Occurrence of glyphosate and acidic herbicides in select urban rivers and streams in Canada, 2007.

INTRODUCTION: Public and scientific concern has grown over the last decade in Canada over the cosmetic use of pesticides in urban centers. With this in mind, a national survey was designed to monitor eight commonly used herbicides in urban rivers and streams across Canada.
MATERIALS AND METHODS: To coordinate sample collections across the country, samples were collected monthly on one of two predetermined dates from April to September, 2007 from 19 sites within 16 watersheds, including 15 sites downstream of urban lands and two reference sites. Water samples were also collected approximately three times from each watershed during or after precipitation events. All samples were collected using a common sampling protocol and all were analyzed using the same analytical laboratories. RESULTS AND DISCUSSION: The herbicides 2,4-D, mecoprop, dicamba, glyphosate and its major metabolite aminomethylphosphonic acid (AMPA) were most frequently detected. Using either herbicide concentrations upstream/downstream of urban centers or bromoxynil and clopyralid as indictors of agricultural inputs of herbicides to streams, it was clear that environmental concentrations of these herbicides downstream of urban areas were linked to urban use in Canada. Herbicide concentrations in streams draining urban areas were greater during or after significant rainfall events and, with the exception of glyphosate, were significantly greater in the Province of Ontario. Herbicide concentrations were not correlated to the proportion of the watersheds in urban land use. Also, there was no difference in seasonal patterns of herbicide concentrations across urban centers when grouped in five geographic areas. None of the herbicide concentrations measured exceeded existing Canadian Water Quality Guidelines for the protection of aquatic life.
CONCLUSIONS: This is the first time a national survey of pesticides in urban rivers has been carried out in a consistent fashion across Canada. Concentrations of 2,4-D, mecoprop, dicamba, glyphosate, and AMPA were linked to urban use and frequently detected in all geographic areas. However, geographic differences in concentration suggested differences in usage or stream connectivity patterns among urban centers. Some jurisdictions in Canada have recently restricted cosmetic use of pesticides and it would be interesting to determine whether such restrictions will lead to reduced pesticide concentrations in urban streams.