From a 3-year study of 21 northern Ontario lakes, models were developed to determine the duration of ice cover, predict the timing of spring stratification, and predict the end-of-season mixing depth. The model to determine lake freeze and thaw dates was based on the daily variability of water temperature measured with a data logger suspended 1 and 2 m below the surface. The model to predict the duration of time (days) from lake thaw to lake stratification was developed using the mean May air temperature (degrees Celsius), dissolved organic carbon, and lake surface area (r(2) = 0.79). The end-of-season mixing depth was best predicted using days to stratification and dissolved organic carbon concentration (r(2) = 0.72). By applying a simple climate change scenario model, we were able to show that increased air temperature, rather than increased water clarity, was the most important factor affecting the timing of stratification. In contrast, lake clarity was the most important factor affecting end-of-season mixing depth in small Shield lakes.