The Oligocene to the Quaternary: climate and biology - Climate Change: Biological and Human Aspects - page 162

northern USA today have ancestors that lived for some time in these two lakes. Both
lakes were bounded on their north-eastern side by the Laurentide ice sheet. On their
western side were the upland plains and hills that lie in the shadow of the Rocky
Mountains. The largest meltwater lake, when at its greatest, was Lake Agassiz, at
some 350 000 km 2 , and it is named after the naturalist who coined the term ice age.
These lakes were only temporary and just existed at the end of the glacial because of
the ice dams. They were also dynamic. When the Laurentide ice sheet began to melt
sufficiently, more water was fed into the lakes and part of the Laurentide ice retreated.
The lakes therefore changed shape and size. Many of the lakes were from time to time
connected to each other. The variability of ice aided this. There was not only melting
but also re-freezing. In addition, not only would there be terminal moraines and local
geography but ice dams were also possible. Geologists think that occasionally an ice
dam that had held back water would, once there was sufficient water behind it, float
and so allow a pulse of water to escape. If the ice dam was at a key location then the
volume of water released would be considerable. In North America sometimes the
drainage would be to the north through the Mackenzie River and sometimes south
through the Missouri and Mississippi rivers: throughout these regions there is ample
geological evidence of brief torrential flooding dating from the end of the last glacial.
It could well be that it was one of the larger of these pulses that precipitated the
Younger Dryas with the suddenness we see on the Greenland ECM record. Indeed,
this is the preferred theory, although it should be noted that there are others.
The other thing to note about the Greenland ice-core ECM is that it appears that
the regional climatic changes associated with the brief interstadials after the LGM,
and also the beginning and end of the Younger Dryas itself, took place extremely
quickly: perhaps as fast as 5-50 years. Furthermore, a detailed examination of the
ECM record reveals climatic flickers as short as 1 year: these could be changes in
atmospheric circulation (Taylor et al., 1993) and possibly represent flickers between
two semi-steady states. Such rapid change serves to remind us of the IPCC's warning
about being wary of surprises with regard to future warming. We shall return to this
in Chapter 6.
Before leaving the topic of Greenland's palaeodust record, it is worth remembering
the effect of iron on plankton blooms (see section 1.3). The Greenland ice-core
ECM markedly reflects the transition from a dusty glacial to a less-dusty interglacial
environment. There is no reason to suppose that other types of dust than the calcium
acid-neutralising dust were not similarly affected. The transition from glacial to
interglacial would have also affected iron-containing dust. In a glacial mode more
iron dust would be blown on to the oceans so encouraging algal blooms that draw
down atmospheric carbon dioxide from the atmosphere and so helping to keep the
planet cool (see Chapter 1). Similarly, under wetter, warmer interglacial conditions
some iron-containing dust would be washed out of the atmosphere before reaching
the oceans. This would reduce algal blooms, reduce carbon dioxide drawdown and so
help maintain the atmospheric concentration and keep the planet warmer. These are
other examples of feedback systems (again, see Chapter 1) that regulate the biosphere
much in the way engineers describe a cybernetic system (Figure 4.13).
Although the climate warmed quickly after the last glacial, some other environ-
mental parameters took longer. Sea-level rise took thousands of years. The Baltic Sea