Anaerobic degradation of dicamba and metribuzin in riparian wetland soils

Abstract

In the Atlantic Coastal Plain, riparian wetlands are thought to play an important role in removal of mobilized agricultural chemicals from groundwater. Degradation potential for dicamba (3,6-dichloro-2-methoxybenzoic acid) and metribuzin (4-amino-6-(1,1-dimethylethyl)-3-(methylthio)-1,2,4-triazin-5 (4 H)-one) were evaluated in riparian wetland soils of the Nomini Creek Watershed, VA, using serum bottle microcosms under nitrate reducing (NO 3 amended) and non-nitrate reducing conditions. Soil samples were collected from two surface horizons (referred to as ponded and terrestrial, 1–15 cm) and the subsurface (25–45 cm) of a Bibb (coarse-loamy, siliceous, acid, thermic Typic Fluvaquent) sandy loam in May (soil temperature 16.4°C), September (19.9°C) and November (13.5°C) of 1993 from 12 locations equidistant along a 250 m section of the riparian stream. Metribuzin degradation potential was determined for soils collected in May only. Dicamba was demethylated to 3,6-dichlorosalicylic acid under either nitrate reducing or non-nitrate reducing conditions, while metribuzin was degraded to diketometribuzin (DK), deaminated diketometribuzin (DADK) and deaminated metribuzin (DA). The presence of metabolites DA and DK was confirmed by gas chromatography coupled with mass spectrometry. The rates of dicamba and metribuzin degradation were generally more rapid in the surface than in the subsurface soil microcosms. In contrast to dicamba, the disappearance of metribuzin was slower. Although the degradation experiments were conducted in microcosms under laboratory conditions, they indicate that these riparian wetland soils possess limited potential to degrade dicamba and metribuzin.