Geoscience Reference
In-Depth Information
replace them, plays a key role in maintaining the conveyor belt pattern of poleward traveling surface
currents in the North Atlantic ocean. The sinking is due in part to the high salinity of the surface
waters, as saltwater is denser than freshwater. The large amount of freshwater released into the North
Atlantic when the ice sheets were melting toward the end of the last Ice Age, roughly twelve thousand
years ago, is believed to have led to a freshening of North Atlantic surface waters, inhibiting their
sinking and thereby shutting down this ocean circulation system. The shutdown is believed to have
induced a sudden return to glacial conditions in regions neighboring the North Atlantic before the
ultimate termination of the Ice Age. Current climate models predict an analogous, though far weaker,
response to the rapid melting of remaining continental ice associated with global warming. Indeed, it
is this thinking that became popularized, albeit in a caricatured manner, by the movie
The Day After
Tomorrow
, where global warming is depicted as paradoxically leading to a sudden new Ice Age in
the Northern Hemisphere.
The evidence supporting a prominent role for the conveyor belt in generating millennial
oscillations was highly controversial however. Though Bond's team had offered some suggestive
evidence, there seemed to be scant additional support that the putative millennial cycle had been
operating during the modern interglacial period (the Holocene period of the past twelve thousand
years). Broecker's hypothesis that the medieval warm period and Little Ice Age were part of such a
natural millennial cycle seemed more tenuous still.
The “segment length curse” issue would come again to the fore the following year. Ed Cook of
Lamont Doherty was one of the seminal contributors to the science of dendroclimatology and had
brought considerable statistical rigor and insight to this field. Along with Jan Esper of Switzerland
and others, he had been working on a new tree ring-based reconstruction of Northern Hemisphere
novel method that was rather aggressive in its effort to retain long-term trends, but at the price of
potential pitfalls.
13
The Esper et al. reconstruction appeared to differ substantially from earlier reconstructions,
including our own. In particular, it suggested larger temperature fluctuations from century to century
and featured a warmer medieval warm period, closer to current warmth. The study got a fair amount
of media attention, and I was asked to do an interview on NPR's
Science Friday
with Ira Flatow. At
the time, I was on vacation on the island of Chincoteague, Virginia, with my fiancé and had to use a
makeshift landline phone for the interview. Cook, in the magnanimous fashion anyone who knows him
might expect, recommended that Flatow interview me about the paper, knowing full well that I had
some misgivings about the new findings.
I felt that the study was a solid and useful contribution to the scientific discourse, but believed
there were important caveats. First, the tree ring data used mainly represented summer temperatures
in the continental interiors, and there were good reasons to believe that temperature variations in
these regions are especially large (a point that the authors acknowledged). We know, for example, that
explosive volcanic eruptions have the greatest cooling impact on continental interiors and in summer.
This could plausibly explain why the greatest discrepancies between their estimate and ours were
during periods of frequent volcanic activity. Second, their tree ring reconstruction didn't record any
of the warming since the mid-twentieth century, so it could not be used to compare medieval and
current warmth. Third, their temperature scaling was somewhat arbitrary, in my view, and arguably
inflated the apparent amplitude of past temperature fluctuations.
14
Finally, I had some reservations
