years. The warming phases correspond to Heinrich events, which are depositions of
iceberg-rafted rock debris in North Atlantic sediments. Not only temperature proxies
change in D–O events, but there are strong increases in ice δ^18 O, implying strong
reductions in glacier volume, freshening of surface layers and cyclic reduction of
deep-water formation and thermohaline circulation. Possibly, they were key factors in
the longer-term CO 2 drawdown. Bond events (e.g. Bond 1997) in the Holocene, eight
since the end of the Wisconsin glaciations, with the last about 1400 years ago, are
identified from rock debris in Atlantic sediments. They have roughly the same
frequency as D–O events, but are not obviously correlated with temperature shifts. In
contrast, some warming events during the previous interglacial, and evident in
Greenland cores, were multi-decade pulses up to 14°C (GRIP Members 1993).
Holocene Bond events may generate sea-surface freshening and for a time reduce
thermohaline circulation, but that is not established. Neither D–O nor Bond events are
fully explained in terms of cause. A likely candidate is cycling between ice build-up
and ice-sheet instability. Immediate ecological consequences of these events are not
well characterized. The lesson that they offer for present-day concerns is that climate
can shift dramatically in very short intervals, whatever the causes.
(^) That covers some of the data and thinking about the glacial cycling process. Data
gathering and thinking about process interactions continue. Much of the research is
motivated by the hope that fully understanding those mechanisms will give insight
about the likely impact of anthropogenic increases in greenhouse gases.
Unfortunately, as pointed out to a vast public audience by (former US Vice President)
Al Gore, we are now so far beyond the remarkably rigid, glacial cycling between 185
and 290 ppmv, that the main insight may well be that those mechanisms will no
longer have the same effects. Ruddiman (2007b) has suggested that we should have
seen substantial cooling as long as 8 kyr ago, but we have not. He attributes that to the
possible greenhouse effects, starting then, of increasing CO 2 and methane derived
from opening of soils and felling of forests for agriculture and development of rice
culture in paddies. One effect of enhanced greenhouse gases, comparable to their
“feedback” effect in the glacial eras, will be continued melting of glaciers. Both
Greenland and Antarctica are in or moving toward net ice loss. Since the bulk of ice
in both places sits on land, the rise in sea level that melting can generate may inundate
island and coastal areas to depths of tens of meters on decadal to centennial time
scales. The possible amounts are recalculated often enough that we do not cite them.
Iron fertilization of the Sea to Counteract Global
Warming
(^) As discussed elsewhere, primary productivity over substantial stretches of ocean