Page 261 - Earth's Climate Past and Future

CHAPTER 13 • Climate During and Since the Last Deglaciation 237

Arctic 13-4 Positive Feedbacks to Deglacial Melting
Ocean
Climate scientists basically agree that rising summer
insolation values caused by changes in Earth’s orbital
tilt and precession set in motion the melting of the great
northern hemisphere ice sheets near 17,000 years ago.
But we still face an important question first explored in
Chapter 11: How did so small an insolation maximum
Labrador Sea melt so much ice so quickly?
Hudson The answer to this question must be that positive
Bay
feedbacks accelerated the loss of ice. These feedbacks
must have been working most effectively when sea level
was rising (ice was melting) most rapidly (see Figure
13–4). The first of the two fast-melting phases (between
17,000 and 14,000 years ago) should have been a time of
especially active feedbacks, because the rapid rise in sea
level (loss of ice volume) at that time occurred well
before summer insolation had reached a peak.
Gulf of Mexico
Several of the bedrock interaction processes described
FIGURE 13-8 Routes of meltwater flow During deglacia- in Chapter 11 are in evidence during this interglacia-
tion, the direction of drainage of the North American ice sheet tion. The negative δ O pulse in Norwegian Sea cores
18
changed, first southward to the Gulf of Mexico early in the
(see Figure 13–5) is evidence that a substantial part of
deglaciation, then east to the Atlantic Ocean (briefly) during
the marine ice sheet over the Barents Sea melted early
mid-deglaciation, and finally north into Hudson Bay and the
in the deglaciation. Because its base lay below sea level,
Arctic Ocean late in the deglaciation. (Adapted from J. Teller,
it may have been vulnerable to early destabilization.
“Meltwater and Precipitation Runoff to the North Atlantic, Arctic,
In addition, two observations point to early thinning
and Gulf of Mexico from the Laurentide Ice Sheet and Adjacent
of the Laurentide ice sheet on North America. The
Regions During the Younger Dryas,” Paleoceanography 5 [1990]:
major influx of icebergs to the North Atlantic Ocean
897–905.)
early in the deglacial sequence arrived at a time when
Laurentide ice had not melted back far from its glacial
position but when global ice volume was rapidly
(between 11,000 and 10,000 C years ago), however, decreasing. This observation suggests that ice streams
14
the negative δ O pulse in the Gulf of Mexico weakened delivered large amounts of ice to the marine margins of
18
because the meltwater was flowing eastward into the the ice sheet but with little loss of area.
Atlantic through the St. Lawrence region of eastern In addition, the elevations of moraine deposits along
Canada. the southern ice lobes of the Laurentide ice sheet indi-
One criticism of this hypothesis is that the Younger cate thin profiles during the early-middle parts of the
Dryas episode occurred at the same time that the rate of deglaciation, consistent with the idea of ice that was slid-
global melting was slowing by a factor of 4 or 5 or more ing on a lubricated base in that region. Ice that flowed to
(see Figure 13–4). With the overall flow of meltwater to the southern margins of the ice sheets across land would
the oceans sharply diminished, it is hard to argue that a have melted relatively rapidly in the relative warmth of
diversion of flow would have greatly lowered the salin- the bedrock depression left behind by the once-thicker
ity of the North Atlantic. More recently, a careful inves- ice. Again, the interior of the ice sheet could have been
tigation of drainage patterns from lakes near the ice thinned without any major retreat of the margins.
margin failed to find any evidence of an unusual outflow The mid-deglacial melting pause and the coincident
of fresh water directly to the Atlantic during the onset Younger Dryas cooling may represent an interval when
of the Younger Dryas event. positive feedback processes slowed their impacts on the
Earlier suggestions that the Younger Dryas cooling climate system. Because summer insolation was still ris-
had a global expression have also proven to be incor- ing, melting continued but more slowly. The second
rect. A small cooling evident in Antarctic ice cores that step of rapid ice melting and sea level rise after 11,500
was once interpreted as correlating with the Younger years ago presumably reflects the return of the various
Dryas event is not correlative. Instead, at least part positive feedbacks to a more active role.
of the Antarctic was undergoing a slow warming Another important feedback process was the rising
throughout the Younger Dryas interval. The origin concentrations of greenhouse gases. As the ice sheets
of the Younger Dryas cooling remains an enigma. melted, the CO and methane levels rose in near
2