Geology Reference
In-Depth Information
geomorphic markers. Second, some landscape
responses are described, using examples of
growing faults and folds, deforming forearcs,
and tilted mountain ranges.
approaches. As described in Chapter 3,
uranium-series (U-Th) dates on aragonitic mate-
rial have provided ages of surfaces several
hundred thousand years old. With improved
techniques and measurement capabilities,
uranium-series dates have become increasingly
precise, such that dates of 100 ka may have meas-
urement uncertainties of
<
1% (Edwards
et al.
,
1997), and chronologies of sea-level changes
have been extended back to greater than 200 ka
(Andersen
et al.
, 2010; Edwards
et al.
, 1997).
In the ideal case, we have a reliable global
(eustatic) sea-level curve (sea level relative to
present versus time; Fig. 2.5) and a flight of
dated terraces. The age and the present
elevation of each terrace (relative sea level)
with respect to the position of the correlative
eustatic sea level at the time of terrace forma-
tion can be used to define amounts and rates
of rock uplift through time. But, what if the
terraces are undated, or only one of them has
an approximate age? Can they still be used to
define the uplift rate? In this situation, one
typically assumes a steady uplift rate (or assigns
one, if a single terrace is dated), and then deter-
mines graphically (or numerically, on the com-
puter) how well the observed heights of
terraces correlate with the predicted position of
terraces based on the eustatic sea-level curve
and the constant apparent sea-level change
(Fig. 9.4A).
In this graphical matching technique, a terrace
should exist at an appropriate height above sea
level for each of the high sea-level peaks defined
in the eustatic curve. If the elevation of each
sea-level highstand matches with that of a pre-
served terrace, then the assumption of constant
uplift appears warranted. If the match is poor,
then a different uplift rate can be tried. If the
match is still unsatisfactory, the rate of uplift
might be varied through time in order to obtain
a satisfactory match (Fig. 9.4B). If one permits
the uplift rate to vary without constraint,
however, a perfect, but probably meaningless,
match can always be obtained. The simplest and
often most convincing approach, therefore,
assumes a constant uplift rate. If rates are varied
through time to obtain a match, then a geologi-
cally reasonable rationale for the proposed
Calibrating rates of deformation
Marine terraces
Several approaches can be employed to define
the vertical deformation pattern along a coastline
using either abrasion platforms or coral terraces
(Figs 2.1 and 2.2). A rising coastal landmass is
like a strip chart that records and uplifts the
geological record of each successive sea-level
highstand. As in all rate studies, one needs
knowledge of the initial and final elevation of a
marker, as well as the time it took to traverse that
vertical distance. The key here is that the landmass
is assumed to be rising steadily (with good
enough data, this assumption can be tested),
whereas sea level is varying widely over glacial-
interglacial cycles that have periods on the order
of tens of thousands to one hundred thousand
years. The abrasional or constructional platform
that is created at time zero is a nearly planar,
gently sloping surface. Subsequent emergence of
a platform implies that the rock mass has moved
upward relative to sea level, and submergence
implies the opposite. In order to know which has
moved, the sea or the land, other information
(e.g., eustatic records) must be brought to bear.
On the other hand, any warping of this surface
can be used directly and unambiguously to
document relative movement of one part of the
coastal landmass with respect to another.
Depending on what is already known about the
landscape, one can either use the terraces to
deduce a sea-level curve, or use the terraces to
deduce the rate of uplift of the landmass. In any
case, in order to document rates of deformation,
one must be able to date the platforms.
The determination of the absolute ages for
former sea levels has been something of an
industry for a large number of researchers for
several decades. Unfortunately, most of the
sea-level highstands that we would like to date
are older (
>
40 ka) than can be dated using
14
C
