Great Lakes evaporate away
Recent water-level declines in the Lake Michigan–Huron system correlate with an increase in evaporation over the past century and a half.
While the U.S. states and Canadian provinces that border the Great Lakes fight over shipping capacity and the right to keep or export the freshwater resource, policy makers and scientists are watching lake levels in the region dip. New research published in ES&T (DOI: 10.1021/es070664+) suggests that the water levels of Lakes Michigan and Huron may be plunging faster than previously thought, a trend obscured by increased rainfall.
The two interconnected lakes are often treated as a single system. The St. Clair River drains the lakes to Lake Erie, and eventually out into the Atlantic Ocean. Shipping needs on the St. Clair led to past dredging, which could be responsible for much of the lakes' slipping levels, some researchers say. The independent U.S.–Canada International Joint Commission (IJC), which advises on policy and management for the Great Lakes, has begun to examine possible reasons behind drops in lake levels in the Lake Michigan–Huron system. IJC expects to report results by spring 2009.
In the meantime, researchers from Duke University and the Great Lakes Environmental Research Laboratory (GLERL), run by the U.S. National Oceanic and Atmospheric Administration, have weighed in on possible culprits. In the ES&T paper, Cynthia Sellinger of GLERL and colleagues examined records collected from 1860 to 2006, including water-level data, precipitation data, and proxies for evaporation. Statistical analyses showed a slight correlation between 11-year sunspot cycles and increased precipitation on an almost-decadal cycle.
But more important, the researchers argue that higher rainfall since the early 1970s is masking the drop in water levels expected from evaporation in the Michigan–Huron system. This long-term underlying decline "seems to be closely related to changes in evaporation," says coauthor Craig Stow of GLERL. Evaporation made a dramatic switch in 1977 from less than –1 millimeter per year (mm/yr) to almost 4 mm/yr, according to the team's modeling. "We see some patterns but not enough [to be] consistent with climate change," Sellinger says, referring to both increased evaporation and decreased precipitation.
Past research has failed to establish a link between precipitation and sunspots, comments Gavin Schmidt of NASA's Goddard Institute for Space Studies. The correlation suggested here is probably from decadal noise in the data, he says. Schmidt believes that the evaporation changes are the more interesting findings. He would like to see comparisons with climate-model projections to assess expected precipitation and evaporation for the region in the future.
Overall, the team's statistical analysis is too simple, and the researchers neglected to take into account human impacts that have long affected lake levels, other hydrologists and climate researchers say. "There have been changes in the period that they use, [such as] dredging in the St. Clair River and gravel mining early in the century," says David Fay, manager of Environment Canada's Great Lakes–St. Lawrence Regulation Office. Because of these anthropogenic changes, "you can't just take water levels and analyze them for climate trends," he says.
Another factor for some shorelines is glacial rebound—the uplift of the land by as much as 20 centimeters after glaciers of the last ice age, 10,000 years ago, receded, notes Frank Quinn, former research hydrologist at GLERL and a current participant in the IJC study. Still, he says, "my contention is that the current low levels are due to changes in water supply," mainly between the upper lakes and Lake Erie. Quinn agrees that the longer-term natural hydrological cycles of the lakes and current low levels are not captured by the relatively simple data analysis in the new research.
"People tend to view lake levels from the perspective of human life spans," rather than longer geologic timescales, comments Douglas Wilcox of the U.S. Geological Survey. He points out that low lake levels, expected as part of 30-year paleohydrological cycles, did not occur in the 1930s and 1960s on Lake Superior "because they were regulated out," in other words, changed by human water use. As for Lakes Michigan and Huron, the current lows have lasted longer than expected, but Wilcox suggests they may match longer-scale natural patterns. He agrees with the authors' conclusions: "If [lake levels] do go lower and [the lows] last even longer, it seems that there is good evidence that climate change is rearing its head. Only time will tell."
