Temporal trends of three phenoxy-herbicides (2,4-D, dichlorprop and MCPA) a
nd bromoxynil were determined in the South Tobacco Creek, which drains a 70
km(2) agricultural watershed near Miami, Manitoba, over a 4-yr sampling pr
ogram. The quantities of all pesticides used in the watershed and their app
lication times were accurately recorded with the cooperation of landowners.
Herbicide concentrations in South Tobacco Creek ranged From <0.01 ng/L to
a maximum concentration of 680 ng/L (for 2,4-D) and reflected local applica
tion times. Spring melt contributions were important when no significant ru
noff events occurred late in the previous growing season. Discharge of thes
e herbicides in the South Tobacco Creek was calculated to be 0.01% of their
quantities used within the watershed. No correlation was found between the
water quality parameters: total organic C, total N, and total P and herbic
ide concentrations in South Tobacco Creek. Total N and P, however, were cor
related to water pow, indicating runoff and ground water sources rather tha
n atmospheric. We concluded that detailed temporal patterns of herbicides a
nd nutrients could only be established if high sampling frequency was maint
ained during such critical periods such as application periods and during m
ajor runoff events. An unexpected finding was that herbicide concentrations
were not related to runoff losses, but instead they did correspond to elev
ated levels in precipitation and air measured within the watershed.