In conjunction with operational forest herbicide spray programs in Ontario, Canada, chemical and biological monitoring studies were conducted in 51 different wetlands to quantify the probability and magnitude of contamination by a glyphosate herbicide formulation (Vision). Wetlands were classified as oversprayed, adjacent, or buffered in relation to the operational target spray blocks. Results show that vegetated buffers significantly mitigated against exposure and thus potential for acute effects. Aqueous concentrations of glyphosate in buffered wetlands were below analytical limits of quantitation (0.02 mg acid equivalent [a.e.]/L) in 14 of 16 cases, with mean concentration (0.03 +/- 0.02 mg a.e./L) significantly (p < 0.05) less than that of either adjacent (0.18 +/- 0.06 mg a.e./L) or oversprayed wetlands (0.33 +/- 0.11 mg a.e./L). Biomonitoring with caged amphibian larvae showed no significant differences among mean mortality (48 h) of either Rana pipiens (p = 0.194) or Rana clamitans larvae (p = 0.129) exposed in situ to Vision under these various wetland conditions. Percent mortality was not significantly (p = 0.05) correlated with exposure concentrations for either amphibian species tested. Results suggest that exposures typically occurring in forest wetlands are insufficient to induce significant acute mortality in native amphibian larvae.

The effects of Vision (glyphosate, 356 mg acid equivalents (a.e.)/L) on mortality, avoidance response, and growth of larval amphibians (Rana clamitans and Rana pipiens) were investigated using in situ enclosures deployed in two forest wetlands of northern Ontario, Canada. In addition to untreated controls, Vision was applied to yield initial concentrations ranging from 0.29 to 14.3 mg a.e./L (0.94-46.1 mg/L of Vision). Resultant 96-h median lethal concentration (LC50) values ranged from 2.70 to 11.5 mg a.e./L (8.71-37.1 mg/L of Vision) depending on the species or site involved. Substantial mortality and incidences of abnormal avoidance response occurred only at concentrations exceeding the expected environmental concentrations (EEC) (1.43 mg a.e./L, or 4.61 mg/L of Vision) as calculated by Canadian regulatory authorities. The concentration dependence of larval growth rate and maximum size varied depending on site and species. Mean growth rates and maximum sizes exposed to 1.43 mg a.e./L (EEC) treatments were the same or greater than controls. Experimental site and biotic/abiotic factors therein, such as pH and suspended sediments, substantially affected the expression of Vision herbicide toxicity in the amphibian larvae tested. Overall, results suggest that the silvicultural use of Vision herbicide in accordance with the product label and standard Canadian environmental regulations should have negligible adverse effects on sensitive larval life stages of native amphibians.

Effects of Release herbicide (triclopyr butoxyethyl ester, [TBEE]) on mortality, avoidance response, and growth of larval amphibians (Rana clamitans, Rana pipiens) were investigated using in situ enclosures deployed in two forest wetlands in northern Ontario, Canada. Release was applied at nominal concentrations ranging from 0.26 to 7.68 mg TBEE acid equivalents (AE)/L. No significant deleterious effects of this herbicide on larval growth were detected. However, concentration-dependent mortality and abnormal avoidance response were observed. Most mortality occurred within 96 h following treatment. Median lethal concentration (LC50) values for each species and experimental site ranged from 2.79 to 3.29 mg AE/L, while median effective concentration (EC50) values (abnormal avoidance response) ranged from 1.67 to 3.84 mg AE/L. The LC10 and EC10 endpoints approximated aqueous concentrations (0.59 mg AE/L) expected under direct aerial overspray scenarios, indicating a potential risk of impacts for a small proportion of native amphibian larvae. However, given the low frequency and limited use of this herbicide formulation in Canadian forestry, these risks are considered negligible. Changes in usage patterns would require concurrent chemical and biological monitoring of operational spray programs to accurately quantify the probability and magnitude of real-world exposures and to relate these exposure levels to concentration-response relationships including those described in this study.

No abstract