Geoscience Reference
In-Depth Information
to this point based on radiometric (K/Ar) age estimation of the polarity change in
lava flows. Earlier reversals have also been used.
According to Wright (1999):
''Another major advancement in refining the oxygen isotope-based
stratigraphy came with the observation that the climate/ d 18 O calcite changes
matched orbital insolation patterns. Hays et al. (1976) compared climate records
from the Southern Ocean with the insolation curves and demonstrated that the
climate changes were paced by insolation changes and, therefore, that they could
be used to explain the cyclic nature of climate change during the past 2 million
years. One implication of the ''Pacemaker'' discovery is that the calculation of
past insolation cycle variations could be used as the basis for a numerical time-
scale. Imbrie et al. (1984) developed what is now called the SPECMAP d 18 O
record by averaging d 18 O calcite records from various localities to reduce noise.
The resulting d 18 O curve was assigned ages by tuning (i.e. adjusting the d 18 O
patterns to match the predicted patterns based on the current astronomic calcula-
tions for orbital variations).''
However, the degree of correlation between solar insolation and features in
the SPECMAP d 18 O record lies somewhat in the eye of the beholder. A simplistic
approach is to align the isotope record to the insolation curve, but with a time lag
for the isotope record. Other methods depend on simple models for ice buildup as
described in Section 9.6.
Yet, of equal importance is the fact that once a solar insolation model is used
to assign a chronology to the sediment core, the resultant dated SPECMAP d 18 O
record loses some of its value in testing the astronomical theory because of
circular reasoning. Whether orbital tuning was a major advancement remains
arguable.
Because benthic and planktic data records tend to be noisy, the common
practice is to create a stack, which is an average of data taken at many sites. Such
stacks would presumably average out local noise and leave a smoothly varying
residual signal representing global climate change. For example, the stack pre-
sented by Lisiecki and Raymo (2005) contained benthic d 18 O records from 57
globally distributed sites covering the past 5.3 million years. These sites were well
distributed in latitude (60 Nto50 S), longitude (but predominantly in the North
and South Atlantic Oceans), and depth in the Atlantic and Pacific, as well as two
sites in the Indian Ocean. The problem in aligning the data is that it is dicult to
estimate the sedimentation rate at each site and independently arrive at a chronol-
ogy for the data. One has a curve of variable d 18 O vs. depth in the sediments at
each site, but it is dicult to convert depth to age. Therefore, following the usual
practice, L&R assumed that the paleoclimate signals contained in all 57 of the
records provided the same basic underlying isotope data, but with differing
variable sedimentation rates. If this assumption is correct, the features (peaks,
valleys, abrupt changes in slope) of all the records should correspond to the same
occurrences in the paleoclimate. Therefore, their first step in producing the stack
Search WWH ::




Custom Search