Herbicide dynamics in prairie wetlands
Prairie wetlands are affected by agricultural activities, in particular herbicide contamination of wetland ecosystems. The environmental fate of herbicides in wetlands is determined by their persistence and transport. Currently, little is known about the persistence of commonly used herbicides in wetland water and sediment. The objective of this dissertation was to determine the importance of wetland sediments in influencing the fate of commonly used herbicides in prairie wetlands. Sediment sorption is an important dissipation pathway for herbicides. The effects of land use on physicochemical properties of wetland sediment, and the associations between these properties and herbicide sorption characteristics for four herbicides (trifluralin, atrazine, 2,4-D and glyphosate), were examined for 17 wetlands. The sorption of herbicides in sediment increased in the order of 2,4-D < atrazine < glyphosate < trifluralin. Overall, sediments from wetlands that were recently cultivated had lower total organic carbon (TOC), total inorganic carbon (TIC), electrical conductivity (EC), exchangeable cation (EXCAT), cation excheageable capacity (CEC), and sorption coefficient (Kd) values (2,4-D, trifluralin, and atrazine) than sediments from semi-permanent and riparian ephemeral wetlands. Although TOC content was correlated to the sorption of 2,4-D, trifluralin and atrazine, riparian wetland sediments despite having a lower TOC content than semi-permanent wetland sediments, had the highest herbicide sorption capacity for 2,4-D, trifluralin and atrazine. To further examine the link between land use and herbicide sorption by sediment, a multi-residue analytical method was developed to quantify seven sulfonylurea herbicides commonly used in crop production (thifensulfuron-methyl, tribenuron-methyl, ethametsulfuron-methyl, metsulfuron-methyl, rimsulfuron, nicosulfuron and sulfosulfuron) in sediment from 17 wetlands. Sediment was extracted with deionized water using pressurized liquid extraction and the resultant extracts were cleaned-up using Oasis HLB solid-phase extraction cartridges. Quantification and confirmation were performed using liquid chromatography interfaced with positive ion electrospray tandem mass spectrometry, and multiple reaction monitoring. Calibration curves were linear with correlation coefficients greater than 0.994 and limits of quantification ranged from 1.0 to 2.0 µg kg-1. Ethametsulfuron-methyl, sulfosulfuron and metsulfuron-methyl, the three most environmentally persistent of the seven sulfonylurea herbicides monitored in this study, were most frequently detected in wetland sediment. The concentrations of sulfonylurea herbicides were higher in sediments from the semi-permanent wetland, which received surface runoff from a larger catchment. To assess the relationship between herbicide dissipation and sorption to sediment, seven commonly used herbicides (glyphosate, dicamba, bromoxynil, 2,4-D, MCPA, mecoprop-P and dichlorprop) were applied to one half of an ephemeral (E) and a semi-permanent (SP) prairie wetland to mimic a direct overspray event, a worst-case scenario for wetland contamination. Water and sediment samples collected over a 77-d study period (early June to late August) were analyzed for herbicide concentration; aminomethylphosphonic acid (AMPA), the degradation product of glyphosate, was also monitored. Glyphosate dissipated rapidly in the water column of each wetland with DT50 values of 3.7 d for wetland E and 6.9 d for wetland SP. The mass of AMPA in each wetland increased with a concomitant decrease in the mass of glyphosate, suggesting that glyphosate degradation was occurring in the water column. In addition, glyphosate was also lost from the water column via sorption to bottom sediment, as evidenced by its detection in sediment samples up to 42-d post-treatment (39.0 – 224.5 µg kg-1). The DT50 of the other six acid herbicides ranged from 2.3 d (bromoxynil) to 31 d (dichlorprop). The two chiral herbicides, mecoprop-P and dichlorprop, were the most persistent acid herbicides in the water column. Sorption to sediment was also an important dissipation route for these six herbicides in water, especially in wetland E. After 77 d in the semi-permanent wetland and 56 d in the ephemeral wetland, the concentrations of bromoxynil, dicamba and 2,4-D were below the Canadian Water Quality Guidelines for the Protection of Aquatic Life. Concentrations of the more persistent herbicides (mecoprop-P and dichlorprop) remained at levels above the guidelines. Use of bromide ion as a conservative tracer indicated that some of the water loss from both wetlands was via infiltration. Because there was a strong correlation between the decrease in bromide ion and herbicide mass in the water column, it is possible some herbicides were lost from the water column along with the infiltrating water. Infiltration to wetland margins during this part of the growing season would have been largely driven by the riparian vegetation surrounding both wetlands. Overall, this thesis demonstrated the importance of sediment in the dissipation of currently used herbicides in wetland ecosystems. Both laboratory sorption studies and whole-wetland experiments provided insights on the sorptivity of herbicides in sediment as well as their persistence in water and sediment.
DegreeDoctor of Philosophy (Ph.D.)
SupervisorPennock, Dan; Farenhorst, Annemieke
CommitteeCessna, Allan; Waiser, Marley; Elliott, Jane; Muir, Derek; Farrell, Rich
Copyright DateApril 2010