Quantitative Stratigraphy, Splining and Geologic Time Scales - Geomathematics: Theoretical Foundations, Applications and Future Developments - page 355

Paleozoic, Mesozoic and Cenozoic; the corresponding numbers for GTS2012 were
increased to 36, 12 and 13 GSSP's, respectively (Gradstein et al. 2012 , Table 2.5).
In general, precise age dates are determined for relatively many rock samples taken
at or close to the boundary on which a GSSP is defined. Especially for the
Mesozoic, several stage boundary ages could be based on such local data sets. It
is clearly advantageous for all earth scientists to use a single time scale with a
common set of stratigraphic subdivisions. However, in some parts of the world it
may be difficult to recognize all stages for a period. As a regional companion to
GTS 2004, Cooper et al. ( 2004 ) produced the “New Zealand Geologic Time Scale”
with 72 regional stages. Almost half of these have their own boundary definitions
and boundary stratotypes (SSPs), not all of which can be readily correlated to the
GSSPs.
Many different geologic time scales have been published after Holmes's first
scale of 1911. Twelve different time scale methods and their usage in twenty-eight
selected time scales are reviewed in Gradstein et al. ( 2012 , eds., Fig. 1.7). The two
methods of dating already mentioned are rate of radioactive decay of elements, and
tuning of cyclic sequences to orbital time scale. The other methods listed include
Holmes's original maximum thickness of sediments per time period method, and
Harland et al.'s equal duration of stages hypothesis (GTS82). Virtually all time
scales use stratigraphic reasoning and biostratigraphic/geomagnetic calibration.
Approximate constancy of spreading in ocean floor segments is helpful in the
Jurassic and Cretaceous. Holmes's constant sedimentation rate hypothesis was
refined for calculation of some later time scales to the assumption that zone duration
can be proportional to zone thickness. Likewise, Harland et al.'s ( 1982 ) equal
duration of stages hypothesis was refined in some later time scales to the equal
duration of (sub-)zones hypothesis. Trends in the 87 Sr/ 86 Sr stable isotope scale also
have been used in some time scales.

9.5.2 Differences Between GTS2012 and GTS2004

For several chronostratigraphic boundaries in Table 9.14 there is a statistically
significant discrepancy between the GTS2004 and GTS2008 estimates. Such dif-
ferences are mainly due to improvements in accuracy of radiometric methods over
the past 10 years. Accuracy and precision can be discussed in the context of the
various methods of time scale estimation that were used. Virtual certainty was
achieved in GTS2012 for the estimates of the Cenozoic stage boundaries that are
based on astrochronology. The relatively large discrepancy in age for the base of the
Pleistocene in Table 9.14 is due to redefinition of this boundary.
As pointed out in Sect 1.2 , the time scales of the Neogene (23.0

2.59 Ma) and

Paleogene (66.0 23.0 Ma) periods now are entirely based on astronomical cali-

brations. Other differences between GTS2012 and GTS2004 also are seen in
Table 9.14 . The differences for several age estimates are significant in that they
exceed the widths of the 95 % confidence intervals representing precision. Such
lack of accuracy (systematic differences) is mainly due to significant improvements