Diffuse geographic distribution of herbicides in northern prairie wetlands

The concentrations of herbicides in water from wetlands an landscapes where herbicides are not used should be less than on farms with moderate (conven tional farms) and intense (minimum-till farms) herbicide use. In general, t his hypothesis was not supported for wetlands situated in the Boreal Plains Ecozone of central Saskatchewan, Canada. The overall detection frequency o f 10 commonly used herbicides was not significantly different among wildlif e habitat with no pesticide use (44.4%), farms with no pesticide use (51.68 ), conventional farms (54.9%), and minimum-till farms (56.5%, chi (2) = 5.6 4, p = 0.13). The herbicides (4-chloro-2-methylphenoxy) acetic acid (MCPA), 2,4-dichlorophenoxyacetic acid (2,4-D), bromoxynil, dicamba, mecoprop, and diclorprop accounted for 87% of all detections. In general, detection freq uencies and concentrations of individual herbicides were similar on all lan d-use types. For example, the mean concentration of 2,4-D in water on the f our land types ranged from 0.12 +/-: 0.104 to 0.26 +/- 0.465 mug/L, and MCP A ranged from 0.08 +/- 0.078 to 0.19 +/- 0.166 mug/L. However, in the year of application, mean concentrations of MCPA and bromoxynil, but not 2,4-D, were significantly higher by about twofold in wetlands situated in fields w here these herbicides were applied compared with all other wetlands. We pro pose that many agricultural pesticides are rapidly lost to the atmosphere a t the time of application by processes such as volatilization from soil and plant evapotranspiration. Then, the herbicides used throughout the region may be directly absorbed to the surface of wetlands from the atmosphere, or they become entrained in local convective clouds, and are redistributed by rainfall in a relatively homogenous mixture over the agricultural landscap e. The low levels of individual herbicides we found in most of the wetland waters would not cause chronic effects to aquatic biota.