Concerns over increased water temperature of the Speed River as it flows through the City of Guelph in Southern Ontario and an observed relationship between summer stream temperatures and low dissolved oxygen levels in the river prompted an investigation into potential stream temperature management practices. Two mechanistic stream temperature models, SNTEMP and CE-QUAL-W2, were applied to the Speed River in order to gauge the effectiveness of various stream temperature management options. Calibrated versions of both models performed well (0.2 degrees C<mean absolute error of SNTEMP<1.8 degrees C; 0.5 degrees C<mean absolute error of CE-QUAL-W2<1.4 degrees C). However, CE-QUAL-W2 performed more consistently spatially and temporally. Air temperature and relative humidity were found to be the most sensitive parameters in both models. Management alternatives considered in this study included modifying discharge from upstream dams, removal of in-stream impoundments, allowing the growth of adequate riparian vegetation to provide shade, and reducing stream width during low-flow periods. Of the various management practices investigated, model results suggest that the removal of in-stream impoundments would be the most effective management alternative to reduce summer stream temperatures. Management alternatives involving removal of in-stream impoundments were projected to reduce minimum stream temperatures by up to 2.2 degrees C, while those not involving the removal of in-stream impoundments were projected to reduce minimum temperatures by less than 0.5 degrees C.