Geoscience Reference
In-Depth Information
moraines are often studded with large boulders that can be
sampled for cosmogenic exposure dating [e.g., Phillips et al.,
1990; Gosse et al., 1995a, 1995b]. In principle, cosmogenic
exposure dating yields direct estimates of moraine ages;
other Quaternary dating methods give only maximum or
minimum age estimates, except in rare cases.
The Younger Dryas (12.9
landform degradation. This stabilization would not have
happened until the glaciers retreated from their moraines at
the end of the Younger Dryas. However, a glacier model
forced by Greenland ice core temperature records produce a
closely spaced complex of moraines distributed over the
Younger Dryas interval [Va c c o e t a l . , 2009; Alley et al.,
2010]. Farther from Greenland, where the event was rela-
tively small, these moraines should be compressed into a
single moraine dating to the end of the event.
The analytical precision of cosmogenic exposure dating
with beryllium-10 is often very good, suggesting that the
method can identify moraines associated with abrupt climate
changes such as the Younger Dryas. Con
11.7 ka [Alley et al., 1993;
Barrows et al., 2007; Walker et al., 2008]) is an example of
an abrupt climate change whose geographic extent has been
traced partly with cosmogenic exposure dating of moraines.
Various proxy records [e.g., Mangerud et al., 1974] show that
the Younger Dryas produced strong cooling around the North
Atlantic, with weaker negative temperature anomalies else-
where in the Northern Hemisphere and warming in the South-
ern Hemisphere [e.g., Broecker et al., 1989; Denton et al.,
2005; Alley, 2007; see also Chiang and Bitz, 2005; Lowell et
al., 2005; Broecker, 2006]. Climate modeling studies with
North Atlantic freshening simulate temperature anomaly pat-
terns consistent with the data [e.g., Vellinga and Wood, 2002].
After the first successful exposure dating studies of mo-
raines [Phillips et al., 1990; Gosse et al., 1995a, 1995b],
glacial geomorphologists used this new tool to look for the
Younger Dryas signal. Moraines dating to the Younger Dryas
were identi
-
cation
of Younger Dryas moraines probably requires an uncertainty
of 10% of the event
dent identi
s duration or about 100 years. Measure-
ments of beryllium-10 concentrations often have uncertain-
ties of ~3% [e.g., Gosse et al., 1995a, 1995b; Owen et al.,
2003; Kelly et al., 2008]. Thus, the 1
'
analytical uncertain-
ties of beryllium-10 exposure dates from Younger Dryas
moraines should be about 360 years (3% of 12.0 kyr). If
measurement error were the only source of uncertainty in
exposure dating, we would need about 13 samples from a
single moraine to reduce this 360 year uncertainty to 100
years [Bevington and Robinson, 2003, equation 4-19, Figure
1]. The uncertainty of the weighted mean is appropriate only
where the dates are normally distributed and have a scatter
consistent with the measurement uncertainties of the dates;
more exposure dates are needed where these conditions are
violated. Production rate estimates also contribute to the
overall uncertainty of exposure dating, as we discuss below.
However, the measurement of nuclide concentrations is
only one step in the exposure dating process, and all steps
contribute to the total uncertainty of exposure dating. These
steps are as follows: (1) collecting the samples [Gosse and
Phillips, 2001; Briner, 2009]; (2) processing the samples
[Kohl and Nishiizumi, 1992; Bierman, 1994]; (3) measuring
the nuclide concentrations in the processed samples using
accelerator mass spectrometry [Muzikar et al., [2003] (this
step also yields the analytical uncertainties of the dates);
(4) calculating the apparent exposure times of the samples
from the nuclide concentrations [Lal, 1991; Gosse and Phil-
lips, 2001; Balco et al., 2008]; and (5) estimating the age of
the moraine from the exposure dates [Applegate et al., 2008,
2010]. Determining the climatic significance of moraines
once their ages are known is also a crucial part of the
exposure dating process, but this additional step is beyond
the scope of this chapter.
Here we indicate potential problems in the selection of
samples for exposure dating and the calculation of exposure
dates from nuclide concentrations (steps 1 and 4, above).
Brie
σ
ed in the Alps [Ivy-Ochs et al., 1999, 2006,
2007; cf. Kelly et al., 2004] but also far from the North
Atlantic (e.g., western North America and New Zealand
[Gosse et al., 1995b; Ivy-Ochs et al., 1999]). Given the likely
distribution of Younger Dryas cooling, the age assignments
for these additional sites are suspect.
Past workers have taken exposure dates falling at any time
within the Younger Dryas interval as evidence for Younger
Dryas cooling in the region [e.g., Gosse et al., 1995b; Ivy-
Ochs et al., 1999; see also Denton and Hendy, 1994]. This
criterion re
ects uncertainties in the exposure dating method
and lingering doubt about when moraines should have been
deposited during the Younger Dryas.
First-order glaciological considerations suggest that the
ages of true Younger Dryas moraines should cluster around
the end of the Younger Dryas, but recent modeling studies
[Vacco et al., 2009] complicate this simple picture. Changes
in glacier margin positions lag temperature change, with a
response time and equilibrium length change that vary
among glaciers [Oerlemans, 2005]. Thus, there was likely a
delay between the initial Younger Dryas cooling and glacier
advances. Glaciers with short response times reached equi-
librium with the new temperature quickly, whereas glaciers
with long response times perhaps never did. Until the gla-
ciers reached their maximal positions, any moraines depos-
ited at their margins would not be preserved [Gibbons et al.,
1984]. In any case, exposure dating records the time of
landform stabilization, assuming no inheritance or long-term
y, geomorphic process modeling suggests that sampling
Search WWH ::




Custom Search