Sorption and mineralization of S-metolachlor and its ionic metabolites in soils and vadose zone solids: Consequences on groundwater quality in an alluvial aquifer (Ain Plain, France)

Highlights

• We study the sorption and mineralization of S-metolachlor and its ionic metabolites.

• We examine the fate of the three molecules in soils and solids of vadose zone.

• We study the transfer of S-metolachlor and its metabolites into alluvial aquifer.

• We demonstrate the link between the recharge and temporal fluctuations of groundwater quality.

Abstract

This study characterizes the transfer of S-metolachlor (SMOC) and its metabolites, metolachlor ethane sulfonic acid (MESA) and metolachlor oxanilic acid (MOXA) to the alluvial aquifer. Sorption and mineralization of SMOC and its two ionic metabolites were characterized for cultivated soils and solids from the vadose (unsaturated) zone in the Ain Plain (France). Under sterile soil conditions, the absence of mineralization confirms the importance of biotic processes in SMOC degradation. There is some adsorption and mineralization of the parent molecule and its metabolites in the unsaturated zone, though less than in soils. For soils, the MESA adsorption constant is statistically higher than that of MOXA and the sorption constants of the two metabolites are significantly lower than that of SMOC. After 246 days, for soils, maximums of 26% of the SMOC, 30% of the MESA and 38% of the MOXA were mineralized. This partly explains the presence of these metabolites in the groundwater at concentrations generally higher than those of the parent molecule for MESA, although there is no statistical difference in the mineralization of the 3 molecules. The laboratory results make it possible to explain the field observations made during 27 months of groundwater quality monitoring (monthly sampling frequency). The evolution of both metabolite concentrations in the groundwater is directly related to recharge dynamics; there is a positive correlation between concentrations and the groundwater level. The observed lag of several months between the signals of the parent molecule and those of the metabolites is probably due to greater sorption of the parent molecule than of its metabolites and/or to degradation kinetics.

Introduction

Groundwater quality is increasingly monitored in Europe, where various levels of nitrate and pesticide and/or metabolite contamination have been confirmed (Loos et al., 2010, Stuart et al., 2012). The Groundwater Daughter Directive (2006/118/EC — Official Journal of the European Union, 2006) to the Water Framework Directive (WFD) particularly requires that measures be taken to prevent or limit the input of pollutants into groundwater and compliance with good chemical status criteria (based on EU nitrate and pesticide standards).

In France, the common occurrence of atrazine and its degradation product deethylatrazine (DEA) in surface- and ground-waters led to the withdrawal of atrazine from the market in 2003. Various alternatives for weed control in maize include the use of chloroacetanilides (metolachlor, acetochlor and, until recently, alachlor). In North America, metolachlor, alachlor, acetochlor and their ionic degradation products are commonly found in groundwater because these molecules have been widely used for many years (Barbash et al., 2001, Hladik et al., 2008, Kolpin et al., 1998, Kolpin et al., 2004, Postle et al., 2004).

In France, the use of racemic metolachlor (rac-metolachlor) was banned on 30/12/2003 (JORF, 2002) and S-metolachlor (SMOC) was added to Annex I of Directive 91/414/EEC. The passage from one substance to another was motivated by the high weed-killing efficiency of SMOC compared to that of rac-metolachlor (Shaner et al., 2006). This led to a reduction in the doses applied by farmers. In fact, the proportions of the four isomers in the marketed product were modified. SMOC is a mixture of 80–100% S-isomers and 20–0% R-isomers. While rac-metolachlor was first marketed in 1977, many countries have now switched from using rac-metolachlor to SMOC (the United States in 1997, Canada in 1998, South Africa in 1998 and Australia in 1999, Ma et al., 2006, and Switzerland in 1997, Buser et al., 2000).

Although rac-metolachlor has been extensively studied, specific data for SMOC are scarce (Bedmar et al., 2011). Some studies do, however, compare rac-metolachlor and SMOC. They generally concern toxicity (Liu et al., 2012, Xu et al., 2010, Ye et al., 2010), and occasionally their behavior in the environment. In terms of adsorption, Shaner et al. (2006), working on 5 soils, did not observe any significant differences in behavior between rac-metolachlor and SMOC, sorption being linearly correlated with the soils' organic carbon and clay contents. In a similar way, when studying the dissipation of molecules in the soil, these authors did not observe any significant differences between the 2 molecules with respect to non-enantioselective degradation, as suggested by Klein et al. (2006). Ma et al. (2006), however, reported a shorter half-life for SMOC than for rac-metolachlor (14.5 days vs. 20.5 days).

It should be noted that “classical” analytical techniques measure all isomers together. Only the use of specific equipment, such as a chiral column, makes it possible to distinguish between them (Klein et al., 2006, Müller et al., 2001). Although several authors have tried to distinguish between the isomers and quantify their respective proportions in surface waters as a result of a change in farming practices (Buser et al., 2000, Kurt-Karakus et al., 2010), most of these studies, and notably those of groundwater, involve the quantification of the parent molecule without any distinction being made of specific isomers.

Degradation experiments carried out in the laboratory have shown that metolachlor ethane sulfonic acid (MESA or ESA) and metolachlor oxanilic acid (MOXA or OXA) are the most prevalent metabolites. White et al. (2010), working on a loamy sand and searching for 8 metolachlor metabolites have shown that MESA and MOXA are predominant. Hydroxymetolachlor was also detected in more than 30% of the soil samples but at concentrations 8 and 10 times lower than those of MESA and MOXA, respectively, and only during the first 28 days of the experiment.

At the European level, as part of the general evaluation of SMOC use, the EU recommends that Member States pay particular attention to the possibility of groundwater contamination by the active substance and its major metabolites when the active substance is applied in areas deemed sensitive in terms of soil and/or climate conditions (SANCO, 2004). However, we believe that few European studies of metabolites in groundwater have been published (Swiss National Monitoring Network, OFEV, 2009; alluvial groundwater in France, Baran et al., 2010). Although the first pan-European reconnaissance of the occurrence of polar organic persistent pollutants in European groundwater revealed the presence of metolachlor in 20.7% of the 164 water samples from 23 European countries, its metabolites were not sought (Loos et al., 2010). The presence of metolachlor metabolites in groundwater is mentioned essentially in North American studies on study sites (Steele et al., 2008) or through regional campaigns (Hladik et al., 2008). Moreover, there are few data relative to temporal monitoring of these substances in groundwater, although pesticide and metabolite concentrations can vary rapidly and significantly as a result, notably, of aquifer hydrodynamics (Baran et al., 2007, Baran et al., 2008, Rowden et al., 2001). In general, the very few available data on the sorption and degradation of the two ionic metabolites come from a few experimental studies (Krutz et al., 2004, Krutz et al., 2006) or modeling studies (Bayless et al., 2008, Webb et al., 2008).

The objectives of this project were therefore, 1) to characterize, under controlled laboratory conditions, the sorption and mineralization of SMOC and its metabolites, MESA and MOXA, in both agricultural soils and solids from the vadose or unsaturated zone of an alluvial aquifer, 2) to understand the temporal dynamics of the groundwater quality by monitoring at least 2 hydrological cycles with a monthly time step, and 3) to compare field data with laboratory data in order to improve our understanding of transfer of the 3 molecules into groundwater.