.0 m in the luvisol and 0.8 m in the calcisol where the limestone substratum was shallower. The cores were sent to the laboratory and cut, after the outer layer had been removed, into segments corresponding to depth intervals of 0 to 5, 5 to 10, 10 to 20, 20 to 30 cm, etc. Each sample was dried at 40°C for 3 d then ground to 2 mm. For each layer, a composite sample was made by mixing equal weights of the 16 individual samples. This composite sample was used for pesticide determination whereas analyses for bromide were done on each of the 16 individual samples.
For both soils, all 16 cores of a given sampling were not exactly the same length, with the depth of the soil profile varying by 10 to 20 cm within the sampled area. For the deepest layers, therefore, there are in some cases fewer than 16 individual samples, but there are always at least 4.
An analysis of variance (ANOVA) was done with Mscore data, a modified mobility index used, for example, by Keller et al. (1998), for bromide using the values of the 16 cores taken at each sampling date in each of the two soil types. Post hoc comparisons focusing on the differences between the two soil types for a similar date were done with Scheffé, Duncan, and Newman–Keuls tests. All statistical tests were performed with STATISTICA 5.5 software (StatSoft, 2000).
Determination of Bromide, Acetochlor, and Acetochlor Oxanilic and Sulfonic Acid Metabolites
Bromide was extracted from 10 g of dried ground soil in 10 mL of a CaCl2 10–4 M solution by stirring for 40 min, followed by centrifugation at 4000 rpm and filtering at 0.45 µm. The extracts were then analyzed in an ion chromatograph (Model 4500i; Dionex, Sunnyvale, CA) equipped with an As14 column. The detection limit is 0.05 mg kg–1. Bromide recovery from the soil samples spiked between 0.7 and 15.4 mg kg–1 was greater than 90%, regardless of the matrix type and the duration of contact between soil and bromide, except for the surface layer of the luvisol for which the recovery was slightly lower.
The two major acetochlor metabolites (Feng, 1991), ESA and OA, were extracted with a water and acetonitrile mixture. The degradates were then analyzed by reverse phase liquid chromatography with detection by electro-spray ionization–mass spectrometry (ESI–MS) in single ion monitoring and negative modes. The average recoveries in the soil extractions for spiking levels of 2 and 80 µg kg–1 range from 70 to 120% for both metabolites, with relative standard deviations being lower than 15%. The detection limits are 1 and 2.0 µg kg–1 for the ESA and OA compounds, respectively (Dagnac et al., 2002).
Acetochlor extractions were done by accelerated solvent extraction (ASE) with acetone for 5 min at 10.1 MPa and 60°C. Analysis was done by chromatography–ion trap mass spectrometry in MS–MS (Dagnac et al., 2001). The detection limit is 0.2 µg kg–1 for recovery rates of between 86 and 102% for spiking levels between 10 and 120 µg kg–1.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The bromide analyses done on the 16 cores taken during each sampling show a high variability between cores in both the measured concentrations at each depth and the overall profiles. The variability in concentration can be linked to the heterogeneity of the tracer application and/or to local variations in infiltration mode. The differences in the various profiles (not shown) are attributed to the variability in the ratio of rapid preferential flow to slower transport through microporosity.
The results obtained for the 16 cores enable us to determine the average amount of bromide in each of the layers on each of the sampling days, and to express this as a percentage of the average dose applied (Tables 2 and 3). Values exceeding 100% obtained for the calcisol are due to application heterogeneity and the natural variability of the bulk density, which was considered to be a constant value in the calculations. Indeed, the soil density was not corrected for the presence of stones (impossible to quantify in the samples taken). The bromide mass balances show a sharp decrease with time of the quantities in the profiles. Aside from plant uptake (Steenhuis et al., 1990; Jemison and Fox, 1991) and knowing that bromide does not decay or become bound, bromide losses can only be explained by lateral flow or leaching. Lateral flow can be excluded due to the absence of any less-permeable soil layer (none observed in any of the 96 cores taken in either plot during the study). Furthermore, the presence of bromide in the deepest horizons at concentrations very much greater than those measured before treatment tend to prove that there is leaching in both soils and that it reaches the deepest layers sampled (Fig. 1 and 2) .
|
|
In addition to visual criteria, statistical data processing of Mscore mobility indexes highlights the differences in hydraulic functioning between the two soils. The chi square test at the 0.05 significance level showed that all but one (calcisol, Date 1) of the 12 Mscore populations for bromide (two soil types x six dates) had a normal distribution. Levene's test at the 0.05 significance level showed that the variance was homogeneous in the two soil types. At the 0.01 significance level, differences in Mscores for bromide were found between luvisol (Date 3) and calcisol (Date 3) and between luvisol (Date 4) and calcisol (Date 4) using the Scheffé post-hoc test. Duncan and Newman–Keuls tests added a significant difference between luvisol (Date 2) and calcisol (Date 2), also significant at the 0.01 level. The leaching of bromide must therefore be considered to be greater in the luvisol than in the calcisol for Dates 3 and 4, and also for Date 2 if we rely on Duncan and Newman–Keuls tests.
Acetochlor
Acetochlor was detected down to a depth of 60 to 70 cm in the luvisol and 50 to 60 cm (maximum depth sampled) in the calcisol 7 d after application, proof that the parent molecule was leached (Fig. 1a, 2a). On the other hand, after 21 d for the calcisol and 27 for the luvisol, acetochlor was detected only in the 0- to 5- and 5- to 10-cm layers. These results contradict those of Konda and Pásztor (2001) who did not detect acetochlor in the 5- to 20-cm layer of a luvisol, the molecule being detected only in the 0- to 5-cm layer and not after Day 140. The detection limit of their analytical method, 40 times greater than ours, might explain this absence of detection in the lower layer and after 140 d for the 0- to 5-cm layer.
Acetochlor was still detected in the surface layer (0–5 cm) 344 d after application even though the increase in bromide concentrations in the deeper layers suggests that water continued to infiltrate from the surface, particularly between Days 160 and 230. After 21 or 27 d, it seems that acetochlor is either no longer or little leachable below a depth of 10 cm, or that the quantities involved result in contents below our detection limits.
The evident leaching in the present case resulted in a decrease in acetochlor contents with time due to mechanisms other than degradation. It is, therefore, not possible to calculate a DT50 (time in days for 50% disappearance) that is due solely to degradation. Only dissipation constants can be estimated. To compare what happens to acetochlor in the two soils regardless of the differences in applied doses, dissipation constants were calculated for the 0- to 10-cm layer. Acetochlor was indeed detected at all dates in this layer only, and was present deeper only at Date 1 and at concentrations negligible compared with those of the 0- to 10-cm layer. The initial concentration was estimated from the measured average dose of applied acetochlor. The concentrations measured on the six dates follow first-order kinetics. The dissipation constants obtained are 3.5 and 3.1 d for the calcisol and the luvisol, respectively. These values must, however, be used with caution because they are shorter than the time lapse between application and the first sampling, the date of which was determined at the beginning of our study, based on available literature data. The dissipation constants obtained are lower than those calculated in field conditions by Mueller et al. (1999) for various 8-cm-thick silty surface layers (6.3 d on the average), or by Vaughan et al. (1999) for a 30-cm-thick layer of sandy surface soil (18 d). Mueller et al. (1999) suggested that the low values, similar to our results, are probably due to leaching of acetochlor to deeper layers. Our values, also lower than the DT50 of 7 and 8 to 15 d measured in the laboratory by Feng (1991) and Vaughan et al. (1999), respectively, confirm the hypothesis of the in situ existence of other processes (e.g., leaching or the formation of bound residues) in addition to degradation that result in a decrease in the acetochlor concentrations with time. However, although they were unable to explain it, Vaughan et al. (1999) obtained DT50 of 2 to 4 d in the laboratory for various temperature and humidity conditions and for a relatively deep soil (30–76 cm). These values are lower than those they obtained for the surface layer, where degradation is usually considered to be greatest due to a greater microbial activity. A high disparity in acetochlor degradation rates is therefore reported in the literature.
In addition to leaching, a combination of degradation, volatilization, and the formation of bound residues (the relative proportions of which cannot be determined) leads to a significant decrease in the quantity of acetochlor detected in the calcisol and the luvisol after Day 7 (27.8 and 21.3% of the average dose applied, respectively; Tables 2 and 3). The formation of bound residues might be predominant. Indeed, Nemeth-Konda et al. (2002) have shown that 47 to 64% of the acetochlor adsorbed on a luvisol was not desorbable with three successive desorptions by CaCl2 after 24 h of adsorption. Various laboratory studies of different types of soils have demonstrated the formation, in 7 d, of bound acetochlor residues in proportions in some cases greater than 20% of the quantity applied (N. Simmons, personal communication, 2001). Heyer and Stan (1993) performed percolation experiments in soil columns under controlled laboratory conditions with alachlor, a molecule in the same family as acetochlor, and suggested that most of the alachlor loss after application was due to the formation of bound residues rather than to volatilization.
Although volatilization of acetochlor probably does occur, as for metolachlor and alachlor (Blanchoud et al., 2002), it is undoubtedly more limited than leaching, degradation, or the formation of bound residues. Wienhold et al. (1993), based on experiments performed in volatilization chambers, showed that for alachlor (a molecule whose physicochemical properties are similar to those of acetochlor, notably its Henry constants: 2.1 x 10–3 and 4.27 x 10–3 Pa m3 mol–1 for alachlor and acetochlor, respectively; Institut National de la Recherche Agronomique, 2002) the volatilization rate after 35 d ranges from less than 2 to 32%. The greatest losses were observed when the chemical was applied in the coated granulate form and at a high temperature, 35°C. When it was applied at a temperature of 15°C, in either the commercial formulation or in granulate form, less than 2% was lost. In our study, the average daily temperature was below 15°C during the 10 d following the application and remained below 21°C during the 35-d period following application. Probably not more than a few percent of acetochlor was, therefore, lost due to volatilization.
As for degradation, the presence of ESA and OA degradates is a clear indication of the key role played by degradation in controlling the fate of acetochlor in the soils studied, as illustrated and discussed below.
Degradation Products
Both ESA and OA were detected in the two soils 7 d after treatment (Fig. 1a, 2a). The onset of acetochlor degradation was therefore rapid, as observed in laboratory incubation experiments (Feng, 1991) and field tests (Mueller et al., 1999). Expressed as a percentage of the average applied dose of acetochlor (Tables 2 and 3), the sum of the percentages of degradates is much lower in the luvisol than in the calcisol after 7 d, whereas it is greater thereafter. This lower percentage could be due to greater preferential flow paths in this soil, as suggested by the bromide data (Fig. 1a, 1b, 1c and 2a, 2b, 2c) and by the fact that the percentage of acetochlor measured on Day 7 is also lower in the luvisol than in the calcisol. These preferential flow paths would enable greater leaching of acetochlor and its derivatives beyond the first meter in the luvisol than in the calcisol. The influence of preferential flow would have been all the greater since it rained on the 4 d following the application (>7 mm each day), the first rain falling only 12 h after application. These rainfalls, and the resulting leaching, probably limited the sorption processes in the upper layers.
For the two soils, except for the period between Days 7 and 21 or 27, the percentage of measured metabolites in relation to the applied dose of acetochlor decreases with time (Tables 2 and 3). This decrease could be due to various factors. The ESA and OA metabolites might degrade into other metabolites not sought in the study and for which we have little information. Indeed, using labeled molecules, Kotoula-Syka et al. (1997) showed, in the laboratory on sandy soils, that up to 15% of the applied acetochlor could be completely mineralized in 48 d, thus suggesting that degradation does not produce only the two derivatives identified here. The ESA and OA metabolites could also become bound to the solid matrix due to the formation of non-extractable residues. Finally, the degradates could be leached to layers below those sampled (i.e., to a soil horizon deeper than 1 m and to consolidated limestone). Such leaching is probably not insignificant as indicated by the fact that ESA was detected in the 50- to 60-cm layer of the calcisol and luvisol on Days 7, 21 or 27, and 56, and that both ESA and OA were detected at deeper levels in the luvisol on Day 55 than on Days 21 or 27 (Fig. 1a, 1b, 1c and 2a, 2b, 2c). Since the metabolites were detected in the deepest horizons sampled, it is quite likely that they were also present at even deeper levels. This hypothesis is supported by the results of measurements made on 88 wells in the United States (Kalkhoff et al., 1998; Kolpin et al., 1998), which showed that acetochlor degraded to mobile metabolites that were then transported to ground water before complete mineralization of the parent compounds. We can, therefore, assume that the ESA and OA metabolites are prone to leaching.
The relative proportions of the acetochlor degradates in the two soil types are also different. In the calcisol (Fig. 1), the ESA concentrations are greater than or equal to the OA concentrations in all the layers, and the OA metabolite was no longer detected at the surface on Day 56. In the luvisol (Fig. 2a, 2b, 2c), on Days 7, 27, and 56, the OA concentrations were equal to or much higher than the ESA concentrations in the top 10 cm, whereas in the deeper layers it was the ESA concentrations that were highest and OA was, in some cases, not even detected. Thereafter (data for Days 160 and 230 not shown), the trend was less marked.
The differences observed between the two plots might be related to the intrinsic properties of the ESA and OA molecules or to the effect of soil type on the formation or fate of the derivatives. The disappearance of OA in both soils with time could indicate that this compound is less stable, more mobile, or has a greater proportion of non-extractable residues than ESA. By analogy with metolachlor and following the reasoning of Phillips et al. (1999), the presence of a carboxyl group might make OA less stable than ESA. Despite an absence of published data, the chemical stability of OA is known to be lower than that of ESA (N. Simmons, personal communication, 2001), and the environmental conditions encountered in the luvisol (pH, for example) compared with those in the calcisol could favor the persistence of the OA derivative. Although it was assumed that weather conditions were the same for the two plots, the different soil textures and structures probably resulted in profile variations in terms of moisture content and temperature—two parameters that are predominant in the degradation of agrochemicals, notably chloroacetanilides (Bouchard et al., 1982; Nègre et al., 1992; Vischetti et al., 1998; Vaughan et al., 1999). These two factors alone could explain the differences observed between the two acetochlor metabolites in the two soil types, the differences being related to the soils' different degradation paths or rates. Phillips et al. (1999) studied the evolution of concentrations of metolachlor ESA and OA degradates in water in a tile drain. They concluded that, regardless of soil properties, under certain hydraulic conditions, notably heavy rainfall and, therefore, rapid circulation, more metolachlor and OA were leached than at other times during the year because there was less contact time with the soil surface and therefore less opportunity for degradation of the metolachlor and metolachlor OA. Preferential flow causing more rapid infiltration of part of the substances, more pronounced in the luvisol than in the calcisol, could therefore explain the differences in the ESA to OA ratio in the two soils in our study. The intensity of the sorption processes could also be different in the two soils and for the two products. Adsorption, different for the two metabolites, could lead to different relative proportions depending on the soil, and indirectly influence degradation. Indeed, Mueller et al. (1999) point out that sorbed pesticides are known to degrade less than the dissolved fraction. We know that the clay content decreases sharply below 30 cm in the calcisol, whereas it continues to increase with depth by a factor of three in the luvisol (Table 1). Furthermore, Weber and Peter (1982) and Wang et al. (1999) have shown that clay, for low organic matter content, can contribute significantly to sorption. Sorption in the luvisol might, therefore, be greater than in the calcisol, leading to the observed differences in degradate concentrations between the two soil types.
Mass Balance
Mass balances determined for the entire soil profile for the sum of acetochlor, ESA, and OA are expressed as percentages of the average applied dose (Tables 2 and 3).
For the calcisol, the sum of metabolites and the parent molecule decreased rapidly from around 45 to 1% between Days 7 and 160. Thereafter, it varied little and remained at about 1%. The ratio of the sum of the two derivatives to the parent molecule varied with time. On Day 7, there was more acetochlor than derivatives, whereas on Days 21 and 56, the sum of the derivatives was at least five times greater than the amount of acetochlor. On Days 160, 230, and 344, acetochlor was again predominant and was the only molecule detected on Day 344. The mass balance shows that regardless of the date, the proportion of ESA in the soil profile is always greater than that of OA, by a ratio of up to 87 (on Day 56).
For the luvisol, the sum of the metabolites and the parent molecule varied little between Days 7 and 27 when it represented about 26% of the dose of applied acetochlor, but had decreased sharply to 2.5% by Day 160, and represented less than 1% of the applied dose after 230 d. Like in the calcisol, the proportion of acetochlor in relation to the derivatives varied with time. The parent molecule was predominant on Days 7, 230, and 344, whereas on Days 27, 56, and 160, the proportion of the two derivatives was at least four times greater than that of acetochlor. On Day 344, only acetochlor was detected. The proportion of OA was 1.3 to 1.5 times greater than that of ESA at the beginning (Days 7, 27, and 56), was equal to it on Day 230, and less on Day 160.
In the two soil profiles, 7 d after application, the mass balances show that less than 45% of the initial dose of acetochlor was detectable in the form of the parent molecule or the metabolites. The even lower percentage (26%) measured in the luvisol might be due to a greater leaching of the parent molecule and/or of its degradates. Although the ESA to OA ratios are different and vary in different ways in the two soils, the two molecules present some similarities in the two soils with, notably, a detection limited in time. Likewise, although acetochlor is still detected 344 d after application, more than 1% of the applied dose is present in the soil profile for only a few months. The fact that acetochlor and its two metabolites were detected for only a relatively short period was probably due to leaching, the formation of bound residues, or degradation into other metabolites not sought in this study.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Although the two derivatives were detected in the two soils, their fates appear to be different, at least in the surface layers. Indeed, during the first few months a greater proportion of OA than ESA was detected in the 0- to 5- and 5- to 10-cm layers of the luvisol, as opposed to what occurred in the calcisol. The results obtained here do not, however, enable us to determine in which of the two soils studied there is the most transfer toward the aquifer. In addition to differences in the physicochemical interactions between the molecules and the two soils leading to differences in sorption, degradation, formation, and persistence of the two metabolites, soil hydrodynamics also play a great role in the contamination attenuation capacity. Nevertheless, ground water contamination by acetochlor, ESA, and OA appears possible in this study site.
Degradation and sorption experiments are now being performed in the laboratory. Their results will, when compared with field observations and calculated dissipation constants, enable us to determine the relative magnitude of the leaching, degradation, and bound residues formation processes. They will also help to validate or invalidate the hypothesized soil-type effect. All the data will then be integrated into a solute transfer model and we will attempt to model the data measured in situ, the hydrodynamic fitting gaining greatly from the fact that a conservative tracer was monitored along with the pesticides.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal |
