Other data sources - Ice Ages and Interglacials: Measurements, Interpretation, and Models - page 191

Figure 6.3. Vostok d D chronologies during the deglaciation of about 135,000 years ago as
derived by seven different investigations (L&W).
Nevertheless, these arguments are somewhat subjective and the treatment
of the two sites as synchronous must be regarded as an assumption rather than
a fact. L&W also noted that differences in chronology of about 5,000 years arise
at different sites at Greenland alone, so they suggested that one should consider
the agreement of diverse sites to within about 5,000 years as evidence of
synchronicity.
L&W developed a procedure for transferring a chronology from a well-dated
paleoclimate record (in this case, Devil's Hole) to one that is not independently
dated (in this case, Vostok). The underlying assumption was that both records are
proxies for the same physical phenomenon and the paleoclimatic conditions
forcing the two records can be considered to have occurred contemporaneously at
both locations. The procedure identifies where significant state changes of compar-
able relative magnitude are under way in each dataset, and then utilizes a visual
examination of the geometry to relate corresponding climatic excursions in the two
records. Figure 6.4 shows how key points were related from the Devil's Hole age
scale to the Vostok depth, and Figure 6.5 shows the resultant age vs. depth curve
derived for Vostok. From this result, L&W prepared the comparison of Vostok
and Devil's Hole data shown in Figure 6.6 . In this figure, the black curve is the
Vostok data based on the Devil's Hole chronology, and the blue curve is the
Devil's Hole result. The red curve is based on extrapolating the Vostok age-depth
curve to greater depths where the data are suspect because layering was disturbed
by dynamic ice processes. Nevertheless, the red curve seems to be reasonable.
However Carl Wunsch has emphasized the subjectivity inherent in comparing
noisy curves and cautioned that such procedures ''can easily lead to unwarranted,
and incorrect, inferences if simple stochastic superposition is confused with
deterministic causes'' (see Section 3.2.10 for further details). Nevertheless, Figure
6.6 is fairly convincing to this writer.