JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 12 October 2005
Published in J Environ Qual 34:1933-1943 (2005)
DOI: 10.2134/jeq2004.0460
© 2005 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
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 (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Beulke, S.
Right arrow Articles by Walker, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Beulke, S.
Right arrow Articles by Walker, A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Beulke, S.
Right arrow Articles by Walker, A.
Related Collections
Right arrow Batch Studies
Right arrow Pesticides
Right arrow Solute Transport Models
Right arrow Agricultural Pesticides

TECHNICAL REPORTS

Bioremediation and Biodegradation

Evaluation of Simplifying Assumptions on Pesticide Degradation in Soil

Sabine Beulkea,*, Wendy van Beinuma, Colin D. Brownb, Matthew Mitchellc and Allan Walkerc

a Cranfield Centre for EcoChemistry, Cranfield University, Silsoe, Bedford, MK45 4DT, UK
b Environment Department, University of York, Heslington, York, YO10 5DD, UK, and Central Science Laboratory, Sand Hutton, York, YO41 1LZ, UK
c Warwick HRI, University of Warwick, Wellesbourne, Warwick, CV35 9EF, UK

* Corresponding author (s.beulke{at}csl.gov.uk)

Received for publication November 26, 2004.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
There is evidence that degradation of pesticides in simple laboratory systems may differ from that in the field, but it is not clear which of the simplifications inherent in laboratory studies present serious shortcomings. Laboratory experiments evaluated several simplifying assumptions for a clay loam soil and contrasting pesticides. Degradation of cyanazine [2-(4-chloro-6-ethylamino-1,3,5-triazin-2-ylamino)-2-methylpropiononitrile] and bentazone [3-isopropyl-1H-2,1,3-benzothiadiazin-4(3H)-one 2,2-dioxide] at fluctuating temperature and moisture was predicted reasonably well based on parameters derived from degradation under constant conditions. There was a tendency for slower degradation of cyanazine and bentazone in soil aggregates of 3 to 5 mm in diameter (DT50 at 15°C and 40% maximum water holding capacity of 25.1 and 58.2 d, where DT50 is the time for 50% decline of the initial pesticide concentration) than in soil sieved to <3 mm (DT50 of 19.1 and 37.6 d), but the differences were not significant for most datasets. Degradation of cyanazine, isoproturon [3-(4-isopropylphenyl)-1,1-dimethylurea], and chlorotoluron [3-(3-chloro-p-tolyl)-1,1-dimethylurea] was measured in soil amended with different amounts of lignin. The effect of lignin on degradation was small despite considerable differences in sorption. The DT50 values of cyanazine, isoproturon, and chlorotoluron were 16.2, 18.6, and 33.0 d, respectively, in soil without lignin and 19.0, 23.4, and 34.6 d, respectively, in soil amended with 2% lignin. Degradation of bentazone and cyanazine in repacked soil columns was similar under static and flow conditions with 50.1 and 47.2% of applied bentazone and 74.7 and 73.6% of applied cyanazine, respectively, degraded within 20 d of application. Thus, the assumptions underpinning laboratory to field extrapolation tested here were considered to hold for our experimental system. Additional work is required before general conclusions can be drawn.

Abbreviations: DT50, time for 50% decline of the initial pesticide concentration • Kd, sorption coefficient • Koc, sorption coefficient normalized to organic carbon content • mwhc, maximum water holding capacity • Q10, factor by which degradation increases when temperature increases by 10°C


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
A FUNDAMENTAL UNDERSTANDING of the behavior of pesticides in the environment is required to minimize adverse side-effects and to ensure that agricultural systems can be maintained. One of the most important processes influencing the fate of pesticides is their degradation in soil. For example, parameters describing degradation in models to simulate pesticide behavior have been shown to be extremely sensitive (Dubus et al., 2003). Data related to pesticide degradation are often derived from simple laboratory experiments assuming that results obtained can be extrapolated to field conditions. This is based on a number of simplistic assumptions: (i) degradation in structured soil with fluctuating moisture and temperature in the field can be characterized by the results from laboratory studies with disturbed soil and constant temperature and moisture regimes, (ii) degradation of a pesticide dissolved in the soil solution and sorbed to soil particles occurs at the same rate and degradation is thus independent of sorption, and (iii) degradation under flow conditions in the field is identical to that in static systems.

A review by Beulke et al. (2000) showed an overall tendency for predictions based on laboratory degradation data to overestimate persistence in the field. This was mainly attributed to differences between soil conditions in the laboratory and the field. Simplifying assumptions regarding pesticide degradation have been evaluated in a number of studies. Walker et al. (1992) and Jurado-Exposito and Walker (1998) investigated degradation of pesticides under constant and fluctuating conditions of temperature or moisture. Degradation of organic chemicals in disturbed and undisturbed soil or in aggregates of different sizes has been compared by Aden (2003). The influence of sorption and mass transfer on degradation has been evaluated by a combination of controlled experiments and modeling (Ogram et al., 1985; Guerin and Boyd, 1992) or by comparison of degradation in soils amended with different amounts of strongly sorbing material (Cantwell et al., 1989; Guo et al., 1999). An alternative method is the analysis of correlations between sorption and degradation properties in different soils (Walker et al., 1985) or in a single soil that has been in contact with the pesticide for different lengths of time (White et al., 1999). Several studies investigated degradation of organic chemicals in static vs. flowing systems or degradation at different flow rates (Langner et al., 1998; Guo et al., 1999; Heistermann et al., 2003). However, the majority of studies reported in the literature evaluated only selected hypotheses and coherent datasets are scarce. The experimental design and the soils and pesticides used for individual studies vary considerably. Conflicting results were often obtained.

The aim of the research presented here was to evaluate a range of assumptions for a single soil and contrasting pesticides. Laboratory degradation studies were performed to investigate: (i) degradation under constant and fluctuating temperature and moisture conditions, (ii) the influence of soil aggregate size on pesticide degradation, (iii) the influence of sorption on degradation, and (iv) degradation under static and flow conditions. The assumptions were evaluated using different test conditions in the laboratory. No comparison was made with measured field behavior.

It is hypothesized that the simplifying assumptions evaluated here are a major cause of differences between degradation in the laboratory and the field. The importance of these assumptions relative to other simplifications is, however, unknown and depends on the situation at hand. For example, the preparation of soil samples for laboratory studies (e.g., drying, sieving) may have a significant effect on degradation for some pesticides. In a preliminary study, we found no difference between disturbed and undisturbed soil for the pesticides tested in this study (data not shown). Also, laboratory studies using individual or a small number of soil samples are not fully representative, for example, of the spatial and temporal variability in degradation encountered in the field that results from a variation in soil properties and microbial populations.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Details of Soil, Pesticide, and Analysis
A clay loam soil of the Salop series (USDA classification: Typic Endoaqualf, 25% clay, 2.3% organic carbon, pH [H2O] 7.4, maximum water holding capacity [mwhc] 0.294 mL g–1 soil) was used for the experiments. Soil was taken from the top 15 cm of an agricultural field in Warwickshire, UK, and stored at 4°C in moist conditions for a maximum of 40 d before use in experiments, whereupon it was air-dried until sieving was possible and passed through a 3-mm sieve.

The herbicides bentazone and cyanazine were chosen for this study. Differences between degradation in the laboratory and the field were found for bentazone by Huber and Otto (1994). The average of 27 laboratory DT50 values (time for 50% decline of the initial pesticide concentration) at 20 to 23°C was 46 d. Degradation was faster in 10 field studies in Germany and the United States with an average DT50 value of 12 d. No information was available on laboratory to field extrapolation for the triazine herbicide cyanazine. However, findings by Müller et al. (2003) for atrazine suggested that differences between laboratory and field degradation may occur for triazine compounds. The authors found faster degradation of atrazine in the field than could be predicted from results obtained under controlled conditions in the laboratory.

A commercial formulation of the herbicide cyanazine (Fortol; 45.5% w/w suspension concentrate [SC]; American Cyanamid Co., Princeton, NJ) was used in the experiments in combination with bentazone (Basagran, 87% w/w SC; BASF plc, Ludwigshafen, Germany) or with isoproturon (Alpha IPU 500; 46% w/w SC; Makhteshim Chemical Works Ltd, Beer-Shiva, Israel) and chlorotoluron (Alpha Chlorotoluron 500; 43.9% w/w SC; Makhteshim Chemical Works Ltd). For experiments investigating the decrease in pesticide residues with time, soil was sampled at intervals and frozen until analysis. Herbicide residues were extracted by shaking with acetonitrile on a wrist-action shaker for 1 h (10 g soil + 20 mL acetonitrile unless otherwise stated). The samples were allowed to settle for 1 h and the clear supernatant was analyzed for herbicide residues by high performance liquid chromatography (HPLC). A DX600 (Dionex, Sunnyvale, CA) HPLC equipped with a PDA100 photodiode array detector and a Discovery C-18 column (150 mm long x 4.6-mm i.d., 5-µm particle size; Supelco, Bellefonte, PA) was used in the study to investigate degradation under static and flow conditions. A Kontron 300 series HPLC (Kontron Instruments, Everett, MA) with Lichrocart RP-18 (5 µm) column (250 mm long by 4.6-mm i.d.) was used in the remaining studies. In the studies into degradation under static and flow conditions the solvent system was acetonitrile to water to orthophosphoric acid (70:29.75:0.25 by volume) at a flow rate of 1 mL min–1. In the remaining studies the ratio was 75:24.75:0.25. Detection was by UV absorbance at 240 nm (Kontron 332) for chlorotoluron and isoproturon and 230 nm for cyanazine and bentazone. The limit of quantification was 50 µg L–1 for chlorotoluron, bentazone, and isoproturon and 60 µg L–1 for cyanazine. Average recovery rates of chlorotoluron, isoproturon, cyanazine, and bentazone from soil were 93, 92, 96, and 99%, respectively.

Degradation under Constant and Fluctuating Temperature and Moisture Conditions
Soil was sieved to <3 mm. A total of 6.4 kg soil (dry weight) was treated with suspensions of the commercial formulations of cyanazine and bentazone to give target herbicide concentrations of 15 mg active ingredient kg–1 soil. This concentration is equivalent to an aeric mass of 2.25 kg ha–1 assuming a bulk density of 1.5 g cm–3 and a penetration depth of 1 cm. This agrees well with typical field application rates for the pesticides. Each batch was mixed by passing several times through a sieve (3-mm mesh) and then allocated to the following treatments.

Constant Conditions
Eight 390-g (soil dry weight) soil samples were incubated at constant temperature and moisture content in the dark (all combinations of 15 or 25°C and 40 or 70% mwhc; two replicates per treatment). The containers were weighed, loosely capped, and transferred to incubators at the required temperatures. Moisture contents were maintained by the addition of distilled water as necessary (usually once per week).

Fluctuating Conditions
Eight 390-g dry weight soil samples were incubated with variable environmental conditions in the dark (two replicates per treatment). Four samples were incubated with constant soil moisture content (40 or 70%) but with an alternating temperature cycle of 3.5 d at 15°C and 3.5 d at 25°C. The remaining four samples were incubated at constant temperature (15 or 25°C) but with variable soil moisture. The moisture content was adjusted initially to 70% mwhc and the containers were capped with nylon mesh, which permitted slow loss of water from the soil. The containers were shaken daily to ensure a uniform moisture distribution and weighed every 3 to 4 d. When soil moisture content had fallen to 40% mwhc, the soil was rewetted to 70% mwhc (approximately every 2 wk at 15°C and weekly at 25°C). The fluctuations in temperature and moisture tested approximate conditions near the soil surface in late spring or summer for the Atlantic maritime climate of the UK.

Samples were taken for extraction and analysis 0, 3, 7, 10, 14, 17, 21, 24, and 28 d after treatment. First-order DT50 values were calculated by minimizing the sum of squared residuals with Microsoft Excel Solver (Microsoft, 2000).

Degradation under fluctuating conditions was predicted using the equations below:

[1]

[2]

[3]

[4]
where C(t) = concentration at time t (mg kg–1), C0 = initial concentration (mg kg–1), k(T, M) = degradation rate constant under actual temperature and moisture conditions (d–1), t = time (d), kref = degradation rate constant at reference temperature and moisture (d–1), Q10 = factor by which degradation increases when T increases by 10°C (–), T = actual temperature (°C), Tref = reference temperature (15°C), M = moisture (% mwhc), Mref = reference moisture (40% mwhc), and B = moisture exponent (–).

The temperature dependence of the first-order degradation rate constant was calculated based on the Q10 approach, which is commonly used in pesticide fate models (see Eq. [2]). This method is based on Van't Hoff's finding that the rate of biological reactions increases by a factor of approximately 2 when temperature increases by 10°C. The factor (referred to as Q10, Eq. [3]) was adjusted to the conditions of this study (see below). The Q10 approach results in a relationship between the degradation rate constant and temperature that is similar in shape to the Arrhenius equation. An alternative is the O'Neill equation, which accounts for the fact that degradation only increases up to a temperature optimum and decreases thereafter. The estimation of the optimum was not possible in this study, because only two temperatures were included. Thus, the O'Neill equation does not give an advantage over the Q10 approach in this study. The equation of Walker (1974) was used to describe moisture dependence of degradation (Eq. [2]). The DT50 values derived under constant incubation conditions were used to derive the parameter Q10 (Eq. [3]) and the moisture exponent B (Eq. [4]). The actual degradation rate k(T, M) was calculated for each day of the simulation period using actual daily temperature and moisture conditions. The ModelMaker software Version 4.0 (FamilyGenetix, 2000) was used for this purpose. Daily water contents were approximated by linear interpolation between dates for which measurements were available.

Influence of Aggregate Size on Degradation
Air-dried soil (water content 9.5% w/w) was sieved to isolate aggregates of 3 to 5 mm in diameter and treated with the herbicides cyanazine and bentazone (target initial concentration 15 mg active ingredient kg–1). Moisture was adjusted to 40% maximum water holding capacity and the soil was then mixed with the pesticides by passing several times through a sieve (3-mm mesh). Duplicate portions were incubated at either 15 or 25°C in the dark. Samples were taken for extraction and analysis 0, 3, 7, 10, 14, 17, 21, 24, and 28 d after treatment. The study was performed alongside the experiment described above to allow a comparison with degradation in soil sieved to <3 mm.

Influence of Sorption on Degradation
Degradation of cyanazine, isoproturon, and chlorotoluron was measured in soil amended with different amounts of hydrolytic lignin (Aldrich product #37 107-6; Gillingham, Dorset, UK; organic carbon content = approximately 67%). Bentazone was not included because sorption of this pesticide did not change significantly with lignin amendment.

The addition of 1 and 2% lignin increased the organic carbon content of the soil from 2.3 to 3.0 and 3.6%, respectively. Sorption of pesticides in soil amended with 0, 1, or 2% lignin was investigated before the degradation study. Lignin was suspended in water by sonication before mixing into the soil, which was then left to stand for 48 h. Pesticide solution (20 mg active ingredient in 20 mL 0.01 M CaCl2) was added to 10 g moist soil. The suspension was mixed for 24 h and then centrifuged at 5000 x g for 3 min. Blanks without soil were treated in the same way. The concentration of the pesticides in the supernatant was measured by HPLC. The amount sorbed was calculated by difference between the concentration in solution and that in the blank samples. Sorption coefficients (Kd) were calculated as the ratio of the concentration sorbed (mg kg–1) and in solution (mg L–1). Normalized coefficients (Koc values) were calculated as Kd/organic carbon content x 100.

Soil (500 g soil for each of the three replicates) amended with 0, 1, or 2% lignin was left to stand for 48 h before adding a solution of commercial formulations of the pesticides at a target initial concentration of 20 mg active ingredient kg–1. This concentration is equivalent to 3 kg ha–1 assuming a bulk density of 1.5 g cm–3 and a penetration depth of 1 cm. After thorough mixing, soil moisture was adjusted to 40% mwhc and the soil was incubated at 15°C in the dark. Each treatment was tested in three replicates. Samples were taken for extraction and analysis 0, 3, 7, 14, 21, 28, and 56 d after application.

Degradation under Static and Flow Conditions
Degradation of bentazone and cyanazine was measured in unsaturated repacked soil columns. The pesticides were mixed evenly into the soil. The soil solution in half of the columns was circulated by applying suction at the bottom and reapplying the effluent at the top. There was no suction or leaching for the remaining columns.

Moist soil was sieved to 6 mm and treated with suspensions of commercial formulations of cyanazine and bentazone to give an initial concentration of 10 mg kg–1 soil. The soil was mixed thoroughly and stored in a refrigerator overnight before packing into columns. Samples were taken from each batch to determine initial pesticide concentrations and water contents. A layer of fine sand (30 g) was added to the bottom of glass columns with an inner diameter of 5 cm and a narrow outlet filled with glass wool. Soil was added in five layers of 4 cm at a time, with 100 g soil per layer. The water content was adjusted to 24% w/w and a layer of very coarse sand (25 g) was placed on top. Eight columns were set aside. The remaining eight columns were connected to an air pump that applied a continuous suction at 13.3 kPa to the column outlets. Effluent from the columns was collected in air traps and immediately dripped onto the soil column using a peristaltic pump. Marprene and Teflon tubing was used in the pump and between the pump, air trap, and column.

The columns were kept at 20°C in the dark with moisture contents adjusted when necessary. The soil appeared to hold the soil water more strongly with time, so extra water had to be added to the columns to sustain circulation of the soil solution at constant suction. Demineralized water (20 mL) was added to the columns 2 d after starting the circulation giving a water content of almost 30% by weight. One week later, an additional volume of 5 mL water was added to ensure continued flow. The flow rate was, therefore, variable throughout the experiment (2–15 mm d–1). Water was also added to the columns that were incubated under static conditions to ensure the moisture conditions were similar for the two treatments. The flow rates in this study are representative of rainfall events of low to medium intensity under European conditions.

Four columns from each treatment were sampled 9 and 20 d after application. Columns were split into five soil layers and two sand layers. Each layer was weighed and extracted with acetonitrile (90 mL acetonitrile were added to 98 g moist soil containing 24 mL soil solution) before analysis. Extracted soil was dried to calculate the moisture content of the individual soil layers.

Statistical Analysis
The goodness of first-order fits to the degradation data was characterized using a combination of visual assessment and a {chi}2 test as recommended by FOCUS (2006). The {chi}2 test considers the deviations between measured and predicted values relative to the uncertainty of the measurements:

[5]
where P = predicted value, O = observed value, = mean of all observed values, and err = measurement error percentage. If {chi}2 is larger than a tabulated value, then the model is not appropriate at the chosen level of significance. The measurement uncertainty is expressed as a percentage error that is used to scale the observed mean. The true error is unknown. It is thus recommended to determine the smallest error value at which the test is passed at the 5% significance level. A fit that results in an error level of 15% is considered acceptable although this is not an absolute cut-off criterion and a visual assessment must always be made.

Statistical analyses were performed to evaluate whether the tested treatments had a significant effect on degradation. A nonlinear regression analysis was performed with GenStat 7.1 (Payne et al., 2003) where the decline in pesticide residues with time could be described with a first-order model. A set of three nested models was fitted to the data and compared using sequential analysis of variance. First, a common first-order curve was fitted to the combined data for all treatments. Next, separate parallel lines with an identical degradation rate constant, but different intercepts, were fitted to the data for each of the treatments. Then, separate lines with different degradation rate constants and intercepts were fitted to the data. We tested whether the fit using different degradation rate constants for each treatment was significantly better (p < 0.05) than the fit using identical rate constants. If that was the case, then the degradation rate constant differed significantly between the two treatments. Datasets that were not well described by first-order kinetics were analyzed using an additive model in GenStat 7.1. An additive model is a nonparametric regression model, where a smoothing spline is fitted to the data, which summarizes the trend of the measured residues (% of initial) as a function of time. In this case, it is not necessary to pre-define a parametric model equation. An analysis of variance was performed to evaluate whether the data were described better by separate splines for each treatment than by a model with different intercepts but identical shape (parallel splines).


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Degradation under Constant and Fluctuating Temperature and Moisture Conditions
First-order DT50 values of cyanazine and bentazone are given in Table 1. Temperature had a strong effect on degradation. An increase in temperature by 10°C resulted in a decrease in the DT50 value by a factor of 1.7 for both compounds (Q10 value; average of both moisture conditions). Degradation of cyanazine was faster at 70% mwhc than in the drier soil. The average moisture exponent B was 0.68. Mean DT50 values of bentazone were similar at both water contents.


View this table:
[in this window]
[in a new window]
  		Table 1. First-order DT50 values (time for 50% decline of the initial pesticide concentration) of cyanazine and bentazone in a clay loam soil with different aggregate sizes under constant and fluctuating temperature and moisture conditions and (in parentheses) percentage error levels at which the {chi}2 test is passed.{dagger}

 
Degradation of cyanazine and bentazone under varying incubation conditions is shown in Fig. 1 and 2 , respectively. Simulations are based on degradation under constant conditions (15°C, 40% mwhc), actual daily temperatures, and moisture contents and the equations to account for temperature and moisture effects given above. Overall, degradation under fluctuating conditions was predicted reasonably well from that under constant conditions. Degradation of cyanazine at temperatures varying between 15 and 25°C was initially faster than expected at 70% mwhc, but not at 40% mwhc. Results are in agreement with those of Jurado-Exposito and Walker (1998) for isoproturon and alachlor, but not for propyzamide. The authors found that degradation of propyzamide under fluctuating conditions was initially slower than expected, but increased markedly approximately 4 wk after treatment. Kubiak (1986) found that degradation of metamitron was faster under variable temperature and moisture conditions than under constant conditions. This is probably due to a change in the activity of degrading microorganisms. Similar results were obtained by Aden (2003) for metazachlor.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 1. Measured degradation of cyanazine under fluctuating temperature or moisture conditions and simulated degradation based on DT50 values (time for 50% decline of the initial pesticide concentration) measured under constant conditions. The term mwhc is maximum water holding capacity.

 


View larger version (21K):
[in this window]
[in a new window]
  		Fig. 2. Measured degradation of bentazone under fluctuating temperature or moisture conditions and simulated degradation based on DT50 values (time for 50% decline of the initial pesticide concentration) measured under constant conditions (lines). The term mwhc is maximum water holding capacity.

 
Influence of Soil Aggregate Size on Pesticide Degradation
Figure 3 compares degradation of cyanazine and bentazone in 3- to 5-mm aggregates with that in soil sieved to less than 3 mm. Measurements on the day of application in the larger aggregates were highly variable. This is probably due to incomplete mixing of the pesticide solution into the aggregates. The entire batch of soil was mixed at each sampling interval and it is assumed that the amount of mixing was sufficient from the second sampling onward. Thus, concentrations measured after the day of application are in line with each other, but clearly deviate from the initial measurements. The initial sampling date was, therefore, not included in the curve fitting for 3- to 5-mm aggregates. Results are presented for incubation of the soil at 25°C in Fig. 3. First-order DT50 values for both incubation temperatures are given in Table 1.



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3. Degradation of (a) cyanazine and (b) bentazone in <3-mm or 3- to 5-mm aggregates. Solid and dashed lines represent first-order fits to data for <3-mm and 3- to 5-mm aggregates, respectively (Day 0 data excluded from curve fitting). The term mwhc is maximum water holding capacity.

 
There was a tendency for slower degradation in soil with larger aggregates (Fig. 3). The average DT50 value of cyanazine in the larger aggregates at 25°C exceeded that in the smaller aggregates by a factor of 1.7 (Table 1). The effect was somewhat smaller at 15°C (factor of 1.3). The effect of aggregate size on the DT50 of bentazone was similar at both incubation temperatures (factor of 1.5). However, the observed differences were only significant for cyanazine at 25°C (p = 0.007; measurements on the day of application excluded from statistical analysis).

Pesticide degradation parameters required as model input are often derived from laboratory studies using sieved, disturbed soil. Sieving changes the structure of the soil by breaking up soil aggregates and may influence pesticide availability and microbial activity, thereby affecting pesticide degradation. There is evidence that microorganisms are concentrated at the surface or in the outer regions of soil aggregates (Priesack and Kisser-Priesack, 1993). The smaller pores inside the aggregates are too small to allow significant colonization. Oxygen and nutrients may be depleted by microorganisms at the surface leading to small concentrations inside the aggregates. Kilbertus (1980) found that soil bacteria mainly occur in pores of 2 to 3 µm in diameter. Only 6% of the inner volume of aggregates was colonized by microorganisms. As a result, a significant proportion of solution retained in pores may be inaccessible to microorganisms at natural field moisture levels and pesticide inside intact soil aggregates may, thus, be protected from degradation. Although pesticides are concentrated at the surface of soil aggregates shortly after application, a considerable proportion may move into the aggregates via convection and diffusion. Finer sieving is expected to improve the contact between the pesticide and the degrading microflora and thereby enhance degradation. On the other hand, sieving increases the number of readily accessible sorption sites that may cause a delay in degradation.

Influence of Sorption on Pesticide Degradation
The Kd values of bentazone in a preliminary experiment were 0.01, 0.06, and 0.02 L kg–1 following addition of 0, 1, and 5% lignin to the soil, respectively. This herbicide was, therefore, not included in the study. Amendment of the clay loam soil with 2% lignin increased the Kd value of cyanazine and chlorotoluron by a factor of 2.3 and that of isoproturon by a factor of 3.1 (Fig. 4a) . The Koc values for the three pesticides also increased with increasing amounts of lignin (Fig. 4b). This implies that a larger amount of chemical is sorbed per unit of carbon in the soils amended with lignin than in untreated soil.



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 4. (a) Sorption coefficients (Kd values) of cyanazine, isoproturon, and chlorotoluron in soil amended with different amounts of lignin. (b) Sorption coefficients normalized to organic carbon content (Koc values).

 
The percentage of applied pesticide extractable with organic solvents on the day of application significantly decreased with increasing amounts of lignin (Fig. 5a) . This suggests that a considerable proportion of the pesticides was rapidly incorporated into the lignin where it was bound strongly. Pesticide extractable with aqueous CaCl2 solution on the day of application was assumed to characterize the potentially available fraction in soil solution. Results are expressed as a percentage of total extractable mass in Fig. 5b. The available proportion decreased with increasing amounts of lignin for chlorotoluron and isoproturon, but not for cyanazine.



View larger version (53K):
[in this window]
[in a new window]
  		Fig. 5. (a) Cyanazine, isoproturon, and chlorotoluron extractable with organic solvent from soil amended with different amounts of lignin (as a percentage of applied pesticide = 20 mg kg–1; Day 0). (b) Percentage extractable with aqueous solution from soil amended with different amounts of lignin (as a percentage of total extractable residues; Day 0).

 
The initial amount extracted was set to 100% for each of the three treatments and the relative decline in extractable residues taken as a measure of degradation. This is valid if the non-extractable mass of pesticide remains constant over the course of the experiment or if degradation in the non-extractable fraction is the same as that in the extractable fraction. It was not possible to verify these assumptions within the scope of this study. The methodology adopted here is consistent with that used by other researchers (Guo et al., 1999). First-order DT50 values and {chi}2 error levels are presented in Table 2. Only small differences were found between the DT50 values in soil with and without lignin. Figure 6 shows residues of cyanazine, isoproturon, and chlorotoluron extractable with organic solvent at different times from application. The decline in residues of cyanazine and isoproturon extracted with organic solvent was initially slightly slower in soil amended with 2% lignin. Degradation of chlorotoluron appeared to be slightly slower in the soil without lignin that in the soil amended with 1% lignin. The observed differences were, however, not significant at the 5% level according to statistical analysis using an additive model. An additive model was chosen because the curves initially deviated somewhat from first-order kinetics (Fig. 6). Nonetheless, the first-order fits are acceptable according to draft guidance from FOCUS (2006) (Table 2). It should be noted that we compared the rates of decline in extractable residues. The total mass of pesticide degraded was smaller in soil with than without lignin.


View this table:
[in this window]
[in a new window]
  		Table 2. First-order DT50 values (time for 50% decline of the initial pesticide concentration) of cyanazine, isoproturon, and chlorotoluron in a clay loam soil amended with different amounts of lignin and (in parentheses) percentage error levels at which the {chi}2 test is passed.{dagger}

 


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 6. Residues of cyanazine, isoproturon, and chlorotoluron extractable with organic solvent for soil amended with different amounts of lignin. Data are expressed as % of the initial amount extractable and are the means of three replicates.

 
In contrast to our results, a number of studies reported in the literature have shown a negative effect of sorption on degradation. Guo et al. (1999) added the pesticide aldicarb to soil amended with varying amounts of activated charcoal (AC). Samples were taken at intervals and extracted with methanol and the rate of decline in extractable residues was interpreted as degradation. Sorption increased with increasing amounts of AC whereas degradation rates decreased. The lowest amount of AC added resulted in an increase in sorption of aldicarb by a factor of 2.2, which is similar to the increase in sorption found in the present study. Similar results on the inhibitory effect of AC on pesticide degradation were found by Cantwell et al. (1989) for imazaquin and imezethapyr. Moyer et al. (1972) found a strong influence of AC on degradation of atrazine and chlorthiamid, but linuron degradation was not affected.

Although inverse relationships between degradation and sorption are often found, degradation of sorbed chemicals cannot be excluded. Mathematical analysis of the data presented by Guo et al. (1999) suggested that although degradation of aldicarb and 2,4-D occurred primarily in soil solution, sorbed pesticide was not completely protected from degradation. Experimental and modeling work by Guerin and Boyd (1993) suggested that the bacterium Pseudomonas putida had access to a portion of naphthalene sorbed to soil. Degradation of sorbed molecules by microorganisms attached on solid surfaces may occur for some compounds. Microorganisms are generally more abundant at or near soil particle surfaces (Stotzky, 1986). Sorption may, thus, concentrate the pesticide in regions of greatest microbial activity, thereby facilitating degradation. On the other hand, pesticide is often sorbed on the walls of small pores inside soil aggregates or organic macromolecules, which are inaccessible for microorganisms. Degradation mediated by chemical reactions can be accelerated by enhanced adsorption due to catalytic effects of solid surfaces. Adsorption-catalyzed hydrolysis has been found for a number of pesticides including chlorotriazines (Hance, 1970). The effect of sorption on degradation thus depends on many factors including soil physical and microbial characteristics and the properties of the chemical.

Degradation under Static and Flow Conditions
Total amounts of bentazone and cyanazine remaining 9 and 20 d after application and their distribution within the column under static and flow conditions are shown in Fig. 7 . There was a strong difference in the vertical distribution of bentazone following incubation under static or flow conditions 9 d after treatment even though the pesticides were mixed evenly into the soil at the start of the experiment. Concentrations of bentazone increased with increasing depth under static conditions. This may be due to faster degradation in the top soil layer where oxygen contents are likely to be larger than in deeper layers. Another possible reason is that movement of the weakly sorbed compound to depth occurred after the addition of 20 mL water 3 d after the start of the experiment. Concentration profiles of bentazone were variable for columns maintained under flow conditions (Fig. 7). Two of the four replicate columns showed smaller concentrations in the top and bottom layer than in the intermediate layers. Concentrations in the remaining two columns decreased with increasing depth. The difference between the two treatments in the vertical distribution of bentazone 20 d after application was smaller than that observed after 9 d for most of the replicate columns. Despite differences in the vertical distribution of bentazone following incubation under static or flow conditions, total amounts degraded after 20 d were very similar. By 20 d after application, 50.1 and 47.2% of the applied bentazone had degraded under static and flow conditions, respectively. Cyanazine was evenly distributed in the soil columns throughout the experiment (Fig. 7). There was no difference between total amounts of cyanazine degraded within 20 d under static (74.7% of initial) or flow conditions (73.6% of initial).



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 7. Vertical distribution of cyanazine and bentazone, 9 and 20 d after uniform mixing into the soil of replicate soil columns incubated under static (dashed lines) and flow conditions (solid lines). Individual replicates are shown.

 
Degradation of organic chemicals in static vs. flowing systems or degradation at different flow rates has been investigated in a number of studies reported in the literature. There was no consistent effect of water flow on degradation. Guo et al. (1999) found an increase in degradation rate constants of aldicarb in leaching columns compared to batch incubation studies by factors of 1.6 to 2.2 depending on the amount of activated carbon added to the soil as an amendment. Enhanced degradation rates at increased water flow velocity were also found by Guo and Wagenet (1999) for alachlor. Heistermann et al. (2003) compared degradation of metsulfuron-methyl in batch experiments using disturbed soil with that under flow conditions in undisturbed microlysimeters. Degradation of metsulfuron-methyl in the microlysimeters (DT50 = approximately 17 d) was much faster than in batch studies (DT50 = 66 d). Faster degradation under flow conditions has been attributed to a number of causes including (i) improved availability of the substrate for degrading microorganisms due to enhanced mixing, (ii) improved aeration and oxygen supply, (iii) improved availability in the soil solution under flow conditions due to non-equilibrium sorption (Guo and Wagenet, 1999), and (iv) removal of products of microbial reactions (Kelsey and Alexander, 1995).

In contrast, Langner et al. (1998) found that measured apparent first-order rate constants of 2,4-D obtained decreased with increasing flow velocity in soil columns. The degradation rate constant obtained at a very low velocity was similar to that measured in a static batch experiment. Slower degradation under flow conditions was also observed by Kelsey and Alexander (1995) for p-nitrophenol. Processes that may lead to a limited degradation in dynamic systems or systems with increased flow velocity include (i) inhibited biomass accumulation with high water flow velocities, (ii) inhibited desorption of nutrients at high velocities, and (iii) limitation of transformation reactions caused by shorter times for contact between chemical and microbial cells or enzymes.

In the present study, the rate of degradation of bentazone and cyanazine was similar under static or flow conditions. In contrast, differences between the two systems were found in the studies reported above. This may be partly due to differences in the experimental setup. The present study was performed under flow conditions relevant to those in the field. Flow ranged between 2 and 15 mm d–1 (0.016 and 0.130 pore volumes per day) and suction was applied at the bottom of the cores to maintain unsaturated conditions. Bentazone and cyanazine were mixed evenly into the soil and the soil solution was circulated in the column. Soils incubated under static conditions were placed in identical glass columns at the same packing density and the same total water content was maintained in both systems. This allowed separation of the effect of water and solute flow on degradation from the effect of differences in the vertical distribution of the pesticide or from differences in soil aeration. The experiment was performed over 20 d and the total amount of pesticide remaining was measured after extraction of the soil. In contrast, the conditions used in the experiments reported in the literature often differed from field conditions. Steady state saturated water flow conditions were established in many studies and flow velocities exceeded those in the present study by one to two orders of magnitude. The effects observed under these conditions are expected to be much larger than those under realistic conditions in the field. The pesticide was applied to untreated soil as a pulse (Guo and Wagenet, 1999; Guo et al., 1999) or as a continuous influent solution of chemical (Kelsey and Alexander, 1995; Langner et al., 1998). The residence times of water chosen for the column experiments reported in the literature are short with a maximum of approximately 3 d. Degradation rate constants under flow conditions were often determined by fitting mathematical models to breakthrough curves whereas those under static conditions were calculated from total residues extracted from the soil. In some of the studies, the experimental setup (packing density, soil moisture) for the batch experiments differed from that for the column experiment.


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
There is evidence that degradation of pesticides in simple laboratory systems may differ from that in the field, but it is not clear which of the simplifications inherent in laboratory studies present serious shortcomings. This study evaluated the hypotheses that (i) degradation in structured soil with fluctuating moisture and temperature in the field can be characterized by the results from laboratory studies with disturbed soil and constant temperature and moisture regimes, (ii) degradation is independent of sorption, and (iii) degradation under flow conditions is identical to that in static systems. Overall, these hypotheses were confirmed in our test system comprising a clay loam soil and contrasting pesticides. This suggests that the simplifying assumptions evaluated in this study are not serious shortcomings for our test system and that degradation in the field can be characterized using standard laboratory studies. However, a review of the relevant literature showed that the experimental setup used, the properties of the compound tested, and the relative importance of other factors influencing degradation may influence the outcome of the study. Additional work with a wider range of pesticides, test conditions, and simplifying assumptions is required before general conclusions on laboratory to field extrapolation can be drawn.


   ACKNOWLEDGMENTS
 
We are grateful for the assistance of Chris Fryer with analysis for the investigation of degradation under static and flow conditions.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
This work was funded by the British Biotechnology and Biological Sciences Research Council (BBSRC) under Research Grant 63/D14743.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2005 34: 0. [Full Text]  



This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
L. J. Krutz, D. L. Shaner, C. Accinelli, R. M. Zablotowicz, and W. B. Henry
Atrazine Dissipation in s-Triazine-Adapted and Nonadapted Soil from Colorado and Mississippi: Implications of Enhanced Degradation on Atrazine Fate and Transport Parameters
J. Environ. Qual., May 1, 2008; 37(3): 848 - 857.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
D. L. Shaner and W. B. Henry
Field History and Dissipation of Atrazine and Metolachlor in Colorado
J. Environ. Qual., January 9, 2007; 36(1): 128 - 134.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
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 (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Beulke, S.
Right arrow Articles by Walker, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Beulke, S.
Right arrow Articles by Walker, A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Beulke, S.
Right arrow Articles by Walker, A.
Related Collections
Right arrow Batch Studies
Right arrow Pesticides
Right arrow Solute Transport Models
Right arrow Agricultural Pesticides

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