Conventional wisdom and climate history
Steven M. Colman*
Large Lakes Observatory and Department of Geological Sciences, University of Minnesota Duluth, Duluth, MN 55812

T

he Younger Dryas interval, a cold snap that chilled many parts of the world for 1,500 years or so in the midst of the last deglaciation (Ϸ13,000–11,500 years ago), is perhaps the best known and most studied paleoclimate event of the last 2 million years. Only a few years ago, it was well accepted that a change in the drainage routing of the huge proglacial lake that fronted the North
American ice sheet occurred at just about the same time as the beginning of the Younger Dryas cold period. This and other coincidences in timing, as well as considerations of the effects this event might have had on ocean circulation, led many to believe that the meltwater rerouting caused, or triggered, the
Younger Dryas cold interval. Over the last few years, all of this conventional wisdom was thrown into turmoil by a few new observations and age determinations. Now, in this issue of PNAS,
Carlson et al. (1) provide a new set of data about meltwater discharge at the start of the Younger Dryas, as well as new detail regarding events within this period. They also suggest that the conventional wisdom about the inception of the Younger Dryas may not be as flawed as has been suggested recently.
The Younger Dryas began and ended abruptly, at least as indicated in ice core records from Greenland, where temperature initially may have fallen by 15°C, with transitions no longer than a few decades (2); most of the final (warming) transition may have occurred in just a few years (3). At just about the time of the inception of the Younger Dryas, a major change in the routing of meltwater from the Laurentide Ice Sheet in
North America seems to have occurred.
The outlet of glacial Lake Agassiz, which fronted the Laurentide Ice Sheet across a vast section of the continental interior, appeared to have switched from Mississippi River drainage (and thence to the Gulf of Mexico) eastward to the Laurentian Great Lakes (and thence to the St. Lawrence River and the North Atlantic Ocean). The switch in outlets was accompanied by a major
(Ͼ40 m) initial drawdown of Lake
Agassiz during its Moorhead Phase, perhaps eventually reaching as much as
150 m below the southern outlet (4).
The switch was also accompanied
(within the uncertainties of radiocarbon dating) by an abrupt change in oxygen isotopes in the Gulf of Mexico (docu-

mented in many studies, most recently in ref. 5), interpreted as resulting from an increase in seawater salinity that accompanied the removal of Agassiz drainage down the Mississippi. Indeed, together with a plausible route for the eastern outflow, the evidence for an eastward switch in drainage seemed compelling. Ocean climate modeling studies (e.g., see references in ref. 6) suggested that the estimated increase in freshwater input to the North Atlantic would be sufficient to suppress Atlantic meridional overturning circulation, distinctly cooling the region where the
Younger Dryas is best documented, as well as other areas. Oceanic proxy evidence for a slowdown or cessation of

The Younger Dryas began and ended abruptly with transitions no longer than a few decades.
Atlantic meridional overturning circulation (in the past loosely called ‘‘thermohaline circulation’’ or ‘‘the conveyor belt’’) during the Younger Dryas is relatively strong (1, 7).
Nevertheless, the inference that eastward routing of Lake Agassiz discharge was the cause of the Younger Dryas has not been without its problems. Isotopic evidence for a freshening of the North
Atlantic off the mouth of the St. Lawrence has been equivocal at best (8, 9).
In addition, the effect of Lake Agassiz inflow on the sediments of the Laurentian Great Lakes has not been clear; sedimentological and isotopic evidence from Lake Michigan was