an introduction to earthquake geodesy

AN INTRODUCTION TO EARTHQUAKE GEODESY :
Another Effort for Earthquake Hazard Monitoring
Andreas H., H.Z.Abidin, Irwan M., D. Darmawan, D.A. Sarsito, M. Gamal
Geodesy Research Group
Department of Geodetic Engineering, Institute of Technology Bandung,
Jl. Ganesha 10 LABTEX IX C telp/FAX +62 22 253 4286, Bandung Indonesia
Abstract
Earthquake is one of catastrophic event produce losses to people’s life as well as
infrastructure damages. Aceh Earthquake following by tsunami was one recent and
biggest examples of earthquake’s tragedy in the last 40 years. Almost 300,000
peoples were killed and leaving very serious damages on infrastructures. Earthquake
happen when the earth’s crust fails in response to accumulated deformation.
Geodetic measurement document the crustal deformation leading to those failures
and the deformation resulting from them, providing unique insight into the physical
processes involved [ Hudnut et.al 1994 ]. Geodetic measurement (e.g. GPS and InSAR)
with space and time domain (continuous or periodic) may detected ground
deformation as well as the accumulation of them. Geodetic measurement may also
constrains physical model of the processes that cause earthquake event. With
geodetic measurement we may saw clearly inter-seismic phase of earthquake
mechanism, pre-seismic signal also sometimes recorded, and well recorded co-seismic
and post-seismic signals. Inter-seismic pattern is now being included as one of
parameter in estimation the probability of earthquake hazard. As a conclusion, the
geodetic measurements became another effort for earthquake hazard monitoring.
These geodetic measurements have a title as Earthquake Geodesy.
Keyword : Earthquake geodesy, GPS, InSAR, deformation, earthquake cycle
I. INTRODUCTION
An earthquake is a shaking of the ground
caused by the sudden breaking and
shifting of large sections of the earth's
crust. Why the crust could break because
they fails in response to accumulated
deformation (see figure 1). The
accumulation of deformation result from
ongoing processes such as aseismic
deformation of subcrustal rock associated
with relative plate motion due to
convection energy from a mantle.
The large energy released as a big
earthquake may occurred when the
earth’s crust fail to response the
maximum accumulation of deformation
and shown us a pictures of catastrophic
event produce losses to people’s life as
well as infrastructure damages. Many
records have documented the number of
big earthquakes followed by fatalities
and damages ( see Table 1). Aceh
Earthquake following by tsunami was one
recent and biggest examples of
earthquake’s tragedy in the last 40 years.
Almost 300,000 peoples were kill and
leaving very serious damages on
infrastructure. Earthquake happen
almost without warning and they
followed a cycle so will strike back
sometimes in a future.

Figure 1. Illustration of accumulation of deformation leading to failure respond of the earth’s
crust when it can no longer resisted to the accumulated energy from ongoing processes such as
aseismic deformation of subcrustal rock associated with relative plate motion due to convection
energy from a mantle, produce an earthquake and sometimes followed by tsunami. Image with
courtesy of J. Mori on KAGI lecture’s note. [Mori 2004 ]
Table 1
Record examples of big Earthquakes (sometimes followed by tsunami) leaved notes of terrible
numbers of fatalities ( Mori 2004);( Vigny 2005 )
no Earthquakes event Fatalities
1 Earthquake in Lisbon Portugal followed by 70.000 peoples were killed
tsunami on November 1, 1755
2 Earthquake in Tokyo Japan ( 7.9 Mw )
99.331 peoples were killed
on September 1, 1923
3 Earthquake in Sanriku Japan
3.064 peoples were killed
on March 3, 1933
4 Earthquake in Fukui Japan
3.769 peoples were killed
on June 48, 1948
5 Earthquake in Thangsan China ( 7.8 Mw ) 240.000 peoples were killed
on July 28, 1976
6 Earthquake in Armenia on year 1988 25.000 peoples were killed
7 Earthquake in Iran on year 1990 40.000 peoples were killed
8 Earthquake in Kobe Japan ( 6.9 Mw )
5096 peoples were killed
on January 17, 1995
9 Earthquake in Aceh Sumatran subduction zone
( Mw 9.0 ) followed by tsunami on December
26, 2004
Almost 300.000 peoples were
killed
Records of such big event has proven that
a single earthquake can cause thousands
of deaths and not to mention thousands
others became injured. Although
earthquakes are uncontrollable, the
losses they cause can be reduced by
develop earthquake hazard monitoring
and mitigation programs (e.g. conducting
many researches to analyzed earthquake
mechanism, matching land use to risk,
building structures that resist earthquake
damage, developing emergency response
plans, and other means). As a result of
technologies, in the early years ninety
has been introduced another tools for
earthquake hazard monitoring namely
earthquake geodesy which will be
explained more in this paper.

II. SEQUENCE AND MECHANISM OF THE
EARTHQUAKE
Until today, the mantle convection still
believed as an appropriated theory to
explain the movement of the plates and
the source mechanism of the earthquake.
The energy from convection steering the
plate to move one another. One block of
plates and the others would converge,
diverge, or move side by side. At the
plates interface where two plates merge
and locked by friction, then this earth’s
crust area will experienced a
deformation. As the energy from the
mantel will continuously forced the
plates to move consistently, a
consequences the deformation in the
plates interface will increasingly
accumulated. Within few ten years to
hundred years when the accumulation of
deformation reached maximum stage,
this earth’s crust may fails in response to
those accumulated deformation and
produce the sudden breaking and shifting
of large sections of the earth's crust,
making the ground to shake and its called
earthquake.
For a moments, the energy was released
by the earthquake ( mainshock and
aftershock), made the deformation
return to zero. The earthquake has
unlock the friction. But after few months
to few years when this dynamic earth has
return to equilibrium stage, the plates
interface will experienced new
deformation. Again, the merged plates
could lock by the friction and the
deformation will beginning to
accumulated leading to failure respond
of this earth’s crust producing new
earthquake. This sequence called
earthquake cycles. In detail the
earthquake cycles divided into interseismic,
co-seismic, and post-seismic
sequence. Pre-seismic is now still being
analyzed before officially included as
part of all sequence.
inter-seismic
At the inter-seismic sequence, the two
plates interface is locked by friction. The
upper plate is accumulating elastic
deformation at a slow rate (~1cm/yr).
This loading phase can last for centuries
Co-seismic
The earthquake releases in one moment
deformation accumulated for centuries.
At that stage, the upper plate
“rebounds”. In the subduction zone, the
whole system being below sea level, this
giant “kick” in sea water generates a
Tsunami
Post-seismic
Return to equilibrium and steady loading
can take years.
Figure 2. Illustration of earthquake cycle in
subduction zone ( Vigny et.al 2005 )

III. GEODETIC DATA AND EARTHQUAKE
STUDIES ( EARTHQUAKE GEODESY )
As mention previously, earthquake
happen when the earth’s crust fails in
response to accumulated deformation.
Geodetic measurement document the
crustal deformation leading to these
failures and the deformation resulting
from them, providing unique insight into
the physical processes involved [ Hudnut
et.al 1994 ]. Geodetic measurement
with space and time domain (continuous
or periodic) may detected ground
deformation as well as the accumulation
of them. Geodetic measurement may
also constrains physical model of the
processes that cause earthquake event.
With geodetic measurement we may saw
clearly inter-seismic phase of earthquake
mechanism, pre-seismic signal also
sometimes recorded, and well recorded
co-seismic and post-seismic signals.
Advances in Global Positioning System (
GPS ) measurement and data analysis
technology, and the increasing number
and improving quality of GPS receivers
deployed in active tectonic area, plus the
occurrence of another most exciting new
development Interferometric Syntetic
Aperture Radar (InSAR), have contributed
steadily towards understanding
earthquake sources and inter-seismic
deformation, and hence toward
improving hazard evaluation. Interseismic
pattern is now being included as
one of parameter in estimation the
probability of earthquake hazard. As a
conclusion, the geodetic measurements
became another effort for earthquake
hazard monitoring. These geodetic
measurements have a title as Earthquake
Geodesy.
technique we have for identifying the
location of future earthquakes in some
areas, because elastic rebound requires
elastic strain accumulation prior to
earthquakes. With geodetic measurement
we may saw clearly inter-seismic phase
of earthquake mechanism and also give
the better constrain to the location of
predicted future earthquake.
Many of the early studies of inter-seismic
deformation using geodetic data ( GPS )
took place in southern California. These
studies were also instrumental in
developing geodetic GPS methods and in
characterizing the precision and accuracy
of the technique (e.g. Larson & Agnew
1991; Feigl et al 1993). Feigl et al (1993)
presented a comprehensive review of
GPS data collected in central and
southern California between 1986 and
1992, as well as VLBI data collected
between 1984 and 1991. They also
discussed methods for obtaining and
combining highly precise GPS solutions
for the determination of station
velocities in regional-scale networks.
Daily solutions with phase ambiguities
fixed to integers (where possible) were
estimated with very loose constraints on
the station coordinates.
III.1 Geodetic Data and Inter-seismic
Deformation
Detection of slow inter-seismic strain
accumulation is probably the best
Figure 3. Inter-seismic deformation along the
coast of California. Image with courtesy of
Prof Thomas Herring. [Herring 2002]

III.2 Geodetic Data and Pre-seismic
Deformation
Just a few days before Tonangkai
earthquake happen in December 1944,
spirit leveling result has shown preseismic
deformation signals (mogi 1984).
The tilt start to changing four day before
the mainshock. This was the amazing
discovery that the earthquake tell us
something before they coming. With this
precursor then the scientist though the
earthquake may be known before and
would give the early warning. But this
signal turn out to be the one and the only
best signal it ever have. Many years
come, the earthquakes happen with
given no pre-seismic signal at all.
Figure 4. spirit leveling result has shown preseismic
deformation signals three to four
days before the mainshock exist (mogi 1984).
In the GPS era, in Arequipa Peru, preseismic
deformation signal has occurred
before the mainshock come. Position in
daily solutions start to give an anomalies
result in few months to a few days before
the mainshock happen. But again in many
well recorded continuous GPS signal in a
period of earthquake there are currently
more negative result then positive result
in pre-seismic deformation signals. Preseismic
deformation signal is now still
being analyzed whether its just an
anomaly signal in some case of
earthquake or we may saw well in any
certain characteristics of plates
interface.
III.3 Geodetic Data and Co-seismic
Deformation
Advances in Global Positioning System (
GPS ) measurement and data analysis
technology, and the increasing number
and improving quality of GPS receivers
deployed in active tectonic area, plus the
occurrence of InSAR Technology has
achieved many well recorded co-seismic
deformations (e.g. Figure 5-6). GPS
measurements can be related to the
earthquake source process through
Volterra’s formula (e.g. Aki & Richards
1980) for displacement at the Earth’s
surface due to slip on a surface of
displacement discontinuity in an elastic
medium. GPS measurements of surface
displacement can thus be inverted to
determine the geometry of earthquake
rupture(s). This determination is
particularly important for earthquakes
that do not rupture the ground surface,
or when seismic data, including
aftershock distributions, do not clearly
determine the rupture geometry.
In estimating the source geometry, an
earthquake is typically represented by
one or more rectangular dislocations with
spatially uniform slip. Once the fault
geometry is known, it is possible to
determine the distribution of slip on the
fault surface, and further on it help us
for better understanding the co-seismic
mechanism and the stress transfer
mechanism concerning the evaluation to
next earthquake potential occurrence
and the constrain of predicted area of
future earthquake.
Below given some examples of co-seismic
captures in some areas suffering the
earthquake (e.g. Kurile island 1994, Aceh
2004, and Chi-chi Earthquake 1999 )

III.4 Geodetic Data and Post-seismic
Deformation
Figure 5. Horizontal displacement vectors
from the M8.1 Kurile Islands (Hokkaido-Toho-
Oki), Japan, earthquake, October 4, 1994.
Displacements are relative to a station 1100
km away. Error ellipses show 99% confidence
regions. After Tsuji et al (1995).
Figure 6. Co-Seismic from the M9.0 Mega
thrust earthquake Aceh, December 26, 2004.
More then 2 meter displacements occurred in
city of Banda Aceh. Irwan et al (2005).
Figure 7. Average co-seismic deformation
interferogram from Chi-chi earthquake 1999
(Liu.et.al 2004)
GPS will play an increasingly important
role in improving our knowledge of postseismic
processes. Traditional seismic
instruments are insensitive to postseismic
processes, with the exception of
aftershocks. Strainmeters and tiltmeters
record short-term transients following
earthquakes, but only geodetic survey
measurements are likely to have the
spatial coverage and long-term stability
needed to resolve post-seismic strains
with characteristic times of years to
decades. With given the precision of the
measurements and the relative ease of
GPS surveying, we can expect a
revolution in our understanding of postseismic
processes in the upcoming
decade (Segall & David, 1997)
The 1989 Loma Prieta and 1992 Landers
earthquakes are the first to yield
significant post-seismic strain signals.
Following the Loma Prieta earthquake,
Savage et al (1994) found evidence for a
decaying transient with a characteristic
time of 1.4 years, from the first 3.3 years
of data.
Advanced various data sets of GPS and
others geodetic measurements, including
interferometric SAR images, could help
clarify competing interpretations of postseismic
deformation. The fact has shown
that the energy of an earthquake would
never reached a hundred percent by the
mainshock, some cased shown only fifty
percent released by the mainshock while
post-seismic hold the rest which can be
released up to several years and even
more. With continuous GPS networks we
can also look forward to the ability to
resolve rapid post-seismic signals, which
were previously only measurable with
strain and tilt meters.

Figure 8. Example of post-seismic deformation from Hector Mine Earthquake. The signals
still continue to change exponentially after more than four years since the mainshock, to
return to equilibrium and steady loading at the plates interface. graphic with courtesy of Prof
Thomas Herring. [Herring 2002]
IV. CONCLUSION
Detection of slow inter-seismic strain
accumulation is probably the best
technique we have for identifying the
location of future earthquakes in some
areas, because elastic rebound requires
elastic strain accumulation prior to
earthquakes. With geodetic measurement
we may saw clearly inter-seismic phase
of earthquake mechanism and also give
the better constrain to the location of
predicted future earthquake. Interseismic
pattern is now being included as
one of parameter in estimation the
probability of earthquake hazard. Preseismic
signal also sometimes recorded
among others mostly negative result.
Pre-seismic deformation signal is now
still being analyzed whether its just an
anomaly signal in some cased of
earthquakes or we may saw well in any
certain characteristics of plates
interface. Well recorded co-seismic
deformations can be related to the
earthquake source process. Further on
this records can thus be inverted to
determine the geometry of earthquake
rupture(s). This determination is
particularly important for earthquakes
that do not rupture the ground surface,
or when seismic data, including
aftershock distributions, do not clearly
determine the rupture geometry. Once
we found the geometry of the rupture it
help us for better understanding the coseismic
mechanism and the stress
transfer mechanism concerning the
evaluation to the next earthquake
potential occurrence and the constrain of
predicted area. Geodetic measurements
will play an increasingly important role in
improving our knowledge of post-seismic
processes. Advanced various data sets of
GPS and others geodetic measurements,

including interferometric SAR images,
could help clarify competing
interpretations of post-seismic
deformation. The fact has shown that the
energy of an earthquake would never
reached a hundred percent by the
mainshock, some cased shown only fifty
percent released by the mainshock while
post-seismic hold the rest. As a last
sentence of conclusion, the geodetic
measurements became another effort for
earthquake hazard monitoring.
REFERENCES
Aki K, Richards PG. 1980. Quantitative
Seismology: Theory and Methods. San
Francisco: Freeman. 932 pp.
´Arnad´ottir T, Segall P. 1994. The 1989
Loma Prieta earthquake imaged from
inversion of geodetic data. J. Geophys.
Res. 99(B11): 21835–55 fis. 37(6):1515–38
Blewitt G. 1993. Advances in Global
Positioning System technology for
geodynamics investigations: 1978–1992. In
Contrib. Space Geodesy Geodyn.: Technol.
Geodyn., Geodyn. Ser., ed. DE Smith, DL
Turcotte, 25: 195–213. Washington, DC:
Am. Geophys. Union. 213 pp.
Feigl KL, AgnewDC, BockY, Dong D, Donnellan
A, et al. 1993. Space geodetic
measurement of crustal deformation in
central and southern California, 1984–1992.
J. Geophys. Res. 98(B12):1677–712
Herring T. Plate Tectonic. Lecture notes,
MIT, 05/14/02
Hudnut KW, Bock Y, Cline M, Fang P, Feng Y,
et al. 1994. Co-seismic displacements of
the 1992 Landers earthquake sequence.
Bull. Seismol. Soc. Am. 84:625–45
Hudnut KW, Shen Z, Murray M, McClusky S,
King R, et al. 1996. Co-seismic
displacements of the 1994 Northridge,
California, earthquake. Bull. Seismol. Soc.
Am. 86(1B): S19–36
Hudnut KW (1994). Earthquake Geodesy and
Hazard Monitoring. U.S. National Report to
IUGG 1991-1994. Rev.Geophys, vol 33
Suppl.,@ 1995 American Geophysical Union
Larson KM, Agnew DC. 1991. Application of
the Global Positioning System to crustal
deformation measurements. 1. Precision
and accuracy. J. Geophys. Res.
96(B10):16547–65
Liu et.al. Pre-and co-seismic ground
deformation of the 1999, Chi-chi, Thaiwan
earthquake, measured with SAR
Interferometry. Computer & Geoscience
30(2004) 333-343.
Murray M, Marshall GA, Lisowski M, Stein RS.
1996. The 1992 M D 7 Cape Mendocino,
California, Earthquake: coseismic
deformation at the south end of the
Cascadia megathrust. J. Geophys. Res.
101:17707–25
Mori J. (2004). Earthquake Prediction.
Lecture notes on KAGI 21 Summer School.
Institute of Technology Bandung, Indonesia.
August 2004
Savage JC, Lisowski M, Svarc JL. 1994.
Postseismic deformation following the 1989
(M D 7:1) Loma Prieta, California,
earthquake. J. Geophys. Res. 99:13757–65
Savage JC, Svarc JL. 1997. Postseismic
deformation associated with the 1992 Mw D
7:3 Landers earthquake, southern
California. J. Geophys. Res. In press
22(13):1669–72
Segall P., J.L Davis. GPS Application for
Geodynamic and Earthquake studies. Annu
Rev. Earth Planet Sci. 1997 25 :361-36
Copyright 1997 by Annual Reviews Inc. all
right reserved
Shen ZK, JacksonDD,FengY, ClineM,KimM, et
al. 1994. Postseismic deformation following
the Landers earthquake, California, 28
June 1992. Bull. Seismol. Soc. Am. 84:780–
91
Stein .S, Wysession M. An Introduction to
Seismology, Earthquake, and Earth
Structure. Book Published by Blackwell
Tsuji H, Hatanaka Y, Sagiya T, Hashimoto M.
1995. Coseismic crustal deformation from
the 1994 Hokkaido-Toho-Oki earthquake
monitored by a nationwide continuous GPS
array in Japan. Geophys. Res. Lett.
22(13):1669–72
Vigny et.al (2005). Report on Banda Aceh
Mega-Thrust Earthquake, December
26,2004. SEAMERGES PROJECT 2005
Williams C, Arnadottir T, Segall P. 1993.
Coseismic deformation and dislocation
models of the 1989 Loma Prieta earthquake
derived from Global Positioning System
measurements. J. Geophys. Res. 98:4567–
78