Abstract:

Numerous long-term trends in the rate-of-change in monthly mean Great Lakes water levels are identified for the period 1860 —1998. Statistically significant trends are found for 2,4,5,and 7 months of the year for Lakes Superior, Michigan-Huron, Erie, and Ontario, respectively. Many of the trends translate into large changes in net water flux (600 —1700 m 3 s -1 ). In each case, significant positive trends are roughly offset by negative trends during other times of the year. Together with similar trends in monthly lake level anomalies (deviations from the annual mean), these trends indicate important changes in the seasonal cycle of Great Lakes water levels. Specifically, Lakes Erie and Ontario are rising and falling (on an annual basis) roughly one month earlier than they did 139 years ago. Maximum lake levels for Lake Superior are also slightly earlier in the year,and the amplitude of the seasonal cycle of Lake Ontario is found to increase by 23%over the 139-year period. Some of the changes are consistent with the predicted impacts of global warming on spring snowmelt and runoff in the Great Lakes region. Other potential contributors to the observed trends include seasonal changes in precipitation and human-induced effects such as lake regulation and changes in land use.