Aging Effects on the Availability of Herbicides to Runoff Transfer

Dec 29, 2006 - Aging Effects on the Availability of Herbicides to Runoff Transfer ... in the top soil (0−2 cm) to the average concentrations in runo...
0 downloads 17 Views 378KB Size
Download PDF
Environ. Sci. Technol. 2007, 41, 1137-1144

Aging Effects on the Availability of Herbicides to Runoff Transfer XAVIER LOUCHART* AND MARC VOLTZ INRA, Laboratoire d’e´tude des Interactions Sol-Agrosyste`me-Hydrosyste`me, UMR LISAH Agro-M-INRA-IRD, 2 place Viala, 34060 Montpellier Cedex 1, France

Realistic estimation of sorption parameters is essential to predict long-term herbicide availability in soils and their contamination of surface water and groundwater. This study examined the temporal change of an effective partition coefficient Kdeff for the herbicides simazine, diuron, and oryzalin from a 0.12 ha field experiment during 7 vineyard growing seasons. Kdeff is the ratio of solvent extractable herbicide concentrations in the top soil (0-2 cm) to the average concentrations in runoff water and is considered to assess the effective availability of herbicides to runoff transfer. Kdeff increased largely with aging time since application, from values similar to those of the literature (determined in 24 h batch conditions, Kdref), up to 88, 164, and 30 times these initial values for simazine, diuron, and oryzalin respectively. The seasonal variation of Kdeff values between years and compounds could be adequately described by a unique model, taking into account the cumulative rainfall since application and Kdref of each compound. This simple model was able to represent the influence of the soil moisture content and its changes in the different biological and physicochemical processes that may contribute to the (bio)available, sorbed, or entrapped state of any of the studied herbicides with aging time under Mediterranean climate.

Introduction The intensive use of pesticides in agriculture results in contamination of surface waters in many regions throughout the world. The degree and the pattern of pesticide contamination of surface waters will depend mainly on the availability of the applied pesticides to be transferred in and by water. Adsorption-desorption interactions of pesticides with soil determine their availability in soil for transport and degradation processes. They are commonly characterized by the partition coefficient Kd, which is the ratio of the amount of sorbed pesticide remaining in the soil to that released into solution as directly measured using batch techniques. Although the sorption parameter Kd reflects some of the physical and chemical properties of the pesticides, the soilwater-pesticide system exhibits a much more complex behavior (1, 2). The desorption of pesticide from soil under natural and dynamic conditions most often cannot be characterized and modeled by a linear isotherm with an unvarying parameter. In fact, the batch method based on 24 h equilibrium will inherently tend to over-estimate shortterm sorption but under-estimate long-term sorption (2). This method and alternative methods that accelerate and * Coresponding author e-mail: [email protected]; phone: +33.(0)4.99.61.23.79; fax: +33.(0)4.67.63.26.14. 10.1021/es061186q CCC: $37.00 Published on Web 12/29/2006

 2007 American Chemical Society

facilitate the estimation of Kd values (3, 4) do not assess the long-term evolution of Kd. Aging of pesticides in soil appears to significantly affect their sorption-desorption behavior. For instance, many authors have observed in laboratory conditions at equilibrium an increase of pesticide sorption on soils with incubation time of a few weeks to a few months: the most recent observations concerned mainly herbicides like triazines (59), ureas (10-14), or other herbicide classes (15-20). Similar results were reported for fungicides (21, 22) and insecticides (23, 24). In field conditions, under natural weathering with residence times of many months, an increase in sorption on soils of a few herbicides was also observed (25-28). Consequently, it was suggested (9, 14, 29, 30) that aging of pesticides in soils results in a decline in their availability for transport in soil and for uptake by plants and microorganisms. But, there is little experimental evidence that enables quantification of the exact impact of aging on pesticides’ availability to water. Studies are required that examine pesticide availability in natural flow conditions where chemical partition between solid and liquid phases may not necessarily reach the equilibrium since contact times between the two phases are limited due to renewal of liquid phase by the flowing water. Moreover, studies should also examine the change of pesticide availability over long time periods since aging of herbicides in the soil can develop over several months, and in turn, significantly modify the patterns of water contamination during runoff events. To our knowledge, only McCall and Agin (16) studied these processes, but temperature and soil moisture content were held constant during the 300-day incubation experiment. Further work should therefore analyze the variation with time of sorption parameter in situ as influenced by time-varying physical and environmental factors such as soil moisture content and soil temperature, and by the specific physicochemical properties of pesticides. This paper describes an analysis of increased sorption with time using an extensive set of in situ experimental data. We evaluated under Mediterranean soil and climate conditions the long-term effects of aging on the availability to transport by overland flow of three soil-applied herbicides of different chemical classes. To this aim, we analyzed the herbicide contents in soil and runoff water observed over 7 years on a vineyard. Herbicide availability was examined by the ratio, Kdeff, of the solvent-extractable contaminant content in the top soil to the average concentration in runoff water. The in situ estimated partition coefficient Kdeff reflects the effective availability of pesticides to desorption by varying water flux during a runoff event.

Materials and Methods Site and Soil. The fate and transport of herbicides were monitored over several years on a 0.12 ha vineyard (Cinsault, Vitis vinifera) located in southern France (43°30′ N, 3°19′ E), 60 km west from Montpellier and belonging to the longterm Environmental Research Observatory OMERE (http:// sol.ensam.inra.fr/omere/). The soil of the site was a calcaric cambisol (31). The main soil properties of the 0-5 cm layer are 1.44 g cm-3 bulk density, pH (H2O) 8.5, 1.06% organic C, 18.6% clay fraction (0-2 µm), 55.4% silt fraction (2-50 µm), 26% sand fraction (50-2000 µm), and 24% CaCO3. Studied Compounds and Field Applications. Three preemergent herbicides frequently used in the past decade in vineyards in South France to control weeds were studied: simazine (6-chloro-N,N′-diethyl-1,3,5-triazine-2,4-diamine), diuron (N′-(3,4-dichlorphenyl)-N,N-dimethylurea), and oryzVOL. 41, NO. 4, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1137

FIGURE 1. Temporal variation of herbicide contents expressed as percentage of the initial herbicide content measured just after application (Co) in the topsoil (0-2 cm) for simazine, diuron, and oryzalin for all studied years. Error bars represent the standard error of herbicide residue in topsoil. alin (3,5-dinitro-N4,N4-dipropylsulfanilamide). They belong to three different chemical classes and have significantly varying physicochemical properties representative of the variability of the actual herbicides used in vineyards. The range of Kd values (derived from the Koc in literature (32, 33) with Kd ) Koc foc where foc is the organic C fraction of the studied soil) is 1.1-2.9, 4.2-5.1, and 6.4-11.7 L kg-1 for simazine, diuron, and oryzalin, respectively. Afterward, Kdref will refer to the median of these Kd values for each compound. All herbicides were sprayed manually as a mixture by the farmer over the entire soil surface at varying application rates and on different dates (Table S-1 in the Supporting Information). Simazine was applied from 1997 to 1999, oryzalin was applied from 1998 to 1999, and diuron was applied from 1995 to 2004, but the field was not monitored in 1996, 2000, and 2001. Devices and Sampling. Detailed methodology and sampling strategy can be found in previous works (34-36). During the study period, the content of herbicides in the topsoil layer (0-2 cm) was regularly measured to study their dissipation. Soil samples were taken weekly after application and then monthly. During all runoff events, the herbicide concentrations in runoff water at the outlet of the field were monitored by automatic water sampling driven by both discharge and time. In all, 80 runoff events were observed 1138

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007

with a number of water samples ranging from 3 to 47 according to the duration and intensity of the event. Analytical Methods. The methods used to extract and quantify simazine, diuron, and oryzalin in both soil and water samples are described in previous papers (34-36). Calculations. The effective partition coefficient, Kdeff, was calculated for the time of each runoff event t as follows:

Kdeff(t) ) Cs(t)/Cw(t) where Cs(t) is the field average herbicide contents in the 0-2 cm topsoil ([µg kg-1]) and Cw(t) is the average herbicide concentrations in water during the runoff event ([µg L-1]). Since the soil sampling dates rarely coincided with the occurrence of runoff events, Cs(t) was estimated by linear interpolation between the previous (t1) and the following (t2) soil sampling dates. Its associated standard error (σs) was computed with a Taylor series expansion of the total variance:

σs(t)2 ) [(t2 - t)2σs(t2)2 + (t - t1)2σs(t1)2]/(t2 - t1)2 Cw(t) was calculated as the arithmetic mean of all samples collected by the automatic sampler during the runoff event. Its associated standard error, σw(t), was also computed.

TABLE 1. Estimated Field Half-Life (DT50, in Days) and Main Characteristic Values of the Calculated Kdeff (L kg-1) for the Studied Compounds: Minimum and Maximum Values, Temporal Rank of Min and Max Values among the Runoff Events of the Year and Corresponding Days since Application year

DT50 (R 2)

min Kdeff (stda)

max Kdeff (stda)

days after rank application min/max min/max

simazine 1997 1998 1999 average

30 (0.94) 39 (0.83) 21 (0.80) 30

3.3 (( 1.6) 2.7 (( 1.2) 2.3 (( 0.6) 2.7

144.5 (( 12.8) 1-10/10 29/253 242.5 (b) 1-10/10 46/312 119.1 (b) 1-4/4 36/215 168.7

diuron 1995 1997 1998 1999 2002 2003 2004 average

49 (0.89) 9.8 (( 2.0) 25 (0.96) 3.9 (( 1.0) 37 (0.91) 4.0 (( 1.9) 18 (0.80) 5.7 (( 1.1) 17 (0.95) 1.5 (( 1.8) 21 (0.86) 5.4 (( 0.2) 15 (0.97) 10.7 (( 0.7) 26 5.85

157.9 (( 48.9) 199.9 (( 6.4) 109.0 (b) 100.5 (b) 239.9 (( 52.7) 114.4 (c) 84.1 (c) 143.7

1-9/9 1-12/13 1-11/11 1-10/11 1-15/15 2-9/11 1-8/10

17/249 29/253 46/304 36/181 30/282 60/268 9/179

oryzalin 1998 30 (0.97) 1999 18 (0.84) average 24

8.3 (( 2.8) 162.1 (( 32.0) 1-9/9 6.7 (( 0.9) 204.3 (b) 1-8/8 7.5 183.2

46/198 36/215

a Values in parentheses are the standard error. b Standard error of concentration in the topsoil could not be calculated (single mixed sample). c Standard error of concentration in runoff water could not be calculated (single mixed sample).

Finally, when possible we calculated the standard error of the Kdeff values according to the following equation:

σKdeff(t)2 ) [1/Cw(t)]2 σs(t)2 + [-Cs(t)2/Cw(t)2]2 σw(t)2 All calculations and model fitting (see Results section) were made with Matlab software (37). All together, the experiments provided sets of 24, 80, and 17 pairs of simazine, diuron, and oryzalin concentrations in soil and runoff water, respectively. Datasets were different for each compound because all chemicals were not applied each year (Table S-1).

Results and Discussion Dissipation of Herbicides in the Topsoil. The changes with time of herbicide content in topsoil are shown in Figure 1 and estimated field half-lives (DT50) are given in Table 1. For all compounds and all studied years, the dissipation of herbicides exhibited two distinct phases. The first phase, which lasted 90-100 days, corresponded to an initial dissipation of the herbicides. Within this period, at least 90% of the applied herbicide disappeared or was not longer extractable by organic solvent, as indicated by the DT50 values. The second phase, extending from 90 to 100 days after application to the end of the year, exhibited a significantly slower dissipation rate than the first phase. In most cases, two subperiods could be distinguished during the second phase (except for simazine in 1998, diuron in 1995 and 1998): the start of the second phase was marked by a stabilization of herbicide content in topsoil, whereas in the last part the herbicide content decreased slightly and reached a few percent of the amount initially applied (Figure 1). Thus, two-three distinct phases could be distinguished each year. Main Characteristics of Kdeff Temporal Variation. The change of Kdeff with time since application is presented in Figure 2 and main characteristic values are given in Table 1.

TABLE 2. Equations and Fitted Parameters for Kdeff (L kg-1) or ln(Kdeff) as a Function of Either the Time since Application (t in Days) or the Cumulative Rainfall since Application (Rc in mm) (Equations Were Fitted for Simazine (N ) 24), Diuron (N ) 80), and Oryzalin (N ) 17) Respectively) RMSEa

k1 (b)/ln(Kdref)

k2 (b)

compound

R2

simazine diuron oryzalin

0.43 0.42 0.47

eq 1: Kdeff ) k1 + k2t0.5 32.6 0 (-23.8) 23.1 0 (-10.1) 20.5 0 (-52.7)

simazine diuron oryzalin

0.80 0.61 0.83

eq 2: ln(Kdeff) ) k1 + k2t0.5 19.5 0 (-0.74) 19.0 1 (0.54) 11.7 0.81 (0.09)

0.29 (0.223) 0.21 (0.169) 0.30 (0.228)

simazine diuron oryzalin

0.84 0.76 0.77

eq 3: ln(Kdeff) ) k1 + k2Rc0.5 17.2 0.79 (0.32) 14.9 1.39 (1.10) 13.5 1.80 (1.22)

0.18 (0.147) 0.14 (0.122) 0.17 (0.122)

simazine diuron oryzalin

eq 4: ln(Kdeff) ) ln(Kdref) + k2Rc0.5 0.84 17.3 0.69 0.19 (0.17) 0.75 15.0 1.53 0.13 (0.13) 0.74 14.4 2.2 0.14 (0.12)

a Root-mean-square error. 95% confidence intervals.

b

2.02 (-0.735) 2.84 (1.713) 4.05 (-2.506)

Values in parentheses are those of the

For all compounds and all studied years, Kdeff increased rapidly (10 times within 1 month after the first runoff event, Figure 2) and steadily from the application onward. In every instance, the minimum Kdeff values were those calculated for the first runoff event. At the end of the growing season, Kdeff increased up to 44-88 times its observed value just after application for simazine, 8-164 for diuron, and 19-30 for oryzalin (Table 1). The smaller the observed initial value of Kdeff for a compound the larger the increase, that is, 62.6, 24.5 and 24.4 times for the average values for simazine, diuron and oryzalin respectively. Average Kdeff values of the first runoff event agree well with reported Kdref values. It can therefore be assumed that for the first runoff event (i) the availability of the compounds in the topsoil is high, (ii) the herbicides readily desorb into runoff water and the partition between soil and water reaches the same equilibrium as that obtained within 24 h period used in batch sorption studies. Thus, the calculation of Kdeff for the first runoff event after application can be an appropriate method to estimate Kdref and vice versa. During a natural weathering period of 179-312 days (Table 1), Kdeff increased much more than Kd reported from previous studies. For example, for studies carried out under constant and controlled climatic conditions in laboratory, the increase of Kd was between 1.6 and 6.6 depending on the compound, the type of soil and the incubation time (1, 16, 38). For soil samples collected in the field after a long residence time under natural climatic conditions, Kd did not increase more than 7.8 times over 175 days for isoproturon (26), and 42 times over 15 months for atrazine and metolachlor (27). From these results, the following conclusions can be drawn: (i) The increase in sorption may be greater when the soil samples are aged under natural conditions as compared to those incubated in controlled conditions. (ii) The factor of increase of Kdeff may not be directly proportional to the aging time of the compounds in the soil. (iii) Kdeff increased much more than the Kdref, which indicates that aging influences comparatively more the short-term pesticide desorption properties as observed when contact times last only a few to tens minutes than the desorption characteristics observed after 24 h equilibrium. This highlights the utility of a partition coefficient like Kdeff to quantify the susceptibility of herbicides to extraction and removal by runoff water. VOL. 41, NO. 4, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1139

FIGURE 2. Temporal variation of Kdeff values (L kg-1) for simazine, diuron, and oryzalin since application for all studied years. Modeling Kdeff with Aging Time. The variation of Kdeff with time can be divided into three distinct phases (Figure 2) that coincide very nearly with the phases described for the dissipation (Figure 1). The first one (from application to 6090 days) corresponds to a rapid increase of Kdeff. During the second phase (from 60-90 to 150-200 days) Kdeff did not increase or decreased slightly as compared to the latest values of the first phase. This second phase corresponds to the summer period with small rainfall. It is only visible for the years in which a sufficient number of runoff events occurred and enabled estimation of Kdeff during summer, which was not the case for simazine and even less for oryzalin. About 150-200 days after application, Kdeff increased again but more slowly than during the former phase. The phases seem to follow an exponential increase, and therefore the whole evolution cannot be described mathematically with a simple linear equation. To take into account the changes of sorption parameters with time, various authors have considered the following time-dependent empirical equation:

Kd ) k1 + k2t1/2

(1)

where k1 and k2 are empirical parameters and t is the time since application (21). It is not possible to ascribe a physical significance to this equation, although a relationship with t1/2 suggests the possible involvement of a diffusion controlled 1140

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007

process (39). Due to the different phases of Kdeff variations and the increase in Kdeff of 2 orders of magnitude, the fitting of the values with eq 1 is quite unsatisfactory (Table 2, Figure S-1a in the Supporting Information). The equation that best fitted our data is of the form

ln(Kdeff) ) k1 + k2t1/2

(2)

Equation 2 explained more than 61% of the total variance (Table 2, Figure S-1b). Some of the unexplained variance may be caused by the high intra- and inter-annual variability in the climatic conditions. Influence of Climatic Conditions on the Availability of Herbicides. Figure 3 presents the daily temperature and water deficit that reflects the soil moisture content of the topsoil. Among the studied years, three phases can also be observed that corresponded to those for herbicide content in the topsoil (Figure 1) and those for Kdeff values (Figure 2). During the first phase (0-60 days), the temperature did not exceed 20 °C and no water deficit was observed. From days 60 to 90 (May-June approximately), the temperature exceeded 20 °C and the water deficit increased rapidly and reached almost its maximum 150-200 days after application. During the summer period under Mediterranean climate, the soil moisture content falls below 10%, but can return temporarily to values close to saturation during intense storms, which produces drying and wetting cycles. The third phase (from

FIGURE 3. Daily water deficit (mm) as calculated by cumulative daily precipitation minus cumulative daily Penman evapotranspiration minus cumulative daily runoff since application. Symbols correspond to the runoff events for which Kdeff were calculated (a). Daily temperature (°C) vs days since application (b). ∼200 days) is characterized by the more frequent and substantial rainfall events in fall (except for 1998) that reduced the water deficit at the soil surface. Herbicide dissipation and Kdeff are highly correlated with the variation phases of soil moisture content and temperature. This is hereafter analyzed in relation to the biological and physicochemical processes that could be involved. After herbicide application in March-April, the rapid increase of Kdeff suggests that the water concentrations are decreasing at a faster rate than the soil concentrations. During this period, soil moisture content and temperature could have enabled substantial biodegradation of the compounds, which is recognized as the main source of degradation processes and rapid dissipation for the studied herbicides (35, 36, 40), in particular when the soil is wet or has been rewetted (41, 42). Biodegradation occurs predominantly in the bulk aqueous phase, or at least when the internal pore spaces of the soil particles where chemicals are sorbed do not exclude microbes (>1 µm) (43). Thus, in our field conditions, biodegradation may rapidly lead to a decline of the available fraction of herbicides that could desorb into runoff water. Another part of this fraction could also have become unavailable due to abiotic processes that have slower kinetics than biodegradation but become greater with aging time (29). It is assumed that aging involves redistribution of the herbicides from weaker to stronger adsorption sites; slow sorption; diffusion and sequestration or entrapment into sites

within the soil matrix that are not readily accessible by even the smallest of micro-organisms; covalent bond formation between the compounds and the soil organic matter (SOM); or a combination of these processes (30, 44-46). During the summer period (second phase), the dry conditions slow down the microbial activity and thus biodegradation (36, 47). Herbicide concentrations in soil and water may decrease at a similar rate (stabilization of Kdeff). The alternation of swelling and shrinking processes, caused by the drying and wetting cycles, may also alter severely the macromolecular structure of SOM (48) and may have significant impacts on the alteration of herbicide sorption capacity and strength (49, 50). The persistence of these alterations during short rewettings in summer may be related to the slow kinetics (many days or weeks) of the swelling process until complete recovery (48, 51). Thus, dry conditions and drying and wetting cycles may stabilize the proportions of available (desorbable into water) and sorbed (solvent extractable) herbicides. During the third phase, the more frequent precipitations may be favorable again for the diffusion and slow desorption of herbicides. But, these conditions did not enable recovery of the same kinetics of Kdeff increase as during the first phase, since rewetting of SOM only leads to a partial recovery of the diffusional pore space (50) and the slow kinetics of hydration process restrict the diffusion and liberation of herbicides. The availability of herbicides for mobilization to runoff water and thereby for biodegradation appears to be rate-diffusion VOL. 41, NO. 4, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1141

FIGURE 4. Temporal variation of Kdeff values (L kg-1) for simazine, diuron, and oryzalin for all studied years vs cumulative rainfall since application. Lines represent the fit of eq 3. limited, which is well considered by the parabolic diffusion law of eq 2 (Table 2). This suggests that Kdeff values should reach a maximum after a certain aging time, which is consistent with the idea that the structural expansion and the physical alteration of SOM can be irreversible and therefore can cause irreversible sorption (52, 53). In summary, the soil moisture conditions that occur at a given time, but also those that prevailed previously, apparently influence the aging process of pesticides in soil and their available fraction to water. Modeling Kdeff with Cumulative Rainfall. Modeling the long-term evolution of Kdeff should therefore take the soil moisture conditions into account. In semi-arid zones like South-France, the change of topsoil moisture content is mostly dependent on rainfall. The cumulative rainfall since application (Rc) reflects implicitly both time and soil moisture conditions since application. Moreover, it is currently available data. We therefore assumed that Rc could adequately explain the variable rates of increase of Kdeff with time. We therefore replaced t by Rc in eq 2:

ln(Kdeff) ) k1 + k2Rc1/2

(3)

Equation 3 improved the fit largely for diuron and slightly for simazine as compared to eq 2 (Table 2, Figure S-1c). As there was a low number of runoff events recorded for oryzalin 1142

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007

during the two monitored years, particularly for the second phase of Kdeff, Rc did not provide any benefit as compared to t in eq 2. It must be stressed that for all compounds the increase of Kdeff is now continuous and almost uniform between the years, except for 1998 which was particularly “dry” (Figure 4). No plateau phase or transition phase can any longer be observed for diuron and simazine. This result highlights the role of soil moisture content and its rapid changes for the changes of herbicide availability and the possibility in our conditions to correlate the soil moisture content and the amount of rainfall. Influence of Herbicide Properties. The evolution pattern of Kdeff was almost the same regardless of the herbicide. The differences between the physicochemical characteristics of the compounds (Koc, solubility, polarity) were less pronounced when comparing the Kdeff values calculated for each herbicide, particularly the maximum values (Table 1). The fitted parameters k1 (eq 3) for the three compounds were in the same order as their respective Kdref. The values of k1 were close to ln(Kdref), which enables eq 3 to be simplified to the following:

ln(Kdeff) ) ln(Kdref) + k2Rc1/2

(4)

Thus, by normalizing the Kdeff values of each compound by the corresponding Kdref value, a simple and unique model

gives good results whatever the herbicide (Table 2, Figure S-1d). In other pedo-climatic conditions, the rates of mobilization of herbicides from the topsoil to runoff water were inversely correlated with their sorption capacities with organic matter (54). In laboratory, significant differences between the long-term extractable amounts of several herbicides were also observed after incubation in constant conditions (7). In the conditions of our study, the soil moisture content and its changes prevail on the physicochemical properties of the herbicides regarding the variation of their availability in the topsoil for water transfer. Taking into account the soil moisture content (Rc), and to a lesser extent the chemical properties (Kdref), our model allows us to quantify the availability of the three herbicides and to reproduce its changes with time. The validity of this model should be tested with other pesticides, particularly ionic compounds. Further studies are also needed to better understand and directly model the role of soil moisture changes on the aging process.

Acknowledgments This research was supported by funds provided by the French National Institute for Agricultural Research (INRA), the French Ministry of Research (ORE-FNS), the French National Program in Hydrological Research (PNRH), the LanguedocRoussillon Region, and by research contracts with Rhoˆne Poulenc and Dow AgroSciences.

Supporting Information Available Table S-1 (main physicochemical properties of herbicides, application rates, and date), Figure S-1 (estimated vs calculated Kdeff values according to the different equations used). This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Koskinen, W. C.; Rice, P. J.; Anhalt, J. A.; Sakaliene, O.; Moorman, T. B.; Arthur, E. L. Sorption-desorption of “aged” sulfonylaminocarbonyltriazolinone herbiides in soil. J. Agric. Food Chem. 2002, 50, 5368-5372. (2) Wauchope, R. D.; Yeh, S.; Linders, J. B. H. J.; Kloskowski, R.; Tanaka, K.; Rubin, B.; Katayama, A.; Kordel, W.; Gerstl, Z.; Lane, M.; Unsworth, J. B. Pesticide soil sorption parameters: theory, measurement, uses, limitations and reliability. Pest Manage. Sci. 2002, 58, 419-445. (3) Gawlik, B. M.; Sotiriou, N.; Feicht, E. A.; Schulte-Hostede, S.; Kettrup, A. Alternatives for the determination of the soil adsorption coefficient, Koc, of non-ionicorganic compounds a review. Chemosphere 1997, 34, 2525-2551. (4) Yazgan, M. S.; Wilkins, R. M.; Sykas, C.; Hoque, E. Comparison of two methods for estimation of soil sorption for imidacloprid and carbofuran. Chemosphere 2005, 60, 1325-1331. (5) Garcı´a-Valca´rcel, A. I.; Tadeo, J. L. Influence of soil moisture on sorption and degradation of hexazinone and simazine in soil. J. Agric. Food Chem. 1999, 49, 3895-3900. (6) Kottler, B. D.; White, J. C.; Kelsey, J. W. Influence of soil moisture on the sequestration of organic compounds in soil. Chemosphere 2001, 42, 893-898. (7) Pa¨tzold, S.; Bru ¨ mmer, G. W. Influence of microbial activity and soil moisture on herbicide immobilization in soils. J. Plant Nutr. Soil Sci. 2003, 166, 336-344. (8) Pignatello, J. J.; Ferradino, F. J.; Huang, L. Q. Elution of aged and freshly added herbicides from soil. Environ. Sci. Technol. 1993, 27, 1563-1571. (9) Sharer, M.; Park, J.-H.; Voice, T. C.; Boyd, S. A. Aging effects on the sorption-desorption characteristics of anthropogenic organic compounds in soil. J. Environ. Qual. 2003, 32, 1385-1392. (10) Altfelder, S.; Streck, T.; Richter, J. Effect of air-drying on sorption kinetics of the herbicide chlortoluron in soil. J. Environ. Qual. 1999, 28, 1154-1161. (11) Gaillardon, P. Influence of soil moisture on long-term sorption of diuron and isoproturon by soil. Pestic. Sci. 1996, 47, 347354.

(12) Renaud, F. G.; Leeds-Harrison, P. B.; Brown, C. D.; Van Beinum, W. Determination of time-dependent partition coefficients for several pesticides using diffusion theory. Chemosphere 2004, 57, 1525-2535. (13) van Beinum, W.; Beulke, S.; Brown, C. D. Pesticide sorption and desorption by lignin described by an intraparticle diffusion model. Environ. Sci. Technol. 2006, 40, 494-500. (14) Walker, A.; Rodriguez-Cruz, M. S.; Mitchell, M. J. Influence of ageing of residues on the availability of herbicides for leaching. Environ. Pollut. 2005, 133, 43-51. (15) Krieger, M.; Merritt, D. A.; Wolt, J. D.; Patterson, V. L. Concurrent patterns of sorption-degradation for oryzalin and degradates. J. Agric. Food Chem. 1998, 46, 3292-3299. (16) McCall, P. J.; Agin, G. L. Desorption kinetics of picloram as affected by residence time in soil. Environ. Toxicol. Chem. 1985, 4, 37-44. (17) Xing, B.; Pignatello, J. J. Time-dependent isotherm shape of organic compounds in soil organic matter: Implications for sorption mechanism. Environ. Toxicol. Chem. 1996, 15, 12821288. (18) Boivin, A.; Amellal, S.; Schiavon, M.; Van Genuchten, M. T. 2,4Dichlorophenoxyacetic acid (2,4-D) sorption and degradation dynamics in three agricultural soils. Environ. Pollut. 2005, 138, 92-99. (19) Bresnahan, G. A.; Koskinen, W. C.; Dexter, A. G.; Cox, L. Sorptiondesorption of ‘aged’ isoxaflutole and diketonitrile degradate in soil. Weed Res. 2004, 44, 397-403. (20) Menasseri, S.; Koskinen, W. C.; Yong Yen, P. Sorption of aged dicamba residues in soil. Pest Manage. Sci. 2003, 60, 297-304. (21) Beigel, C.; Barriuso, E.; Di Pietro, L. Time dependency of triticonazole fungicide sorption and consequences for diffusion in soil. J. Environ. Qual. 1997, 26, 1503-1510. (22) Burton, E. D.; Phillips, I. R.; Hawker, D. W. Tributyltin partitioning in sediments: Effect of aging. Chemosphere 2006, 63, 73-81. (23) Ahmad, R.; Kookana, R. S.; Megharaj, M.; Alston, A. M. Aging reduces the bioavailability of even a weakly sorbed pesticide (carbaryl) in soil. Environ. Toxicol. Chem. 2004, 23, 2084-2089. (24) Cox, L.; Koskinen, W. C.; Yen, P. Y. Sorption-desorption of imidacloprid and its metabolites in soils. J. Agric. Food Chem. 1997, 45, 1468-1472. (25) Barriuso, E.; Koskinen, W. C.; Sorenson, B. Modification of atrazine desorption during field incubation experiments. Sci. Total Environ. 1992, 123/124, 333-344. (26) Beck, A. J.; Harris, G. L.; Howse, K. R.; Johnston, A. E.; Jones, K. C. Spatial and temporal variation of isoproturon residues and associated sorption/desorption parameters at the field scale. Chemosphere 1996, 33, 1283-1295. (27) Pignatello, J. J.; Huang, L. Q. Sorptive reversibility of atrazine and metolachlor residues in field soil samples. J. Environ. Qual. 1991, 20, 222-228. (28) Scribner, S. L.; Benzing, T. R.; Sun, S.; Boyd, S. A. Desorption and bioavailability of aged simazine residues in soil from a continuous corn field. J. Environ. Qual. 1992, 21, 115-120. (29) Alexander, M. Aging, bioavailability, and overestimation of risk from environmental contaminants. Environ. Sci. Technol. 2000, 34, 4259-4265. (30) Gevao, B.; Jones, K.; Semple, K.; Craven, A.; Burauel, P. Nonextractable pesticide residues in soil. Environ. Sci. Technol. 2003, 37, 138A-144A. (31) FAO-UNESCO. Soil map of the world at 1:5000000, 2nd ed.; Food and Agriculrure Organization: Rome, 1989. (32) Hornsby, A. G.; Wauchope, R. D.; Herner, A. E. Pesticides Properties in the Environment; Springer-Verlag: New York, 1996. (33) Tomlin, C. D. S. The Pesticide Manual, 11th ed.; British Crop Protection Council: Bracknell, UK, 1997. (34) Lennartz, B.; Louchart, X.; Andrieux, P.; Voltz, M. Diuron and simazine losses to runoff water in mediterranean vineyards. J. Environ. Qual. 1997, 26, 1493-1502. (35) Louchart, X.; Voltz, M.; Andrieux, P.; Moussa, R. Herbicides transport at field and watershed scales in a Mediterranean vineyard area. J. Environ. Qual. 2001, 30, 982-991. (36) Louchart, X.; Voltz, M.; Coulouma, G.; Andrieux, P. Oryzalin fate and transport in runoff water in Mediterranean vineyards. Chemosphere 2004, 57, 921-930. (37) The Mathworks. Using Matlab, version 6; Natick, MA, 2000. (38) Park, J.-H.; Feng, Y.; Yong Cho, S.; Voice, T. C.; Boyd, S. A. Sorbed atrazine shifts into non-desorbable sites of soil organic matter during aging. Water Res. 2004, 38, 3881-3892. (39) Crank, J. The Mathematics of Diffusion, 2nd ed.; Oxford University Press: London, 1975. VOL. 41, NO. 4, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1143

(40) Cullington, J. E.; Walker, A. Rapid biodegradation of diuron and other phenylurea herbicides by a soil bacterium. Soil Biol. Biochem. 1999, 31, 677-686. (41) Baer, U.; Calvet, R. Fate of soil applied herbicides: experimental data and prediction of dissipation kinetics. J. Environ. Qual. 1999, 28, 1765-1777. (42) Shelton, D. R.; Sadeghi, A. M.; Karns, J. S.; Hapeman, C. Y. Effect of wetting and drying of soil on sorption and biodegradation of atrazine. Weed Sci. 1995, 43, 298-305. (43) Haws, N. W.; Ball, W. P.; Bouwer, E. J. Modeling and interpreting bioavailability of organic contaminant mixtures in subsurface environments. J. Contam. Hydrol. 2006, 82, 255-292. (44) Hatzinger, P. B.; Alexander, M. Effect of aging of chemicals in soil on their biodegradability and extractability. Environ. Sci. Technol. 1995, 29, 537-545. (45) Pignatello, J. J.; Xing, B. Mechanisms of slow sorption of organic chemicals to natural particles. Environ. Sci. Technol. 1996, 30, 1-11. (46) White, J. C.; Quinones-Rivera, A.; Alexander, M. Effect of wetting and drying on the bioavailability of organic compounds sequestered in soil. Environ. Toxicol. Chem. 1998, 17, 23782382. (47) Strandberg, M. T.; Scott-Fordsmand, J. J. Field effects of simazine at lower trophic levels - a review. Sci. Total Environ. 2002, 296, 117-137. (48) Schaumann, G. E.; LeBoeuf, E. J. Glass transitions in peat: their relevance and the impact of water. Environ. Sci. Technol. 2005, 39, 800-806.

1144

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007

(49) Louchart, X.; Voltz, M.; Lennartz, B. Sorption behaviour of diuron under Mediterranean climate. Agron. Sustain. Dev. 2005, 25, 301-307. (50) Lennartz, B.; Louchart, X. Effect of drying on the desorption of diuron and terbuthylazine from natural soils. Environ. Pollut. 2006, in press (doi:10.1016/j.envpol.2006.04.040). (51) Belliveau, S. M.; Henselwood, T. L.; Langford, C. H. Soil wetting processes studied by magnetic resonance imaging: correlated study of contaminant uptake. Environ. Sci. Technol. 2000, 34, 2439-2445. (52) Lu, Y.; Pignatello, J. J. History-dependent sorption in humic acids and a lignite in the context of a polymer model for natural organic matter. Environ. Sci. Technol. 2004, 38, 5853-5862. (53) Sander, M.; Lu, Y. F.; Pignatello, J. J. Conditioning-annealing studies of natural organic matter solids linking irreversible sorption to irreversible structural expansion. Environ. Sci. Technol. 2006, 40, 170-178. (54) Leu, C.; Singer, H.; Stamm, C.; Muller, S. R.; Schwarzenbach, R. P. Variability of herbicide losses from 13 fields to surface water within a small catchment after a controlled herbicide application. Environ. Sci. Technol. 2004, 38, 3835-3841.

Received for review May 17, 2006. Revised manuscript received October 11, 2006. Accepted November 13, 2006. ES061186Q