Including spatial variability in Monte Carlo simulations of pesticide leaching

Modeling the fate of p,p'-DDT in water and sediment of two typical estuarine bays in South China: Importance of fishing vessels' inputs

Antifouling paint applied to fishing vessels is the primary source of dichloro-diphenyl-trichloroethane (DDT) to the coastal marine environments of China. With the aim to provide science-based support of potential regulations on DDT use in antifouling paint, we utilized a fugacity-based model to evaluate the fate and impact of p,p'-DDT, the dominant component of DDT mixture, in Daya Bay and Hailing Bay, two typical estuarine bays in South China. The emissions of p,p'-DDT from fishing vessels to the aquatic environments of Hailing Bay and Daya Bay were estimated as 9.3 and 7.7 kg yr(-1), respectively. Uncertainty analysis indicated that the temporal variability of p,p'-DDT was well described by the model if fishing vessels were considered as the only direct source, i.e., fishing vessels should be the dominant source of p,p'-DDT in coastal bay areas of China. Estimated hazard quotients indicated that sediment in Hailing Bay posed high risk to the aquatic system, and it would take at least 21 years to reduce the hazards to a safe level. Moreover, p,p'-DDT tends to migrate from water to sediment in the entire Hailing Bay and Daya Bay. On the other hand, our previous research indicated that p,p'-DDT was more likely to migrate from sediment to water in the maricultured zones located in shallow waters of these two bays, where fishing vessels frequently remain. These findings suggest that relocating mariculture zones to deeper waters would reduce the likelihood of farmed fish contamination by p,p'-DDT.

Fine scale spatial variability of microbial pesticide degradation in soil: scales, controlling factors, and implications

Pesticide biodegradation is a soil microbial function of critical importance for modern agriculture and its environmental impact. While it was once assumed that this activity was homogeneously distributed at the field scale, mounting evidence indicates that this is rarely the case. Here, we critically examine the literature on spatial variability of pesticide biodegradation in agricultural soil. We discuss the motivations, methods, and main findings of the primary literature. We found significant diversity in the approaches used to describe and quantify spatial heterogeneity, which complicates inter-studies comparisons. However, it is clear that the presence and activity of pesticide degraders is often highly spatially variable with coefficients of variation often exceeding 50% and frequently displays non-random spatial patterns. A few controlling factors have tentatively been identified across pesticide classes: they include some soil characteristics (pH) and some agricultural management practices (pesticide application, tillage), while other potential controlling factors have more conflicting effects depending on the site or the pesticide. Evidence demonstrating the importance of spatial heterogeneity on the fate of pesticides in soil has been difficult to obtain but modeling and experimental systems that do not include soil's full complexity reveal that this heterogeneity must be considered to improve prediction of pesticide biodegradation rates or of leaching risks. Overall, studying the spatial heterogeneity of pesticide biodegradation is a relatively new field at the interface of agronomy, microbial ecology, and geosciences and a wealth of novel data is being collected from these different disciplinary perspectives. We make suggestions on possible avenues to take full advantage of these investigations for a better understanding and prediction of the fate of pesticides in soil.

European scenarios for exposure of soil organisms to pesticides

Standardised exposure scenarios play an important role in European pesticide authorisation procedures (a scenario is a combination of climate, weather and crop data to be used in exposure models). The European Food Safety Authority developed such scenarios for the assessment of exposure of soil organisms to pesticides. Scenarios were needed for both the concentration in total soil and for the concentration in the liquid phase. The goal of the exposure assessment is the 90th percentile of the exposure concentration in the area of agricultural use of a pesticide in each of three regulatory European zones (North, Centre and South). A statistical approach was adopted to find scenarios that are consistent with this exposure goal. Scenario development began with the simulation of the concentration distribution in the entire area of use by means of a simple analytical model. In the subsequent two steps, procedures were applied to account for parameter uncertainty and scenario uncertainty (i.e. the likelihood that a scenario that is derived for one pesticide is not conservative enough for another pesticide). In the final step, the six scenarios were selected by defining their average air temperature, soil organic-matter content and their soil textural class. Organic matter of the selected scenarios decreased in the order North-Centre-South. Because organic matter has a different effect on the concentration in total soil than it has on the concentration in the liquid phase, the concentration in total soil decreased in the order North-Centre-South whereas the concentration in the liquid phase decreased in the opposite order. The concentration differences between the three regulatory zones appeared to be no more than a factor of two. These differences were comparatively small in view of the considerable differences in climate and soil properties between the three zones.

Propagation of uncertainties in soil and pesticide properties to pesticide leaching

In the new Dutch decision tree for the evaluation of pesticide leaching to groundwater, spatially distributed soil data are used by the GeoPEARL model to calculate the 90th percentile of the spatial cumulative distribution function of the leaching concentration in the area of potential usage (SP90). Until now it was not known to what extent uncertainties in soil and pesticide properties propagate to spatially aggregated parameters like the SP90. A study was performed to quantify the uncertainties in soil and pesticide properties and to analyze their contribution to the uncertainty in SP90. First, uncertainties in the soil and pesticide properties were quantified. Next, a regular grid sample of points covering the whole of the agricultural area in the Netherlands was randomly selected. At the grid nodes, realizations from the probability distributions of the uncertain inputs were generated and used as input to a Monte Carlo uncertainty propagation analysis. The analysis showed that the uncertainty concerning the SP90 is 10 times smaller than the uncertainty about the leaching concentration at individual point locations. The parameters that contribute most to the uncertainty about the SP90 are, however, the same as the parameters that contribute most to uncertainty about the leaching concentration at individual point locations (e.g., the transformation half-life in soil and the coefficient of sorption on organic matter). Taking uncertainties in soil and pesticide properties into account further leads to a systematic increase of the predicted SP90. The important implication for pesticide regulation is that the leaching concentration is systematically underestimated when these uncertainties are ignored.

Effect of pesticide fate parameters and their uncertainty on the selection of 'worst-case' scenarios of pesticide leaching to groundwater

BACKGROUND

For the registration of pesticides in the European Union, model simulations for worst-case scenarios are used to demonstrate that leaching concentrations to groundwater do not exceed a critical threshold. A worst-case scenario is a combination of soil and climate properties for which predicted leaching concentrations are higher than a certain percentile of the spatial concentration distribution within a region. The derivation of scenarios is complicated by uncertainty about soil and pesticide fate parameters. As the ranking of climate and soil property combinations according to predicted leaching concentrations is different for different pesticides, the worst-case scenario for one pesticide may misrepresent the worst case for another pesticide, which leads to 'scenario uncertainty'.

RESULTS

Pesticide fate parameter uncertainty led to higher concentrations in the higher percentiles of spatial concentration distributions, especially for distributions in smaller and more homogeneous regions. The effect of pesticide fate parameter uncertainty on the spatial concentration distribution was small when compared with the uncertainty of local concentration predictions and with the scenario uncertainty.

CONCLUSION

Uncertainty in pesticide fate parameters and scenario uncertainty can be accounted for using higher percentiles of spatial concentration distributions and considering a range of pesticides for the scenario selection.

Enhanced degradation and soil depth effects on the fate of atrazine and major metabolites in Colorado and Mississippi soils

The aim of this report is to inform modelers of the differences in atrazine fate between s-triazine-adapted and nonadapted soils as a function of depth in the profile and to recommend atrazine and metabolite input values for pesticide process submodules. The objectives of this study were to estimate the atrazine-mineralizing bacterial population, cumulative atrazine mineralization, atrazine persistence, and metabolite (desethylatrazine [DEA], deisopropylatrazine [DIA], and hydroxyatrazine [HA]) formation and degradation in Colorado and Mississippi s-triazine-adapted and nonadapted soils at three depths (0-5, 5-15, and 15-30 cm). Regardless of depth, the AMBP and cumulative atrazine mineralization was at least 3.8-fold higher in s-triazine-adapted than nonadapted soils. Atrazine half-life (T1/2) values pooled over nonadapted soils and depths approximated historic estimates (T1/2 = 60 d). Atrazine persistence in all depths of s-triazine-adapted soils was at least fourfold lower than that of the nonadapted soil. Atrazine metabolite concentrations were lower in s-triazine-adapted than in nonadapted soil by 35 d after incubation regardless of depth. Results indicate that (i) reasonable fate and transport modeling of atrazine will require identifying if soils are adapted to s-triazine herbicides. For example, our data confirm the 60-d T1/2 for atrazine in nonadapted soils, but a default input value of 6 d for atrazine is required for s-triazine adapted soils. (ii) Literature estimates for DEA, DIA, and HA T1/2 values in nonadapted soils are 52, 36, and 60 d, respectively, whereas our analysis indicates that reasonable T1/2 values for s-triazine-adapted soils are 10 d for DEA, 8 d for DIA, and 6 d for HA. (iii) An estimate for the relative distribution of DIA, DEA, and HA produced in nonadapted soils is 18, 72, and 10% of parent, respectively. In s-triazine-adapted soils, the values were 6, 23, and 71% for DIA, DEA, and HA, respectively. The effects of soil adaptation on metabolite distribution need to be confirmed in field experiments.

Agronomic and environmental implications of enhanced s-triazine degradation

Novel catabolic pathways enabling rapid detoxification of s-triazine herbicides have been elucidated and detected at a growing number of locations. The genes responsible for s-triazine mineralization, i.e. atzABCDEF and trzNDF, occur in at least four bacterial phyla and are implicated in the development of enhanced degradation in agricultural soils from all continents except Antarctica. Enhanced degradation occurs in at least nine crops and six crop rotation systems that rely on s-triazine herbicides for weed control, and, with the exception of acidic soil conditions and s-triazine application frequency, adaptation of the microbial population is independent of soil physiochemical properties and cultural management practices. From an agronomic perspective, residual weed control could be reduced tenfold in s-triazine-adapted relative to non-adapted soils. From an environmental standpoint, the off-site loss of total s-triazine residues could be overestimated 13-fold in adapted soils if altered persistence estimates and metabolic pathways are not reflected in fate and transport models. Empirical models requiring soil pH and s-triazine use history as input parameters predict atrazine persistence more accurately than historical estimates, thereby allowing practitioners to adjust weed control strategies and model input values when warranted.

Solute transport in eroded and rehabilitated prairie landforms. 2. Reactive solute

The impact of varying soil, landscape, and climate conditions on the off-site transport of pesticides must be determined to develop improved pesticide management practices. This study quantified the rate of S-metolachlor dissipation after fall and spring application in eroded and rehabilitated landforms in which topsoil was moved from the lower slope to the upper slope. Fall-applied metolachlor provided no control of annual grasses because approximately 80% was removed from the root zone during the winter and early spring, presumably by leaching and runoff. S-Metolachlor dissipated in the spring with a DT(50) of 24-29 days. These results suggest that fall-applied metolachlor may not provide economic weed control and presents an increased risk of water contamination. Although landscape position and bulk soil movement within the landform had a large impact on soil properties, no significant differences in metolachlor dissipation between different landscape positions and between eroded and rehabilitated landforms were observed.

A novel approach to evaluate the production kinetics of extracellular polymeric substances (EPS) by activated sludge using weighted nonlinear least-squares analysis

This paper develops a novel and convenient approach for evaluation of production kinetics of extracellular polymeric substances (EPS) by activated sludge. In this approach, the weighted least-squares analysis is employed to calculate approximate differences in EPS concentration between model predictions and data. An iterative search routine in the Monte Carlo method is utilized for optimization of the objective function by minimizing the sum of squared weighted errors. Application of the approach in this work shows that the fraction of substrate electrons diverted to EPS formation (k(EPS)) is 0.23 g COD(EPS) g(-1) CODs with a bacterial maximum growth rate of 0.32 h(-1). The obtained parameters are found to be reasonable as they are generally bounded. The validity of this approach is confirmed by both the independent EPS production tests and the EPS data reported in literature. It also corrects the overestimation of cellular production and identifies that k(EPS) is the key parameter in EPS production kinetics. Furthermore, this approach could estimate the kinetic parameters accurately using few data sets or even one set, which becomes very attractive for the processes where data are costly to obtain.