Geology Reference
In-Depth Information
isostatic compensation (and different subsidence
or uplift) everywhere around the world. This
poorly known isostatic response, therefore,
makes it difficult to predict the site-specific
effects of global sea-level change. Nonetheless,
the use of local sea level as a reference surface
can be exploited to assess vertical deformation.
Because tide gauges have been routinely used
to record mean sea-level heights at numerous
localities worldwide, a large database exists
from which vertical changes can be extracted.
Where tide-gauge records are available along an
irregular, embayed coastline, a three-dimensional
reconstruction of the regional uplift pattern can
be determined. In order to use tide-gauge data
to define vertical changes of the shoreline, two
corrections are typically applied. First, changes
attributed to
eustatic
sea level - that is, those
changes that are due to changes in the volume
of ocean water or in the shape of the oceans and
result in changes in average sea level - are
removed. At the time scale of years to decades,
the volume of the oceans is affected by growth
and decay of glaciers and by thermal expansion
or contraction of the water column. At present,
melting glaciers and warming temperatures
appear to be increasing ocean volume at a rate
that causes mean sea level to rise at about
2 mm/yr (Douglas, 1991). Second, local
oceanographic effects resulting from changes
in salinity, ocean temperature, atmospheric
pressure, and ocean currents are removed. A
remarkable inter-annual sea-level variability can
result from these factors (Fig. 5.10), and they
often cause local sea-level changes that are
far greater than those attributable to tectonic
effects. The oceanographic correction is usually
determined through principal component
analysis of a regional set of tide gauges (Savage
and Thatcher, 1992). The objective of this
analysis is to identify those components, termed
common mode signals
, of the observed sea-level
variation that, after the eustatic effect is removed,
are shared among all of the stations. Following
scaling for each site, the oceanographic
correction is removed from the locally measured
tidal record. The resultant data are then
interpreted to represent changes in relative sea
level (Fig. 2.4) due to vertical rock movement.
100
Oceanographic
Correction
50
0
-50
Nankai Trough
-100
1950
55
60
65
70
75
80
1985
Time (year)
Fig. 5.10
Oceanographic correction from the
southwest coast of Japan.
Note that both total variability (
∼
150 mm) and
year-to-year variability are large and unsteady compared
to the rate of tectonic deformation. Modified after
Savage and Thatcher (1992).
The use of tide-gauge data to reconstruct a
regional pattern of vertical deformation is well
illustrated by a study from the southwestern coast
of Japan, where an excellent record spanning
1950-85 has been analyzed (Savage and Thatcher,
1992). In 1944 and 1946, two large (
M
s
=
8) earth-
quakes occurred offshore along the Nankai
Trough subduction zone. Given the nearly con-
tinuous tidal record since those earthquakes,
questions concerning interseismic crustal defor-
mation can be investigated. For example, does
interseismic strain accumulate at a steady rate,
or is there an interval of rapid, post-seismic
deformation immediately after an earthquake
that is followed by more steady deformation?
Removal of eustatic and oceanographic signals
(Fig. 5.10) from 27 Japanese tide-gauge records
reveals clear site-to-site contrasts in the rate of
vertical deformation (Fig. 5.11). Even without
removal of the oceanographic effects, different
long-term trends are visible in these data.
Appropriate corrections, however, suggest that
nearly constant rates of uplift or subsidence
have been sustained at many sites during the
past 30 years, but that the decade immediately
following the earthquakes was characterized
by more rapid rates of deformation (Fig. 5.11).
These data suggest that an interval of more
rapid post-seismic deformation that may indeed
persist for years or decades following a major
