Anda di halaman 1dari 201

ASSESSMENT AND MITIGATION OF SEISMIC RISK

IN THE EASTERN CARIBBEAN REGION

EUCENTRE

Via Ferrata 1, 27100 Pavia, Italy

March 2010
SUMMARY

The project “Assessment and mitigation of seismic risk in the Eastern Caribbean region”,
funded for the year 2009 by the Council of Milan for “Contribution to solidarity and
International cooperation”, has, as main objective, the implementation of a
multidisciplinary methodology which integrates the aspects of seismic monitoring,
seismic hazard, vulnerability and exposure of the buildings stock for the Eastern
Caribbean Countries, especially for the islands of Dominica, Barbados and Trinidad.

The European Centre for Training and Research in Earthquake Engineering


(EUCENTRE, Pavia - Italy) is the project coordinator with the Institute for
Advanced Study (IUSS, Pavia - Italy) and has, as Italian partner, the Italian Institute
of Geophysics and Volcanology (INGV, Milan - Italy), while the “University of West
Indies” (UWI, St Augustine - Trinidad) and the “Council of Caribbean Engineering
Organizations” (CCEO, Barbados) as Caribbean partner.

The main activities developed in this study corrently underway are illustrated in the
following points:

¾ Hazard analysis has been implemented using a logic tree approach which
allowed to systematically take into account model-based (i.e. epistemic)
uncertainty and its influence onto the computed ground motion parameters.
A critical review of past hazard studies performed in the region was
undertaken as a preliminary task; the standard Cornell-McGuire approach
based on the definition of appropriate seismogenic zones has been adopted
and has produced the updating of the maps on seismic hazard for four return
periods, 95, 475, 975 and 2475, and for 22 structural periods, which are 0,
0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.6, 0.75, 0.9, 1, 1.25, 1.5, 1.75, 2,
2.25, 2.5, 2.75 and 3 sec.; (“SECTION A - Probabilistic Seismic Hazard
Assessment”);
¾ Setting-up of a feasibility study for the installation of a modern digital strong-
motion seismic network which includes the study for the data re-
transmission; aim of this study is the improvement of the seismic monitoring
of the Eastern Caribbean Countries (“SECTION B - Strong Motion Network:
feasibility study and data transmission”);
¾ Preliminar evaluation through the use of remote sensing techniques in order
to approximately evaluate the level of seismic vulnerability for the buildings
which belongs to some important urban aggregates in the Eastern Caribbean
Islands (“SECTION C - Telecommunications and Remote Sensing”).

The new funds granted by the Council of Milan for the year 2010 will permit to
extend the research work, improving the results obtained in the first phase of this
project.
SECTION A

Probabilistic Seismic Hazard Assessment

by

Carlo Giovanni Lai1, Francesca Bozzoni1, Walter Salazar2, Laura Scandella1,


Mirko Corigliano1, Richard Robertson2, Lloyd Lynch2 and Joan
Latchman2

1 European Centre for Training and Research in Earthquake Engineering (EUCENTRE), Pavia,
Italy.
2 The University of the West Indies (UWI), St Augustine, Trinidad.
INDEX

1. INTRODUCTION............................................................................................................................1

1.1 THE AREA OF RESPONSABILITY ...................................................................................................1

1.2 DEFINITION OF THE STUDIED AREA ..........................................................................................2

2. CRITICAL REVIEW OF PAST SEISMIC HAZARD ASSESSMENT STUDIES ...............7

2.1 GENERAL SEISMICITY .................................................................................................................10

2.1.1 Data sources for the catalogue...........................................................................................12

2.2 SEISMIC HAZARD ESTIMATES .....................................................................................................15

2.2.1 First studies...........................................................................................................................15

2.2.2 Studies in the 1990’s: seismic hazard for the Caribbean, Mexico and the Americas ..22

2.2.3 Most recent studies..............................................................................................................27

2.3 CONCLUSIONS ..............................................................................................................................32

3. THE CONSTRUCTION OF THE EARTHQUAKE CATALOGUE .................................33

3.1 THE UPDATING OF THE EARTHQUAKE CATALOGUE .............................................................34

3.1.1 The databases used as sources ...........................................................................................34

3.1.2 Merging of databases catalogues........................................................................................36

3.1.3 Evaluation of differences in magnitude estimates of the databases sources ...............38

3.1.4 Homogenization of different scales of magnitude ..........................................................40

3.1.5 Catalogue 1997-2009 ...........................................................................................................46

4. SEISMOTECTONIC SETTING AND DEFINITION OF SEISMIC SOURCES ............49

4.1 SEISMOGENIC SOURCES ..............................................................................................................51


4.1.1 Zone 1: Volcanic Island-arc............................................................................................... 51

4.1.2 Zone 2-5: Subduction in the Lesser-Antilles.................................................................. 52

4.1.3 Zones 6-8: Puerto Rico and Virgin Islands ..................................................................... 53

4.1.4 Transition Zones 9 and 10A.............................................................................................. 54

4.1.5 Zone 10B: East of Trinidad............................................................................................... 54

4.1.6 Zone 11: North of Paria Peninsula................................................................................... 55

4.1.7 Zone 12: Trinidad Faults.................................................................................................... 55

4.1.8 Zones 13 and 14: El Pilar fault.......................................................................................... 55

4.1.9 Zone 15: South of Trinidad ............................................................................................... 56

5. PROBABILISTIC SEISMIC HAZARD ANALYSIS ............................................................... 61

5.1 METHOD OF CALCULATION ...................................................................................................... 61

5.1.1 Cornell-McGuire Approach............................................................................................... 61

5.2 PROCESSING OF THE EARTHQUAKE CATALOGUE ................................................................. 62

5.2.1 Declustering process........................................................................................................... 62

5.2.2 Estimation of completeness periods ................................................................................ 70

5.2.3 Gutenberg-Richter recurrence relationships ................................................................... 76

5.3 ADOPTED GROUND MOTION PREDICTION EQUATIONS ....................................................... 82

5.4 COMPARISON OF ATTENUATION RELATIONSHIPS WITH AVAILABLE STRONG MOTION


DATA ................................................................................................................................................... 89

5.5 PROBABILISTIC SEISMIC HAZARD ANALISYS AND RESULTS ................................................. 107

5.5.1 Treatment of Uncertainties in PSHA Computations ................................................... 107

5.6 RESULTS ..................................................................................................................................... 111

5.6.1 Dominica Island ................................................................................................................ 124

5.6.2 Barbados Island ................................................................................................................. 126

5.6.3 Trinidad Island .................................................................................................................. 129

5.6.4 Considerations about the obtained results..................................................................... 132


BIBLIOGRAPHY..............................................................................................................................133

APPENDIX I......................................................................................................................................141
1.INTRODUCTION

Nowadays we observe the growth of big cities in seismically active areas all around the
world which requires very precise evaluation of seismic hazard in order to prevent loss of
life, property damage, and social and economic disruption due to earthquakes.

One of these regions is the Caribbean islands which are located in a very active seismic
zone on the Caribbean/South American/North American plate boundaries. The
population of the Caribbean remained very low for a long time up until the 20th century,
so there are not so many studies about seismic hazard in the region. In recent times, rapid
growth of tourism in the Caribbean has resulted in increasing construction. This requires
new investigations in seismic hazard assessment of the region; since new seismic
instrumental data are available it is a very good time to make these investigations.

Seismic hazard was ignored in Caribbean region until the 1950’s, but when a series of
damaging earthquakes occurred from 1950-1970, this attracted scientist’s attention to the
seismic hazard problem in the region.

1.1 THE AREA OF RESPONSABILITY


Primarily it is necessary to delineate the area of the eastern Caribbean, which will be
studied in this work. In particular we define our area of responsibility (AOR), that
encompasses the islands between St. Martin/Anguilla and Trinidad; the following figure
shows the AOR.
A2

Figure 1.1 Area of responsibility (AOR) including the islands between St. Martin/Anguilla and
Trinidad, exclusive of the French territories (Guadeloupe and Martinique).

In this chapter, we will present some areas investigated in past seismic hazard assessment
studies and we will chose the zone object of this work.

1.2 DEFINITION OF THE STUDIED AREA


Figure 1.2 e Figure 1.3 show the area of responsibility (pink) with areas studied
respectively in the work of Taylor et al. (1978) and in the study of Shepherd, Lynch and
Tanner (1993).
A3

Figure 1.2 Area of responsibility (pink) with area studied in the work of Taylor et al. (1978)
A4

Figure 1.3 Area of responsibility (pink) with area studied in the work of Shepherd, Lynch and
Tanner (1993)

We chose to modifyed the areas involving in cited previous works in order to obtain a
zone whose boundaries have a distance greater then 300 km from analysed islands. We
enlarge the area selected in the study of Shepherd, Lynch and Tanner (1993). Table 1
show the coordinates of the rectangular area object of study.

Table 1.1 Our studied area


Studied Area
22,5 N - 68.3 W
22,5 N - 56 W
7 N - 68.3 W
7 N - 56 W

In the following figure (Figure 1.4), we show an image of our studied area.
A5

Figure 1.4 Our studied area (green rectangule)


A6
2.CRITICAL REVIEW OF PAST SEISMIC HAZARD
ASSESSMENT STUDIES

Researcher ignored the seismic hazard in the Caribbean region until the 1950’s, but when
a series of damaging earthquakes occurred from 1950-1970, this attracted scientist’s
attention to the seismic hazard problem in the region.

The production of a countors maps represents the fundamental element in a seismic


hazard analysis ; it has to be comprehensible to the users. The earliest attempts to do this
were by way of earthquake intensity scales of which the most known is the Modified
Mercalli scale; for example, in the case of Trinidad and Tobago, Robson (1962), Tomblin,
Key and Imbert (1968) used this concept, but they were not widely accepted by the target
groups.

In 1968 Cornell (1968) introduced two important concepts into this field: the first was
the concept of probabilistic seismic hazard, expressed as the level of some parameter of
ground motion which had a particular return period at a particular site; the second was
the concept of effective peak ground acceleration. If the intervals between earthquakes
are assumed to follow a Poisson process and the return period is assumed to be T years
then the probability that there will be one or more events within any period of t years life
time is:

t

P( t ) = 1 − e T (2.1)

Thus the phrase “seismic hazard” is used in the sense of “probabilistic seismic hazard”.

Algermissen and Perkins (1976) adopted a probability level of 10% and a period of
interest of 50 years so that their measure of earthquake hazard was the level of ground
acceleration (PGA) with shows that such en event has return period of approximately 475
years.

The first application of these concepts to the Caribbean region were made by Pereira and
Gay (1978) who computed hazard maps for Trinidad and Tobago and Jamaica. Taylor et
al.. (1978) extended this work to include the Lesser Antilles. The most serious common
problems of these works were basically the following: the magnitudes and locations of
A8

the earthquakes used to estimate levels of hazard were not well known. For the particular
case of Trinidad and Tobago Sheperd and Aspinall (1983) improved the quality of both
locations and magnitudes of the earthquakes used and produced an improved hazard map
for Trinidad and Tobago.

Since 1990 a large groups of researchers from Caribbean, Mexico, Central America and
South America produced a series of hazard maps for the whole of this region using the
concepts introduced by Cornell (1968) and Algermissen and Perkins (1976). In those
studies magnitudes of all recorded earthquakes within the region were re-computed and
expressed using the moment-magnitude scale of Kanamori (1970); the results relevant to
the Trinidad and Tobago region were published by Shepherd et al.. (1997), Tanner and
Shepherd (1997) and Shedlock (1999). A very comprehensive discussion of the
methodology used is that by McQueen (1997).

The most recent seismic hazard estimates which encompass the Caribbean region were
presented by Shepherd and Lynch (2003) for Trinidad and Tobago and by Tanner and
Shedlock (2004) for Mexico, the Caribbean and Central and South America.

Figure 2.1 contains time distribution of the most important past seismic hazard
assessment studies; histograms represent annual number of works in which seismic
hazard studies involving the Caribbean region have been developed.
A9

2008
2007
2006
Annual number of works regarding with the seismic hazard studies of the Caribbean region 2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
4

Figure 2.1 Time distribution of the most important past seismic hazard assessment studies for the
Caribbean region
A10

2.1 GENERAL SEISMICITY


Seismicity in the Caribbean is concentrated along the Caribbean/South American/North
American (and associated smaller) plate boundaries.

The Tectonic Setting of the Caribbean is illustrated in Figure 2.2, which shows the
approximate Caribbean Plate Boundaries; as can be seen, all of the Caribbean countries
lie close to these boundaries.

Figure 2.2 Tectonic Setting of the Caribbean (after Jordan, 1975)

The Caribbean Plate is moving eastward respect to the adjacent North American and
South American Plates at a rate of approximately 20 mm per year (Gibbs, 2001).

The Caribbean plate is overriding the North and South American plates on the east
(Figure 2.2); on the west, the Cocos plate is subducting beneath the Caribbean plate.

Figure 2.3 shows Lesser Antilles subduction zone.


A11

Figure 2.3 Lesser Antilles subduction zone (Byrne, Davis and Sykes, 1988)

Seismicity recorded in the Caribbean region is concentrated along the Antilles arc, from
Hispaniola to Trinidad (Figure 2.4), where the Caribbean plate is overriding the North
and South American plates. Most of the large and moderate earthquakes have been
shallow intraplate earthquakes near the plate boundaries, although some researchers have
interpreted a few large earthquakes as plate interface events (Dewey and Suàrez, 1991).

Figure 2.4 Plottage of earthquakes with moment magnitude (MW) greater than 4.6 that occurred
during the years 1900 through 1994 in Mexìco, Central America and the Caribbean
(Shedlock, 1999).
A12

An in-depth investigation on the origin of seismicity in the Caribbean arc will be


presented in the following chapter when we will discuss the definition of appropriate
seismogenic zones on which the standard Cornell-McGuire approach is based.

2.1.1 Data sources for the catalogue


There are two distinct sets of data sources for the catalogue of the Caribbean region.

The first set consists of reports of earthquakes felt in region since the first European
settlements in the early sixteenth century; the historical catalogue begins in 1532
(Shepherd, 1993). Until about 1907, these reports have been collated and analyzed by
number of authors notably Robson (1964) and Grases (1971); for Trinidad, considered as
an observatory par excellence because located at a seismotectonic crossroads, Vogt (2008)
analyzed several felt or damaging earthquakes in Trinidad and Tobago and presented new
original archive findings that modify significantly the picture of the Trinidad seismicity as
found in the catalogue of Robson.

The second set consists of recorded data; the instrumental recording of earthquakes
began in the early 20th century, but the deployment of a regional radio-linked seismograph
network occurred in 1976; thus, the instrumental data set can be divided into two
segment.

Before about 1976 reliable earthquake location depended mainly on teleseismic


observations combined with those from a very small number of high-quality seismograph
stations within the region; the lower magnitude limit for reliable earthquake location
during this period was about 5. A seismograph station was established in Port of Spain in
1899 as part of John Milne’s world-wide seismograph network and continued in
operation until some time during the First World War. Instruments of the Milne type
were designed primarily to record distant earthquakes and responded poorly to near
earthquakes; the accuracy of epicentral location of the only two earthquakes with
epicentre close to Trinidad was no better than ± 50 km; depth control was not-existent.
A permanent, single-component short-period station was established in 1953 but it was
not until 1964, with the completion of the World-Wide Standardzed Seismograph System
Network (WWSSN) which includes a station (TRN) in Trinidad, that the accuracy of
location of earthquakes and the detection threshold improved to the point at which
accurate, quantitative seismicity estimates could be made.

The second segment of the instrumental record consist of data collected by two dense
networks for radio-linked short-period seismograh stations which have been continuously
operated within the regions by the Seismic Research Centre of University of the West
Indies (SRU) and by the Institut de Physique du Globe of the University of Paris (IPG).
A13

A third high-quality network is operated on the periphery of the region by the Fundación
Venezolana de Investigaciones Sismológicas (FUNVISIS). The operation of this
networks has resulted in much more precise location of all earthquakes and in a lowering
of the lower magnitude threshold for reliable detection to about M=3.

All events which have occurred since 1976 have been relocated using appropriate regional
travel-time tables (Shepherd, 1993).

Since September 15, 2000, six digital strong-motion accelerographs have been established
in Trinidad; five of these instruments have operated continuously since then. Their
important advantages consist of having an improved dynamic range and being linear to
ground motion over the range of frequencies of interest in earthquake strong ground
motion studies (Shepherd and Lynch, 2003).

According to information from http://www.uwiseismic.com website at the moment


there are several seismic sub- networks in the eastern Caribbean. Figure 2.5 represents the
seismic network which covers all the eastern Caribbean, this network is maintained by the
Seismic Research Centre of the University of West Indies combined with other networks
managed by cooperating agencies. On the figure green stars shows the seismic network
managed by the Fundación Venezolana de Investigaciones Sismológicas (FUNVISIS) and
Universidad de Oriente (UDO); those which operated by Institut de Physique du Globe
de Paris (IPGP) in the French Antilles are shown in blue. Though each sub-network
contains mainly short period stations that record only the vertical component, it also
includes at least one broadband three-component station.
A14

Figure 2.5 Seismic networks in the eastern Caribbean area (http://www.uwiseismic.com)


A15

2.2 SEISMIC HAZARD ESTIMATES

2.2.1 First studies


In 1978, the First Caribbean Earthquake Engineering Conference took place in Port of
Spain, Trinidad; during this event some researchers (Pereira and Gay, Taylor et al..,
Tomblin, Rothwell) presented the first hazard estimates for the Caribbean region. Pereira
and Gay (1978) computed hazard maps for Trinidad and Tobago and Jamaica; Taylor et
al.. (1978) extended the work of Pereira and Gay to include Lesser Antilles. They
estimated the level of earthquake hazard in the eastern Caribbean employing the peak
horizontal acceleration with 90 percent probability of non-exceedence in any 50-year
period; the methods used were based on the method of earthquake hazard analysis
introduced by Cornell (1968) but the quantitative results varied enormously both in the
overall level of hazard and in the way in which it varied throughout the region.

Pereira and Gay (1978) derived relationships between return period T and acceleration
for Port of Spain, Trinidad, Tobago and Barbados (in the eastern Caribbean). The
authors defined two zones of activity (Figure 2.6):

- El Pilar Zone with a width of approximately 100 km, chosen in order to


encompass all the earthquake events, thought to be associated with the system
of faulting at the plate boundary in this region; a central line was drawn for the
purpose of applying the Cornell’s linear source model.
- Lesser Antillean Fault Zone which defined a source of shallow earthquakes
and one of intermediate depth events; the researchers chose a linear fault to
represent shallow events aligned almost N-S.
A16

Figure 2.6 Representation of fault zones activity (Pereira and Gay, 1978)

The catalogue used contained data on reported earthquake events for a period of 74 years
1899-1973; the researchers applied the Cornell analytical model with recalculated
magnitudes MS using the relationship of Vanek et al.. (1962) and, for the special case of
earthquakes prior to 1910, recorded on Milne instruments, a relation developed by
Ambraseys; they obtained an over estimation of ground parameters with this second
approach. The attenuation law used is Esteva’s relationship (1974):

y = b1 e b2 M ( R + RO ) −b3 (2.2)

where y is the ground motion parameter, RO, b1, b2 and b3 are constants, M is the
magnitude and R is the focal distance (km). This law had been developed for shallow
events; this represents a limit of this work: ground motion parameters for intermediate
depth and deep events do not appeared to conform to a predictable attenuation law when
Pereira and Gay developed their analysis. Further the degree of uncertainty involved in
the use of Esteva’s laws was high.

They obtained hazards curves, which are presented in the original paper (since the paper
is very old and the pictures are bad quality it was decided not to include them within this
review).
A17

Taylor et al. (1978) proposed a preliminary analysis of seismic hazard in the Lesser
Antilles and Trinidad and Tobago; earthquake source zones were developed from a
review of the tectonics and structure of the region and an analysis of strain release,
developed by Allen et al. (1965) for the areas of Southern California, using the earthquake
magnitude records for the period 1904-1974. They created strain release maps, like the
one represented in Figure 2.7 which shows that seismic energy release takes place
throughout the island arc over the seventy-one years considered in this work; Figure 2.7
shows the resulting contours based on epicentral location only, but the researchers
developed maps in order to investigate the effect of depth strain release.

Figure 2.7 Strain release maps (Taylor et al., 1978)

Taylor et al. separated the “shallow” zone and the “deep”, because, they said, these two
source had different magnitude-frequency relationship; they chose rectangular elements,
consistent with the requirements of the computer programme used (Cornell and Grigori,
1974), in order to schematize the source zones (Figure 2.8).
A18

The magnitude-frequency relationships used were developed from the following sources:
Gutenberg-Richter (1904-1955), Sykes and Ewing (1956-1964), Bulletins of the
International Seismological Centre (1965-1970) and Local Earthquake Solutions by
Seismic Research Unit (1965-1974); in the first and in the second relationship above the
magnitudes are surface wave magnitudes (MS) whereas the third and the fourth are body
wave magnitudes (mb), in which MS had been converted using the formula of Geller and
Kanamori (1977).

Figure 2.8 The source zones used in the “ shallow” zone and in the “ deep” zone (Taylor et al.,
1978)

The attenuation relationship used is McGuire’s law (1974):

y = ae bM ( R + 25 −C )ε (2.3)
where y is the peak ground acceleration, M is Richter magnitude, R is focal distance (km),
a, b and c are regression constants and ε is residual error term.

Results obtained in this study are presented in the following figure (Figure 2.9):
A19

Figure 2.9 Results obtained by Taylor et al. (1978)

Generally, all of these early attempts to produce seismic hazard maps suffered from a
number of problems of which the most serious was that the magnitudes and locations of
the earthquakes used to estimate levels of hazard were not well known.

Shepherd and Aspinall (1983) improved the quality of both locations and magnitudes of
the earthquakes used and produced an iso-acceleration map for the Trinidad and Tobago
region.

The establishment of station TRN and other station in the south-eastern Caribbean
(1964) contributed to correct a deficiency in location capability, but Shepherd and
Aspinall (1983) designed criteria to ensure that the earthquakes of the period 1964-1976
could be located with an accuracy not worse than ± 10 km in epicentral coordinate and
± 20 km in depth; the criteria were that the earthquake should had been recorded by at
least ten seismograph stations including at least one in each of the four quadrants of
azimuth from the epicentre and at least one station within 150 km of the epicentre.
Starting by hypocentres computed by the International Seismological Centre the
earthquakes were re-located by variant of the method of Joint Hypocentral
Determination specially developed for use in the Caribbean region. The catalogue used in
the analysis covered the period 1964-1979 and the magnitudes were body-wave
magnitudes mb.

Shepherd and Aspinall (1983) noted that the different definitions of the source zone in
the previous seismic hazard estimates produced divergent results; thus they presented a
detailed account of the neo-tectonics of south-eastern Caribbean, which conducted them
to identify three major source zones (Figure 2.10):
A20

- a zone of shallow (depth less than 70 km), probably strike-slip earthquakes;


- a deeper (depth greater than 70 km) zone of probable thrust earthquakes;
- a diffuse zone of probable normal faulting.

Figure 2.10 Definition of major source zone near Trinidad and Tobago (Shepherd and Aspinall 1983)

The researchers applied the truncated exponential form of the frequency-magnitude


relationship (introduced by Cornell) developed by Cornell and Vanmarcke (1969): they
estimated maximum magnitudes for earthquakes in each of the three source zones, using
the concept of the seismic moment of an earthquake suggested by Molnar (1979).

Shepherd and Aspinall used McGuire’s attenuation low (1974):

y = b1e b2 M f ( R ) (2.4)
where y is the peak ground acceleration (cm2/s) and R the hypocentral distance (km); the
researchers calibrated constants for events in source zones 1 and 3 and constants for
events in source zone 2 since the source zones showed different intensities effects.

They estimated the level of earthquake hazard in Trinidad and Tobago as the peak
horizontal bedrock acceleration with 90 percent probability of non-exceedence in any 50-
year period (see Figure 2.11).
A21

Figure 2.11 Estimated level of earthquake hazard in Trinidad and Tobago (Shepherd and Aspinall
1983)

None of the hazard maps presented by these authors was generally accepted as a basis for
seismic regionalization, because in the mid-1980’s it was felt that the current state of
knowledge did not permit the construction of definitive isoacceleration and iso-velocity
maps.

The Council of Caribbean Engineering Organizations (CCEO) then appointed a


subcommittee to examine the data bases and the methodologies used, to consult with
different authors and to recommend levels of later forces to be applied to earthquake-
resistant design in the Caribbean. The subcommittee reported to the CCEO regional
conference on earthquake and wind design in Port of Spain in 1983 (Faccioli et al., 1983).
These recommendations, which contained a simple approach to earthquake design based
on the United States Uniform Building Code (UBX), were then incorporated into the
Caribbean Uniform Building Code.

The figures with the estimated peak ground accelerations (in cm/s2) contained in
Caribbean Uniform Building Code were presented with the warning that they would been
substantial redacted in the near future.
A22

2.2.2 Studies in the 1990’s: seismic hazard for the Caribbean, Mexico and the
Americas
In 1990 began International Decade for Natural Disaster Reduction (IDNDR): a large
group of scientists from the Caribbean, Mexico, Central America and South America
produced a series of hazard maps for the whole of this region using the concepts
introduced by Cornell and Algermissen and Perkins. A particular feature of this work was
that magnitudes of all recorded earthquakes within the region were re-computed and
expressed using the moment-magnitude scale of Kanamori (1977). The results relevant to
the Trinidad and Tobago region were published by Shepherd et al. (1997), Tanner and
Shepherd (1997) and Shedlock (1999). A very comprehensive discussion of the
methodology used is that by McQueen (1997). McQueen (1997) carried out an evaluation
of seismic hazard in the Caribbean and Central America using three different probabilistic
methods of seismic hazard estimation. The methods she used were the source zone or
Cornell-type (two different computer programmes), the extreme value method (Gunibel,
1958; Makropoulos and Burton, 1986) and the historic parametric method. In her
evaluation of the results she found all methods gave similar results, but concluded that
the historic parametric method seemed more stable under varying conditions of
computation. The mentioned authors followed her recommendation and used historic
parametric method, about which we propose the following brief illustration. The first
step of this method is the development of a uniform earthquake catalogue for the region;
next, appropriate attenuation relations are identified or developed. Estimates of ground
motion from every earthquake in the catalogue are calculated at every site in the region
based on the chosen attenuation function(s). Return periods for exceedance of a range of
values of ground motion are tabulated, and curves are fitted to these data.

Thus, the earthquake catalogue served as the source characterization for the seismic
hazard map: no geological information was used and no source zones were drawn
(Shedlock, 1999).

Before we present the seismic hazard studies of previous mentioned authors, we focus on
the features of their work regarding magnitudes estimates and completeness of seismicity
catalogues evaluation.
A23

Magnitudes estimates

“Since good results depend to the first order on the catalogue upon which the
computations are based, in the absence of a top quality catalogue with magnitude
estimates on a uniform scale (preferably moment magnitude), seismic hazard estimates to
modern standards are not possible” (Shedlock, 1999). All magnitudes of the earthquakes
occurred in the Caribbean region were converted to moment magnitude (MW) through a
multi-step process (Tanner and Shepherd, 1997; Shedlock, 1999); as a first step in their
conversion scheme, Tanner and Shepherd (1997) obtained the seismic moment (M0) of as
many of earthquakes in the catalogue as possible. For most of the large earthquakes that
occurred prior to the 1980’s, M0 has been derived from the relationship:

M 0 = µAd (2.5)
where µ is the shear modulus (rigidity), A is the area of the fault plane and d is the
average slip during the earthquake (Aki and Richards, 1980; Tanner and Shepherd, 1997)
then converted M0 to MW using the relationship 2.6:

M W = 2 3 log M 0 − 10.73 (2.6)


derived by Kanamori (1977). Tanner and Shepherd (1997) developed a hierarchical
scheme, summarized by Shedlock (1999), to assign MW values to all the earthquakes in
the catalogue:

- MW calculated from M0, where M0 was determined directly from M0 from the
Aki and Richards (1980) relationship.
- MW calculated from M0, where M0 was determined directly from digital,
broadband seismic records.
- MW from M0, where M0 was taken directly from the Harvard University CMT
catalogue.
- M W ≈ M S where MS has been reliably determined for MS>6.6 and
M W = 2 3 log M S + 2.34 for M S ≤ 6.6.
- MW calculated from mb through a two-step process of first converting to mb to
MS using M S = 1.74 mb − 3.95 then using one of the relationship given
above.
- MW calculated from other magnitude scales through a two-step process of first
converting the other magnitude to MS using one of the several known
relationship, then using one of the relationship given above.
A24

Completeness of catalogue

Completeness of seismicity catalogues is always an issue. Periods of completeness for the


pre-instrumental period were advanced by Shepherd and Lynch (1992), who had
described the seismicity of the Caribbean up to the year 1900; the results are summarised
in Table 2.1.

Table 2.1 Periods of completeness; earthquakes of different intensities (after Shepherd and Lynch,
1992).

Intensity Period of completeness Number (to 1900)


≥ MMVI 1820 -1900 73
≥ MMVII 1800 - 1900 47
≥ MMVIII 1750 - 1900 30

For the instrumental period, a number of methods had been suggested to test the
completeness of catalogues. The most widely used test of completeness would seem to be
the so-called Stepp test (Tanner and Shepherd, 1997), in which the standard deviation of
the estimate of the mean rate of earthquake occurrence for a given magnitude range is
plotted against time on a log-log graph. The data are considered to be complete so long
as the graph follows a linear trend. The authors presented this analysis for the whole
catalogue of Latin America and the Caribbean (Figure 2.12).

Figure 2.12 Annual numbers of 1900-1993 earthquakes classified by range of magnitude (Tanner and
Shepherd, 1997)
A25

The recorded seismicity increases approximately linearly until 1964, when the WWSSN
and national networks throughout the region began to be deployed. The curve for MW ≥ 6
is least changed by the introduction of more uniform earthquake recording and reporting.
The curve for MW ≥ 5 continues to increase approximately linearly. The dramatic jump in
the curve for MW ≥ 4 clearly indicates that the catalogue is incomplete for these events
prior to 1964. Although the period of completeness for an earthquake of M W ≥ 5 is the
least clear of all three, Tanner and Shepherd (1997) assumed that the catalogue is
complete for this magnitude from 1930 onwards and M W=5 has been chosen as the
lower bound magnitude for earthquakes to be included because earthquakes with MW<5
do not usually cause damage throughout the region. Aftershocks were removed by a
straightforward application of the single link cluster analysis method developed by Davis
and Frolich (1991), but, because of uncertainties surrounding dates and locations, it is
possible that some duplicates remained in the macroseismic catalogue. However,
estimated periods of completeness by magnitude compare in Table 2.2.

Table 2.2 Estimated periods of completeness by magnitude for the unified catalogue (after Tanner
and Shepherd, 1997)

Intensity Period of completeness


MW ≥ 7.5 1900 - 1993
MW ≥ 5.75 1930 - 1993
MW ≥ 4.0 1964 - 1993

Tanner and Shepherd (1997) in the second part of their work produced seismic hazard
maps for Latin America and the Caribbean using the created catalogue which contained
all earthquakes with MW ≥ 4 that occurred between 1900 and 1994. Any earthquakes with
magnitudes that could not be converted to MW using the above hierarchy were excluded
from catalogue. For the period July, 1976 December, 1991 all events had been relocated
using the Joint Hypocentral Determination and regional travel-time tables (Shepherd et
al., 1987). The results obtained specifically for the Caribbean region are also published by
Shepherd et al. (1997) and by Shedlock (1999), who attended to seismic map of North
and Central and the Caribbean.

About attenuation law, they applied the relation of Aspinall et al. (1994), who examined
number attenuation relationships and concluded that the equation developed by
Woodward-Clyde (1982) for subduction zone settings best fit scene in the Trinidad-
Tobago region. Although this was intended for subduction zones this was been suggested
for consideration by the project office for all the Caribbean, largely because it agreed well
better than other relations with the limited data available. The relation used is the
following:
A26

ln a = c1 + c 2 M − c3 ln( D + c 4 ) (2.7)
where a is PGA, D is the distance within the context of the “Singh rupture zone”, M is
the magnitude and the constants are: c1 = 5.347 , c 2 = 0.5 , c 3 = 0.85 and
c 4 = e 0.463M .

For shallow (with depths of 15 km or less) events of the whole region examined, they
decide to use the Joyner and Boore (1993) relationship.

The authors (Tanner and Sheperd, 1997; Shepherd et al., 1997; Shedlock, 1999) proposed
the first systematic, detailed Caribbean seismic hazard map (Figure 2.13), which depics
PGA (in cm2/s) with a 10% chance of exceedance in 50 years; hazard levels had been
computed on a grid of 0.25 degrees.

Figure 2.13 Caribbean seismic hazard map (Tanner and Shepherd, 1997; Shepherd et al., 1997;
Shedlock, 1999)

Tanner and Shepherd (1997) used the results McQueen (1997) obtained with the historic
parametric catalogue as the basis of comparison. The results expressed in term of PGA
values were not the same; Tanner and Shepherd identified the cause in the different
attenuation relations used respectively by the authors.
A27

2.2.3 Most recent studies


Most recent hazard estimates for the Caribbean region are Shepherd and Lynch
microzonation of Trinidad and Tobago (2001; 2003) and Tanner and Shedlock seismic
hazard maps of Mexico, the Caribbean, and Central and South America (2004).

In the end we present Douglas and Mohais (2008) work, which is interesting, because
they presented a specific analysis of the ability of attenuation law for subduction
earthquakes to estimate observed earthquake ground motions on the islands of the Lesser
Antilles.

Shepherd and Lynch (2003) work is divided in two parts: the first produced in the 2001
and the second in the 2003; the project developed by the Seismic Research Unit of the
University of the West Indies (UWI) with the purpose to produce seismic hazard
assessment and microzonation of Trinidad and Tobago began in the 2000 when six digital
strong-motion accelerographs had been established in Trinidad. The method used for
assessment is the historic parametric method as applied by Tanner and Shepherd (1997);
two crucial input in this method are: a complete and homogenous earthquake catalogue
for the region affected and the attenuation relationship.

The data of the catalogue used by Shepherd and Lynch (2003) covered the period from
1965 to 2003 because for this period data are complete for magnitudes greater than 4.0
and the accurancy with which epicentres are determined has improved considerably over
that period.

One of the most important original objectives of the work (Shepherd and Lynch, 2003)
consisted in developing a attenuation relationship for region analysed using data recorded
by digital strong-motion accelerographs; since these data which covered the period
September 2000 to February 2001 were not enough, the authors used the relationships of
Boore et al. (1993, 1997) for shallow events (developed from studies of a very suite of
shallow earthquakes in western North America) and, for deeper events, chose Youngs et
al. (1997) law instead of Climent et al. (1994) because the first one fit better the
observations recorded. Results obtained were presented as countered maps of the
Spectral Ground Acceleration at 0.2 seconds (Figure 2.14) and at 1.0 seconds (Figure
2.15) with 2 % probably of being exceeded in any 50-years period.
A28

Figure 2.14 Maps for Trinidad and Tobago of the Spectral Ground Acceleration (gals) at 0.2 seconds
with 2% probably of being exceeded in any 50-years period (Shepherd and Lynch, 2003)

Figure 2.15 Maps for Trinidad and Tobago of the Spectral Ground Acceleration (gals) at 1.0 seconds
with 2 % probably of being exceeded in any 50-years period (Shepherd and Lynch, 2003)
A29

For Shepherd and Lynch the level of earthquake hazard in Tobago and in the north-
eastern corner of Trinidad were much higter than was previously thought; this idea could
not derive from direct comparisons with previous works, but it derived from
considerations about the choice of attenuation relationship which fit observed levels of
ground acceleration better.

They examined also the effect of surface geology on strong ground motion both by direct
measurements in the field and by studying results obtained elsewhere in the world and
identified areas in Trinidad where ground amplification was likely to be important.

The authors estimated response spectra for a range and focal depths and recommended
two normalized response spectra, one for hard rock and one for soft ground, for use in
earthquake resistant design.

Tanner and Shedlock (2004) seismic hazard values were calculated for Mexico, the
Caribbean, and Central and South America using the historic earthquake occurrence
method as applied by Tanner and Shepherd (1997). They extended catalogue
completeness analyses developed by Tanner and Shepherd (1997) to cover the period
1900-1997 (Figure 2.16).

Figure 2.16 Numbers of events per year from the unified catalogue and for MW=4, 5, and 6 (Tanner
and Shedlock 2004)

Events selected for their computations resulted from the extraction of all earthquakes
with MW<5 and removal of aftershocks; thus, the period of completeness chosen defined
an interval of 68 years (1930–1997) for all events with a magnitude of 5 or greater.
A30

About attenuation functions, Tanner and Shedlock used three equally weighted
relationships to calculate ground motions from shallow ( ≤ 15 km) earthquakes plus a
subduction zone relationship for intermediate/deep earthquakes; they considered the
following authors of attenuation relations: Boore et al. (1997) who determined a suite of
attenuation relations that provided estimations of ground motions from shallow
earthquakes with 5.5 ≤ MW ≤ 7.5 within a horizontal distance of 80 km or less; Campbell
(1997) who developed a law that may be used to estimate ground motions from shallow
earthquakes with MW ≥ 5.5 within a rupture distance of 60 km or less; Sadigh et al. (1997)
who developed relations suitable for application when a MW ≥ 5.5 shallow earthquake
occurs within 100 km or less of a site; Youngs et al. (1997) who developed a suite of
attenuation relations that provide estimations of ground motions from interface and
intraslab subduction zone earthquakes with MW ≥ 5.5 within rupture distances of 10 km
to 500 km.

These authors developed these relations using strong motion records recorded from 164
subduction zone earthquakes between 10 and 229 km deep throughout the world. About
16% of these earthquakes had hypocentral depths greater than 100 km, with just four
events at depths greater than 175 km. Tanner and Shedlock could not locate any reports
of damage from earthquakes as deep or deeper than 175 km throughout the region, so
they simply imposed 175 km as the upper bound distance for earthquakes to be included
in our calculations. The YCSH97 relationships for interface events were assumed for
earthquakes at depths between 15 and 50 km and the intraslab relationships were used to
calculate ground motions from earthquakes deeper than 50 km.

Tanner and Shedlock (2004) presented maps depicting PGA and 0.2 and 1.0 seconds
spectral accelerations (SA) with 50%, 10%, and 2% chances of exceedance in 50 years for
rock sites; in Figure 2.17, the map of PGA (%g) with a 2% chance of exceedance in 50
years. The author compared the obtained results with seismic map of North and Central
and the Caribbean developed by Shedlock (1999), but no consideration about the
Caribbean region are reported.
A31

Figure 2.17 PGA (%g) with a 10% chance of exceedance in 50 years (Tanner and Shedlock 2004)

Recent specific studies about attenuation relationship

Douglas and Mohais (2008) presented a quantitative analysis of the ability of eight
published empirical ground-motion prediction equations (GMPEs) for subduction
earthquakes (interface and intraslab) to estimate observed earthquake ground motions on
the islands of the Lesser Antilles (specifically Guadeloupe, Martinique, Trinidad, and
Dominica). Recorded ground motions from the selected subduction earthquakes (in total,
over 300 records from 22 earthquakes, happened in the period 1996-2008, from various
seismic networks) had been compared with ground motions estimated by these eight sets
of equations: Atkinson and Boore (2003), Crouse (1991), García et al. (2005), Kanno et al.
(2006), Lin and Lee (2008), McVerry et al. (2006), Youngs et al. (1997), and Zhao et al.
(2006).
A32

One of the most important findings was that the recent ground-motion models
developed from Japanese data (Kanno et al. 2006; Zhao et al. 2006) provided quite good
predictions of observed earthquake ground motions and their variabilities in the Lesser
Antilles. Other recent GMPEs derived for Mexico (García et al. 2005), Taiwan (Lin and
Lee 2008), and New Zealand (McVerry et al. 2006) were ranked poorly by the method
used (Scherbaum et al., 2004) for dataset considered.

All of obtained results, however, were based on comparisons between ground motions
from magnitudes and distance ranges generally outside those used to derive the tested
attenuations relationships; thus, Douglas and Mohais suggested that any seismic hazard
assessment for this region selects a number of attenuation law so that the epistemic
uncertainty in ground-motion prediction is not underestimated; the ranking of the
different models presented here could be useful as one component of a scheme to weight
the predictions from the various GMPEs.

2.3 CONCLUSIONS
Caribbean islands are located in very active seismic zone; increased growth of population
and the natural beauty has attracted tourists from all around the world, which has lead to
increase of construction, it is necessary to estimate seismic hazards in order to avoid
damage and loss of life in case of large earthquake.

Most of previous studies about seismic hazard in Caribbean were reviewed in this report.
As can be seen, not much investigation has been performed in Caribbean area; it means
that there is a lot of work to do, especially now, when new methods and techniques are
available. We see that the most recent studies were done before 2004, which is already 5
years old. Nowadays there are more qualitative seismic data available, so more precise
estimation might be performed.
3.THE CONSTRUCTION OF THE EARTHQUAKE
CATALOGUE

A preliminary, fundamental step of a seismic hazard study consists in the construction of


an earthquake catalogue; this catalogue has to be composite, homogeneous well-defined
and obtained from a merger of the available regional databases accounting for both
historical and instrumental seismicity in a consistent framework.

In the previous Chapter, we present past seismic hazard study involving Eastern
Caribbean; now we refer in particular to Tanner and Shepherd (1997) work, in which they
describe the methods used to construct the earthquake catalogue for the Latin America
and the Caribbean (IPGH catalogue). The Seismic Research Unit of the University of the
West Indies (SRU/UWI) provide us the MANAGE software system and the IPGH
catalogue. Thus, we have selected all events with Moment Magnitude Mw greater than 4.0
happened in the studied zone during the period between 1530-1996; they are about 2000.
The updating of IPGH catalogue to cover the period 1997-2009 consists of a merger of
the available databases; all examined sources are presented in Figure 3.1.
Database
CERESIS Hipocentros
Catálogos Regional “Hipocentros”
CERESIS Intensidades
Catálogos Regional “Intensidades”
ISC
International Seismological Centre

USGS/NEIC (PDE)
U. S. Geological Survey Earthquake Database
USGS/NEIC (NOAA)
U. S. Geological Survey Earthquake Database
USGS/NEIC (NGDC)
U. S. Geological Survey Earthquake Database
USGS/NEIC (SISRA)
U. S. Geological Survey Earthquake Database
USGS/NEIC (Centennial)
U. S. Geological Survey Earthquake Database
NGDC
National Geophysical Data Center
ANSS
Advanced National Seismic System

Figure 3.1 All examined sources


A34

3.1 THE UPDATING OF THE EARTHQUAKE CATALOGUE

3.1.1 The databases used as sources


The updating of IPGH catalogue to cover the period 1997-2009 consists of a merger of
the available databases; not all examined sources presented in Table 1 were used because
of the period covered by the sources themselves. Figure 3.1 indicates period covered by
each considered databases.

Database Catalogue

CERESIS-Hypocentros 1471-1991
CERESIS-Intensity 1520-1981
ISC 1900-2009
USGS-NEIC (PDE) 1973-2009
USGS-NEIC (NOAA) 1530-1986
USGS-NEIC (NGDC) 1900-1979
USGS-NEIC (SISRA) 1471-1981
USGS-NEIC (Centennial) 1900-2002
NGDC 1520-2009
ANSS 1898-2009

Figure 3.1 Period covered by each considered databases

Thus we select only the following database to update the catalogue until 2009:

I. ISC 3
The ISC catalog is a catalog that includes a regional list of shallow and
intermediate earthquakes.

II. USGS-NEIC (PDE) 4


USGS-NEIC catalog contains all earthquakes located by USGS NEIC and its
predecessors. In particular the sub-catalog PDE (Preliminary Determinations of
Epicenters) includes determining the magnitude and hypocenters made by USGS

3
http://www.isc.ac.uk/index.html
4 http://neic.usgs.gov/neis/
A35

NEIC. The publication is called “preliminary” because it is normally produced


few months after the events occur while the “final” computation of hypocenters
is considered to be the Bulletin of the International Seismological Centre (ISC),
which is produced about two years after the earthquakes occur.

III. USGS-NEIC (Centennial) 5


The USGS-NEIC Centennial catalog, originally compiled by Engdahl and
Villaseñor (2002) until 1999 and updated and extended by NEIC since 1900 to
April 2002, is a global catalog of the locations and magnitudes of more
significant earthquakes. For the most recent years (1964-present), it considers
only the events with magnitude 5.5 or greater. For the period prior to 1964, the
magnitude limit is considered equal to 6.5.

IV. ANSS 6
The ANSS catalog (Advanced National Seismic System) is a composite catalog
created by merging the master earthquake catalogs and then removing duplicate
solutions for the same event. The ANSS earthquake catalog grew out of the
efforts of the CNSS (Council of the National Seismic System). It was previously
called the CNSS earthquake catalog.

V. NGDC 7
The NGDC catalog (National Geophysical Data Center) contains all the
earthquakes considered “destructive”, that is all those events that have moderate
damage, ten or more deaths, magnitude 7.5 or greater and intensity X or greater.
This catalog should theoretically coincide with the NEIC (NOAA) as it is based
on the same criteria and, in origin, comes from the same source but noting the
number of events, can be seen that the NGDC catalog contains more number of
earthquakes.

From those databases we have extracted the events happened in the studied area during
the period between 1997 and 2009; to obtain all events with Mw greater then 4.0, the
selection of the minimum value of the other types of magnitude (like MS, MB, ML) has
been based on the correlation used in Tanner and Shepherd (1997) work; for example, in
the case of MS, the minimum adopted value is 2.9. For the other types of magnitude the

5 http://earthquake.usgs.gov/research/data/centennial.php
6 http://www.ncedc.org/anss/
7 http://www.ngdc.noaa.gov/
A36

minimum value is 4.0. In the following figure (Figure 3.3) we report the number of
selected events from used databases.

Database N° events with Mw ≥ 4

ISC 530

USGD-NEIC (PDE) 1200

USGD-NEIC (Centennial) 13

ANSS 619

NGDC 8

Total 2370

Figure 3.3 Number of selected events from used databases

3.1.2 Merging of databases catalogues


The databases catalogues were first placed in chronological order and then merged;
considerable time was spent on the problem of multiple solutions to the same events.
Each events of the merged catalogue was examined to determine whether or not it
represented multiple solutions to the same event. This control was performed in two
steps: during the first step, we identified multiple solutions, characterized by the same
values of data (year, month and day), hour and minute and the same location; during the
second step we created a hierarchy for the used databases to define the location estimate
of each events. The chosen hierarchy of databases is the following: ISC, USGS-NEIC
(PDE), ANSS, USGS-NEIC (Centennial) and NGDC. Thus, for example, when ISC and
USGS-NEIC(PDE) give different estimates of depth or coordinates for the same event,
we chose the one provided by ISC; this hierarchy is not followed when in the main
source there is unknown value of depth. In the created catalogue, the following different
types of magnitude appear:

¾ Local Magnitude, or Richter Magnitude ML


¾ Surface-wave Magnitude MS
¾ Body-wave Magnitude mb
¾ Duration Magnitud MD
A37

¾ Moment Magnitude MW
The macroseismic intensity, measured by the MMI scale (Modified Mercalli Intensity)
appear only for two events in the NGDC catalogue.

We clarify that while ISC, USGS-NEIC (PDE), ANSS and NGDC give a single estimate
of the same type of magnitude for each event USGS-NEIC (Centennial) reports different
magnitudes, depending on the source referenced. In the case of USGS-NEIC
(Centennial), in order to obtain a global catalog with many multiple dates, we have chosen
to refer to the average value of each type of magnitude. This choice was made after
calculating, for each event and for each type of magnitude M, the maximum difference
∆max and the absolute average absolute deviation δ through the following relations:

1
∆ max = max ( x i − x j ) δ= ∑ x −x (3.1)
n
where xi; xj are two estimates of the same M (when the number of M is ≥ 2, the
procedure was repeated); x is the mean value of the same M for each event. The
objective of these operations was to determine whether it should be taken into account
the average value or the maximum value of each type of magnitude. As it can be seen
from results reported in Figure 3.4, the dispersion between the estimated magnitude is
substantial especially in the case of mb, so we have chosen to consider the average values
of each estimate of magnitude.

MS MW mb
Standard Standard Standard
Average Average Average
Deviation of Deviation of Deviation of
∆max ∆max ∆max
∆max ∆max ∆max
0 0 0.04 0.05 0.23 0.05
δav δmax δav δmax δav δmax
0 0 0.03 0.071 0.16 0.21

Figure 3.4 Evaluation of dispersion between the different estimates of the same magnitude in the
USGS-NEIC (Centennial) catalogue; determination of average value and standard
deviation of the maximum differences ∆max, and average and maximum value of all
average absolute deviations δ for each event

In the operation of merging databases catalogue, we have been noted different estimates
of the same type of magnitude; we chose to keep this feature and for each event listed in
our catalogue we could have different estimates of the same type of magnitude.
A38

We report that for two events contained only in the NGDC catalogue there were not the
type of magnitude so we consulted CMT catalogue 8 in order to obtain magnitude
estimates for those events.

3.1.3 Evaluation of differences in magnitude estimates of the databases


sources
As we have previously pointed out, for each event listed in most catalogues, we could
have different estimates of the same type of magnitude. In order to evaluate the
differences between the various catalogues and to associate with each event an
uncertainty, for each event and for each type of magnitude we have calculated average
absolute deviation δ of the values of magnitude referred to their average value, according
to the report:

1
δ=
n
∑ x−x (3.2)

where x is the mean value of the same magnitude for each event.

Subsequently we examined all those events characterized by a deviation δ >0.2 and we


have checked if was determined by errors arising from manual integration or by real
differences between the catalogues. If the event was characterized by real discordant
estimates of the same magnitude ( δ >0.2), when there were several values of other types
of magnitude, the type of magnitude in which we have observed the significant deviation
has been rejected. If the event was characterized by a single type of magnitude estimated
from only two discordant sources, it was chosen to maintain greater magnitude, in favor
of safety.

The elaborations performed on the average absolute deviation δ values have helped to
have limited differences between the values of the same type of magnitude for each
event. This can be seen from Figure 3.5 and Figure 3.6, which respectively show the
correlation between the values of mb magnitude reported in the ISC catalogue and in the
USGS-NEIC (PDE) catalogue for all available events (254), and the correlation between
the mb values listed in the catalogues ISC and ANSS for available events (323). In both
cases the coefficient of determination R2 is very close to unity; these values indicate a
good correspondence between the data included in catalogues.

8 http://www.globalcmt.org
A39

Correlation mb fromISC and fromUSGS-PDE

6,0
y = 0,9583x + 0,3537
R2 = 0,8265
5,5

5,0

mb,PDE

4,5

4,0

3,5
3,5 4,0 4,5 5,0 5,5 6,0
mb,ISC

Figure 3.5 Correlation between the values of mb magnitude provided by the ISC catalogue and those
provided by the USGS-NEIC (PDE) catalogue; the correlation was obtained on 254
events

Correlation mb fromISC and fromANSS


6,0
y = 0,923x + 0,4959
R2 = 0,8044
5,5

5,0

mb, ANSS

4,5

4,0

3,5
3,5 4,0 4,5 5,0 5,5 6,0
mb, ISC

Figure 3.6 Correlation between the values of mb magnitude provided by the ISC catalogue and those
provided by the ANSS catalogue; the correlation was obtained on 323 events
A40

To confirm the stability and the minimal dispersion of the data, in Figure 3.7 the linear
correlations mb,ISC - mb,USGS-NEIC (PDE) and mb,ISC - mb,ANSS are compared: there is a
substantial coincidence of the two lines.

We focus the analysis only on mb magnitude because it is the type of magnitude with the
significantly higher number of events with different mb estimates.

Correlation mb from ISC, USGS-PDE and ANSS

6,0

USGS-PDE
5,5

ANSS
5,0

mb

4,5

4,0

3,5
3,5 4,0 4,5 5,0 5,5 6,0
mb, ISC

Figure 3.7 Regression lines obtained by the correlations mb,ISC - mb, USGS-NEIC (PDE) and mb,ISC -
mb,ANSS

Despite the contained variability of the values of magnitude, in the formulation of the
global catalogue we have been chosen to consider, for each event and for each type of
magnitude, the maximum value of magnitude provided by different sources.

3.1.4 Homogenization of different scales of magnitude


Despite the large number of collected events, only for 42 events the cited sources report
the Moment Magnitude MW value. Estimates of the other four types of magnitude are
more numerous: we have 20 events with magnitude MS, 716 events with magnitude mb,
12 events with magnitude ML and 574 events with magnitude MD.
A41

In order to obtain for each event the Moment Magnitude MW, which is necessary to apply
the attenuation laws, it was necessary to find an empirical correlation between the
estimates of the magnitude MS, ML, mb and MD and the estimation of MW.

The research of these correlations was based on the maximum values of mb and MS
estimated for each event; in the case of ML and MD, no pairs of magnitude ML-MW appear
in our catalogue and there is only one pair of magnitude MD-MW.

The research of the correlation between the Surface wave Magnitude MS values and those
of Moment Magnitude MW was made on the basis of 12 pairs of magnitude MS-MW to
which correspond events with MS values between 4.6 and 6.8. The limitation of the range
of application of the relation found (4.6 ≤ MS ≤ 6.8), however, is not derived from an
effective restriction of the empirical correlation, but depends only on the limited available
data. Figure 3.8 shows that it is possible to establish a linear relationship between MW and
MS expressed by the following equation:

M W = 0,6789M S + 2,2192
(3.3)
R 2 = 0,8215
The correlation developed here has been compared in Figure 3.8 with the linear
correlation proposed by Tanner and Shepherd (1997); they are substantially coincident.
Figure 5 shows also the bilinear correlation proposed by Scordilis (2006); this relation
does not fit well the data contained in the present case; on the other hand, the author
himself says that the correlation proposed by him can be considered valid only for large-
scale studies, while for regional studies is necessary to refer to relations that take account
of local magnitude, as was done in this work.

After these considerations, we decided to use our above correlation.


A42

Correlation Ms-Mw
7,5
Linear Correlation based on
y = 0,6789x + 2,2192 our data
7,0
R2 = 0,8215

6,5

6,0

Mw 5,5
Bilinear Correlation developed
byScordilis (2006)
5,0

4,5
Linear Correlation developed
byTanner and Shepherd (1997)
4,0

3,5
3,5 4,0 4,5 5,0 5,5 6,0 6,5 7,0 7,5

Ms

Figure 3.8 Correlation between MW and MS obtained from local data for earthquakes with magnitude
4.6 ≤ MS ≤ 6.8 (solid line). The dashed lines indicate the linear correlation proposed by
Tanner and Shepherd (1997) and the bilinear correlation proposed by Scordilis (2006)

Regarding the correlation between mb and MW, Figure 3.9 shows a roughly linear
distribution, although the data are more dispersed than in the previous case. The relation
mb - MW, derived from 30 pairs of magnitude mb - MW corresponding to events with 4.0
≤ mb ≤ 6.2, is the following:

M W = 1,1302mb - 0,3383
(3.4)
R 2 = 0,7967
The correlation developed here has been compare in Figure 3.9 with the linear correlation
proposed by Tanner and Shepherd (1997) and the bilinear correlation proposed by
Scordilis (2006).

The linear correlation proposed by Tanner and Shepherd (1997) and our correlation are
different; we chose to use our correlation for two reasons: because it fits better our
available data which regard the studied area (Eastern Caribbean) and because Tanner and
Shepherd (1997), who proposed their relations for Latin America and Caribbean, say in
their report that relationship fit is extremely poor and a wide range of constants could be
substituted in their linear regression without changing the quality of fit substantially.
A43

In the cases of MD and MW, it is not possible to determine correlation ML-MW and MD-
MW, so a bibliographic research was made.

Correlation mb-Mw

7,5
Linear Correlation
y = 1,1302x - 0,3383 based on our data
7,0 R2 = 0,7967

6,5

6,0

Mw 5,5
Bilinear Correlation developed
byScordilis (2006)
5,0

4,5
Linear Correlation developed
byTanner and Shepherd (1997)
4,0

3,5
3,5 4,0 4,5 5,0 5,5 6,0 6,5 7,0 7,5

mb

Figure 3.9 Correlation between MW and mb obtained from local data for earthquakes with magnitude
4.0 ≤ MS ≤ 6.2 (solid line). The dashed lines indicate the linear correlation proposed by
Tanner and Shepherd (1997) and the bilinear correlation proposed by Scordilis (2006)

In particular, for MW, Figure 3.10 shows the sources and the correlations analyzed, while
in Figure 3.11 you can observe a comparison between the correlations proposed by
various sources analyzed (dashed lines). The comparison shows that the relationships
proposed are very different because each of them has been obtained for a specific region.

This result is consistent with studies conducted by Uhrhammer and Collins (1990),
Uhrhammer et al. (1996), Papazachos et al. (1997) and Margaris and Papazachos (1999)
which showed that the ML estimates can be very different for a same event depending on
the characteristics of the Wood Anderson seismograph used in particular according to the
“effective magnification” that characterizes it (Anderson and Wood, 1924, 1925).

Figure 3.10 shows also the solution adopted by Tanner and Shepherd (1997), who
assumed ML=MW if ML is less than about 7, according to Kanamori (1978).
A44

CORRELATION SOURCE STUDIED ZONE


MW = 1.17 + 0.436ML + 0.059 ML2 Bollinger et al. (1993) Western United States
MW = 0.997ML - 0.050 Uhrhammer et al. (1996) California
Grünthal and Wahlström Northern and Northwestern
MW = 0.67+0.56ML+0.046ML2
(2003) Europe
Joshi G. and M. L. Sharma
MW = 0.962ML - 0.235 Area around Delhi
(2008)
MW = 0.97ML + 0.58 Papazachos et al. (1997) Greece
MW = ML + 0.57 Ristau et al. (2005) Western Canada
Wahlström and Grünthal
MW = 1.2+0.28ML+0.06ML2 Sweden, Finland and Denmark
(2000)
Figure 3.10 Bibliographic sources used to convert ML in MW

In our catalogue there are only 12 events with magnitude ML with maximum value of 4.4,
so we decided to assume ML=MW.

In the case of Duration Magnitude, from bibliographic research conduct we extract the
correlations developed by Chen and Tsai (2008) for Taiwan and those proposed by
Pasyanos et al. (1996) for Northern and Central California; they are shown in Figure
3.12, in which there are both expression developed by Chen and Tsai (2008), one for
analogical examined recording and one for digital examined recording, and both relations
proposed by Pasyanos et al. (1996) based on different kind of available data.
A45

Correlation ML-Mw

7.5

7.0

6.5

6.0

Mw 5.5

5.0

4.5

4.0

3.5
3.5 4 4.5 5 5.5 6 6.5 7 7.5
ML

Lineare (Tanner and Shepherd (1997))


Lineare (Bollinger et al. (1993))
Lineare (Uhrhammer et al. (1996))
Lineare (Grünthal and Wahlström(2003))
Lineare (Joshi G. and M. L. Sharma (2008))
Lineare (Papazachos et al. (1997) )
Lineare (Ristau et al. (2005) )
Lineare (Wahlströmand Grünthal (2000))

Figure 3.11 Correlation between MW and ML obtained from used bibliographic sources

Events with only MD values are 429, but MD assumes maximum value of 4.9; thus, we
decided to use Linear correlation in light green developed by Pasyanos et al. (1996): it
represents the intermediate line considering the range of our data (4.0 ≤ MD ≤ 4.9); the
expression is the following:

log M 0 = 1,51M D + 16,26

Where M0 is determined with Kanamori (1978) relation M W = 2 3 log M 0 − 10,7 .


A46

Correlation MD-Mw

7,5

7
Linear Correlations developed
by Chen and Tsai (2008)
6,5

Mw 5,5
Linear Correlations developed by
Pasyanos (1996)
5

4,5

3,5
3,5 4 4,5 5 5,5 6 6,5 7 7,5
MD

Figure 3.12 Correlation between MW and ML obtained from bibliographic sources used and the
unique pair of MW-ML of our catalogue.

3.1.5 Catalogue 1997-2009


The determination of MW is based on a hierarchy identified through the critical analysis of
the correlations (presented in the previous paragraph) reliability.

If the magnitude MW appeared directly in one of the sources, this value was primarily
used; otherwise, we chose MW obtained from our linear correlation MW - MS if MS value is
included in the range 4.6-6.8, in which our correlation has been calibrated. In alternative
MW is calculated using MW - mb correlation if mb value is included in the range 4.0-6.2; the
last choice is represented by other kind of magnitude, like ML and MD, whose correlations
with MW were extracted from bibliographic sources. This hierarchy is substantially
according to the one proposed by Tanner and Shepherd (1997). In conclusion the global
catalogue for the Eastern Caribbean cover the period between 1530-2009 and includes
more then 3000 events with MW greater than 4.0 (Figure 3.13). APPENDIX I contains all
events of the Catalogue 1530-2009 with MW ≥ 4.5, used in the following analyses.
A47

Figure 3.13 Events contained in our catalogue for the period 1530-2009 with Mw greater than 4.0
selected for the studied area
A48
4.SEISMOTECTONIC SETTING AND DEFINITION
OF SEISMIC SOURCES

An updated study on the seismotectonic setting of Caribbean region is of foremost


importance. It clarifies the origin of seismicity in the Caribbean which is mostly but not
exclusively due to subduction activity. The interplay and complexities between shallow
crustal, intraplate and subduction seismicity of Caribbean region is thoroughly
investigated together with their association with fault movements whenever this
information is available. This study constitutes the basis for the definition of seismic
sources required for the implementation of Cornell-McGuire approach when computing
the seismic hazard.

The main features of Caribbean region seismicity have been already presented in the
previous chapter: it appears evident as the the Caribbean plate is overriding the North
and South American plates on the east, while on the west, the Cocos plate is subducting
beneath the Caribbean plate.

Caribbean Plate is moving eastward respect to the adjacent North American and South
American Plates at a rate of approximately 20 mm per year (Gibbs, 2001).

Seismicity in the Caribbean is concentrated along the Caribbean/South American/North


American (and associated smaller) plate boundaries.

An in-depth investigation on the origin of seismicity in the Caribbean arc will be


presented in this chapter in order to discuss the definition of appropriate seismogenic
zones on which the standard Cornell-McGuire approach is based.

We divided our studied area in 15 seismogenic zones, shown in Figure 4.1; they are
delineated in detail in the following paragraphs. It is evident how some zones are
overlapped by other zones, so Figure 4.2 shows a 3D vision of particular sections in
order to clarify the general chosen scheme.
A50

Figure 4.1 Seismogenic zones in which our studied area, that encompasses Eastern Caribbean
Islands (brown), has been divided
A51

Figure 4.2 The 3D vision of some particular sections ot the seismogenic zones in which our studied
area, that encompasses Eastern Caribbean Islands (brown), has been divided

4.1 SEISMOGENIC SOURCES

4.1.1 Zone 1: Volcanic Island-arc


The upper-crustal seismicity concentrates within the upper 35 km of the Caribbean
continental plate in the Lesser Antilles Arc (Boynton et al. 1987), with epicenters plotting
from the island of Grenada to Anguilla within a nearly continuous belt of 100 km width
along both, the axis of the principal active volcanoes and the inland and offshore shallow
faults which run parallel to the Subduction Trench. Within this zone the magnitudes are
moderate, reaching a maximum value of about 6.6 through historic times. Zone 1
comprises an area from North of Martinique to Anguilla covering the Leeward Islands,
and we characterize it with a higher seismic activity than the Windward Islands at the
South part of the Lesser Antilles Arc (from Grenada to Saint Lucia). Volcanic disasters in
the region over the past 300 years suggested major explosive eruptions in the Soufriere in
St. Vincent (1718, 1812, 1902, 1979), Mt. Pelé in Martinique (1902), Soufriere in
A52

Guadeloupe (1976-1977) and from 1995 to present in Montserrat volcanoes (SRC,


2009a). However, the moderate shallow earthquakes do not necessarily occur in
conjunction with volcanic eruptions and frequently appears in clusters with no discernible
mainshock (swarms).

Bernard & Lambert (1988) suggested that the evaluation of seismic hazard must also take
into account these shallow-moderate earthquakes as the ones occurred on 1851 and 1897
in Guadeloupe (~5.5-6.0 Mw), March 16th 1985 (6.4 Mw) at South of Nevis, and the
event occurred on November 21st 2004 with a magnitude of 6.3 (Mw) in the North-West
of Dominique near the Les Saintes Islands. The fault plane solutions in this zone yield
both, normal and strike-slip focal mechanisms. We observe a marked lower level of
seismicity in the Windward Islands as compared with the other zones of the Eastern
Caribbean; our seismic catalogue lists only two upper-crustal events (depth < 20 km) in
the Windward Islands dated on September 8th 1972 (4.5 Mw) and May 19th 1990 (4.7
Mw) on Grenada and Saint Lucia respectively, confirming the quiescence characteristic of
the Lesser Antilles Arc South region.

4.1.2 Zone 2-5: Subduction in the Lesser-Antilles


The volcanic island-arc lays about 300 km from the Eastern Caribbean Trench, where the
North American plate begins to submerge underneath the Caribbean plate reaching
depths of 200 km below the islands generating earthquakes as large of magnitude 8.0 Mw.
We include in Zones 2 and 3 all the shallow focus earthquakes (depth ≤ 50 km) along the
inclined inter-face seismic zone that yields underthrust focal mechanisms (Byrne et al.
1988). Convergence between the Caribbean and North American plates occurs at a rate
of about 37 mm/yr (Sykes et al., 1982; McCann 1985). The focal mechanisms of deeper
intra-plate events (>50 km) indicate that there is a normal faulting resulting from initial
flexure of the down going Atlantic slab (Zone 4 and 5) with an average of westward
dipping angle of 50º (Bengoubou-Valeruis et al, 2008). Zone 2 and 4 cover the latitudes
from 14.8ºN to 20.0ºN and it is characterized with a higher seismic activity than Zone 3
and 5 (from 11.0ºN to 14.8ºN latitude). Bengoubou-Valeruis et al. (2008) and Russo et
al. (1993) attributes the differences of the seismic activity to the following reasons: a)
changes in the tectonic structures mapped by Feuillet et al. (2002), b) enough sediments
to lubricate or decouple the two plates in the Subduction Zone, c) strengthening caused
by thick accretionary prism overbudden which lies above the shallow reach of the
subduction zone; the quiescent area coincides with the deepest part of the Barbados
accretionary wedge. The upper-crustal seismic activity level observed along the volcanic-
island arc reflects also the differences observed in the seismic activity in the Subduction
Zone.
A53

The largest interface event listed in the catalogue dates on October 10th 1974 (7.3 Mw)
with its epicenter located between Antigua and Barbuda (Zone 2). The largest intra-plate
earthquakes within Zone 4 occurred on February 8th 1843 and on April 5th 1690, with a
magnitude of 8.0 and 7.5 (Mw), respectively, with both epicenters located to the west of
Antigua and Barbuda Islands. Other intraplate big events occurred in Zone 5 around
Martinique on January 11th 1839, on December 3rd 1906 and November 29th 2007 with
magnitudes 7.3-7.4 (Mw).

4.1.3 Zones 6-8: Puerto Rico and Virgin Islands


The Puerto Rico and the Virgin Islands region is considered as a microplate that is
surrounded by the obliquely subducting North America plate, the Caribbean plate and
several major faults as the Mona Canyon to the East and Abnegada Passage to the West
(McCann, 1985, Jansma et al. 2000) and the Muertos Trough to the South. Puerto Rico
accommodates approximately 16.9 mm/yr of deformation relative to North America and
2.4 mm/yr relative to the Carribbean Plate (Clinton et al., 2006). We divided the area in
three seismogenic sources as follows.

Zone 6

This zones includes the Puerto Rico Trench area within depth less than 50 km including
the megathurst faulting along the plate interface of the subducting North American Plate
southward deepening. Also this zone comprises the left lateral strike slip faulting that is
subparallel to the Puerto Rico trench North and North-West of Puerto Rico including the
Septentrional fault. On July 29th 1943 an earthquake ruptured the Puerto Trench with a
magnitude of 7.5 (Mw). This seismogenic zone covers the North of Puerto Rico and
Virgin Islands.

Zone 7

This zone comprises the shallow faults (less than 50 km depth) inland Puerto Rico and
offshore namely, Mona Canyon, South Lajas Fault, Great Northern and Southern Puerto
Rico fault zone, the Anegada Trough and Sombrero Seismic Zone (Clinton et al., 2006).
This seismogenic source has produced earthquakes of magnitude 7.5 and 7.3 (Mw) in the
Anegada and Mona Passage in 1867 and 1918 respectively, yielding normal faulting in a
broad zone of active crustal extension and accompanied by destructive tsunamis (Mueller
et al., 2003). The absence of volcanism in Puerto Rico and the Virgin Islands suggests
that this zone is not an extension of the island-arc Lesser Antilles structure (Molnar &
Sykes, 1969).
A54

Zone 8

This zone includes the intra-plate subduction seismicity generated by the bending of
North-American slab with depth greater than 50 km. Besides, recent research suggests
the existence of the subducted Caribbean slab confirmed by low velocity anomalies
beneath the Island (Mendoza & McCann, 2006). Then this seismogenic area comprises
the subduction intra-plates slabs of North America southward dipping and the Caribbean
northward dipping beneath the microplate. The largest earthquake listed in our catalogue
within this zone dates on March 24th 1916 (7.4 Mw).

The Muertos Trough offshore southern Puerto Rico constitutes the thrust-trench locus
convergence between the microplate and the Caribbean Plate northward deepening
(Mendoza & McCann, 2009). We exclude this seismic zone of our analysis since the rates
of activity in this zone are poorly known (Mueller et al. 2003) and it seems that based on
the knowledge of the seismic history, the motion along the Muertos Trough appears to
be a small fraction of the Puerto Rico Trench (McCann, 1985). We confirm this
suggestion with the epicentral maps elaborated for this work, the historical literature
consulted to compile our catalogue and the low slip rate suggested in this boundary.

4.1.4 Transition Zones 9 and 10A


We define these seismogenic zones as the intersections amongst the transform faults and
subduction zones with the Lesser Antilles Arc located at the North and the South of the
Eastern Caribbean. Zone 10 includes the shallow seismic activity in the South part of the
island of Tobago which we consider within the Caribbean-South American plate
boundary (Latchman 2009, Weber, 2009 and Burmester et al 1996). Russo et al. (1993)
suggest also that the northern boundary of the Eastern Caribbean-South America plate
may lie as far north as the southern end of the Grenada basin. Moderate but shallow
earthquakes occurred South of Tobago on 1982 (4.8 Mw) and 1997 (6.7 Mw) with right-
lateral strike slip and normal faulting mechanism, respectively (Morgan et al., 1988;
Latchman 2009). The transition Zone 9 is characterized by a low-seismicity level in the
boundary zone between the Lesser Antilles arc and the Puerto Rico Trench yielding
mainly normal focal mechanisms.

4.1.5 Zone 10B: East of Trinidad


Russo & Speed (1992) suggested that the earthquakes located in this zone are consistent
with the detachment and bending-flexure of the South American slab moving toward the
collision zone. The zone covers mainly normal faulting mechanism with ENE-WSW
striking planes and strike slip faults with an average depth of 45 km. The maximum
magnitude reported in the catalogue for this zone is 6.7 Mw (March 10th 1988).
A55

4.1.6 Zone 11: North of Paria Peninsula


This zone constitutes a subducting detached oceanic lithosphere with depth ranging from
50 to 300 km and represents one of the most active seismogenic sources in the Eastern
Caribbean (Russo et al. 1993; SRC, 2009b). The largest reported earthquakes occurred in
October 21th 1766 and January 10th 1888 with magnitude of 7.5 and 7.0 Mw
respectively. The focal mechanisms indicate that there is a normal faulting resulting from
the initial flexure of the down going slab with a steeply NW-dipping of 60º. However,
mixed-motion earthquakes with thrust and strike slip indicated bending of the subducting
slab at deeper depths.

4.1.7 Zone 12: Trinidad Faults


This zone includes the faults mapped in Trinidad namely, the Northern-Range and
Central Range, and Darien Ridge and Arima and Los Bajos Fault, characterized by
earthquake with depth less than 50 km (SRC, 2009b). Weber (2009) employed far and
near field geodesy and palaeoseismology to search fossil earthquakes on the Central
Range Fault, the principal active dextral strike-slip in Trinidad. He concluded that the
Central Range fault was locked and stored and released significant elastic motion in the
recent past, and that it could be locked today and might constitute a major seismic threat
in Trinidad. He suggests a slip rate of 12 mm/year and that several meters of more of
motion could be stored up in the fault. In the south, he suggests a slip rate of 6 mm/year
in Los Bajos Fault with a dextral motion. Regional tectonic-geological studies conclude
that el Pilar Fault might right-step into Central Trinidad (i.e. Flinch et al., 2000), however,
Weber (2001, 2009) affirms that the N68ºE oblique trending in the Central Range Fault is
not associated with el Pilar Fault 90º trending of pure wrenching.

The Northern Range and the Arima Fault comprises a complex fault system with lateral
strike-slip, thrust and normal faulting. On December 2nd and 3rd 2004 events with a
magnitude of 5.8 and 5.4 (Mw) occurred in the central north-east of Trinidad; fault plane
solutions suggest mainly a normal motion with a component of right-lateral strike slip.
The location of these earthquakes and the correspondent focal mechanisms coincides
with the Northern Range normal fault dipping southward mapped by Algar & Pindell
(1993) beneath the Caroni Swamp area.

4.1.8 Zones 13 and 14: El Pilar fault


These zones comprise the boundary between the Caribbean and the South American
plate. The events that have their origin in the fault are shallow - less that 50 km depth -
and they are characterized mainly by right lateral strike slip mechanism in the northern
coast of South America. The Caribbean Plate is moving about 20 mm/yr in an easterly
direction relative to South America (Perez et al. 2001). However, thrust focal mechanism
A56

also takes place in this region reflecting the oblique collision at crustal levels between the
Caribbean and the South American Plate. We observed a high level seismic output in
Zone 14 that extends from 63.5º W to 62.3ºW longitude covering the Araya-Paria
Isthmus, and a moderate seismicity level in Zone 15 that extends from 67.0º to 63.5º W
longitude covering the vicinity of Caracas to the Araya region. The maximum magnitude
listed in our catalogue occurred on October 4th 1957 (6.4 Mw) in Zone 14 and on
September 1st 1530 (8.0 Mw) in Zone 15.

4.1.9 Zone 15: South of Trinidad


Russo et al. (1993) defined this zone as a passive margin edge in the Foreland basin in
North of South America continent, covering events with strike slip, mixed thrust and
strike slip, and thrust mechanism around the Orinoco-Delta region in Venezuela, with an
average depth of 50 km and a maximum magnitude of 6.6 (Mw).

Table 4.1 summurises the main characteristics of the selected seismic zone, which are
depth mean value, type and dominant focal mechanism.

Table 4.1 Main characteristics of the seismogenic zones

Main Focal
Depth (km) Type
Mechanism
ZONE 1 19.1 Volcanic Normal and Strike-Slip
ZONE 2 29.6 Interface Thrust (Inverse)
ZONE 3 29.4 Interface Thrust (Inverse)
ZONE 4 86.0 Intraplate Normal
ZONE 5 97.9 Intraplate Normal
ZONE 6 32.3 Interface Thrust and Strike-Slip
ZONE 7 28.4 Shallow Normal
ZONE 8 74.5 Intraplate Normal
ZONE 9 24.4 Transition Normal and Strike-Slip
ZONE 10 43.9 Transition/Intraplate Normal and Strike-Slip
ZONE 11 99.5 Intraplate Normal
ZONE 12 32.5 Crustal Normal and Strike-Slip
ZONE 13 23.3 Crustal Strike slip and Thrust
ZONE 14 14.7 Crustal Strike slip and Thrust
ZONE 15 57.3 Crustal Strike slip and Thrust
A57

Figure 4.3 and Figure 4.4 illustrates the geometrical configuration for both, shallow and
deep zones covering the Subduction, upper-crustal volcanic island-arc, transform and
intra-plate faulting and transitions zones. This configuration comprises the Subduction
Trench to the East, and the deepest part of the Atlantic Plate to the West. The
geometrical delimitation for shallow seismicity in the arc includes the Islands, related
epicenters, and main geological structures such as volcanoes and seismic faults.
A58

Figure 4.3 Shallow seismogenic zones with the shallow events of the constructed Catalogue
A59

Figure 4.4 Deep seismogenic zones with the deep events of the constructed Catalogue
A60
5.PROBABILISTIC SEISMIC HAZARD ANALYSIS

A Probabilistic Seismic Hazard Analysis (PSHA) provides an estimate of the frequency of


exceeding specified levels of ground motion at a site by integrating the contributions of
earthquakes of all possible magnitudes and locations (or distances from the site where
hazard is being computed) in a consistent manner. This study utilized the capabilities of
the classical Cornell-McGuire approach (Cornell, 1968; McGuire, 1976) based on the
definition of seismogenic zones.
All the parameters used in this approach introduce epistemic uncertainty in hazard
analysis, namely, seismogenic zoning scenarios, procedures for estimating completeness
periods of the earthquake catalogue for different magnitude classes, ground motion
attenuation relationships and the maximum magnitude in computations. These
uncertainty are incorporated in the computation through an expedient tool, called the
logic-tree. The logic-tree permits a combination of various alternatives in different steps
of the hazard computations based on weights assigned to the alternatives.
After a brief outline of the method (Paragraph 5.1) the following paragraphs are
dedicated to the description of the data (catalogue, zonation, etc.) elaborated and used for
the seismic hazard assessment for the selected methods.

5.1 METHOD OF CALCULATION

5.1.1 Cornell-McGuire Approach


PSHA method was initially developed by Cornell (1968) and its computer form was
developed by McGuire (1976 and 1978). In the Cornell–McGuire approach, tectonic and
geological information are combined with a seismic catalogue to define regions within
which earthquakes occur at rates which are defined by simple recurrence relations and are
assumed to have the same probability of occurring at any location. The incorporation of
geological and tectonic information is particularly useful when the seismic data is poor
and in regions where return periods are long compared to the length of the earthquake
record. The result of this method is the reduction of a whole catalogue of earthquakes
into three parameters, namely, the activity rate, b-value of the Gutenberg–Richter
recurrence relationship and the assumed value of the maximum magnitude. These
parameters are combined with a ground motion attenuation relationship through
A62

probabilistic methods to define a single ground motion parameter (e.g. PGA, spectral
acceleration, etc.), which corresponds to a specified probability of exceedance. The
method is computationally efficient and is particularly useful when there is large
uncertainty regarding the locations of future earthquakes. Conversely, the method results
in the ‘smoothing’ of the hazard over the source areas, which may result in significant
spatial variations in hazard not being revealed.

5.2 PROCESSING OF THE EARTHQUAKE CATALOGUE


The composite catalogue constructed in Chapter 3 need to be processed in order to
remove dependent events based on magnitude–time–distance parameters appropriate for
characterising aftershock or foreshock earthquake sequences, by a process termed
declustering. Removal of dependent events ensures that the Poisson assumption, that the
recurrence of earthquakes is a memory-less process (inherent to the classical Cornell-
McGuire approach to PSHA), is not violated.

Secondly, historical earthquake records are usually more complete for larger earthquakes
than for smaller ones. Small earthquakes can go undetected for a variety of physical and
demographical reasons. Time windows in which the catalogue is complete have to be
defined.

Finally, The seismicity of each seismogenic zone is then quantified by the standard
Gutenberg-Richter recurrence relationship (Gutenberg and Richter, 1942), which
hypothesizes the existence of an exponential correlation between the mean annual rate of
exceedance of an earthquake of specified magnitude and the magnitude itself.

5.2.1 Declustering process


The Declustering process represents a problematic step in the catalogue processing,
specially for the studied case, because, first of all, the method we will use, that is the
algorithm developed by Gardner and Knopoff (1974), described in the following
paragraph, is calibrated using the data of California, which is characterized by different
seismo-tectonic setting and seismicity compared to the typical ones of the Eastern
Caribbean region; in addition, chosen seismic zonation is complex, with some zones that
overlap other zones; at last, ZONE 1 covers the volcanic arc, so is marked by these
particular features already discussed in the previous paragraph. For these reasons, we
decide to follow those steps:

¾ to apply the Gardner and Knopoff (1974) algorithm to all catalogue events;
A63

¾ to check specifically the most important events that will be lost because of the
decustering process;

¾ to make specific study for the ZONE 1.

We decide to consider only the events of the catalogue with Moment Magnitude greater
or equal to 4.5.

5.2.1.1 Dynamic window method


The declustering algorithm developed by Gardner and Knopoff (1974), who claimed that
the development of foreshock and aftershock events is dependent on the size of main
earthquake event, forms the basis of the dynamic windowing method. The temporal and
spatial window parameters would, in this case, be different, based on the main event’s
magnitude. All events satisfying the criteria, that is, falling within the spatial and temporal
window (mutually inclusive) for a given main shock, are eliminated from the catalogue.
The dynamic windowing method has been utilised in the current study based on the
parameters reported in Table 5.1 and showed in Figure 5. For the relationship between
Magnitude and logarithm of time a quadratic relationship have been used in the present
study, since fit well the data of Gardner and Knopoff (1974).

Table 5.1 Time and distance parameters according to the dynamic window declustering method
(Gardner and Knopoff, 1974)

Magnitude of main event Spatial window, L (km) Temporal window, T (days)


2.5 19.5 6
3.0 22.5 11
3.5 26.0 22
4.0 30.0 42
4.5 35.0 83
5.0 40.0 155
5.5 47.0 290
6.0 54.0 510
6.5 61.0 790
7.0 70.0 915
7.5 81.0 960
8.0 94.0 985
A64

2.5

1.5
Log (L)

y = 0.1238x + 0.983
1 2
R = 0.9997
0.5

0
2 4 6 8 10
M

4
3.5 y = -0.061x2 + 1.0741x - 1.616
3 R2 = 0.9929
2.5
Log (T)

2
1.5
y = 0.4336x - 0.1165
1
R2 = 0.9491
0.5
0
2 4 6 8
M

Figure 5.1 Regression of the time and distance data proposed by Gardner and Knopoff (1974)

5.2.1.2 Lost events for declustering process


Because of the decustering process, we lost some events; for the previously explained
reason, we checked specifically the most important lost events, that are the ones with
magnitude grater than 6.

Those events are shown in Figure 5.2: the labels are referred to each studied event.
A65

Figure 5.2 Lost events because of Gardner and Knopoff (1974) declustering method
A66

The comparison between the declustered catalogue and the original catalogue permitted
to indentify these events that should be reinsert in the catalogue by hand: they are
underlined in Table 5.2 with pink rectangules.

For example, the event 2 in ZONE 3 was eliminated in the declustering algorithm
because of the “near” (in time and in space) presence of an event that belog to another
zone, that is ZONE 4; for this case, we decide to reinsert the event 2.
A67

Table 5.2 Specific analyses for the important lost events because of the declustering algorithm
(Gardner and Knopoff, 1974); in yellow the line referred to the studied event; in gray the
line referred to what the algorithm consider the main event.

LAT LONG DEP MW YEAR MONTH DAY HOUR MIN ZONE


14,5 -61 100 7,8 1839 1 11 9 50 5
1 14,5 -60,5 75 6,6 1839 8 2 6 25 5

2 14,5 -60,5 50 7,0 1906 2 16 17 25 3


15 -61 100 7,5 1906 12 3 22 59 4

3 19 -67 0 6,8 1915 10 11 19 33 6


10,89 -66,8 25 5,8 1915 12 13 2 30 15
18,5 -68 80 7,4 1916 4 24 4 26 8
4 19 -67,5 50 7,0 1917 7 27 1 1 6
11 -62 0 6,4 1918 2 24 23 0 13
19 -62,5 0 6,1 1918 6 11 12 36 2
5 18,5 -67,5 0 7,3 1918 10 11 14 14 7
19,5 -64,5 0 6,4 1919 9 6 9 29 6
13,5 -58 0 6,1 1919 11 6 7 13 3
6 18 -67,5 0 6,6 1920 2 10 22 7 7

7 16,5 -62,5 10 6,2 1935 5 6 19 55 1


19 -65 0 6,0 1935 9 15 4 1 6
16,5 -62,5 100 6,4 1935 11 10 18 27 4

19,25 -67,5 0 7,5 1943 7 29 3 2 6


8 19,25 -67,75 0 6,6 1943 7 30 1 2 6

10,85 -62,77 6 6,7 1957 10 4 5 26 14


10,88 -62,68 10 5,6 1957 10 6 0 54 14
11,02 -62,97 0 4,9 1957 10 14 8 17 14
18,8 -64,33 33 5,0 1957 10 23 4 38 7
10,86 -63,11 0 5,0 1957 12 18 2 11 14
9 10,46 -62,54 22 6,1 1957 12 25 16 26 14

17,42 -61,18 42 6,7 1967 12 24 20 3 2


10 17,61 -61,25 5 6,3 1967 12 24 21 32 2

15,78 -59,63 1 7,0 1969 12 25 21 32 2


17 -61,59 71 4,1 1969 12 25 21 56 4
15,77 -59,63 42 4,6 1969 12 25 22 13 2
16,23 -59,7 32 4,5 1969 12 25 22 16 2
16,19 -59,72 12 4,9 1969 12 25 22 17 2
15,84 -59,66 4 5,4 1969 12 25 22 26 2
11 16,07 -59,79 6 6,1 1969 12 25 22 31 2

12 11,47 -61,18 73,4 6,1 1997 4 2 6 14 10


11,35 -60,83 70,4 4,5 1997 4 2 7 21 10
19,07 -63,09 33 4,5 1997 4 5 21 5 2
11,35 -60,79 10 4,4 1997 4 5 7 6 10
11,1 -61 55,79 5,0 1997 4 5 17 4 10
11,31 -60,9 6,59 4,3 1997 4 5 17 8 10
10,32 -62,22 51,09 4,6 1997 4 7 23 25 12
11 -60,68 7,69 4,6 1997 4 8 9 37 10
11,06 -60,61 10 4,3 1997 4 8 14 13 10
11,06 -61,02 45,2 5,5 1997 4 8 17 11 10
10,97 -60,61 6,59 4,2 1997 4 8 18 41 10
11,02 -60,7 16,79 4,3 1997 4 8 23 46 10
11,06 -61,06 46,29 5,0 1997 4 9 0 50 10
11,14 -60,91 1,39 5,1 1997 4 10 2 23 10
11,02 -60,79 14,1 5,0 1997 4 11 6 57 10
10,94 -60,72 10 4,2 1997 4 12 3 5 10
11,01 -60,66 9,89 4,5 1997 4 16 11 42 10
10,72 -62,45 86 4,2 1997 4 18 4 3 12
11,15 -61,09 47,09 6,7 1997 4 22 9 31 10
11,06 -61,09 67,09 4,7 1997 4 22 9 49 10
13 11,11 -60,99 50,09 6,1 1997 4 22 10 11 10
A68

5.2.1.3 ZONE 1 events


ZONE 1 covers the volcanic arc region, so the events, which happen here, have a
particular behaviour, that hinders the identification of the main events.

So we computed the Gutenberg and Richter relations, with a preliminary evaluation of


the completeness period using the Visual Cumulative method (described in the following
paragraph) for all the events contained in the catalogue, considering different cases:

¾ all the events in ZONE 1 (DECLUSTERING 0);

¾ only the picks in each swarm (DECLUSTERING 1);

¾ only the peaks with magnitude grater than 5 in each swarm (DECLUSTERING
2);

¾ the declustered ZONE 1 catalogue using Gardner and Knopoff (1974) method
(DECLUSTERING 3).

The comparison of the obtained results is shown in Figure 5.3.

0,500
DECLUSTERING 0
0,000
DECLUSTERING 1
-0,500
DECLUSTERING 2
-1,000
DECLUSTERING 3
log(λMw)

-1,500
y = -1,0807x + 5,2847
-2,000
y = -0,9483x + 4,4466
-2,500
-3,000 y = -0,9952x + 4,7502

-3,500 y = -0,6561x + 2,4279


-4,000
4,50 5,00 5,50 6,00 6,50 7,00 7,50
MW

Figure 5.3 Specific study for ZONE 1

The Gardner and Knopoff method results is very different from the others: it probably
conducts to underestimate the hazard. The DECLUSTERING 1 and 2 are linked to the
A69

precision in the magnitude determination that is usually considered equal ± 0.2 (for
moment magnitude). The most conservative results are provided by the solution which
keeps all events of ZONE 1, so we decide that all events of ZONE 1 will be put in the
declusterd catalogue.

5.2.1.4 Conclusions
The declustered composite catalogue now includes 770 events. Figure 5.4 shows the
seismicity of the region before and after the declustering.

Figure 5.4 Maps showing the seismicity of the region (a) before declustering (b) after declustering

We notice that applying the events in ZONE 10 B after the declustering becomes 4, a
number that is not sufficient to compute the following analyses.

So we decide to compute the Gutenberg and Richter relations with a preliminar


evaluation of the completeness period considering all events (contained in the original
catalogue) of ZONE 10 B and to compare the obtained results with the ones which come
out from the same analysis done for ZONE 10 A and done the ZONE 10 A united with
ZONE 10 B.
A70

The comparison of the obtained results is shown in Figure 5.5.

1,000
ZONE10 A-10 B
0,500
0,000 ZONE 10 A

-0,500 ZONE 10 B
-1,000
log(ΗMw)

-1,500 y = -1,0042x + 5,1428


-2,000
y = -1,0522x + 5,2203
-2,500
-3,000 y = -0,9457x + 4,337
-3,500
-4,000
4,50 5,00 5,50 6,00 6,50 7,00 7,50
MW

Figure 5.5 Specific study for ZONE 10 A and ZONE 10 B

The Gutenberg and Richter parameters are very similar for the different analysed cases so
we decide to consider ZONE 10 A and ZONE 10 B like a unique zone in the hazard
computation.

5.2.2 Estimation of completeness periods


A second important step in processing an earthquake catalogue to make it suitable for a
probabilistic seismic hazard analysis is the definition of the time windows in which the
catalogue is presumed to be complete. Catalogue incompleteness exists because, for
historical earthquakes the recorded seismicity differs from the ‘true’ seismicity.
Completeness analyses have been performed separately for all events of the Declustered
Catalogue and for each seismogenic source zone because of the marked spatial
heterogeneity (Albarello et al., 2002) in the compilation of the earthquake catalogues,
particularly for the most ancient events.

5.2.2.1 The Visual Cumulative method (VC)


The Visual Cumulative method (Tinti and Mulargia, 1985) is a simple graphical procedure
based on the observation that if earthquakes of a given magnitude are assumed to follow
a stationary occurrence process, in a complete earthquake catalogue, the average rate of
A71

occurrence of seismic events must be a constant. The procedure involves developing


plots of the cumulative number of events versus the time from the beginning of the
catalogue for different earthquake magnitude classes. The period of completeness for a
given class is considered to begin from the earliest time when the slope of the fitting
curve can be approximated by a straight line. Figure 5.6 shows an example of the
application of this method regarding all zones.

M=4.75- All data M=5.25- All data

350

200
300

250
150
N° of events
N° of events

200

100
150

100
50 1950
1960
50

0 0
1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 1950 1960 1970 1980 1990 2000
Years Years
M=5.75- All data M=6.25- All data

50
60
45

40 50

35

40
30
N° of events
N° of events

25
30

20

20 1910
15
1950
10
10
5

0 0
1860 1880 1900 1920 1940 1960 1980 2000 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000
Years Years
A72

M=6.75- All data M=7.25- All data

30 12

25 10

20 8

N° of events
N° of events

15 6

10 4

1880
1810
5 2

0 0
1750 1800 1850 1900 1950 2000 1900 1920 1940 1960 1980 2000
Years Years

M=7.75- All data M=8.25- All data


2

1.8
6

1.6

5
1.4

1.2
4
N° of events

N° of events

1
3
0.8
1530

2 0.6

0.4
1
0.2

0 1690 0
1700 1750 1800 1850 1900 1950 1550 1600 1650 1700 1750 1800 1850 1900 1950
Years Years

Figure 5.6 Estimation of completeness period with Visual Cumulative Method considering all
seismogenic zones

The obtained results are reported in Table 5.3.


A73

Table 5.3 Estimation of completeness period with Visual Cumulative Method

MW
4.75 5.25 5.75 6.25 6.75 7.25 7.75 8.25
Zone
1 1960 1949 1849 1879 1897 - - -
2 1959 1949 1959 1919 1889 1969 - -
3 1960 1980 1980 1910 1810 1900 - -
4 1970 1980 1970 1930 1830 1910 1690 1840
5 1958 1958 1908 1818 1718 1908 1838 -
6 1950 1980 1980 1870 1920 1910 1940 -
7 1963 1943 1963 1853 1843 1913 1863 -
8 1958 1948 1958 1938 - 1908 - -
9 1960 1990 1990 1930 - - - -
10 1970 1990 1980 - 1980 - - -
11 1969 1959 1969 1829 1819 1879 1759 -
12 1960 1950 1990 1910 - - - -
13 1959 1959 1989 1869 1919 - - -
14 1959 1949 1909 1809 1929 1989 1899 1529
15 1967 1957 1967 1937 1937 - - -
All 1960 1950 1950 1910 1810 1880 1690 1530

5.2.2.2 Stepp’s method


To overcome or at least to minimize the effect of incompleteness, Stepp (1973)
introduced an empirical and statistically simple method based on the stability of the
magnitude recurrence rate. According to Stepp (1973), the entire catalogue of earthquakes
is grouped into magnitude range, say ∆M = 0.5 unit, and in time intervals of 10 years. The
average number of events per year, R(M), are then evaluated for each magnitude class for
increasing time interval lengths, starting with the most recent time interval. The first
window consists of the most recent, e.g. 10 years, the next window would consist of the
recent 20 years, and so on. An analysis of the series of R(M) obtained as above will show
the length of the time window for which R(M) becomes stationary for a given magnitude
range. This interval must be long enough to establish a stable mean rate of occurrence
and short enough so as not to include intervals in which the data are falling in
incompletely period.
A74

The procedure involves determining those fractions of the total time sample in which the
mean rate of occurrence, R(M), is stable for each magnitude class. This subinterval then
represents the minimum period in which it is assumed that there is complete reporting.
For this purpose, Stepp modelled R(M) as a Poisson point process in time, such that, for
a time interval of T years, the variance of R(M) is given by the following equation,

R( M ) (5.1)
S R2 =
T
where SR is the standard deviation of the mean rate, R(M). Assuming stationary, the
statistical properties of R do not change with time, SR will behave as 1 T . The plot of
the standard deviation as a function of T, known as the completeness plot shows the
expected behaviour of SR (refer Figure 5.7). For a particular magnitude class, the period of
completeness is reflected in this figure by a distinct departure of the SR value from the
linearity of the 1 T slope. This period which should be a minimum for a stable R(M)
becomes successively longer with each higher magnitude class.

1
10

0
10
/ T 0.5

-1
10
0.5
σλ = λ

-2
10

-3
10
1 2
10 10
Time (year)

Figure 5.7 Determination of completeness period using Stepp’s method considering all seismogenic
zones

The obtained results are reported in Table 5.4.


A75

Table 5.4 Estimation of completeness period with Stepp’s method

MW
4.75 5.25 5.75 6.25 6.75 7.25 7.75 8.25
Zone
1 1979 1950 1950 1910 1889 - - -
2 1960 1950 1950 1919 1889 1810 - -
3 1960 1960 1960 1910 1900 1900 - -
4 1970 1970 1970 1930 1830 1830 1690 1530
5 1960 1958 1950 1910 1830 1830 1690 -
6 1960 1960 1960 1910 1910 1910 1690 -
7 1963 1963 1963 1910 1843 1843 1690 -
8 1960 1950 1950 1938 1938 1908 - -
9 1960 1960 1960 1930 - - - -
10 1970 1970 1970 1970 1970 - - -
11 1969 1959 1959 1910 1819 1819 1690 -
12 1960 1950 1950 1910 - - - -
13 1960 1960 1960 1910 1910 - - -
14 1960 1950 1950 1910 1810 1810 1690 1530
15 1967 1967 1967 1937 1937 - - -
All 1960 1950 1950 1910 1810 1810 1690 1530

5.2.2.3 Considerations
Comparing the values of Table 5.5 and Table 5.4, it comes out that the evaluation of the
completeness period using the two selected methods gives not so different results, except
for the particular case of ZONE 1.

¾ the developed analyses underline the presence of some anomalies, that is, for
example, in some cases we notice that the completeness period referred to one
zone is grater then the one referred to all data. Basing on those evidences, we
decided to modify the estimates of the completeness periods values using the
following criteria:

¾ for each class of magnitude, the completeness period referred to one zone can
not be grater then the completeness period referred to all data;
A76

¾ for each zone and for the case in which we consider all data, the completeness
period referred to a class of magnitude has not to be greater then the
completeness period of the class characterized by bigger magnitude extreme
values;

¾ for 7.75 and 8.25 magnitude classes, we considered the same completeness
period values estimated for the case which considers all data;

The obtained results are reported in Table 5.5.

Table 5.5 Final estimation of completeness period

MW
4.75 5.25 5.75 6.25 6.75 7.25 7.75 8.25
Zone
1 1960 1950 1950 1910 1897 - - -
2 1960 1950 1950 1919 1889 1810 - -
3 1960 1960 1960 1910 1900 1900 - -
4 1970 1970 1970 1930 1830 1830 1690 1530
5 1960 1958 1950 1910 1830 1830 1690 -
6 1960 1960 1960 1910 1810 1810 1690 -
7 1963 1963 1963 1910 1843 1843 1690 -
8 1960 1950 1950 1938 1938 1908 - -
9 1960 1960 1960 1930 - - - -
10 1970 1970 1970 1970 1970 - - -
11 1969 1959 1959 1910 1819 1819 1690 -
12 1960 1950 1950 1910 - - - -
13 1960 1960 1960 1910 1910 - - -
14 1960 1950 1950 1910 1810 1810 1690 1530
15 1967 1967 1967 1937 1937 - - -
All 1960 1950 1950 1910 1810 1810 1690 1530

5.2.3 Gutenberg-Richter recurrence relationships


The seismicity of each seismogenic zone is then quantified by the standard Gutenberg-
Richter recurrence relationship (Gutenberg and Richter, 1942), which hypothesizes the
existence of an exponential correlation between the mean annual rate of exceedance of an
earthquake of specified magnitude and the magnitude itself. This law of earthquake
A77

occurrence is a simple mathematical statement that larger events are less frequent than
weaker events and that the difference in relative terms follows an exponential law.
Regional seismicity of a seismogenic zone is described by the parameters ‘a’ and ‘b’ of the
Gutenberg-Richter relationship. For any single source zone, the mean annual rate of
exceedance (λM) of earthquakes with magnitude greater than or equal to M satisfies the
Gutenberg-Richter’s relationship, defined by:
log(λ M ) = a − bM (5.2)
where, a and b are constants, which are often estimated from past seismicity data. The
yearly number of earthquake magnitudes greater than or equal to zero is 10a. A value of b
lower than one indicates that the zone is characterized by the occurrence of a relatively
large number of strong earthquakes, whereas a b value greater than one denotes a
situation where the number of large events is relatively small compared to those of
smaller magnitudes. Once this relationship is defined for a specific zone, the number of
earthquakes (λM) within different magnitude ranges expected in the source zone can be
obtained. Gutneberg-Richter relationship can be rewritten as:
α − β ⋅M
λM = e (5.3)

where α = 2.303⋅a e β = 2.303⋅b. The reciprocal of of the annual mean rate of


exceedance (λM) for a specific value of the magnitude MW is the return period.

The standard Gutenberg-Richter relationship covers an infinite range of magnitude (from


-∞ to +∞). For engineering purposes, the effects of very small earthquakes are of little
interest and at the other end of the magnitude scale, the standard Gutenberg-Richter
relationship predicts nonzero mean rates of exceedance for magnitude up to infinity. For
the previous reasons a bounded Gutenberg-Richter have been used in this study, which
can be expressed as (Kramer, 1996):

sup inf
e − β ⋅( MW − MW ) − e − β ⋅( MW
inf
− MW )
λM = ν ⋅ sup
− β ⋅( MW − MWinf )
with MWinf ≤ MW ≤ MWsup (5.4)
1− e
where ν = exp(α − β ⋅ M Winf ) , α = 2.303⋅a and β = 2.303⋅b. M Winf e M Wsup are respectively the
lower and upper bound of magnitude MW considered, while the parameters “a” e “b” are
the same of the sandard Gutenberg-Richter relationship.

Recurrence relationship have been computed after dividing the earthquake magnitude in
different magnitude range of amplitude equal to 0.5. The Gutenberg-Richter recurrence
relationships, based on the results of the completeness period evaluation, have been
A78

estimated. In the present study the value of the maximum magnitude have been
considered is the maximum historical magnitude increased of 0.5.
For the particular case of ZONE 1, we decide to compute the recurrence relationship
also dividing the earthquake magnitude in magnitude range of amplitude equal to 0.25
(assuming valid the evaluation of completeness period done for the range of amplitude
equal to 0.5 because of the low number of available data) and to assume the value which
increases the maximum historical magnitude equal to 0.3. The comparison on the results
obtained with the two different range of magnitude are shown in Table 5.6 (where EI is
“earthquake interval”, that is the inverse value of N in the formula of G-R for a certain
magnitude).

Table 5.6 Gutenberg-Richter parameters for ZONE 1

zone M range a b a @ Mmin λ @ Mmin α β ν Mmin Mmax1 Mmax2


1 0.5 5.163 -1.059 0.398 2.502 11.889 2.438 1.360 4.5 6.6 6.9
1 0.25 4.794 -1.012 0.240 1.736 11.039 2.331 1.297 4.5 6.6 6.9

We decided to consider magnitude range of amplitude equal to 0.25: the range equal to
0.5 could mask some particular features, typical of this zone.

The parameters of the frequency-magnitude recurrence relationship for all the


seismogenic zones are in Table 5.7, while Figure 5.8 shows the Gutenberg-Richter
relationships for all the seismogenic zones.
zone a b a @ Mmin ◊ @ Mmin ◊ ◊ ◊ Mmin Mmax1 Mmax2 EI (Mmin) EI (Mmax1) EI (Mmax2 )
1 4.794 -1.012 0.240 1.736 11.039 2.331 1.297 4.5 6.6 6.9 0.6 76.9 154.7
2 3.651 -0.755 0.254 1.795 8.408 1.738 1.162 4.5 7.3 7.8 0.6 72.4 172.7
3 3.216 -0.725 -0.046 0.899 7.405 1.669 0.592 4.5 7 7.5 1.1 72.2 166.3
4 4.164 -0.821 0.470 2.949 9.587 1.890 1.838 4.5 8 8.5 0.3 253.3 651.7
5 2.941 -0.680 -0.121 0.757 6.772 1.567 0.512 4.5 7.8 8.3 1.3 232.3 508.5
6 4.724 -0.941 0.488 3.076 10.877 2.168 1.789 4.5 7.5 8 0.3 216.8 640.9
7 3.043 -0.705 -0.131 0.739 7.008 1.624 0.493 4.5 7.5 8 1.4 176.8 398.4
8 3.640 -0.810 -0.003 0.993 8.382 1.864 0.623 4.5 7.4 7.9 1.0 224.5 570.2
9 2.961 -0.727 -0.311 0.489 6.819 1.674 0.322 4.5 6.4 6.9 2.0 49.2 113.7
10 2.127 -0.531 -0.261 0.548 4.898 1.222 0.404 4.5 6.7 7.2 1.8 26.8 49.4
11 3.643 -0.783 0.120 1.317 8.388 1.803 0.839 4.5 7.8 8.3 0.8 291.1 716.9
12 2.580 -0.664 -0.407 0.391 5.940 1.528 0.267 4.5 6.4 6.9 2.6 46.6 100.1
13 3.392 -0.747 0.031 1.074 7.810 1.720 0.699 4.5 6.7 7.2 0.9 40.9 96.7
14 2.567 -0.635 -0.289 0.514 5.910 1.461 0.356 4.5 8 8.5 1.9 324.2 673.2
Table 5.7 Gutenberg-Richter parameters for all the seismogenic zones

15 2.825 -0.699 -0.321 0.478 6.505 1.610 0.320 4.5 6.6 7.1 2.1 61.5 137.5
A79
A80

ZONE 1 ZONE 2

0.500 0.500
0.000 y = -1.0122x + 4.7943 0.000 y = -0.755x + 3.6514
2
-0.500 R = 0.9882 -0.500 R2 = 0.9677
-1.000 -1.000
log(λMw)

log(λMw)
-1.500 -1.500
-2.000 -2.000
-2.500 -2.500
-3.000 -3.000
-3.500 -3.500
-4.000 -4.000
4.5 5.0 5.5 6.0 6.5 7.0 7.5 4.5 5.0 5.5 6.0 6.5 7.0 7.5
MW MW

ZONE 3 ZONE 4

0.000 0.500
0.000 y = -0.8209x + 4.1637
-0.500
y = -0.7249x + 3.216 R2 = 0.9929
-1.000 2
-0.500
R = 0.9911
-1.000
-1.500
log(λMw)

log(λMw)

-1.500
-2.000
-2.000
-2.500
-2.500
-3.000 -3.000
-3.500 -3.500
-4.000 -4.000
4.5 5.0 5.5 6.0 6.5 7.0 7.5 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5
MW MW

ZONE 5 ZONE 6

0.000 0.500
-0.500 y = -0.6804x + 2.9409 0.000
-0.500
-1.000 R2 = 0.9511
-1.000 y = -0.9413x + 4.724
-1.500
log(λMw)
log(λMw)

-1.500 2
R = 0.9826
-2.000
-2.000
-2.500 -2.500
-3.000 -3.000
-3.500 -3.500
-4.000 -4.000
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5
MW MW
A81

ZONE 7 ZONE 8

0.000 0.000
-0.500 -0.500 y = -0.8096x + 3.6402
-1.000 y = -0.7055x + 3.0434 -1.000 R2 = 0.9462
-1.500 R2 = 0.9937 -1.500
log(λMw)

log(λMw)
-2.000 -2.000
-2.500 -2.500
-3.000 -3.000
-3.500 -3.500
-4.000 -4.000
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 4.5 5.0 5.5 6.0 6.5 7.0 7.5
MW MW

ZONE 9 ZONE 10

0.000 0.000
y = -0.5306x + 2.127
-0.500 y = -0.7271x + 2.9615 -0.500
R2 = 0.8929
-1.000 2
R = 0.9948 -1.000
-1.500 -1.500
log(λMw)
log(λMw)

-2.000 -2.000
-2.500 -2.500
-3.000 -3.000
-3.500 -3.500
-4.000 -4.000
4.5 5.0 5.5 6.0 6.5 7.0 7.5 4.5 5.0 5.5 6.0 6.5 7.0 7.5
MW MW

ZONE 11 ZONE 12

0.000 0.000
y = -0.7829x + 3.6428 y = -0.6638x + 2.5798
-0.500 2
-0.500
R = 0.9863 R2 = 0.9801
-1.000 -1.000
-1.500 -1.500
log(λMw)
log(λMw)

-2.000 -2.000
-2.500 -2.500
-3.000 -3.000
-3.500 -3.500
-4.000 -4.000
4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 4.5 5.0 5.5 6.0 6.5 7.0 7.5
MW MW
A82

ZONE 13 ZONE 14

0.000 0.000
-0.500 y = -0.7468x + 3.3918 -0.500 y = -0.6347x + 2.5668
-1.000 R2 = 0.9695 -1.000 R2 = 0.9744
-1.500 -1.500

log(λMw)
log(λMw)

-2.000 -2.000
-2.500 -2.500
-3.000 -3.000
-3.500 -3.500
-4.000 -4.000
4.5 5.0 5.5 6.0 6.5 7.0 7.5 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5
MW MW

ZONE 15

0.000
-0.500 y = -0.6991x + 2.8253
-1.000 R2 = 0.987

-1.500
log(° Mw)

-2.000
-2.500
-3.000
-3.500
-4.000
4.5 5.0 5.5 6.0 6.5 7.0 7.5
MW

Figure 5.8 Bounded Gutenberg-Richter frequency-magnitude recurrence relationship for all the
seismic zone

5.3 ADOPTED GROUND MOTION PREDICTION EQUATIONS


Ground Motion Predicted Equations (GMPEs) are a main component for seismic hazard
analyses and they account directly to the aleatory uncertainty.

GMPEs are applied to estimate the ground motion parameters (i.e. peak ground
acceleration, peak ground velocity, peak ground displacement, spectral ordinates, Arias
intensity, macroseismic intensity, etc.) as a function of independent parameters
characterising the earthquake and the site under study, such as magnitude, source-to-site
distance, earthquake type, faulting mechanism, focal depth, local site conditions.
A83

GMPEs are regionally dependent and the choice of an appropriate attenuation


relationship depends on the tectonic setting of the site of interest (e.g. stable continental
region, active tectonic region, subduction zone, etc.). Unfortunately, not even one GMPE
has been developed for Caribbean islands, therefore relations calibrated for other regions
with similar seismic activity were considered.

To account for uncertainty in strong motion prediction, different attenuation


relationships were used in the PSHA, selected on the basis of the following criteria:

¾ model developed for shallow crustal and/or subduction earthquake areas;

¾ attenuation model based on adequate documentation;

¾ period range of the model appropriate for engineering applications;

¾ model well recognized and widely used in similar seismic environment;

¾ adequate range of data in terms of magnitude and distance used to develop the
attenuation relationships.

After a careful examination, various GMPEs were selected to characterize each of the
identified seismic zones described in Section 4.1, depending on the dominant typology of
earthquakes (subduction, crustal or volcanic) and prevalent focal mechanism (see Table
4.1):

• SUBDUCTION ZONES:
− Youngs et al. (1997);
− Atkinson and Boore (2003-2008);
− Zhao et al (2006);
− Kanno et al. (2006);
− Lin and Lee (2008).
All these GMPEs are defined for both interface and inslab events and they take into
account for the focal depth. A summary of the main features of these attenuation
equations are listed in Table 5.8. Douglas and Mohais (2008) shown that Kanno et
al. (2006) and Zhao et al. (2006) relations provide quite good predictions of
observed earthquake ground motions and their variabilities in the Lesser Antilles.
This observation is confirmed also in the present study, as it is shown in the
A84

following through the comparison between estimates of GMPEs and earthquake


recordings.

• CRUSTAL ZONES:
− Kanno et al. (2006)
− Zhao et al (2006);
− Abrahamson and Silva (2008);
− Boore and Atkinson (2008);
− Campbel and Bozogornia (2008);
− Chiou and Youngs (2008).
Except for the relations of Kanno et al. (2006) and Zhao et al. (2006), all the
GMPEs adopted for crustal zones are NGA relations developed in the framework
of Next Generation Attenuation Project (Stewart et al., 2008). These relationships
account for magnitude and distance contributions, focal mechanism, site
amplification, sediments effects, hanging wall and non linear soil behaviour. In the
present work rock response only was analysed, while site and sediments effects,
hanging wall and non linear contributions were neglected.

• VOLCANIC ZONE:
− Sadigh et al. (1997);
− Salazar (2004);
− Zhao et al (2006);
− Kanno et al. (2006);
− Mcwerry et al. (2006);
− Abrahamson and Silva (2008);
− Chiou and Youngs (2008)
Unfortunately few attenuation relationships were specifically developed for volcanic
areas. Mcwerry et al. (2006) proposed for New Zealand an attenuation model which
takes into account the faster attenuation of high frequency ground motions in
volcanic regions than elsewhere, while a GMPE for upper-crustal earthquakes in the
volcanic region of El Salvador was developed by Salazar (2004).
A85

In the following paragraph the accurancy of prediction of the previous listed GMPEs is
tested by the comparison with available recordings.
Table 5.8 Summary of attenuation relationships used in the current study for subduction zones

MAGNITUDE
DISTANCE DEFINITION AND HYPOCENTRE DEPTH AND
ATTENUATION EQUATION APPLICATION DEFINITION AND DATABASE
LIMITATIONS LIMITATIONS
LIMITATIONS
Closest distance to the fault
PGA and PSA Developed from a database including
trace for earthquakes with
for 0.075 to 3 s. Depth ≤ 50 km for interface events worldwide earthquakes in subduction
Youngs et al.. (1997) Mw ≥ 5 available fault models, otherwise
Horizontal hypocentral distance Depth > 50 km for inslab events zones: Alaska, Chile, Cascadia, Japan,
component Mexico, Peru, Solomon Islands
10 ≤ Rrup ≤ 500 km.
Mw ≥ 5 Developed from a database including
PGA and PSA Depth < 50 km for interface events thousands of strong motion recordings
Mw =8.5 for Closest distance to the fault
for 0.33 to 25 s. from events occurring in subduction
Atkinson and Boore (2003- interface events trace Depth ≥ 50 km for inslab events
zones around the world, based on both
2008) Horizontal with Mw > 8.5 Depth=100 km for events with
Rrup ≤ 400 km. Cruise (1991) and Youngs et al. (1997)
component Mw =8 for interface Depth>100 km catalogues and added events: Cascadia,
events with Mw > 8 Japan, Mexico, Central America.
Developed from the Japanese strong
Depth < 25 km for crustal events motion dataset and additional overseas
Shorter distance to the rupture
PGA and PSA data from the Western part of US and
zone for earthquakes with Depth < 50 km for interface events
for 0.05 to 5 s. the 1978 Tabs, Iran evens to constrain
Zhao et al. (2006) 5.0 ≤ Mw ≤ 8.3 available fault models, otherwise Depth ≥ 50 km for inslab events
Horizontal the near-source behaviour (20 overseas
hypocentral distance
component Depth =125 km for events with events of a total of 269 events).
Rrup ≤ 300 km for slab events Depth > 125 This GMPE was used also for the
volcanic zone.
Closest distance to the fault
plane for earthquakes with
available fault models, otherwise Developed from Japanese seismic
PGA, PGV, PSA
hypocentral distane waveform data: earthquake records from
for 0.05 to 5 s. Depth ≤ 30 km for shallow events
Kanno et al. (2006) Mw ≥ 5.5 K-NET and KiKnet databases.
Horizontal 1 ≤ Rrup ≤ 30 km Depth > 30 km for deep events
for shallow events This GMPE was used also for shallow
component
and volcanic zones.
30 ≤ Rrup ≤ 300 km
for deep events
Hypocentral distance
PGA and PSA 4.1 ≤ Mw ≤ 6.7
for intraslab events 40 ≤ R ≤ 600 km Depth ≤ 50 km for interface events
for 0.01 to 5 s.
Lin and Lee (2008) for intraslab events Regional relation based on Chilean data.
Horizontal 5.3 ≤ Mw ≤ 8.1 Depth > 50 km for inslab events
for interface events 20 ≤ R ≤ 300 km
component
for interface events
Table 5.9 Summary of attenuation relationships used in the current study for crustal zones

MAGNITUDE DISTANCE
FAULT DEPTH DEFINITION AND
ATTENUATION EQUATION APPLICATION DEFINITION AND DEFINITION AND DATABASE
CONTRIBUTIONS TO SEISMIC RESPONCE
LIMITATIONS LIMITATIONS
Z_TOR: Depth to the top of the coseismic rupture
plane (used in “depth-to-top rupture model”
PGA and PSA contribution). NGA database plus Kocaeli
for 0.01 to 10 s Rupture distance Z_1.0: Depth to the 1 km/s shear wave velocity and Chi-Chi mainshocks,
Rotation- horizon (used in “soil-depth” contribution, which is Chi Chi and Duzce
Abrahamson & Silva (2008) 5 ≤ Mw ≤ 8.5 0 ≤ Rrup ≤ 200 km null if Z_1.0<200 and VS30 ≥ 1000 m/s). aftershocks.
independent
average horizontal In our analysis the hanging wall contribution is not This GMPE was used also
component considered. for the volcanic zone.
If V S30 ≥ 1100 m/s the soil response is assumed
linear for all the periods.
Z_2.5: Depth to the 2.5 km/s shear wave velocity
horizon (used in “sediments” contribution 0 <
Mw > 4
Z_2.5 < 10 km).
Mw < 7.5 Closest distance to the
PGA and PSA If 2 ≤Z_2.5≤ 3, f_sed=0. Developed from the
for normal events surface projection of
for 0.01 to 10 s. the coseismic rupture Z_TOR: Depth to the top of the coseismic rupture updated PEER strong
Mw < 8
Campbell & Bozorgnia (2008) Horizontal plane plane (used in “style-of-foulting” and “hanging motion database, referred to
for reverse events
component wall” contributions, 0<Z_TOR<15 km. simply as the NGA
Mw < 8.5 0 ≤ Rrup ≤ 200 km
(geometric mean) In our work hanging wall contribution is not database.
for strike-slip events
considered.
If VS30 ≥ 1100 m/s the soil response is assumed
linear for all the periods.
Z_TOR: Depth to the top of rupture (used in
“source effects and hanging wall” contributions )
Mw ≥ 4 Z_1.0: Depth to the 1 km/s shear wave velocity
PGV, PGA,PSA horizon (used in “sediments” contribution), which
Mw ≤ 8.5 Closest distance to the PEER-NGA database
for 0.01 to 10 s. is minimum for Z_1.0=0. Z_1.0= 40 m is
for strike-slip events rupture plane (3551 recordings from 173
Chiou & Youngs (2008) Orientation- suggested. earthquakes).
Mw ≤ 8 0 ≤ Rrup ≤ 200 km
independent In our work hanging wall contribution is not
for reverse and This GMPE was used also
average horizontal considered.
normal faulting events for the volcanic zone.
component
Site effects are null for VS30≥1130 m/s (linear
response).
150≤VS30≤1500 m/s
Table 5.10 Continuation of Table 5.9

MAGNITUDE
DISTANCE DEFINITION FAULT DEPTH DEFINITION AND
ATTENUATION EQUATION APPLICATION DEFINITION AND DATABASE
AND LIMITATIONS CONTRIBUTIONS TO SEISMIC RESPONCE
LIMITATIONS
Developed from NGA
database, calibrated for
Western US and other
Closest horizontal distance It does not account for basin response and
similar tectonically active
PGV, PGA, PSA to the surface projection of depth-to top of rupture model as Abrahamson
region of shallow crustal
for 0.01 to 10 s. the fault plane & Silva (2008) and Campbell & Bozorgnia
Boore & Atkinson (2008) 5.0 ≤ Mw ≤ 8.0 faults. Representative of the
Average horizontal (2008).
RJB ≤ 200 km NGA models and
component If VS30(Rock)>760 m/s site response is linear confidently applicable
and site amplification is null. within Europe (Stafford et
al., 2008)

Table 5.11 Attenuation relationship used in the current study only for the volcanic zone

MAGNITUDE
DISTANCE DEFINITION FAULT DEPTH DEFINITION AND
ATTENUATION EQUATION APPLICATION DEFINITION AND DATABASE
AND LIMITATIONS CONTRIBUTIONS TO SEISMIC RESPONCE
LIMITATIONS
Shorter distance to the
PGA and PSA California earthquake
rupture surface for Strike-slip and normal events are distinguished
(<1994) plus Gazli (USSR,
for 0.07 to 4 s. earthquakes with available from reverse/thrust faulting events
Sadigh et al. (1997) 45.0 ≤ Mw ≤ 8.0+ 1976) and Tabas (Iran,
Horizontal fault models, otherwise (amplification coefficient of 1.2).
1978) earthquakes
component hypocentral distance VS30 (Rock) > 750 m/s
Rrup ≤ 100 km
A89

5.4 COMPARISON OF ATTENUATION RELATIONSHIPS WITH AVAILABLE STRONG


MOTION DATA

In order to attribute to the GMPEs appropriate weight in the logic tree adopted for the
PSHA, seismic response estimated by the attenuation relationships were compared with
strong motion recordings.

We analyzed and corrected the Strong Motion Data that belong to Seismic Research
Centre (SRC) for five stations located in Trinidad namely Chaguaramas (TCHG), Point
Cumana (TPTC), Bringand Hill (TBH), West Moorings (TWMO) and Atlantic (LNG) for
the period 1997-2008, for the N-S, E-W and Up-Down components of motion. The
instruments are K2 Kinemetrics Inc recording accelerations sampled at 0.005s with a
frequency flat response up to 50 Hz. We corrected the acceleration time histories using a
base-line correction and applying a band pass filter depending of the noise-to-signal ratio
and the dynamic amplification factor of the sensors; we performed the integration in the
frequency domain yielding the velocity and displacement histories. The Response Spectra
for 5% damping are also reproduced in this scheme. Since we are interested in evaluating
the seismic hazard considering firm soil, we selected in our analysis only the recordings
that belong to the stations lying on rock site conditions.

As a part of the Strong Motion processing at SRC, we use the data of BRGM (Bureau de
Recherches Géologiques et Minières - France) that belong to the stations located in
Martinique and Guadeloupe Islands. We employed only the records for two important
earthquakes dated on 1999/06/08 M 5.8 and 2004/11/21 M 6.3 in the interface and
volcanic zone respectively.
A90

Recordings of the strong motion events listed in Table 5.12 are available at various rock
stations. Figure 5.9 shows the location of epicentres and rock stations whose recordings
were used for the comparison with GMPEs (Figure 5.10 - Figure 5.21) in terms of PGA
and spectral accelerations at the representative periods of 0.2 and 1 seconds.

Table 5.12 Eastern Caribbean events whose recordings were compared with GMPEs
Date Mw H Earthquake type LAT. LONG.
November 29, 2007 7.4 148.0 intraplate 14.97 -61.26
October 4, 2000 6.1 110.4 intraplate 11.16 -62.29
October 28, 2005 5.5 80.9 intraplate 11.11 -62.04
November 15, 2006 5.2 98.9 intraplate 10.78 -62.65
October 24, 2005 5.1 137.7 intraplate 11.10 -62.39
November 17, 2006 4.9 135.8 intraplate 11.39 -62.24
January 25, 2001 4.6 85.5 intraplate 10.70 -62.57
June 8, 1999 5.8 52.4 interface 15.07 -60.40
December 2, 2004 5.8 48.2 crustal 10.49 -61.45
December 3, 2004 5.4 40.5 crustal 10.54 -61.46
June 21, 2003 5.3 10.0 crustal 10.79 -59.27
November 21, 2004 6.3 21.2 volcanic 15.73 -61.68
A91

Figure 5.9 Map of Eastern Caribbean islands with location of epicentres and recording station of
earthquakes whose recordings were compared with GMPEs estimates
A92

The comparison between GMPEs and recorded data shows that records are generally
within the bounds defined by the GMPEs, but with a relative high dispersion with
respect to the various attenuation relationships. As a consequence the identification of
the GMPEs which better fit the recordings is not a simple task.

For subduction events the GMPE of Atkinson and Boore (2008) shows a high variability
with Mw and it predicts significant lower values at low magnitudes (Mw<6) if it is
compared with the other GMPEs, showing an increasing attenuation of amplitudes at
increasing periods and distances (see Figure 5.12 - Figure 5.17). This behavior is more
marked for interface events than for intraplate ones, also at high magnitudes (see Figure
5.17). These differences become smaller for larger magnitudes (see Figure 5.10 and Figure
5.11). In fact the GMPE of Atkinson and Boore (2008) constitutes a lower bound among
the GMPEs adopted for subduction zones, except at Mw=7.4, where it provides
amplitudes in the average range estimated by the other GMPEs (Figure 5.12). This
relationship is in good agreement also with recordings of earthquakes with Mw less than 5
which are strongly overestimated by the other GMPEs (Figure 5.15 and Figure 5.16). At
higher Mw and at hypocentral distances larger than 100 km, for which recordings are
available, the GMPEs of Zhao et al. (2006) and Kanno et al. (2006) seem to better fit the
data with respect to the others. The relationship of Lin and Lee (2008) is characterized by
a lesser attenuation with distance with respect to the other GMPEs and it intersects them
at different hypocentral distances, depending on the period, magnitude and hypocentral
depth.

Concerning crustal earthquakes, from the comparison of the adopted GMPEs, it emerges
a different behaviour depending on the focal mechanism. Kanno et al. (2006) GMPE
constitute the upper bound for normal mechanism (Figure 5.16 and Figure 5.19),
providing significant overestimated values if compared with the other GMPEs, while for
inverse events Chiou and Youngs (2008) constitutes the lower bound and Kanno et al.
(2006) provides values in agreement with Abrahamson and Silva (2008) and Zhao et al.
(2006) (Figure 5.20). Note that, Chiou and Youngs (2008) relations were not used for
earthquakes with hypocentral depth higher than 19 km, over which the function is not
longer valid. Campbel and Bozogornia (2008) relation shows a lesser attenuation with
distance if compared with the other GMPEs. In fact, it maintains almost the same
inclination with distance, while the slope of the other curves significantly increases at
large distances (> 120-140 km). From the comparison with recordings it comes out that
Zhao et al. (2006) and Abrahamson and Silva (2008) are the relations which better
represent an average behaviour of recordings (unfortunately records from only one
earthquake and one rock station are available), although the choose of the most proper
GMPEs is difficult to identify, due to the dispersion of available data.
A93

Regarding with the attenuation in the volcanic Zone 1 we can stress out that the
attenuation characteristics for the earthquakes attacking the islands with epicenters
offshore but inside of our zone 1 (volcanic island arc) could be different from the
attenuation of an earthquake inside the islands with waves travel just below the volcanoes
at short distances (less than 50 km). Concerning volcanic earthquakes, unfortunately, only
PGA from recordings of November 21, 2004 (M 6.3) are available at rock stations (from
Bengoubou-Valerius et al. (2008), as shown in Figure 5.21. The two attenuation
relationships of McVerry et al. (2006) and Salazar (2004) specifically calibrated for
volcanic regions yield quite similar attenuation with distance in volcanic environments for
New Zeland and El Salvador, Central America, respectively. In fact, the recordings of
November 21, 2004 earthquake with epicenter offshore would be dominated by waves of
critical reflection in the Moho not traveling through the entire volcanic structures,
matching the Sadigh et al. (1997) attenuation relationship for upper-crustal earthquakes
worldwide. It is important to notice that McVerry et al. (2006) and Salazar (2004)
relations are valid for hypocentral distances less than about 50 km, because of the rapid
attenuation of waves due to this type of earthquakes. It is clear that they strongly
underestimate recorded data, while the other GMPEs adopted for the volcanic zone are
in reasonable agreement with PGA values. Unfortunately, only data at distances larger
than about 40 km are available; in fact data shown in Figure 5.21 and recorded from 30 to
40 km are circled since the classification of the respective recording stations as “rock
stations” by Bengoubou-Valerius et al. (2008) is disputed. The GMPE proposed by
Sadigh et al. (1997) seems to fit the data better than the others relations, at distances
larger than 100 km too.

It is important to notice that, the GMPE of Kanno et al. (2006) distinguishes between
shallow (depth ≤ 30 km) and deep earthquakes (depth > 30 km), but it does not account
for the hypocentral depth. In fact for all events with depth larger than 30 km it is
assumed the same equation, which differs from the function developed for all events with
depth less than 30 km. Furthermore, no distinction is made between crustal-shallow and
interface earthquakes, and between crustal-deep and intraplate earthquakes. This means
that, although this equation is in good agreement with the available recordings (for the
particular combination of magnitude, depth and kind of earthquake), it generally tends to
provide larger values if compared with the other GMPEs, both for subduction and
crustal events. On the other hand, Zhao et al. (2006) relation is sensible to both
earthquake typology (subduction or crustal) and focal depth. In the PSHA the Chiou and
Youngs (2008) GMPE will be used for the volcanic zone only, since crustal zones are
characterized by depths generally larger than the validity range (< 19 km).
A94

Concluding, for the PSHA the following GMPEs have been adopted:

• SUBDUCTION ZONES:
i) Youngs et al. (1997);
ii) Atkinson and Boore (2003-2008);
iii) Zhao et al (2006);
iv) Kanno et al. (2006);
v) Lin and Lee (2008).

• CRUSTAL ZONES:
i) Zhao et al (2006);
ii) Kanno et al. (2006);
iii) Abrahamson and Silva (2008);
iv) Boore and Atkinson (2008);
v) Campbel and Bozogornia (2008).

• VOLCANIC ZONE:
i) Sadigh et al. (1997);
ii) Zhao et al (2006);
iii) Kanno et al. (2006);
iv) Abrahamson and Silva (2008);
v) Chiou and Youngs (2008).
Mw =7.4 T =0 sec Mw =7.4 T =0.2 sec
1 1
10 10

0 0
10 10
Acceleration (g)

Acceleration (g)
-1 -1
10 10

-2 -2
10 10

-3 -3
10 2
10 2
10 10
Hypocentral distance (km) Hypocentral distance (km)
Youngs et al. 97 - average
Mw =7.4 T =1 sec " - average +σ
1 " - average - σ
10
Kanno et al., 06 - average
" - average +σ
" - average - σ
0
10 Lin & Lee 08 - average
" - average +σ
Acceleration (g)

" - average - σ
-1 Zhao et al. 06 - average
10 " - average +σ
" - average - σ
Atkinson & Boore 08 - average
-2 " - average +σ
10
" - average - σ
2007-11-29 component=1 station=GSCA
2007-11-29 component=2 station=GSCA
-3 2007-11-29 component=1 station=GBRA
10 2
10 2007-11-29 component=2 station=GBRA
2007-11-29 component=1 station=GGFA
Hypocentral distance (km)
2007-11-29 component=2 station=GGFA
1

Figure 5.10 Comparison between GMPEs and records of November 29, 2007 intraplate earthquake (Mw =7.4) at three rock stations (GSCA, GBRA, GGFA) in terms of PGA and
spectral accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
Mw =6.1 T =0 sec Mw =6.1 T =0.2 sec
0 0
10 10

-1 -1
Acceleration (g)

Acceleration (g)
10 10

-2 -2
10 10

-3 -3
10 2
10 2
10 10
Hypocentral distance (km) Hypocentral distance (km)
Youngs et al. 97 - average
Mw =6.1 T =1 sec " - average +σ
0
10 " - average - σ
Kanno et al., 06 - average
" - average +σ
" - average - σ
Lin & Lee 08 - average
-1 " - average +σ
Acceleration (g)

10
" - average - σ
Zhao et al. 06 - average
" - average +σ
" - average - σ
-2 Atkinson & Boore 08 - average
10
" - average +σ
" - average - σ
2000-10-4 component=1 station=TPTC
2000-10-4 component=2 station=TPTC
-3 2000-10-4 component=1 station=TBH
10 2 2000-10-4 component=2 station=TBH
10
Hypocentral distance (km)
1

Figure 5.11 Comparison between GMPEs and records of October 4, 2000 intraplate earthquake (Mw =6.1) at two rock stations (TPTC and TBH) in terms of PGA and spectral
accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
0
Mw =5.5 T =0 sec 0
Mw =5.5 T =0.2 sec
10 10

-1 -1
10 10
Acceleration (g)

Acceleration (g)
-2 -2
10 10

-3 -3
10 10

-4 -4
10 2
10 2
10 10
Hypocentral distance (km) Hypocentral distance (km)
Youngs et al. 97 - average
" - average +σ
Mw =5.5 T =1 sec
0 " - average - σ
10
Kanno et al., 06 - average
" - average +σ
" - average - σ
-1
10 Lin & Lee 08 - average
" - average +σ
Acceleration (g)

" - average - σ
-2 Zhao et al. 06 - average
10 " - average +σ
" - average - σ
Atkinson & Boore 08 - average
-3 " - average +σ
10
" - average - σ
2005-10-28 component=1 station=ALNG
2005-10-28 component=2 station=ALNG
-4
10 2005-10-28 component=1 station=TCHG
2
10 2005-10-28 component=2 station=TCHG
Hypocentral distance (km) 2005-10-28 component=1 station=TBH
2005-10-28 component=2 station=TBH
1

Figure 5.12 Comparison between GMPEs and records of October 28, 2005 intraplate earthquake (Mw =5.5) at three rock stations (ALNG, TCHG, TBH) in terms of PGA and
spectral accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
0
Mw =5.2 T =0 sec 0
Mw =5.2 T =0.2 sec
10 10

-1 -1
10 10
Acceleration (g)

Acceleration (g)
-2 -2
10 10

-3 -3
10 10

-4 -4
10 2
10 2
10 10
Hypocentral distance (km) Hypocentral distance (km)
Youngs et al. 97 - average
Mw =5.2 T =1 sec " - average +σ
0
10 " - average - σ
Kanno et al., 06 - average
" - average +σ
" - average - σ
-1
10 Lin & Lee 08 - average
" - average +σ
Acceleration (g)

" - average - σ
Zhao et al. 06 - average
-2
10 " - average +σ
" - average - σ
Atkinson & Boore 08 - average
-3
" - average +σ
10 " - average - σ
2006-11-15 component=1 station=ALNG
2006-11-15 component=2 station=ALNG
-4
2006-11-15 component=1 station=TCHG
10 2
2006-11-15 component=2 station=TCHG
10 2006-11-15 component=1 station=TBH
Hypocentral distance (km) 2006-11-15 component=2 station=TBH
1

Figure 5.13 Comparison between GMPEs and records of November 15, 2006 intraplate earthquake (Mw =5.2) at three rock stations (ALNG, TCHG, TBH) in terms of PGA and
spectral accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
0
Mw =5.1 T =0 sec 0
Mw =5.1 T =0.2 sec
10 10

-1 -1
10 10
Acceleration (g)

Acceleration (g)
-2 -2
10 10

-3 -3
10 10

-4 -4
10 2
10 2
10 10
Hypocentral distance (km) Hypocentral distance (km)
Youngs et al. 97 - average
Mw =5.1 T =1 sec " - average +σ
0
10 " - average - σ
Kanno et al., 06 - average
" - average +σ
-1 " - average - σ
10 Lin & Lee 08 - average
Acceleration (g)

" - average +σ
" - average - σ
-2 Zhao et al. 06 - average
10
" - average +σ
" - average - σ
Atkinson & Boore 08 - average
-3
10 " - average +σ
" - average - σ
2005-10-24 component=1 station=ALNG
-4
2005-10-24 component=2 station=ALNG
10 2
2005-10-24 component=1 station=TCHG
10 2005-10-24 component=2 station=TCHG
Hypocentral distance (km) 2005-10-24 component=1 station=TBH
2005-10-24 component=2 station=TBH
1

Figure 5.14 Comparison between GMPEs and records of October 24, 2005 intraplate earthquake (Mw =5.1) at three rock stations (ALNG, TCHG, TBH) in terms of PGA and
spectral accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
0
Mw =4.9 T =0 sec 0
Mw =4.9 T =0.2 sec
10 10

-1 -1
10 10
Acceleration (g)

Acceleration (g)
-2 -2
10 10

-3 -3
10 10

-4 -4
10 10

2 2
10 10
Hypocentral distance (km) Hypocentral distance (km)

Youngs et al. 97 - average


0
Mw =4.9 T =1 sec " - average +σ
10 " - average - σ
Kanno et al., 06 - average
-1 " - average +σ
10 " - average - σ
Lin & Lee 08 - average
Acceleration (g)

-2 " - average +σ
10
" - average - σ
Zhao et al. 06 - average
-3 " - average +σ
10
" - average - σ
Atkinson & Boore 08 - average
-4 " - average +σ
10
" - average - σ
2006-11-17 component=1 station=TCHG
2006-11-17 component=2 station=TCHG
2
10
Hypocentral distance (km)
1

Figure 5.15 Comparison between GMPEs and records of November 17, 2006 intraplate earthquake (Mw =4.9) at the station TCHG in terms of PGA and spectral accelerations at
periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
1

0
Mw =4.6 T =0 sec 0
Mw =4.6 T =0.2 sec
10 10

-1 -1
10 10
Acceleration (g)

Acceleration (g)
-2 -2
10 10

-3 -3
10 10

-4 -4
10 10

2 2
10 10
Hypocentral distance (km) Hypocentral distance (km)
Youngs et al. 97 - average
" - average +σ
0
Mw =4.6 T =1 sec
10 " - average - σ
Kanno et al., 06 - average
" - average +σ
-1
10 " - average - σ
Lin & Lee 08 - average
Acceleration (g)

-2 " - average +σ
10 " - average - σ
Zhao et al. 06 - average
-3 " - average +σ
10
" - average - σ
Atkinson & Boore 08 - average
-4
10 " - average +σ
" - average - σ
2001-1-25 component=1 station=ALNG
2 2001-1-25 component=2 station=ALNG
10 2001-1-25 component=1 station=TBH
Hypocentral distance (km) 2001-1-25 component=2 station=TBH

Figure 5.16 Comparison between GMPEs and records of January 25, 2001 intraplate earthquake (Mw =4.6) at two rock stations (ALNG and TBH) in terms of PGA and spectral
accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
0
Mw =5.8 T =0 sec 0
Mw =5.8 T =0.2 sec
10 10

-1 -1
10 10
Acceleration (g)

Acceleration (g)
-2 -2
10 10

-3 -3
10 10

-4 -4
10 2
10 2
10 10
Hypocentral distance (km) Hypocentral distance (km)
Youngs et al. 97 - average
Mw =5.8 T =1 sec " - average +σ
0
10 " - average - σ
Kanno et al., 06 - average
" - average +σ
-1 " - average - σ
10 Lin & Lee 08 - average
Acceleration (g)

" - average +σ
" - average - σ
-2 Zhao et al. 06 - average
10
" - average +σ
" - average - σ
Atkinson & Boore 08 - average
-3
10 " - average +σ
" - average - σ
1999-6-8 component=1 station=mbra
-4 1999-6-8 component=2 station=mbra
10 2 1999-6-8 component=1 station=gbra
10 1999-6-8 component=2 station=gbra
Hypocentral distance (km) 1999-6-8 component=1 station=GPAA
1999-6-8 component=2 station=GPAA
1

Figure 5.17 Comparison between GMPEs and records of June 8, 1999 interface earthquake (Mw =5.8) at three rock stations (mbra, gbra, GPAA) in terms of PGA and spectral
accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
Mw =5.8 T =0 sec Mw =5.8 T =0.2 sec
0 1
10 10

-1 0
10 10
Acceleration (g)

Acceleration (g)
-2 -1
10 10

-3 -2
10 10

-4 -3
10 2
10 2
10 10
Hypocentral distance(km) Hypocentral distance(km)

Zhao et al. 08 - average


Mw =5.8 T =1 sec " - average +σ
0
10 " - average - σ
Kanno et al. 06 - average
" - average +σ
-1 " - average - σ
10 Abrahamson & Silva 08 - average
Acceleration (g)

" - average +σ
" - average - σ
-2 Campbell & Bozorgnia 08 - average
10 " - average +σ
" - average - σ
Boore & Atkinson 08 - average
-3 " - average +σ
10 " - average - σ
2004-12-2 component=1 station=ALNG
2004-12-2 component=2 station=ALNG
-4 2004-12-2 component=1 station=TCHG
10 2 2004-12-2 component=2 station=TCHG
10 2004-12-2 component=1 station=TBH
Hypocentral distance(km) 2004-12-2 component=2 station=TBH
1

Figure 5.18 Comparison between GMPEs and records of December 2, 2004 crustal earthquake (Mw =5.8) in the Trinidad faults area, at three rock stations (ALNG, TCHG, TBH) in
terms of PGA and spectral accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
Mw =5.4 T =0 sec Mw =5.4 T =0.2 sec
0 0
10 10

-1 -1
10 10
Acceleration (g)

Acceleration (g)
-2 -2
10 10

-3 -3
10 10

-4 -4
10 2
10 2
10 10
Hypocentral distance(km) Hypocentral distance(km)
Zhao et al. 08 - average
Mw =5.4 T =1 sec " - average +σ
-1
10 " - average - σ
Kanno et al. 06 - average
" - average +σ
" - average - σ
Abrahamson & Silva 08 - average
-2
Acceleration (g)

10 " - average +σ
" - average - σ
Campbell & Bozorgnia 08 - average
" - average +σ
" - average - σ
-3
10 Boore & Atkinson 08 - average
" - average +σ
" - average - σ
2004-12-3 component=1 station=ALNG
2004-12-3 component=2 station=ALNG
-4
10 2
2004-12-3 component=1 station=TCHG
10 2004-12-3 component=2 station=TCHG
Hypocentral distance(km) 2004-12-3 component=1 station=TBH
2004-12-3 component=2 station=TBH
1

Figure 5.19 between GMPEs and records of December 3, 2004 shallow earthquake (Mw =5.4) in the Trinidad faults area, at three rock stations (ALNG, TCHG, TBH), in terms of
PGA and spectral accelerations at periods of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
M =5.3 T =0 sec M =5.3 T =0.2 sec
0
w 0
w
10 10

-1 -1
10 10
Acceleration (g)

Acceleration (g)
-2 -2
10 10

-3 -3
10 10

-4 -4
10 2
10 2
10 10
Hypocentral distance(km) Hypocentral distance(km)
Zhao et al. 08 - average
Mw =5.3 T =1 sec " - average +σ
-1 " - average - σ
10
Kanno et al. 06 - average
" - average +σ
" - average - σ
Abrahamson & Silva 08 - average
-2 " - average +σ
Acceleration (g)

10 " - average - σ
Campbell & Bozorgnia 08 - average
" - average +σ
" - average - σ
-3 Chiou & Youngs 08 - average
10 " - average +σ
" - average - σ
Boore & Atkinson 08 - average
" - average +σ
" - average - σ
-4
10 2
2003-6-21 component=1 station=TBH
10 2003-6-21 component=2 station=TBH
Hypocentral distance(km)
1

Figure 5.20 Comparison between GMPEs and records of June 21, 2003 crustal earthquake (Mw =5.8) at the rock station TBH, in terms of PGA and spectral accelerations at periods
of 0.2, and 1 sec. Both horizontal components are plotted: NS (component 1) and EW (component 2)
Mw =6.3 T =0 sec
0
10
Sadigh et al., 97 - average
" - average +σ
" - average - σ
Zhao et al. 08 - average
-1
" - average +σ
10 " - average - σ
Kanno et al., 06 - average
" - average +σ
Acceleration (g)

" - average - σ
Abrahamson & Silva 08 - average
-2 " - average +σ
10
" - average - σ
Chiou & Youngs 08 - average
" - average +σ
" - average - σ
-3
Salazar, 04 - average
10 " - average +σ
" - average - σ
McVerry et al., 06 - average
" - average +σ
" - average - σ
-4 2004-11-21 - PGA from Bengoubou-Valerius et al. 08
10 1 2
10 10
Hypocentral distance (km)
1

Figure 5.21 Comparison between GMPEs and records of November 21, 2004 Les Saintes earthquake (Mw =6.3) in Guadeloupe, along the volcanic arc, at various rock stations in
terms of PGA taken from Bengoubou-Valerius et al. (2008). The circle contours stations for which the identification as rock stations by Bengoubou-Valerius et al. (2008)
is disputed
A107

5.5 PROBABILISTIC SEISMIC HAZARD ANALISYS AND RESULTS


After a number of sequential processes described in the preceding paragraphs, namely,
compilation of the earthquake catalogue, definition of seismic source zones and selection
of appropriate attenuation equations, the PSHA had been executed. The outcome of the
PSHA is composed of the horizontal hazard curves in terms of the PGA and the Uniform
Hazard Spectra (UHS) in accelerations, pertaining to reference return periods of 95, 475,
975 and 2475 years on rock outcropping (or stiff site). The computation of the
probabilistic seismic hazard has been done using the software EZ-FRISK for the points
of a grid with a mesh of 0.025°, that is a point every about 2.8 km (the total number of
points is equal to 2099). Using EZ-FRISK software, we had the opportunity to consider
the adopted attenuation relations truncate at median amplitude plus 3 sigmas.

5.5.1 Treatment of Uncertainties in PSHA Computations


The PSHA methodology adopted in the current study has accounted for the uncertainties
related to the inherent random nature of the various input parameters used to describe
the seismicity and the ground motion attenuation (i.e. “aleatory” variability) and to the
“epistemic” variability. The effects of this variability can be accounted for by defining a
discrete probability density function for all the alternatives and by estimating the “mean”
or “median” value of the ground motion parameter. In the PSHA approach, the “logic-
tree” formulation is often used to incorporate the effects of both aleatory and epistemic
uncertainties (e.g. Coppersmith and Youngs, 1986; SSHAC, 1997). The logic-tree
methodology considers a large number of different probabilistic models and model
parameters and computes the hazard for all the combinations of parameter values defined
by the end branches of the logic-tree. Each of the input parameters is assigned an
appropriate weight to define a discrete probability density function for the frequency.

The controlling parameters of the logic-tree considered in this study are: a) estimation
methods of catalogue completeness periods, b) attenuation relationships and c) the
maximum magnitudes. The logic tree for the horizontal component is composed of a 10
branches for each zone; the total number of branches is 150 (see Table 5.13).

Weighting factors have been assigned to alternative models chosen for the controlling
parameters. Higher values of these factors are attributed to models that are believed
(subjectively) to be characterised by a higher likelihood of being correct.
A108

Table 5.13 Proposed logic tree for the horizontal component of ground motion.
ZONE Mmax ATTENUATION RELATION BRANCH
Abrahamson_Silva08_shallow_normal 0.2 1 0.100
Chiou_Youngs08_SHA_shallow_normal 0.2 2 0.100
Mmax 0.5 Kanno06_interface_shallow 0.15 3 0.075
Sadigh_1997_normal 0.25 4 0.125
Zhao06_SHA_crustal_normal 0.2 5 0.100
1
Abrahamson_Silva08_shallow_normal 0.2 6 0.100
Chiou_Youngs08_SHA_shallow_normal 0.2 7 0.100
Mmax+∆ 0.5 Kanno06_interface_shallow 0.15 8 0.075
Sadigh_1997_normal 0.25 9 0.125
Zhao06_SHA_crustal_normal 0.2 10 0.100
AB08_interface 0.1 11 0.050
Kanno06_interface 0.2 12 0.100
Mmax 0.5 Lin_Lee08_interface 0.2 13 0.100
Youngs97_interface 0.2 14 0.100
Zhao06_interface 0.3 15 0.150
2
AB08_interface 0.1 16 0.050
Kanno06_interface 0.2 17 0.100
Mmax+∆ 0.5 Lin_Lee08_interface 0.2 18 0.100
Youngs97_interface 0.2 19 0.100
Zhao06_interface 0.3 20 0.150
AB08_interface 0.1 21 0.050
Kanno06_interface 0.2 22 0.100
Mmax 0.5 Lin_Lee08_interface 0.2 23 0.100
Youngs97_interface 0.2 24 0.100
Zhao06_interface 0.3 25 0.150
3
AB08_interface 0.1 26 0.050
Kanno06_interface 0.2 27 0.100
Mmax+∆ 0.5 Lin_Lee08_interface 0.2 28 0.100
Youngs97_interface 0.2 29 0.100
Zhao06_interface 0.3 30 0.150
AB08_intraslab 0.1 31 0.050
Kanno06_intraslab 0.2 32 0.100
Mmax 0.5 Lin_Lee08_intraslab 0.2 33 0.100
Youngs97_intraslab 0.2 34 0.100
Zhao06_intraslab 0.3 35 0.150
4
AB08_intraslab 0.1 36 0.050
Kanno06_intraslab 0.2 37 0.100
Mmax+∆ 0.5 Lin_Lee08_intraslab 0.2 38 0.100
Youngs97_intraslab 0.2 39 0.100
Zhao06_intraslab 0.3 40 0.150
AB08_intraslab 0.1 41 0.050
Kanno06_intraslab 0.2 42 0.100
Mmax 0.5 Lin_Lee08_intraslab 0.2 43 0.100
Youngs97_intraslab 0.2 44 0.100
Zhao06_intraslab 0.3 45 0.150
5
AB08_intraslab 0.1 46 0.050
Kanno06_intraslab 0.2 47 0.100
Mmax+∆ 0.5 Lin_Lee08_intraslab 0.2 48 0.100
Youngs97_intraslab 0.2 49 0.100
Zhao06_intraslab 0.3 50 0.150
A109

ZONE Mmax ATTENUATION RELATION BRANCH


AB08_interface 0.1 51 0.050
Kanno06_interface 0.2 52 0.100
Mmax 0.5 Lin_Lee08_interface 0.2 53 0.100
Youngs97_interface 0.2 54 0.100
Zhao06_interface 0.3 55 0.150
6
AB08_interface 0.1 56 0.050
Kanno06_interface 0.2 57 0.100
Mmax+∆ 0.5 Lin_Lee08_interface 0.2 58 0.100
Youngs97_interface 0.2 59 0.100
Zhao06_interface 0.3 60 0.150
Abrahamson_Silva08_shallow_normal 0.2 61 0.100
Boore_Atkinson08_shallow_normal 0.2 62 0.100
Mmax 0.5 Campbel_Bozogornia08_shallow_normal 0.2 63 0.100
Zhao06_interface 0.2 64 0.100
Kanno06_interface_shallow 0.2 65 0.100
7
Abrahamson_Silva08_shallow_normal 0.2 66 0.100
Boore_Atkinson08_shallow_normal 0.2 67 0.100
Mmax+∆ 0.5 Campbel_Bozogornia08_shallow_normal 0.2 68 0.100
Zhao06_interface 0.2 69 0.100
Kanno06_interface_shallow 0.2 70 0.100
AB08_intraslab 0.1 71 0.050
Kanno06_intraslab 0.2 72 0.100
Mmax 0.5 Lin_Lee08_intraslab 0.2 73 0.100
Youngs97_intraslab 0.2 74 0.100
Zhao06_intraslab 0.3 75 0.150
8
AB08_intraslab 0.1 76 0.050
Kanno06_intraslab 0.2 77 0.100
Mmax+∆ 0.5 Lin_Lee08_intraslab 0.2 78 0.100
Youngs97_intraslab 0.2 79 0.100
Zhao06_intraslab 0.3 80 0.150
Abrahamson_Silva08_shallow_normal 0.2 81 0.100
Boore_Atkinson08_shallow_normal 0.2 82 0.100
Mmax 0.5 Campbel_Bozogornia08_shallow_normal 0.2 83 0.100
Zhao06_interface 0.2 84 0.100
Kanno06_interface_shallow 0.2 85 0.100
9
Abrahamson_Silva08_shallow_normal 0.2 86 0.100
Boore_Atkinson08_shallow_normal 0.2 87 0.100
Mmax+∆ 0.5 Campbel_Bozogornia08_shallow_normal 0.2 88 0.100
Zhao06_interface 0.2 89 0.100
Kanno06_interface_shallow 0.2 90 0.100
Abrahamson_Silva08_shallow_normal 0.2 91 0.100
Boore_Atkinson08_shallow_normal 0.2 92 0.100
Mmax 0.5 Campbel_Bozogornia08_shallow_normal 0.2 93 0.100
Zhao06_interface 0.2 94 0.100
Kanno06_interface_shallow 0.2 95 0.100
10
Abrahamson_Silva08_shallow_normal 0.2 96 0.100
Boore_Atkinson08_shallow_normal 0.2 97 0.100
Mmax+∆ 0.5 Campbel_Bozogornia08_shallow_normal 0.2 98 0.100
Zhao06_interface 0.2 99 0.100
Kanno06_interface_shallow 0.2 100 0.100
A110

ZONE Mmax ATTENUATION RELATION BRANCH


AB08_intraslab 0.1 101 0.050
Kanno06_intraslab 0.2 102 0.100
Mmax 0.5 Lin_Lee08_intraslab 0.2 103 0.100
Youngs97_intraslab 0.2 104 0.100
Zhao06_intraslab 0.3 105 0.150
11
AB08_intraslab 0.1 106 0.050
Kanno06_intraslab 0.2 107 0.100
Mmax+∆ 0.5 Lin_Lee08_intraslab 0.2 108 0.100
Youngs97_intraslab 0.2 109 0.100
Zhao06_intraslab 0.3 110 0.150
Abrahamson_Silva08_shallow_normal 0.2 111 0.100
Boore_Atkinson08_shallow_normal 0.2 112 0.100
Mmax 0.5 Campbel_Bozogornia08_shallow_normal 0.2 113 0.100
Zhao06_interface 0.2 114 0.100
Kanno06_interface_shallow 0.2 115 0.100
12
Abrahamson_Silva08_shallow_normal 0.2 116 0.100
Boore_Atkinson08_shallow_normal 0.2 117 0.100
Mmax+∆ 0.5 Campbel_Bozogornia08_shallow_normal 0.2 118 0.100
Zhao06_interface 0.2 119 0.100
Kanno06_interface_shallow 0.2 120 0.100
Abrahamson_Silva08_shallow_strike_slip 0.2 121 0.100
Boore_Atkinson08_shallow_strike_slip 0.2 122 0.100
Mmax 0.5 Campbel_Bozogornia08_shallow_strike_slip 0.2 123 0.100
Zhao06_interface 0.2 124 0.100
Kanno06_interface_shallow 0.2 125 0.100
13
Abrahamson_Silva08_shallow_strike_slip 0.2 126 0.100
Boore_Atkinson08_shallow_strike_slip 0.2 127 0.100
Mmax+∆ 0.5 Campbel_Bozogornia08_shallow_strike_slip 0.2 128 0.100
Zhao06_interface 0.2 129 0.100
Kanno06_interface_shallow 0.2 130 0.100
Abrahamson_Silva08_shallow_strike_slip 0.2 131 0.100
Boore_Atkinson08_shallow_strike_slip 0.2 132 0.100
Mmax 0.5 Campbel_Bozogornia08_shallow_strike_slip 0.2 133 0.100
Zhao06_interface 0.2 134 0.100
Kanno06_interface_shallow 0.2 135 0.100
14
Abrahamson_Silva08_shallow_strike_slip 0.2 136 0.100
Boore_Atkinson08_shallow_strike_slip 0.2 137 0.100
Mmax+∆ 0.5 Campbel_Bozogornia08_shallow_strike_slip 0.2 138 0.100
Zhao06_interface 0.2 139 0.100
Kanno06_interface_shallow 0.2 140 0.100
Abrahamson_Silva08_shallow_strike_slip 0.2 141 0.100
Boore_Atkinson08_shallow_strike_slip 0.2 142 0.100
Mmax 0.5 Campbel_Bozogornia08_shallow_strike_slip 0.2 143 0.100
Zhao06_interface 0.2 144 0.100
Kanno06_interface_shallow 0.2 145 0.100
15
Abrahamson_Silva08_shallow_strike_slip 0.2 146 0.100
Boore_Atkinson08_shallow_strike_slip 0.2 147 0.100
Mmax+∆ 0.5 Campbel_Bozogornia08_shallow_strike_slip 0.2 148 0.100
Zhao06_interface 0.2 149 0.100
Kanno06_interface_shallow 0.2 150 0.100
A111

5.6 RESULTS
The ground motion parameters chosen for the seismic hazard computations are the
spectral accelerations (SA) defined according to the following scheme:

¾ 5 percent damped acceleration response spectrum;

¾ 22 structural periods T = 0, 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.6, 0.75,
0.9, 1, 1.25, 1.5, 1.75, 2, 2.25, 2.5, 2.75 and 3 sec.;

¾ 4 return periods RP = 95 - 475 - 975 - 2475 years;

Uniform hazard spectra have been calculated for bedrock motion (horizontal
component).

The above scheme is consistent with the prescriptions of IBC (2006) and ASCE Standard
7-05 (2006) whose compliance was requested by CCEO (Council of Caribbean
Engineering Organizations) President and APETT (Association of Professional
Engineers of Trinidad & Tobago) representative during the workshop held at the
Institute for Critical Thinking, The University of the West Indies, St. Augustine, Trinidad
on February 25-26, 2009.

Compliance with North American Building Codes to define the seismic hazard is
guaranteed by the computation of the spectral response accelerations at 0.2 seconds
(denoted in IBC, 2006 by the symbol SS) and 1.0 second (denoted in IBC, 2006 by the
symbol S1) for 2475 return period which correspond to 2 percent probability of
exceedance within a 50-yr. period.

The following maps contain the values of PGA (Figure 5.22), 0.2 sec. SA (Figure 5.23)
and 1 sec. SA (Figure 5.24) computed for the Eastern Caribbean region, considering the
four return periods, 95, 475, 975 and 2475 years.

The following paragraphs focus on the islands of Dominica, Barbados and Trinidad,
whose probabilistic spectra are presented for the four different return periods (bedrock
motion, horizontal component). Some zooms on the maps which contains PGA, 0.2 sec.
SA and 1 sec. SA values are also presented for the islands of Dominica, Barbados and
Trinidad; in particular the zooms regard the maps referred to the spectral response
accelerations at 0.2 seconds and 1.0 second for 2475 return period, which are basic data
of the North American Building Codes. The other maps, which contain all the zooms on
the islands of Dominica, Barbados and Trinidad, are in the folder “PSHA_Maps”,
attached in this report.
A112

Figure 5.22 a) Map of PGA values (g) for the 95 return period
A113

Figure 5.22 b) Map of PGA values (g) for the 475 return period
A114

Figure 5.22 c) Map of PGA values (g) for the 975 return period
A115

Figure 5.22 d) Map of PGA values (g) for the 2475 return period
A116

Figure 5.23 a) Map of SA values (g) at 0.2 sec. for 95 return period
A117

Figure 5.23 b) Map of SA values (g) at 0.2 sec. for 475 return period
A118

Figure 5.23 c) Map of SA values (g) at 0.2 sec. for 975 return period
A119

Figure 5.23 d) Map of SA values (g) at 0.2 sec. for 2475 return period
A120

Figure 5.24 a) Map of SA values (g) at 1 sec. for 95 return period


A121

Figure 5.24 b) Map of SA values (g) at 1 sec. for 475 return period
A122

Figure 5.24 c) Map of SA values (g) at 1 sec. for 975 return period
A123

Figure 5.24 d) Map of SA values (g) at 1 sec. for 2475 return period
A124

5.6.1 Dominica Island


Figure 5.25 represents a zoom for Dominica island on the maps referred to the spectral
response accelerations at 0.2 seconds and 1.0 second for 2475 return period, which
represent basic data of the North American Building Codes. The point with coordinates
15.42 N and 61.32 W is rapresentative of Dominica island; its Uniform Hazard Spectra
(USH) is presented in Figure 5.26.

Figure 5.25 a) Maps for Dominica of SA values (g) at 0.2 sec. for 2475 return period
A125

Figure 5.25 b) Maps for Dominica of SA values (g) at 1 sec. for 2475 return period
A126

UHS-Dominica

2 RP=95 years
1.75 RP=475 years
1.5 RP=975 years
1.25 RP=2475 years
SA(g) 1
0.75
0.5
0.25
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3
T (s)

Figure 5.26 Probabilistic spectra for different return periods (95, 475, 975 and 2475 years) for bedrock
motion (horizontal component) computed for Dominica.

5.6.2 Barbados Island


Figure 5.27 represents a zoom for Barbados island on the maps referred to the spectral
response accelerations at 0.2 seconds and 1.0 second for 2475 return period, which
represent basic data of the North American Building Codes. The point with coordinates
13.12 N and 59.57 W is rapresentative of Barbados island; its Uniform Hazard Spectra
(USH) is presented in Figure 5.28.
A127

Figure 5.27 a) Maps for Barbados of SA values (g) at 0.2 sec. for 2475 return period
A128

Figure 5.27 b) Maps for Barbados of SA values (g) at 1 sec. for 2475 return period
A129

UHS-Barbados

2 RP=95 years
1.75 RP=475 years
1.5 RP=975 years
1.25 RP=2475 years
SA (g) 1
0.75
0.5
0.25
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3
T (s)

Figure 5.28 Probabilistic spectra for different return periods (95, 475, 975 and 2475 years) for bedrock
motion (horizontal component) computed for Barbados.

5.6.3 Trinidad Island


Figure 5.29 represents a zoom for Trinidad island on the maps referred to the spectral
response accelerations at 0.2 seconds and 1.0 second for 2475 return period, which
represent basic data of the North American Building Codes. Uniform Hazard Spectra
(USH), presented in Figure 5.30, for Trinidad has been computed for the capital Port of
Spain.
A130

Figure 5.27 a) Maps for Trinidad of SA values (g) at 0.2 sec. for 2475 return period
A131

Figure 5.27 b) Maps for Trinidad of SA values (g) at 1 sec. for 2475 return period
A132

UHS-Trinidad (Port of Spain)

2 RP=95 years
1.75 RP=475 years
1.5 RP=975 years
1.25 RP=2475 years
SA (g) 1
0.75
0.5
0.25
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3
T (s)

Figure 5.30 Probabilistic spectra for different return periods (95, 475, 975 and 2475 years) for bedrock
motion (horizontal component) computed for Trinidad.

5.6.4 Considerations about the obtained results


In this paragraph, some considerations and comparisons between the results obtained in
this study and the past seismic hazard estimates are presented.

Regarding PGA values referred to 475 return period, Figure 5.22 b) shows that the results
obtained in this study for the Estern Caribbean islands are included in a range that
encompasses values from 0.2 to 0.35 g; the same range of PGA values is presented in the
hazard assessment studies of the 1990’s, developed for the Caribbean, Mexico and the
Americas (see Paragraph 2.2.2).

In addition, from all those studies it is clear like the Leewards islands are characterized by
highter seismicity then the Windwards islands.

Unique recent study regarding specifically the Eastern Caribbeam islands was published
in 2003 by Shepherd and Lynch; in this study the maps are referred to the spectral
response accelerations (gals) at 0.2 seconds and 1.0 second for 2475 return period. For
the comparison of the results, we focus on Trinidad island: it comes out that, in the case
of 0.2 sec. SA, this study estimates a range of values between 0.95 and 1.6 g, a little bit
highter then the ones presented in Shepherd and Lynch work (0.6-1.2 g); while for 1 sec.
SA, the values of both the studies are between about 0.25 and 0.45 g.
BIBLIOGRAPHY

Abrahamson M and Silva W., (2008) Summary of the Abrahamson and Sinva NGA ground-
motion relations, Earthquake Spectra, Vol. 24, No. 1, pp. 67-97.

Algar S. and Pindell J. (1993). Structure and deformation of the Northern Range of Trinidad and
Adjacent Areas. Tectonics, vol. 12, pp. 814-829.

Anderson J.A. and Wood H.O. (1924). A torsion seismometer, J. Opt. Soc. Am. Rev. Sci. Inst. 8,
817-822.

Anderson J.A. and Wood H.O. (1925). Description and theory of the torsion seismometer, Bull.
Seism. Soc. Am. 15, 1-72.

Atkinson G.M. and Boore D. M., (2003), Empirical Ground-Motion Relations for Subduction-
Zone Earthquakes and Their Application to Cascadia and Other Regions, Bulletin of the
Seismological Society of America, Vol. 93, No. 4, pp. 1703–1729.

Atkinson G.M. and Boore D. M., (2008), Erratum to Empirical G round-Motion Relations for
Subduction Zone Earthquakes and Their Application to Cascadia and Other Regions,
Bulletin of the Seismological Society of America, Vol. 98, No. 5, pp. 2567–2569.

Bengoubou-Valerius M., Bazin S., Bertil D., Beauducel F., and Bosson A. (2008) CDSA: A New
Seismological Data Center for the French Lesser Antilles Seismological Research Letters,
January 1, 2008; 79(1): 90-102.

Bernard P. and J. Lambert (1988) Subduction and seismic hazard in the Northern Lesser Antilles:
Revision of the Historical Seismicity. Bulletin of the Seismological Society of America, Vol.
78, 6, pp. 1965-1983.

Bollinger G.A. et al. (1993): A comparison of earthquake damage areas as a function of magnitude
across the United States, Bulletin of the Seismological Society of America, Vol. 83, No. 4,
pp. 1064-1080, August.
A134

Boore D. M. and Atkinson G. M., (2008), Ground-motion prediction equations for the average
horizontal component of PGA, PGV, and 5% -damped PSA and spectral periods between
0.01 s and 10.0 s., Earthquake Spectra, Vol. 24, No. 1, pp. 99-138.

Boynton C. H., Westbrook G. K., Bott M. H. P., Long R. E. (1979). A seismic refraction
investigation of crustal structure beneath the Lesser Antilles island arc. Geophysical Journal
of the Royal Astronomical Society 58:371-393.

Burmester R., M. Beck, R. Speed and A. Snoke (1996). A preliminary paleomagnetic pole for mid-
Cretaceous rock from Tobago: further evidence for large clockwise rotations in the
Caribbean-South American plate boundary zone. Earth and Planetary Science Letters, vol.
139, pp. 79-90.

Byrne D. E., D. M. Davis, and L. R. Sykes (1988) Loci and maximum size of thrust earthquakes
and the mechanics of the shallow region of subduction zones, Tectonics, 7(4), 833-857.

Campbell K.W and Bozogorgnia Y., (2008), NGA ground motion model for the geometric mean
horizontal component of PGA, PGV, PGD and 5% damped linear elastic response spectra
for periods ranging from 0.01 to 10 s, Earthquake Spectra, Vol. 24, No. 1, pp. 139-171.

Chen K.P. and Tsai Y.B. (2008). A Catalog of Taiwan Earthquakes (1900–2006) with
Homogenized Mw Magnitudes, Bulletin of the Seismological Society of America, Vol. 98,
No. 1, pp. 483–489, February.

Chiou B.S.J. and Youngs R.R., (2008), An NGA model for the average horizontal component of
peak ground motion and response spectra, Earthquake Spectra, Vol. 24, No. 1, pp. 173-215.

Clinton J., G. Cua, V. Huérfano, C. G. von Hillebrandt-Andrade and J. Cruzado (2006) The
Current State of Seismic Monitoring in Puerto Rico. Seismological Research Letter Vol. 77,
no. 5, pp. 532-543.

Cornell C.A. (1968). Engineering seismic risk analysis, Bull. Seismol. Soc. Am. 58, 1583–1696.

Cornell C.A. and Winterstein, S.R., (1986), Applicability of the Poisson Earthquake-Occurrence
Model. Seismic Hazard Methodology for the Central and Eastern United States, Vol. 1:
Probabilistic Seismic Hazard Methodology. EPRI Research Report, Project P101-38.

Douglas J. and Mohais R. (2008). Comparing predicted and observed ground motions from
subduction earthquakes in the Lesser Antilles. J. Seismol.
A135

Faccioli E. et al. (1983). Recommendations on the level of lateral forces to be used for earthquake-
resistant design in the Caribbean. CCEO Conference on Earthquake and Wind Engineering,
Port of Spain, Trinidad.

Feuillet N., I. Manighetti and P. Tapponnier (2002) Arc parallel extension localization of volcanic
complexes in Guadeloupe, Lesser Antilles. Journal of Geophysical Research 107 (B12), 2,
331.

Flinch J.F., Rambaran, V., Ali, W., De Lisa, V., Hernandez, G., Rodrigues, K. and Sams, R. (2000)
Structure of the Gulf of Paria pull-apart basin (Eastern Venezuela-Trinidad). – In: Mann, P.
(Ed.): Caribbean Sedimentary Basins, Elsevier Basins of the World, Elsevier.

Gardner J. K., Knopoff L. (1974). Is the sequence of earthquakes in southern California, with
aftershocks removed, Poissonian?. Bulletin of the Seismological Society of America, Vol. 64,
No. 5, pp. 1363-1367.

Gibbs T. (2001). Natural Hazards in the Caribbean, in Proceedings USAID/OAS PGDM building
inspector training workshop, Antigua.

Grunthal G. and Wahlstrom (2003). An Mw based earthquake catalogue for central, northern and
northwestern Europe using a hierarchy of magnitude conversions, Journal of Seismology 7:
507-531.

Gusev A.A. (1991): Intermagnitude relationships and asperity statistics, Pure and Applied
Geophysics, Volume 136, Number 4 / December, pp. 515-527

Gusev A.A. and Melnikova V.N. (1992). Relationships between Magnitude Scales for global and
kamchatkan earthquakes, Volc. Seism, Vol. 12(6), pp.723-733.

Gutenberg B. and Richter, C.F., (1942) “Earthquake magnitude, intensity, energy, and
acceleration”, Bulletin of the Seismological Society of America, Vol. 32, No. 3, pp. 163-191.

Gutenberg B. and Richter, C.F., (1954), Seismicity of the Earth and Associated Phenomena.
Princeton University Press, New Jersey, 2º Ed., 310pp.

Gutenberg B. and Richter, C.F., (1956), Earthquake magnitude, intensity, energy and acceleration,
Bulletin of the Seismological Society of America, Vol. 46, pp. 105-145.

Hanks T.C. (1977), Earthquake stress drops, ambient tectonic stresses and stresses that drive plate
motions, Pure Appl. Geophys., 115, 441-458.
A136

Jordan T.H. (1975). The present-day motions of the Caribbean Plate. J. geophys. res. 80. 4433-
4439.

Joschi G.C. and Lal Sharma M. (2008), Uncertainties in the estimation of Mmax, J. Earth Syst. Sci.
117, S2, November, pp. 671-682.

Kanamori H. (1977). The energy release in great earthquakes, J. Geophys. Res., 82, 2981-2987.

Kanamori H. (1983). Magnitude Scale and quantification of earthquakes, Tectonophysics, 93, 185-
199.

Kanamori H. and Anderson D.L. (1975). Theoretical basis of some empirical relations in
seismology, Bull. Seismol. Soc. Amer. , 65, 1073-1096.

Kanamori H. and Hanks T.C. (1979). A moment magnitude scale, J. Geophys. Res., 84: 2348-
2350.

Kanamori H. and Jennings P.C. (1978). Determination of local magnitude, ML, from strong-
motion accelerograrns, Bull. Seismol. Soc. Am., 68: 471-485.

Kanno T., Narita A., Morikawa N., Fujiwara H., Fukushima Y., (2006), A new attenuationn
relation for strong motion in Japan based on recforded data, Bulletin of the Seismological
Society of America, Vol. 96, N. 3, pp. 879–897.

Latchman J. (2009) Tobago and Earthquakes. Ph.D. Thesis. The University of the West Indies.

Lienkaemper J.J. (1984). Comparison of two surface-wave magnitude scales: M of Gutenberg and
Richter (1954) and MS of "Preliminary determination of epicenters”, Bulletin of the
Seismological Society of America 1984 74: 2357-2378.

Lin P.S. and Lee C.T., (2008), Ground-Motion Attenuation Relationships for Subduction-Zone
Earthquakes in Northeastern Taiwan, Bulletin of the Seismological Society of America, Vol.
98, No. 1, pp. 220–240.

Margaris B.N. and Papazachos C.B (1999). Moment-magnitude relations based on strong-motion
records in Greece, Bulletin of the Seismological Society of America 89: 442-455.

McCann W. R. (1985). On the earthquake hazards of Puerto Rico and the Virgin Islands, Bull.
Seism. Soc. Am., v.75, 251-262.

McGuire R., (1993), Computations of seismic hazard, Ann. Geofis., 36 (3-4), 181-200.
A137

McQueen C.M. (1997). An evaluation of seismic hazard in the Caribbean and Central America
using three probabilistic methods. PhD thesis. University of Lancaster, 342 pp.

McVerry G.H., Zhao J.X., Abrahamson A., Somerville P.J., (2006), New Zealand acceleration
spectrum attenuation relations for crustal and subduction zone earthquakes, Bulletin of New
Zealand Society for earthquake engineering, Vol. 39, N. 1.

Mendoza C. and W. McCann (2006) Improving the Seismic Hazard Model for Puerto Rico
through Seismic Tomography and a Reliable Microearthquake Catalog with Recalculated
Magnitudes and Calibrated Hypoentral Error Estimates. Research supported by USGS No.
05HQGR0012.

Molnar P., Sykes L. R. (1969). Tectonics of the Caribbean and Middle America regions from local
mechanisms and seismicity, Geol. Soc. Am. Bull. 80, 1639.

Morgan F.D., Wadge G., Latchman J., Aspinall W.P., Hudson D. and Samstag F. (1988). The
earthquake Hazard alert of September 1982 in southern Tobago. Bull. Seismol. Soc. Am. 78
4, pp. 1550-1562.

Mueller C., Frankel A., M. Petersen and E. Leyendecker (2004) Documentatioin for 2003 USGS
Seismic Hazard Maps for Puerto Rico and the U.S. Virgin Islands. USGS Open File Report
03-379.

Papazachos C.B. et al. (1997). Toward a homogeneous moment-magnitude determination for


earthquakes in Greece and the surrounding area, Bulletin of the Seismological Society of
America 87: 474-483.

Papazachos C.B. et al. (2002). Uncertainties in the estimation of earthquake magnitudes in Greece,
Journal of Seismology 6: 557-570.

Pasyanos M.E et al. (1996). Toward Real-Time Estimation of Regional Moment Tensors, Bull.
Seism. Soc. Amer., 86, 1255-1269.

Pereira J. and Gay D. (1978). An engineering risk analysis for Trinidad and Jamaica, in Proceedings
1st Caribbean Earthquake Engineering Conference, Port-of-Spain, Trinidad, 71-92.

Pérez O.J., Bilham R., Bendick R., Velandia J., Moncayo C., Hoyer M. and Kozuch M. (2001)
Velocity fields across the southern Caribbean Plate boundary and estimates of the
Caribbean/South American plate motion using GPS geodesy 1994-2000. Geophysical
Research Letters 28 (15) , pp. 2987-2990.
A138

Ristau J. et al. (2005). Moment Magnitude – Local Magnitude Calibration for Earthquakes in
Western Canada, Bulletin of the Seismological Society of America, Vol. 95, No. 5, pp. 1994–
2000, October.

Robson G.R. (1964). An earthquake catalogue for the Eastern Caribbean, 1530-1960, Bull.
Seismol. Soc. Am., Vol. 54, No. 2, pp. 785-832.

Russo R. and R. Speed (1992) Oblique collision and tectonic wedging of the South America
continent and Caribbean terranes. Geology, 20, pp. 447-450.

Russo R. M., Silver P. G., Franke M., Ambeh W. B. and James D. E. (1996). Shear-wave splitting
in northeast Venezuela, Trinidad, and the eastern Caribbean. Physics of the earth and
Planetary Interiors 95. pp. 251–275.

Russo R. M., Speed R. C., Okal E. A., Shepherd J. B. and Rowley K. C. (1993). Seismicity and
Tectonics of the Southeastern Caribbean, J. Geophys. Res., 98(B8), 14,299-14,319.

Sadigh K., Chang C.-Y., Egan J.A., Makdisi F., Youngs R.R., (1997). Attenuation relationships for
shallow crustal earthquakes based on Califirnia strong motion data, Seismological Research
Letters, Vol. 68, N. 1, pp. 58-7.

Salazar W. (2004). Evaluation of source, path and site effects on earthquake ground motions using
an inversion technique for subduction and upper-crustal earthquakes in El Salvador, Central
America. Doctoral Thesis, Tokyo, Institute of Technology, Japan. pp. 1-181.

Scordilis E.M. (2006). Empirical global relations converting MS and mb to moment magnitude,
Journal of Seismology, 10: 225-236.

Seismic Research Centre SRC (2009a) Web Page: http://www.uwiseismic.com.

Seismic Research Centre SRC (2009b) Eastern Caribbean Earthquake Source Zones and *b-
values. The University of the West Indies, Trinidad and Tobago, W.I.

Shedlock K.M. (1999). Seismic hazard map of North and Central America and the Caribbean,
Ann. Geofis. 42, 977–997.

Shedlock K.M. and Tanner J.G. (1999). Seismic hazard map of the western hemisphere. Ann.
Geofis. 42, 1199–1214.

Shepherd J.B. (1993). Seismic hazard in the eastern Caribbean. In: McGuire, R.K. (Ed.), The
Practice of Earthquake Hazard Assessment. IASPEI, Denver, pp. 51–55.
A139

Shepherd J.B. and Aspinall W.P. (1983). Seismicity and earthquake hazard in Trinidad and Tobago,
West Indies, Earthquake Eng. Struct. Dyn. 11, 229-250.

Shepherd J.B. and Lynch L.L. (1992). An earthquake catalogue for the Caribbean. Part 1: The pre-
instrumental period 1502:1900. Proceedings of the meeting of the steering committee, Latin
America and Caribbean Seismic Hazard Project. Melbourne, Florida, April 1992 95-158.

Shepherd J.B. and Lynch L.L. (2003). Seismic Hazard Assessment and Microzonation of Trinidad
and Tobago. Final report submitted to the National Emergency Management Agency of
Trinidad and Tobago. Seismic Research Unit, University of West Indies, St. Augustine,
Trinidad, pp. 21.

Shepherd J.B. et al. (1997). Seismic hazard in Latin America and the Caribbean, Seismic Hazard
Maps for the Caribbean, IRDC, Ottawa, vol. 5, pp. 15.

Stafford P. J., Strasser F.O., Bommer,J. J. (2008), An evaluation of the applicability of the NGA
models to ground-motion prediction in the Euro-Mediterranean region, Bull Earthquake
Eng Vol. 6, pp. 149–177, DOI 10.1007/s10518-007-9053-2.

Stepp J.C., (1973), Analysis of completeness of the earthquake sample in the Puget Sound area. In
“Seismic Zoning”, ed. Harding, S.T., NOAA Tech. Report ERL 267-ESL30, Boulder,
Colorado.

Stewart J.P, Archuleta, R.J., Power, M.S., (2008), Special Issue on the Next Generation
Attenuation Project, Earthquake Spectra, Vol. 24, N. 1.

Tanner J.G. and Shedlock K.M. (2004). Seismic hazard maps of Mexico, the Caribbean, and
Central and South America, Tectonophysics 390, 159-175.

Tanner J.G. and Shepherd J.B. (1997). Seismic hazard in Latin America and the Caribbean, Project
Catalog and Seismic Hazard Maps, IRDC, Ottawa, vol. 1, p. 143.

Taylor L.O. et al. (1978). A preliminary analysis of seismic risk in the Lesser Antilles, in
Proceedings 1st Caribbean Earthquake Engineering Conference, Port-of-Spain, Trinidad, 10-
27.

Thatcher W. and Hanks T.C. (1973), Source parameters of southern California earthquakes, J.
Geophys. Res., 78, 8547-8576.

Tomblin J. (1978). Earthquake parameters for engineering design in the Caribbean, in Proceedings
1st Caribbean Earthquake Engineering Conference, Port-of-Spain, Trinidad, 10-27.
A140

Uhrhammer R.A. and Collins E.R. (1990): Synthesis of Wood-Anderson seismograms from
broadband digital records, Bulletin of the Seismological Society of America 80: 702-716.

Uhrhammer R.A. et al. (1996): Determination of local magnitude using BDSN broadband records,
Bulletin of the Seismological Society of America 86: 1314-1330.

Vogt J. (2008). Comprendre et compléter un catalogue de séismes : le cas de Trinidad. In: J.


Frechet, M. Meghraoui and M. Stucchi (eds), Multidisciplinary Advances in Historical
Seismology, Springer.

Wahlstrom R. and Grunthal G. ( 2000): Probabilistic seismic hazard assessment (horizontal PGA)
for Sweden, Finland and Denmark using different logic tree approaches, Soil Dynamics and
Earthquake Engineering 20, 45-58.

Weber J., Dixon T., Demets C., C. Ambeh, W., Jansma, P., Mattioli, G., Bilham, R., Saleh, J. And
Perez, O. (2001) A GPS Estimate of the Relative Motion between the Caribbean and South
American Plates, and Geological Implications for Trinidad and Venezuela. Geology, 29, pp.
75-78.

Weber J.C. (2009). Neotectonics in the Trinidad and Tobago, West Indies segment of the
Caribbean-South American plate boundary. Occasional Papers of the Geological Institute of
Hungary, volume 204.

Youngs R.R., Silva W.J., Humphrey J.R., (1997), Strong ground motion attenuation relationships
for subduction zone earthquakes, Seismological Research Letters, Vol. 68, N. 1, pp. 58-7.

Zhao J. X., Zhang J., Asano A., Ohno Y., Oouchi T., Takahashi T., Ogawa H., Irikura K., Thio H.
K., Somerville P. G., Fukushima Y., and Fukushima Y., (2006), Attenuation Relations of
Strong Ground Motion in Japan Using Site Classification Based on Predominant Period,
Bulletin of the Seismological Society of America, Vol. 96, No. 3, pp. 898–913.
APPENDIX I
A142

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


1 *SISRA 1530 9 1 14 30 10.70 -64.10 10.0 8.0
2 *SAL 1690 4 5 0 0 17.00 -62.00 51** 7.5
3 *SAL 1727 11 9 0 0 14.50 -61.00 75.0 6.6
4 *SISRA 1766 10 21 9 0 11.00 -62.50 100.0 7.8
5 *SISRA 1812 3 26 20 7 10.60 -66.90 6.0 6.4
6 *SAL 1816 12 22 0 0 13.00 -60.00 50.0 6.6
7 *SAL 1822 12 1 0 0 12.00 -61.75 100.0 6.2
8 *SAL 1825 9 21 1 45 11.00 -62.00 75.0 6.6
9 *SAL 1827 12 21 0 0 14.50 -60.75 51** 6.6
10 *SAL 1831 12 4 1 45 11.50 -62.00 100.0 6.4
11 *SAL 1833 2 9 1 10 17.30 -62.75 100.0 6.6
12 *SAL 1834 11 25 0 0 13.00 -61.00 100.0 6.6
13 *SAL 1837 9 22 11 15 13.25 -60.50 75.0 6.2
14 *SAL 1839 1 11 9 50 14.50 -61.00 100.0 7.8
15 *SAL 1839 8 2 6 25 14.50 -60.50 75.0 6.6
16 *SAL 1843 2 8 0 0 17.00 -62.00 51** 8.0
17 *SAL 1844 1 19 5 40 12.00 -61.75 100.0 6.6
18 *SAL 1844 4 16 0 0 18.75 -66.00 33.0 6.6
19 *SAL 1844 8 30 7 0 13.30 -61.00 100.0 6.8
20 *SAL 1846 9 26 16 40 11.50 -62.00 100.0 6.6
21 Bernard&Lambert (1988) 1851 5 16 13 0 16.20 -61.40 10.0 5.5
22 *SISRA 1853 7 15 0 0 12.10 -63.60 14.0 6.7
23 *SAL 1856 8 28 18 0 18.50 -65.00 33.0 6.4
24 *SAL 1860 10 23 0 0 18.50 -67.50 33.0 6.6
25 *SAL 1863 7 10 0 0 10.80 -62.00 75.0 6.2
26 *SAL 1865 8 30 0 0 18.00 -66.50 33.0 6.1
27 *SAL 1867 11 18 20 0 18.50 -65.00 33.0 7.5
28 *SISRA 1874 8 18 3 0 10.50 -63.10 16.0 6.2
29 *SAL 1874 8 26 11 15 19.00 -66.00 50.0 6.4
30 *SAL 1875 12 9 0 0 19.00 -67.00 50.0 6.4
31 *SISRA 1878 4 13 0 11 10.30 -66.80 13.0 6.2
32 *SAL 1886 7 8 18 0 17.00 -62.00 33.0 6.1
33 *SISRA 1888 1 10 13 0 11.30 -62.20 51** 7.0
34 *SAL 1895 5 20 21 44 18.00 -61.00 33.0 6.6
35 *SAL 1897 4 29 14 15 16.20 -61.50 10.0 6.6
36 *SISRA 1900 10 29 9 11 11.00 -66.00 25.0 7.6
37 *MACRO 1906 2 16 17 25 14.50 -60.50 50.0 7.0
38 *GUTE 1906 12 3 22 59 15.00 -61.00 100.0 7.5
39 *SISRA 1910 1 23 18 49 12.00 -60.50 100.0 7.2
40 *MACRO 1911 11 20 9 45 13.50 -61.00 100.0 5.9
41 *GUTE 1914 10 3 17 22 16.00 -61.00 100.0 7.4
42 *GUTE 1915 10 11 19 33 19.00 -67.00 0.0 6.8
43 *SISRA 1915 12 13 2 30 10.90 -66.80 25.0 5.8
44 *GUTE 1916 4 24 4 26 18.50 -68.00 80.0 7.4
45 *GUTE 1917 7 27 1 1 19.00 -67.50 50.0 7.0
46 *ROR 1918 2 24 23 0 11.00 -62.00 0.0 6.4
47 *GUTE 1918 6 11 12 36 19.00 -62.50 0.0 6.1
48 *GUTE 1918 10 11 14 14 18.50 -67.50 0.0 7.3
49 *GUTE 1919 9 6 9 29 19.50 -64.50 0.0 6.4
50 *GUTE 1919 11 6 7 13 13.50 -58.00 0.0 6.1
51 *GUTE 1920 2 10 22 7 18.00 -67.50 0.0 6.6
52 *ROR 1922 5 11 6 45 11.33 -60.05 33.0 6.2
53 *GUTE 1922 12 18 12 35 19.00 -67.00 0.0 6.4
54 *GUTE 1923 3 15 6 3 20.00 -68.00 0.0 6.0
55 *ROR 1923 8 8 12 1 11.00 -62.76 100.0 6.6
56 USGS-NEIC (Centennial) 1925 7 7 17 43 17.36 -60.74 35.0 6.8
57 *GUTE 1925 9 29 17 33 18.50 -62.00 0.0 6.6
58 *ROR 1926 2 1 1 17 10.88 -62.40 5.0 6.6
59 *GUTE 1927 8 2 0 51 19.00 -64.50 0.0 6.6
60 *ROR 1928 9 27 0 44 11.58 -59.49 30.0 6.6
A143

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


61 *ROR 1929 1 17 11 45 10.35 -63.98 0.0 6.9
62 *MACRO 1930 5 15 16 10 17.00 -62.50 33.0 5.9
63 *GUTE 1930 6 25 12 6 19.00 -64.00 0.0 6.4
64 *GUTE 1933 4 4 12 9 17.50 -60.50 0.0 6.0
65 *MACRO 1934 12 12 14 55 16.50 -62.50 10.0 5.9
66 *ROR 1935 4 10 22 32 10.64 -62.77 102.0 6.6
67 *MACRO 1935 5 6 19 55 16.50 -62.50 10.0 6.2
68 *GUTE 1935 9 15 4 1 19.00 -65.00 0.0 6.0
69 *GUTE 1935 11 10 18 27 16.50 -62.50 100.0 6.4
70 *GUTE 1936 3 20 17 48 18.50 -62.00 50.0 6.0
71 *SISRA 1938 4 13 13 53 12.00 -60.50 0.0 6.0
72 *GUTE 1939 3 7 11 20 18.00 -67.00 0.0 6.0
73 *SISRA 1939 4 20 17 46 13.00 -60.50 130.0 5.8
74 *GUTE 1939 6 12 4 5 20.50 -66.00 0.0 6.4
75 *ROR 1939 10 14 6 2 10.67 -63.17 20.0 6.0
76 *GUTE 1939 12 24 18 53 18.00 -68.00 0.0 6.0
77 *ROR 1940 2 27 12 12 8.51 -60.96 0.0 6.2
78 *ROR 1940 6 23 18 59 10.08 -68.19 20.0 6.0
79 *ROR 1940 7 6 3 40 13.31 -61.48 160.0 6.6
80 *GUTE 1941 1 17 12 35 18.50 -63.50 0.0 6.0
81 *GUTE 1941 3 12 2 53 17.00 -61.00 0.0 6.0
82 *ROR 1942 5 6 22 50 10.85 -64.93 10.0 6.2
83 *GUTE 1943 1 23 13 30 18.00 -61.50 110.0 6.1
84 *GUTE 1943 7 29 3 2 19.25 -67.50 0.0 7.5
85 *GUTE 1943 7 30 1 2 19.25 -67.75 0.0 6.6
86 *GUTE 1943 8 8 0 38 19.00 -68.00 0.0 6.0
87 *GUTE 1943 8 15 0 13 19.00 -68.25 0.0 6.0
88 *GUTE 1944 6 3 7 12 20.00 -63.00 0.0 6.1
89 *GUTE 1944 8 9 4 15 18.50 -67.00 60.0 6.0
90 *GUTE 1945 11 23 8 10 10.00 -62.00 100.0 6.6
91 *ROR 1946 5 21 9 16 14.78 -60.40 49.0 6.2
92 *ROR 1946 7 31 0 29 12.07 -59.91 56.0 6.2
93 *SAE 1950 3 9 10 3 15.63 -60.39 0.0 5.4
94 *SAE 1950 5 24 12 54 17.11 -60.26 11.0 5.1
95 *SAE 1950 7 23 23 32 19.44 -61.58 0.0 5.2
96 *SAE 1950 7 26 8 31 19.02 -67.49 0.0 5.0
97 *SAE 1950 10 6 8 16 19.57 -64.30 43.0 5.2
98 *SAE 1950 10 6 11 20 17.76 -68.11 37.0 5.0
99 *SAE 1950 10 19 3 48 19.36 -64.44 47.0 5.2
100 *SAE 1950 10 20 7 44 19.33 -64.58 55.0 5.0
101 *SAE 1950 12 4 16 21 18.32 -64.23 0.0 5.0
102 *SAE 1950 12 27 23 9 16.89 -62.44 6.0 5.3
103 *SAE 1950 12 27 23 19 17.02 -62.46 28.0 5.4
104 *SAE 1950 12 29 20 16 17.12 -62.54 51** 5.1
105 *SAE 1950 12 31 9 46 17.84 -62.62 36.0 5.0
106 *SAE 1951 6 13 1 6 18.20 -61.67 18.0 5.0
107 *SYKES 1951 6 29 3 49 17.01 -66.40 59.0 5.0
108 *SAE 1951 11 12 9 36 16.73 -61.36 61.0 6.1
109 *SAE 1952 7 21 21 39 18.63 -61.51 0.0 5.2
110 *SAE 1952 8 19 14 3 15.92 -61.06 70.0 5.1
111 *SAE 1952 8 20 8 31 18.95 -67.14 0.0 4.8
112 *SAE 1952 8 27 17 1 19.09 -66.58 52.0 5.0
113 *SAE 1952 11 21 6 10 19.33 -67.90 0.0 5.1
114 *SAE 1952 11 21 13 40 18.88 -63.23 0.0 5.2
115 *ROR 1952 12 31 1 38 11.53 -59.23 40.0 5.4
116 *ROR 1953 3 19 8 27 14.00 -61.24 133.0 7.0
117 *SAE 1954 2 9 8 56 19.12 -65.15 68.0 5.0
118 *SYKES 1954 4 7 21 25 19.63 -66.00 9.0 4.8
119 *SAE 1954 9 18 4 55 16.10 -60.16 0.0 4.9
120 *ROR 1954 12 4 18 31 10.79 -61.36 46.0 6.3
A144

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


121 *SISRA 1955 5 26 6 59 10.46 -64.58 0.0 5.0
122 *SAE 1956 2 13 15 33 18.93 -66.30 50.0 5.0
123 *SAE 1956 3 21 7 43 19.41 -64.54 45.0 4.8
124 *SYKES 1957 9 14 23 16 16.16 -61.04 59.0 5.2
125 *ROR 1957 10 2 12 27 10.88 -62.79 10.0 5.9
126 *TRN 1957 10 3 6 39 10.65 -62.82 100.0 5.3
127 *ROR 1957 10 4 5 26 10.86 -62.77 6.0 6.7
128 *ROR 1957 10 6 0 54 10.88 -62.68 10.0 5.6
129 *SISRA 1957 10 14 8 17 11.02 -62.97 0.0 4.9
130 *SAE 1957 10 23 4 38 18.81 -64.33 33.0 5.0
131 *SISRA 1957 12 18 2 11 10.87 -63.11 0.0 5.0
132 *ROR 1957 12 25 16 26 10.46 -62.55 22.0 6.1
133 *TRN 1957 12 30 13 28 10.39 -62.76 1.0 5.0
134 *TRN 1958 1 2 22 35 11.27 -60.87 45.0 5.2
135 *SAE 1958 3 25 18 42 17.99 -64.90 36.0 5.1
136 *SYKES 1958 5 16 20 21 18.13 -68.21 64.0 4.8
137 *SAE 1958 11 24 22 27 17.11 -61.60 51.0 4.5
138 *SAE 1959 1 8 1 33 15.25 -61.26 140.0 6.6
139 *SAE 1959 3 29 5 40 18.43 -65.59 70.0 5.0
140 *SAE 1959 4 25 6 6 19.59 -65.75 0.0 4.8
141 *SAE 1959 5 15 18 37 19.60 -63.57 0.0 4.8
142 *SISRA 1959 5 23 9 55 10.64 -62.97 35.0 5.0
143 *SAE 1959 5 24 0 9 18.90 -64.35 0.0 4.8
144 *SAE 1959 5 26 5 27 17.04 -61.67 46.0 5.5
145 *SAE 1959 7 1 1 13 18.72 -64.52 25.0 4.6
146 *SAE 1959 7 21 9 17 18.87 -68.04 55.0 5.8
147 *SAE 1959 8 1 15 40 20.00 -67.36 0.0 4.6
148 *SAE 1959 8 11 8 9 19.20 -64.80 0.0 4.7
149 *SISRA 1959 8 28 7 48 10.45 -62.15 15.0 5.0
150 *SAE 1959 9 1 10 49 19.93 -65.15 0.0 5.4
151 *SAE 1959 9 2 9 31 20.02 -65.02 0.0 4.8
152 *SAE 1960 4 3 23 57 15.19 -60.35 52.0 5.0
153 *SISRA 1960 5 9 6 45 10.92 -62.96 5.0 4.9
154 *SAE 1960 5 31 11 2 17.72 -61.63 27.0 6.2
155 *SAE 1960 6 2 18 7 17.94 -61.45 0.0 4.9
156 *SISRA 1960 6 18 21 18 10.75 -67.17 0.0 4.6
157 *SAE 1960 9 23 0 20 20.42 -56.67 22.0 4.8
158 *SYKES 1961 1 9 11 7 17.65 -61.27 42.0 5.1
159 *SYKES 1961 1 9 11 11 17.68 -61.07 0.0 5.1
160 *SYKES 1961 1 9 19 22 17.66 -61.32 38.0 5.1
161 *SAE 1961 6 6 0 33 17.65 -61.50 79.0 4.9
162 *SYKES 1961 7 5 5 2 15.09 -60.39 52.0 5.0
163 *SYKES 1961 11 2 23 3 17.13 -62.68 0.0 5.0
164 *SAE 1961 11 7 23 40 17.15 -62.61 0.0 4.8
165 *SAE 1962 1 15 8 22 13.09 -60.29 60.0 4.8
166 *SAE 1962 2 10 19 31 17.95 -62.00 49.0 5.2
167 *SAE 1962 4 7 23 4 14.99 -60.61 65.0 5.2
168 *SAE 1962 5 20 15 1 20.52 -65.91 0.0 5.1
169 *SAE 1962 6 3 17 31 17.85 -61.49 34.0 4.8
170 *SYKES 1962 7 23 22 11 18.77 -65.44 59.0 4.8
171 *SYKES 1962 9 8 13 3 15.93 -60.54 0.0 5.0
172 *SISRA 1962 9 13 14 35 11.69 -61.02 72.0 4.9
173 *SAE 1962 11 24 7 31 10.65 -62.51 24.0 4.9
174 *SAE 1962 12 26 6 12 15.17 -61.34 144.0 5.0
175 *SYKES 1963 1 6 22 33 15.65 -59.61 0.0 5.0
176 *SISRA 1963 1 24 2 52 8.35 -60.93 52.0 5.0
177 *SAE 1963 5 23 7 43 19.20 -64.60 55.0 5.3
178 *SISRA 1963 7 14 5 41 10.44 -62.74 20.0 6.4
179 *ISC 1964 1 5 9 8 16.97 -60.43 40.0 4.5
180 *ISC 1964 1 5 9 13 16.60 -61.10 0.0 4.6
A145

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


181 *SISRA 1964 2 7 7 16 10.30 -67.00 0.0 4.5
182 *ISC 1964 3 14 15 12 15.94 -60.53 21.0 5.4
183 *ISC 1964 7 14 9 55 19.06 -66.44 48.0 4.8
184 *ISC 1964 8 10 1 10 19.00 -67.28 49.0 5.3
185 *SAA 1964 8 10 16 58 9.11 -62.06 58.0 5.4
186 *ISC 1964 8 20 8 37 14.85 -60.47 82.0 5.4
187 *ISC 1964 10 23 1 56 19.80 -56.11 43.0 6.8
188 *ISC 1964 10 23 16 46 19.07 -57.50 25.0 4.8
189 *ISC 1965 2 25 6 28 15.05 -59.89 42.0 4.6
190 *ISC 1965 5 17 7 26 18.18 -67.55 97.0 4.5
191 *SAA 1965 7 11 1 53 10.74 -61.58 17.0 4.5
192 *SISRA 1965 7 15 16 55 11.23 -61.25 10.0 4.9
193 *ISC 1965 7 17 15 52 17.76 -61.58 45.0 4.6
194 *SISRA 1965 7 20 4 28 10.20 -64.45 10.0 4.9
195 *SISRA 1965 7 26 23 6 10.97 -61.94 150.0 4.5
196 *ISC 1965 7 29 8 54 16.59 -60.13 33.0 4.8
197 *SISRA 1965 8 4 2 16 10.10 -61.90 33.0 5.4
198 *SISRA 1965 9 2 8 37 10.51 -61.75 25.0 4.5
199 *USCGS 1965 9 6 4 59 18.60 -67.60 33.0 6.0
200 *SAA 1965 9 18 8 30 10.39 -62.43 33.0 4.5
201 *SISRA 1965 9 23 21 40 8.99 -65.99 10.0 6.0
202 *SISRA 1965 10 3 20 27 10.70 -62.92 60.0 4.5
203 *SISRA 1965 10 12 1 44 8.28 -65.29 10.0 5.1
204 *ISC 1965 10 18 22 48 18.54 -60.83 5.0 4.9
205 *SISRA 1965 11 10 9 24 11.21 -61.77 51** 6.1
206 *USCGS 1965 11 21 23 3 19.30 -67.40 21.0 4.9
207 *SAA 1966 1 9 9 11 11.44 -62.24 163.0 4.8
208 *ISC 1966 1 13 10 30 19.02 -64.74 59.0 4.9
209 *ISC 1966 1 14 8 52 17.82 -62.03 46.0 5.3
210 *ISC 1966 2 1 21 5 17.50 -63.60 10.0 4.9
211 *ISC 1966 2 11 6 36 19.10 -64.81 61.0 4.5
212 *SISRA 1966 2 13 3 43 12.60 -65.50 190.0 4.9
213 *ISC 1966 2 13 6 7 14.11 -61.36 188.0 4.8
214 *SISRA 1966 3 16 16 10 10.46 -63.31 10.0 5.1
215 *ISC 1966 3 20 18 28 21.96 -58.90 10.0 4.8
216 *ISC 1966 4 16 7 12 18.28 -61.86 28.0 4.8
217 *SISRA 1966 5 14 21 18 10.04 -62.72 175.0 4.5
218 *SAA 1966 6 2 6 24 10.40 -62.26 36.0 5.1
219 *ISC 1966 6 12 3 10 18.30 -62.10 10.0 5.3
220 *SISRA 1966 7 9 9 43 10.52 -62.75 25.0 4.5
221 *ISC 1966 7 15 7 59 16.99 -61.49 62.0 5.3
222 *SISRA 1966 8 1 1 46 10.01 -65.31 10.0 5.3
223 *ISC 1966 8 2 6 1 11.09 -62.13 14.0 5.3
224 *SISRA 1966 8 2 8 26 13.72 -61.73 160.0 4.5
225 *SISRA 1966 8 11 18 7 11.05 -60.97 20.0 4.6
226 *SISRA 1966 8 18 17 39 10.41 -63.80 10.0 5.3
227 *ISC 1966 8 28 15 55 19.98 -64.01 0.0 4.6
228 *ISC 1966 9 10 21 58 19.09 -67.94 53.0 5.1
229 *SISRA 1966 9 12 16 38 10.30 -62.25 29.0 5.4
230 *ISC 1966 9 20 9 0 13.00 -59.70 17.0 4.5
231 *ISC 1966 9 24 19 56 9.95 -62.99 68.0 5.3
232 *ISC 1966 10 2 5 33 19.10 -67.20 10.0 4.6
233 *ISC 1966 10 7 1 3 19.01 -64.67 0.0 4.9
234 *ISC 1966 10 8 18 56 17.63 -65.65 0.0 4.8
235 *SAA 1966 10 10 10 3 10.78 -62.59 104.0 4.8
236 *ISC 1966 10 14 1 49 19.26 -67.84 38.0 4.9
237 *SISRA 1966 10 16 11 23 11.34 -64.05 160.0 4.8
238 *ISC 1966 10 31 6 46 19.50 -67.67 33.0 4.8
239 *ISC 1966 10 31 18 23 19.38 -67.75 55.0 4.8
240 *ISC 1966 11 3 11 37 19.13 -67.87 39.0 5.3
A146

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


241 *ISC 1966 11 3 16 24 19.17 -67.92 22.0 6.0
242 *ISC 1966 11 4 10 52 19.25 -67.74 49.0 4.6
243 *ISC 1966 11 9 21 55 19.20 -67.87 42.0 4.9
244 *ISC 1966 11 13 2 51 17.05 -61.94 92.0 5.4
245 *ISC 1966 11 13 11 42 17.24 -61.37 60.0 4.9
246 *ISC 1966 11 22 12 14 19.18 -67.87 37.0 4.6
247 *ISC 1966 12 5 1 59 16.60 -57.40 0.0 5.4
248 *ISC 1966 12 24 2 34 18.69 -64.51 59.0 5.3
249 *ISC 1967 1 2 4 56 19.60 -67.60 96.0 4.5
250 *SISRA 1967 1 4 20 15 10.93 -62.52 94.0 5.4
251 *SISRA 1967 1 13 5 18 13.11 -60.42 20.0 4.5
252 *ISC 1967 1 29 7 1 17.03 -60.88 22.0 4.5
253 *ISC 1967 2 21 4 16 19.14 -67.97 44.0 5.1
254 *ISC 1967 2 21 13 26 13.96 -59.80 0.0 4.9
255 *ISC 1967 3 11 20 15 10.70 -62.37 115.0 4.5
256 *ISC 1967 4 5 18 10 15.10 -60.55 0.0 4.8
257 *ISC 1967 4 5 23 25 18.60 -63.60 1.0 4.5
258 *ISC 1967 4 11 12 42 18.96 -62.59 4.0 4.9
259 *ISC 1967 4 12 19 1 19.40 -63.58 28.0 4.6
260 *SISRA 1967 4 20 9 54 10.94 -62.40 85.0 4.5
261 *SISRA 1967 4 23 19 43 10.10 -62.35 60.0 4.5
262 *SAA 1967 4 26 10 47 11.13 -62.29 131.0 4.6
263 *ISC 1967 5 24 2 10 18.14 -62.26 50.0 4.5
264 *SISRA 1967 5 31 11 38 12.48 -60.31 70.0 5.1
265 *ISC 1967 6 13 20 1 19.10 -66.30 79.0 4.5
266 *ISC 1967 6 14 22 9 19.61 -66.53 0.0 4.6
267 *SISRA 1967 6 15 0 35 10.57 -65.34 10.0 4.5
268 *ISC 1967 6 16 6 47 19.53 -66.30 47.0 4.5
269 *ISC 1967 7 6 18 32 18.96 -61.94 15.0 5.1
270 *SISRA 1967 7 15 0 35 10.60 -65.30 31.0 4.5
271 *ISC 1967 7 25 20 52 19.70 -63.45 32.0 4.5
272 *SISRA 1967 7 30 0 0 10.68 -67.40 26.0 6.6
273 *ISC 1967 9 3 22 45 18.64 -67.67 0.0 5.6
274 *ISC 1967 9 12 0 32 18.50 -63.30 64.0 5.1
275 *ISC 1967 9 21 0 7 19.20 -62.50 0.0 5.3
276 *ISC 1967 9 25 8 51 17.63 -61.61 57.0 4.8
277 *ISC 1967 9 26 11 0 19.50 -60.00 0.0 6.1
278 *ISC 1967 10 14 3 31 17.33 -60.89 42.0 5.3
279 *ISC 1967 10 17 6 5 16.90 -61.10 54.0 4.5
280 *ISC 1967 10 26 12 21 17.72 -60.98 46.0 5.3
281 *ISC 1967 10 26 13 44 17.63 -61.08 59.0 5.3
282 *SISRA 1967 11 3 22 4 11.18 -61.78 87.0 5.1
283 *ISC 1967 11 4 12 32 17.73 -60.93 0.0 4.6
284 *ISC 1967 11 4 15 11 17.83 -61.01 33.0 4.8
285 *ISC 1967 11 28 3 21 18.47 -62.36 52.0 4.8
286 *ISC 1967 11 29 1 23 18.57 -62.39 6.0 4.9
287 *SISRA 1967 11 29 19 42 12.27 -60.37 85.0 4.5
288 *ISC 1967 12 24 20 3 17.42 -61.19 42.0 6.7
289 *ISC 1967 12 24 21 32 17.61 -61.26 5.0 6.3
290 *ISC 1968 3 19 2 19 15.06 -60.47 57.0 4.8
291 *ISC 1968 3 26 17 38 14.90 -60.40 11.0 4.5
292 *ISC 1968 3 29 20 32 18.73 -64.78 61.0 4.5
293 *ISC 1968 4 13 1 15 19.03 -66.86 33.0 5.1
294 *SAA 1968 6 11 15 49 10.86 -62.27 104.0 4.5
295 *SISRA 1968 8 31 1 49 14.78 -60.80 93.0 4.5
296 *ISC 1968 9 1 19 6 18.12 -68.19 107.0 4.5
297 *ISC 1968 9 3 15 37 20.58 -62.30 34.0 5.8
298 *SAA 1968 9 20 6 0 10.79 -62.68 105.0 6.5
299 *SISRA 1968 9 22 15 4 10.87 -62.69 104.0 4.5
300 *SAA 1968 10 5 12 26 10.89 -62.62 102.0 4.5
A147

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


301 *SISRA 1968 10 12 23 21 10.81 -62.63 97.0 4.6
302 *ISC 1968 10 31 14 30 17.96 -67.53 45.0 4.5
303 *SAA 1968 12 2 5 48 10.89 -62.77 116.0 4.5
304 *ISC 1969 1 3 5 14 18.73 -64.75 91.0 4.5
305 *ISC 1969 5 15 20 43 16.75 -61.39 57.0 6.0
306 *ISC 1969 6 30 18 36 20.03 -64.14 25.0 4.9
307 *ISC 1969 7 27 23 49 19.97 -64.24 27.0 4.6
308 *ISC 1969 7 29 0 40 19.95 -64.18 35.0 4.8
309 *ISC 1969 8 1 13 6 18.85 -64.52 53.0 4.8
310 *ISC 1969 8 11 20 16 20.06 -64.29 31.0 4.6
311 *ISC 1969 8 11 20 46 19.99 -64.36 10.0 4.6
312 *SISRA 1969 9 4 12 2 12.04 -64.95 38.0 4.5
313 *ISC 1969 10 14 5 36 19.43 -65.52 40.0 4.5
314 *SAA 1969 10 22 12 52 10.86 -62.49 80.0 5.4
315 *ISC 1969 12 1 22 13 16.68 -60.80 47.0 5.6
316 *SISRA 1969 12 14 22 39 11.51 -65.00 24.0 4.6
317 *ISC 1969 12 25 21 32 15.79 -59.64 1.0 7.0
318 *ISC 1969 12 25 22 13 15.78 -59.64 42.0 4.6
319 *ISC 1969 12 25 22 16 16.24 -59.70 32.0 4.5
320 *ISC 1969 12 25 22 17 16.19 -59.73 12.0 4.9
321 *ISC 1969 12 25 22 26 15.84 -59.67 4.0 5.4
322 *ISC 1969 12 25 22 31 16.08 -59.79 6.0 6.1
323 *ISC 1969 12 26 0 11 16.13 -59.64 0.0 4.6
324 *ISC 1969 12 26 4 30 15.93 -60.00 70.0 4.5
325 *ISC 1969 12 26 5 21 16.08 -59.71 32.0 4.5
326 *ISC 1969 12 26 5 34 16.08 -59.68 25.0 4.9
327 *ISC 1969 12 26 8 46 15.74 -59.59 35.0 4.9
328 *ISC 1969 12 26 10 34 16.14 -59.79 43.0 5.3
329 *ISC 1969 12 26 10 46 16.28 -59.68 9.0 4.8
330 *ISC 1969 12 26 16 43 15.97 -59.53 1.0 4.5
331 *ISC 1969 12 26 20 3 15.79 -59.56 9.0 5.3
332 *ISC 1969 12 26 10 15 15.84 -59.63 33.0 4.6
333 *ISC 1969 12 27 4 5 16.20 -60.20 31.0 4.5
334 *ISC 1969 12 27 4 21 16.21 -59.75 36.0 4.5
335 *ISC 1969 12 27 9 55 16.15 -59.68 23.0 4.8
336 *ISC 1969 12 27 11 39 15.74 -59.65 25.0 4.6
337 *ISC 1969 12 27 13 8 16.15 -59.62 12.0 4.6
338 *ISC 1969 12 27 14 3 16.17 -59.61 9.0 5.3
339 *ISC 1969 12 27 15 43 16.15 -59.68 19.0 5.3
340 *ISC 1969 12 27 16 50 16.12 -59.70 15.0 4.5
341 *ISC 1969 12 28 23 23 16.20 -59.63 13.0 5.1
342 *ISC 1969 12 29 0 51 16.18 -59.74 33.0 5.6
343 *ISC 1969 12 29 1 29 16.19 -59.66 36.0 4.5
344 *ISC 1969 12 29 2 0 16.17 -59.70 35.0 4.5
345 *ISC 1969 12 29 13 55 16.03 -59.75 9.0 5.1
346 *ISC 1969 12 29 13 55 16.10 -59.74 33.0 5.1
347 *ISC 1969 12 29 14 13 16.17 -59.70 20.0 4.9
348 *ISC 1969 12 31 5 41 16.05 -59.59 21.0 4.8
349 *ISC 1969 12 31 6 49 15.90 -59.78 62.0 4.6
350 *ISC 1970 1 1 18 45 15.98 -59.76 26.0 4.6
351 *ISC 1970 1 2 1 19 16.12 -59.66 21.0 4.8
352 *ISC 1970 1 2 7 38 15.84 -59.48 13.0 4.5
353 *ISC 1970 1 3 13 31 15.79 -59.60 33.0 4.6
354 *ISC 1970 1 5 9 9 16.05 -59.62 16.0 5.1
355 *ISC 1970 1 5 9 13 16.17 -59.66 25.0 5.1
356 *ISC 1970 1 5 9 25 16.12 -59.62 13.0 4.6
357 *ISC 1970 1 6 0 40 16.19 -59.65 16.0 4.5
358 *ISC 1970 1 6 12 56 15.86 -59.71 13.0 5.1
359 *ISC 1970 1 7 7 56 15.86 -59.78 37.0 5.6
360 *ISC 1970 1 8 8 43 16.88 -61.14 53.0 4.8
A148

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


361 *ISC 1970 1 8 11 43 15.88 -59.47 5.0 4.6
362 *ISC 1970 2 8 21 39 16.05 -59.70 42.0 4.5
363 *ISC 1970 2 9 11 26 16.84 -61.25 60.0 4.6
364 *ISC 1970 2 13 10 54 18.20 -65.10 63.0 4.6
365 *ISC 1970 2 19 11 51 16.10 -59.67 3.0 4.5
366 *ISC 1970 3 11 23 38 15.99 -59.98 61.0 4.6
367 *ISC 1970 3 17 0 59 16.32 -59.71 33.0 4.5
368 *ISC 1970 4 9 8 4 15.80 -59.45 10.0 4.5
369 *ISC 1970 4 18 6 42 18.68 -62.86 54.0 4.5
370 *ISC 1970 4 23 1 58 16.11 -59.71 41.0 4.5
371 *ISC 1970 5 8 17 41 19.56 -64.71 36.0 4.5
372 *ISC 1970 5 20 17 29 16.20 -59.67 24.0 4.6
373 *ISC 1970 6 13 8 43 19.29 -65.50 60.0 4.5
374 *ISC 1970 7 8 4 49 18.00 -64.67 148.0 6.1
375 *ISC 1970 7 16 7 8 14.53 -61.07 161.0 4.5
376 *ISC 1970 8 16 8 33 19.21 -65.04 42.0 4.5
377 *ISC 1970 8 21 6 35 16.11 -60.00 45.0 4.5
378 *ISC 1970 11 8 23 10 18.69 -64.57 64.0 4.6
379 *ISC 1970 12 3 15 54 19.20 -64.72 36.0 4.5
380 *ISC 1970 12 4 8 37 17.01 -60.95 33.0 4.5
381 *ISC 1971 2 22 6 4 19.27 -67.90 33.0 4.5
382 *ISC 1971 3 20 2 39 14.23 -60.58 82.0 5.1
383 *ISC 1971 6 12 19 18 18.94 -64.28 28.0 4.8
384 *ISC 1971 6 26 15 47 19.02 -68.03 51.0 5.1
385 *ISC 1971 6 27 8 31 19.10 -67.97 34.0 4.9
386 *ISC 1971 6 30 15 50 15.20 -60.56 65.0 4.5
387 *ISC 1971 7 8 5 54 19.21 -67.97 50.0 4.6
388 *ISC 1971 7 8 14 39 19.19 -64.51 33.0 4.5
389 *SISRA 1971 12 26 18 54 12.13 -60.42 53.0 4.5
390 *SAA 1972 1 3 7 25 10.59 -62.79 69.0 4.6
391 *SAA 1972 2 25 20 17 10.86 -62.57 129.0 4.5
392 *SAA 1972 3 1 3 51 11.16 -62.35 138.0 4.5
393 *SISRA 1972 7 24 10 51 10.68 -65.61 7.0 5.6
394 *SISRA 1972 8 29 3 29 11.00 -62.35 84.0 5.3
395 *SISRA 1972 9 8 2 6 12.06 -61.64 18.0 4.5
396 *ISC 1972 10 30 1 50 22.35 -61.97 0.0 4.6
397 *ISC 1972 11 4 1 33 17.36 -60.77 39.0 4.5
398 *ISC 1972 11 4 1 38 17.39 -60.79 38.0 4.5
399 *ISC 1972 11 9 1 12 18.10 -61.83 33.0 4.5
400 *ISC 1973 2 17 16 2 16.96 -61.48 63.0 5.1
401 *ISC 1973 3 22 14 0 15.34 -61.29 151.0 4.8
402 *ISC 1973 4 1 18 25 19.22 -64.36 35.0 4.5
403 *ISC 1973 7 8 16 59 15.92 -60.70 26.0 4.8
404 *ISC 1973 8 5 19 59 14.35 -60.89 113.0 4.5
405 *SISRA 1973 9 7 18 13 11.14 -61.27 75.0 4.5
406 *SISRA 1973 10 7 17 6 9.01 -61.16 52.0 4.6
407 *ISC 1973 10 16 0 37 19.45 -64.28 56.0 4.5
408 *ISC 1973 11 25 21 53 18.87 -64.60 49.0 4.5
409 *SAA 1973 11 27 11 55 10.22 -61.50 56.0 4.5
410 *ISC 1974 4 30 20 12 12.07 -59.93 59.0 4.5
411 *SISRA 1974 6 12 16 25 10.61 -63.47 34.0 6.3
412 *ISC 1974 7 16 7 51 14.35 -59.41 44.0 4.5
413 *ISC 1974 8 23 23 55 19.22 -68.04 15.0 4.6
414 *ISC 1974 9 7 19 40 15.15 -60.65 56.0 5.6
415 *ISC 1974 10 8 9 50 17.37 -61.99 41.0 7.3
416 *ISC 1974 10 8 12 49 17.55 -61.88 44.0 4.8
417 *ISC 1974 10 18 0 26 17.56 -62.27 60.0 4.8
418 *ISC 1974 10 21 19 3 17.14 -62.22 109.0 4.5
419 *SISRA 1974 10 29 3 10 10.58 -63.45 33.0 5.8
420 *ISC 1974 11 20 8 33 14.28 -59.37 45.0 4.9
A149

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


421 *SISRA 1974 12 24 19 37 11.02 -62.47 131.0 4.6
422 *ISC 1974 12 30 20 33 17.42 -62.08 50.0 4.5
423 *SAA 1975 2 9 21 18 11.49 -62.67 165.0 4.6
424 *ISC 1975 3 3 9 43 17.21 -60.99 44.0 4.8
425 *ISC 1975 3 15 9 13 14.45 -59.38 77.0 4.5
426 *ISC 1975 3 18 7 26 17.96 -63.54 122.0 4.5
427 *SISRA 1975 4 5 9 34 13.00 -68.00 0.0 5.3
428 *SISRA 1975 4 15 9 47 9.42 -61.47 52.0 5.5
429 *ISC 1975 6 17 5 1 18.51 -66.32 119.0 4.6
430 *SISRA 1975 6 18 2 36 10.71 -63.34 18.0 4.7
431 *SISRA 1975 7 18 5 17 10.94 -64.46 3.0 5.0
432 *SISRA 1975 7 25 8 28 10.42 -63.63 6.0 4.5
433 *ISC 1975 8 3 10 35 19.68 -63.19 32.0 4.9
434 *SISRA 1975 8 16 4 12 9.97 -64.25 44.0 4.5
435 *SISRA 1975 8 24 1 5 10.75 -62.65 111.0 5.1
436 *ISC 1975 9 14 7 24 17.56 -61.88 36.0 4.5
437 *SISRA 1975 12 5 9 31 10.83 -62.67 114.0 4.6
438 *ISC 1976 1 22 12 6 8.77 -60.33 63.0 5.3
439 *SISRA 1976 2 9 2 33 11.63 -62.34 161.0 4.6
440 *ISC 1976 3 10 9 4 16.84 -61.06 56.0 6.0
441 *ISC 1976 6 11 3 41 17.05 -60.56 36.0 5.1
442 *ISC 1976 6 13 19 6 19.07 -67.91 43.0 5.4
443 *ISC 1976 6 14 4 37 19.25 -67.95 17.0 4.5
444 *ISC 1976 6 16 16 35 19.02 -67.92 49.0 4.8
445 *ISC 1976 7 1 3 38 16.60 -61.24 39.0 4.8
446 *ISC 1976 7 15 0 7 19.24 -64.17 43.0 4.6
447 *SAA 1976 8 23 13 56 11.05 -62.34 88.0 4.8
448 *SISRA 1976 10 13 17 35 10.81 -61.53 63.0 4.6
449 *ISC 1976 10 15 19 50 18.93 -64.49 47.0 4.6
450 *SISRA 1976 11 11 19 31 12.45 -61.23 126.0 4.6
451 *SISRA 1976 11 29 13 2 11.81 -61.33 84.0 4.6
452 *ISC 1976 12 28 2 57 22.10 -63.45 33.0 4.8
453 *SBAC 1977 8 14 4 22 10.87 -62.22 96.0 5.7
454 *ISC 1977 8 31 7 43 15.94 -61.42 110.0 4.5
455 *SBAC 1977 9 3 15 25 10.41 -62.17 14.0 4.6
456 *ISC 1977 9 12 23 17 19.35 -64.20 58.0 4.5
457 *ISC 1977 9 13 18 38 15.58 -60.32 12.0 4.6
458 *SBAC 1977 9 18 17 31 10.49 -63.33 17.0 4.8
459 *SBAC 1977 9 21 16 5 10.47 -62.50 0.0 5.1
460 *ISC 1977 11 28 0 30 15.89 -60.86 68.0 4.6
461 *ISC 1978 1 2 14 3 18.54 -67.75 39.0 4.6
462 *SBAC 1978 1 18 1 17 10.35 -62.25 0.0 4.5
463 *ISC 1978 2 22 12 9 19.55 -63.59 12.0 4.6
464 *ISC 1978 5 29 3 9 17.71 -61.59 53.0 5.6
465 *ISC 1978 6 8 23 43 16.24 -61.69 123.0 4.8
466 *ISC 1978 7 17 9 17 17.63 -61.15 33.0 4.5
467 *SISRA 1978 7 23 0 35 11.22 -64.55 0.0 4.5
468 *SBAC 1978 8 5 15 5 11.54 -61.78 139.0 4.5
469 *ISC 1978 9 7 20 36 19.88 -64.79 14.0 4.8
470 *SISRA 1978 9 28 2 47 10.42 -62.47 123.0 4.5
471 *SISRA 1978 11 14 4 43 10.59 -62.42 117.0 4.5
472 *ISC 1978 11 25 7 57 17.43 -65.53 69.0 4.5
473 *ISC 1979 1 22 4 25 19.10 -64.70 51.0 5.3
474 *ISC 1979 1 22 15 52 19.26 -64.61 38.0 4.6
475 *ISC 1979 1 23 0 14 19.35 -64.63 23.0 4.6
476 *ISC 1979 3 18 22 37 19.85 -64.67 14.0 4.9
477 *ISC 1979 4 12 4 54 17.63 -61.59 46.0 4.5
478 *ISC 1979 5 1 3 36 20.32 -64.77 40.0 4.5
479 *ISC 1979 6 4 7 53 19.59 -65.36 40.0 4.5
480 *SISRA 1979 6 16 12 41 13.88 -61.32 153.0 4.5
A150

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


481 *ISC 1979 7 15 10 0 18.09 -61.44 0.0 4.5
482 *ISC 1979 7 19 14 26 19.04 -67.50 56.0 4.5
483 *SBAC 1979 9 13 10 46 10.88 -61.72 35.0 5.1
484 *ISC 1979 10 3 0 31 19.00 -67.83 57.0 4.6
485 *ISC 1979 10 9 22 23 16.68 -61.23 61.0 5.4
486 *ISC 1979 10 10 18 20 16.71 -61.32 59.0 4.6
487 *ISC 1979 11 3 0 31 17.54 -67.69 9.0 4.9
488 *ISC 1979 11 3 15 11 14.69 -61.26 164.0 4.5
489 *SBAC 1980 1 18 7 25 10.86 -62.38 42.0 4.5
490 *ISC 1980 2 14 17 11 18.65 -64.67 68.0 5.0
491 *SBAC 1980 2 28 19 54 12.05 -60.13 0.0 4.5
492 *SBAC 1980 2 29 3 44 12.02 -60.42 49.0 4.6
493 *SBAC 1980 3 24 21 59 10.85 -62.36 42.0 4.6
494 *ISC 1980 5 24 14 56 19.49 -64.34 43.0 4.5
495 *ISC 1980 5 30 14 59 19.31 -66.90 33.0 4.5
496 *SBAC 1980 6 3 9 50 11.51 -61.82 91.0 4.5
497 *ISC 1980 6 8 23 5 14.17 -60.83 102.0 4.5
498 *SBAC 1980 6 19 20 34 10.68 -60.73 38.0 4.5
499 *SBAC 1980 9 5 11 45 12.20 -59.43 56.0 4.9
500 *ISC 1980 12 29 5 43 18.71 -62.89 42.0 4.9
501 *SBAC 1980 12 31 8 46 10.85 -62.22 95.0 4.9
502 *SBAC 1981 1 9 8 32 16.86 -61.41 34.0 5.3
503 *SBAC 1981 3 2 7 54 12.39 -60.12 19.0 5.4
504 *ISC 1981 3 31 18 4 14.89 -60.54 70.0 4.5
505 *SBAC 1981 6 23 22 57 10.75 -63.33 0.0 4.8
506 *SBAC 1981 7 3 18 19 19.02 -62.88 35.0 4.5
507 *SBAC 1981 8 4 22 0 16.84 -61.35 25.0 4.9
508 *ISC 1981 8 29 0 15 18.61 -65.08 47.0 4.6
509 *SISRA 1981 9 19 12 57 10.35 -62.69 76.0 4.5
510 *SBAC 1981 10 10 22 59 18.66 -62.26 34.0 4.9
511 *SBAC 1981 12 3 5 54 10.77 -62.77 103.0 4.5
512 *SBAC 1981 12 4 21 44 10.31 -61.17 50.0 4.8
513 *SBAC 1981 12 5 12 41 17.98 -60.74 0.0 4.5
514 *SBAC 1981 12 25 12 35 10.85 -62.10 103.0 4.9
515 *SBAC 1982 1 8 1 23 15.05 -61.29 192.0 4.6
516 *SBAC 1982 1 20 15 15 13.80 -60.32 70.0 4.6
517 *SBAC 1982 1 30 2 35 16.79 -61.35 49.0 6.1
518 *SBAC 1982 2 10 19 55 19.67 -63.46 37.0 4.5
519 *SBAC 1982 4 5 2 4 15.12 -61.24 22.0 4.8
520 *SBAC 1982 4 8 23 57 18.46 -63.50 75.0 4.5
521 *SBAC 1982 5 9 19 29 19.42 -60.83 33.0 5.3
522 *SBAC 1982 5 10 1 25 10.65 -62.35 90.0 5.4
523 *ISC 1982 6 11 21 57 18.74 -64.41 50.0 4.9
524 *SBAC 1982 9 20 16 26 11.25 -60.83 0.0 4.8
525 *ISC 1982 9 30 13 35 18.73 -64.34 63.0 4.8
526 *ISC 1982 10 25 17 6 11.32 -62.09 105.0 4.8
527 *SBAC 1982 11 23 17 26 10.50 -63.21 4.0 4.9
528 *SBAC 1982 11 23 19 31 17.51 -62.16 97.0 5.4
529 *SBAC 1983 3 3 11 53 14.87 -60.49 59.0 4.8
530 *SBAC 1983 3 8 17 6 10.87 -62.11 83.0 5.9
531 *SBAC 1983 4 5 15 59 17.89 -61.59 42.0 4.8
532 *SBAC 1983 4 11 8 6 10.45 -62.58 29.0 4.6
533 *SBAC 1983 4 11 8 18 10.35 -62.59 45.0 6.1
534 *SBAC 1983 6 25 1 4 16.59 -60.34 12.0 4.5
535 *SBAC 1983 6 30 2 0 18.78 -62.73 0.0 4.6
536 *SBAC 1983 8 23 4 1 18.83 -62.57 7.0 4.6
537 *SBAC 1983 8 26 21 21 20.51 -59.07 1.0 4.8
538 *SBAC 1983 12 19 5 15 18.98 -67.58 14.0 5.6
539 *ISC 1984 1 14 12 38 17.85 -60.81 14.0 4.5
540 *ISC 1984 1 23 5 59 20.29 -65.95 0.0 4.6
A151

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


541 *SBAC 1984 1 23 21 36 10.65 -62.52 114.0 4.5
542 *SBAC 1984 2 8 8 35 10.71 -62.07 35.0 4.5
543 *SBAC 1984 2 11 13 57 12.13 -59.97 60.0 5.7
544 *SBAC 1984 3 8 16 46 10.96 -60.73 52.0 4.5
545 *ISC 1984 3 19 12 14 19.35 -64.94 45.0 5.2
546 *ISC 1984 3 30 7 59 17.38 -59.63 15.0 5.5
547 *ISC 1984 4 5 1 53 19.37 -64.96 41.0 5.2
548 *ISC 1984 4 18 16 11 16.82 -61.17 58.0 5.0
549 *ISC 1984 5 25 0 59 10.37 -62.42 41.0 4.7
550 *ISC 1984 6 9 9 2 19.67 -64.83 57.0 4.5
551 *ISC 1984 7 7 9 29 16.78 -60.57 46.0 5.0
552 *SBAC 1984 8 10 17 47 19.14 -63.15 55.0 4.6
553 *SBAC 1984 8 20 23 29 10.43 -62.26 0.0 5.1
554 *ISC 1984 8 20 23 55 10.45 -62.45 10.0 5.2
555 *SBAC 1984 10 5 10 30 11.72 -59.90 56.0 5.5
556 *ISC 1984 11 27 18 39 16.16 -59.64 25.0 5.5
557 *SBAC 1985 1 20 13 3 10.71 -62.54 28.0 4.5
558 *ISC 1985 3 16 14 54 16.98 -62.46 20.0 6.4
559 *ISC 1985 3 16 14 57 16.83 -62.40 33.0 5.8
560 *SBAC 1985 3 18 5 7 17.00 -62.31 6.0 4.5
561 *ISC 1985 3 19 10 45 16.82 -62.38 15.0 5.6
562 *SBAC 1985 3 19 16 53 17.18 -62.22 48.0 4.6
563 *SBAC 1985 4 13 4 12 17.09 -62.47 17.0 4.6
564 *SBAC 1985 5 19 15 1 17.91 -64.26 14.0 4.5
565 *ISC 1985 6 26 17 10 18.89 -64.61 48.0 5.9
566 *ISC 1985 7 5 6 17 18.44 -63.06 60.0 5.2
567 *ISC 1985 7 21 13 10 19.11 -67.93 22.0 5.7
568 *ISC 1985 7 30 5 22 19.25 -67.78 59.0 5.0
569 *ISC 1985 8 12 11 4 19.15 -64.63 52.0 4.9
570 *SBAC 1985 8 15 1 33 18.64 -64.91 5.0 4.6
571 *ISC 1985 10 29 6 12 18.92 -67.13 37.0 4.9
572 *SBAC 1985 11 28 0 14 11.71 -61.18 41.0 5.5
573 *SBAC 1986 2 12 23 41 16.98 -62.28 5.0 5.6
574 *ISC 1986 2 13 11 17 19.13 -67.91 0.0 4.7
575 *ISC 1986 2 18 2 58 17.98 -66.49 21.0 4.5
576 *ISC 1986 2 28 1 12 9.64 -61.39 45.0 5.2
577 *SBAC 1986 3 25 7 10 10.40 -62.42 8.0 4.5
578 *ISC 1986 4 7 12 35 10.28 -61.65 57.0 4.6
579 *ISC 1986 4 11 12 59 12.26 -59.38 39.0 5.2
580 *ISC 1986 5 7 16 37 10.32 -62.31 18.0 4.8
581 *ISC 1986 5 27 2 33 19.67 -65.19 10.0 4.6
582 *ISC 1986 6 11 13 48 10.60 -62.93 20.0 6.3
583 *ISC 1986 6 25 6 46 13.07 -58.79 16.0 5.0
584 *ISC 1986 7 11 8 0 10.61 -63.26 21.0 5.1
585 *SBAC 1986 8 3 0 8 17.17 -61.84 108.0 4.5
586 *ISC 1986 8 7 10 53 16.27 -61.65 155.0 4.5
587 *SBAC 1986 8 18 20 55 10.64 -67.15 3.0 5.3
588 *ISC 1986 8 25 17 25 10.24 -67.09 10.0 5.1
589 *SBAC 1986 9 12 3 31 17.68 -59.69 29.0 4.5
590 *SBAC 1986 10 7 6 43 9.89 -68.16 34.0 4.9
591 *ISC 1986 10 15 9 56 16.90 -60.90 45.0 4.6
592 *ISC 1986 11 9 22 27 17.28 -61.40 52.0 5.1
593 *ISC 1986 11 14 17 26 10.75 -63.24 10.0 5.0
594 *ISC 1986 12 11 4 46 17.24 -61.38 53.0 5.2
595 *ISC 1987 2 5 7 36 15.87 -60.71 43.0 4.5
596 *ISC 1987 2 7 22 32 10.51 -62.78 48.0 5.4
597 *SBAC 1987 4 13 13 5 22.16 -62.97 4.0 5.3
598 *ISC 1987 4 18 23 57 16.66 -62.58 187.0 4.5
599 *SBAC 1987 5 11 10 18 13.69 -65.35 260.0 4.6
600 *ISC 1987 5 22 14 26 19.26 -64.78 33.0 4.6
A152

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


601 *ISC 1987 5 30 17 55 17.96 -67.15 3.0 4.9
602 *SBAC 1987 6 1 3 40 11.91 -61.76 151.0 5.2
603 *SBAC 1987 7 4 17 16 10.95 -62.05 41.0 4.5
604 *ISC 1987 7 18 5 7 19.10 -68.00 33.0 4.5
605 *ISC 1987 7 27 3 20 10.27 -62.38 10.0 4.8
606 *SBAC 1987 8 8 15 53 7.51 -66.45 33.0 5.6
607 *ISC 1987 8 12 3 9 14.10 -59.21 37.0 5.7
608 *ISC 1987 9 9 3 28 17.13 -60.36 40.0 4.9
609 *SBAC 1987 11 5 5 28 10.86 -62.31 61.0 4.5
610 *SBAC 1987 11 28 0 21 19.27 -63.59 11.0 4.5
611 *ISC 1988 1 31 7 15 10.81 -62.30 100.0 4.5
612 *ISC 1988 2 2 11 40 17.88 -62.27 10.0 4.7
613 *ISC 1988 2 9 2 22 18.60 -62.90 52.0 5.6
614 *ISC 1988 2 9 19 12 18.60 -62.91 52.0 5.5
615 *ISC 1988 3 10 6 17 10.24 -60.54 54.0 6.7
616 *ISC 1988 3 11 6 14 10.53 -60.48 54.0 5.2
617 *ISC 1988 3 11 16 1 10.18 -60.59 47.0 5.2
618 *ISC 1988 3 12 4 32 10.19 -60.62 33.0 5.8
619 *ISC 1988 3 12 23 0 10.23 -60.22 39.0 4.7
620 *SBAC 1988 3 13 17 0 10.10 -60.32 77.0 4.5
621 *ISC 1988 3 16 5 48 10.24 -60.62 55.0 5.7
622 *SBAC 1988 3 16 14 24 10.26 -60.33 58.0 4.5
623 *ISC 1988 3 20 0 54 10.41 -60.30 51.0 4.5
624 *ISC 1988 3 25 16 20 10.14 -60.69 56.0 5.5
625 *SBAC 1988 3 29 13 37 10.50 -65.02 10.0 5.2
626 *SBAC 1988 4 12 19 41 10.65 -62.76 78.0 5.2
627 *SBAC 1988 4 22 4 3 17.22 -61.37 32.0 6.5
628 *ISC 1988 4 30 21 37 10.40 -61.60 52.0 4.6
629 *SBAC 1988 5 8 3 15 10.09 -60.29 34.0 5.2
630 *SBAC 1988 5 17 7 43 10.36 -62.42 3.0 4.5
631 *ISC 1988 5 19 21 12 16.79 -61.16 60.0 4.8
632 *ISC 1988 5 25 16 25 19.08 -64.49 50.0 5.0
633 *ISC 1988 6 16 7 37 19.59 -64.29 10.0 4.9
634 *ISC 1988 6 24 8 57 10.18 -60.56 24.0 5.9
635 *ISC 1988 6 26 7 3 10.34 -60.59 58.0 5.1
636 *SBAC 1988 7 12 17 59 10.37 -62.25 0.0 5.3
637 *ISC 1988 7 14 21 39 16.89 -61.10 51.0 4.8
638 *SBAC 1988 8 4 7 49 19.80 -60.76 170.0 4.5
639 *ISC 1988 9 2 6 18 10.45 -62.52 48.0 5.1
640 *ISC 1988 9 3 12 3 16.84 -61.08 54.0 5.0
641 *ISC 1988 9 26 3 54 10.89 -65.51 11.0 5.0
642 *ISC 1988 10 26 18 24 17.36 -62.83 15.0 5.0
643 *ISC 1988 11 3 19 42 19.09 -67.26 29.0 6.0
644 *ISC 1988 11 4 2 38 19.24 -67.25 35.0 4.5
645 *ISC 1988 11 21 5 19 15.08 -59.57 15.0 5.1
646 *ISC 1988 12 4 0 41 10.17 -60.37 48.0 4.9
647 *ISC 1988 12 21 8 57 19.40 -66.42 39.0 5.2
648 *SBAC 1989 2 3 13 5 11.73 -60.83 55.0 4.5
649 *ISC 1989 3 11 19 31 17.82 -60.87 21.0 5.2
650 *SBAC 1989 4 15 14 26 8.45 -60.68 47.0 5.4
651 *ISC 1989 4 27 6 32 17.80 -65.63 19.0 4.5
652 *ISC 1989 4 30 0 29 11.21 -68.07 10.0 5.0
653 *ISC 1989 5 4 0 22 11.07 -68.26 10.0 5.5
654 *ISC 1989 5 24 1 22 11.06 -65.09 48.0 4.8
655 *ISC 1989 5 30 12 29 19.23 -64.56 41.0 5.0
656 *SBAC 1989 6 23 8 43 17.86 -65.75 33.0 4.9
657 *ISC 1989 6 29 17 21 13.59 -58.58 6.0 4.9
658 *SBAC 1989 6 29 23 18 11.55 -61.04 82.0 4.5
659 *ISC 1989 9 1 14 2 10.44 -63.84 10.0 4.5
660 *ISC 1989 9 24 10 53 10.12 -59.84 47.0 5.2
A153

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


661 *SBAC 1989 9 24 16 56 10.08 -59.64 19.0 4.5
662 *ISC 1989 11 14 1 13 18.24 -67.86 187.0 4.5
663 *SBAC 1990 1 9 19 9 19.69 -64.75 33.0 4.5
664 *ISC 1990 2 3 7 30 18.69 -64.64 60.0 5.0
665 *ISC 1990 2 4 20 7 18.57 -65.71 125.0 4.9
666 *ISC 1990 2 4 21 33 18.98 -65.82 56.0 4.8
667 *SBAC 1990 2 19 11 2 17.94 -65.47 33.0 4.5
668 *SBAC 1990 2 21 18 20 16.90 -62.19 111.0 5.8
669 *ISC 1990 3 1 20 55 16.72 -60.95 43.0 5.3
670 *ISC 1990 3 14 3 33 10.07 -59.75 30.0 5.4
671 *SBAC 1990 3 18 2 46 10.97 -65.85 33.0 4.5
672 *ISC 1990 3 21 18 51 10.84 -65.38 23.0 5.3
673 *SBAC 1990 4 26 2 50 10.14 -62.86 33.0 4.6
674 *ISC 1990 5 19 17 44 13.79 -60.98 19.0 4.7
675 *SBAC 1990 6 7 15 30 11.00 -62.81 10.0 4.5
676 *ISC 1990 7 12 23 8 14.65 -60.45 24.0 5.8
677 *ISC 1990 7 31 9 15 17.70 -61.29 55.0 4.7
678 *ISC 1990 8 27 7 15 17.78 -61.65 41.0 5.0
679 *ISC 1990 9 17 13 23 18.87 -62.74 0.0 5.4
680 *ISC 1990 10 7 2 53 18.90 -62.56 35.0 5.3
681 *ISC 1990 10 7 7 15 18.92 -62.48 0.0 5.6
682 *SBAC 1990 10 26 11 8 18.50 -59.19 33.0 4.6
683 *SBAC 1990 10 27 12 47 18.51 -63.20 63.0 4.8
684 *SBAC 1990 11 18 20 28 18.06 -62.84 91.0 5.3
685 *SBAC 1990 12 20 7 18 17.83 -60.68 12.0 4.5
686 *SBAC 1990 12 22 20 4 11.53 -62.01 161.0 5.1
687 *ISC 1990 12 26 8 12 18.92 -68.11 46.0 4.5
688 *CMT 1991 2 5 23 34 17.31 -61.46 15.0 5.2
689 *ISC 1991 2 5 23 49 17.71 -61.71 50.0 5.4
690 *ISC 1991 3 26 10 35 10.33 -62.38 23.0 5.1
691 *ISC 1991 3 26 21 53 10.33 -62.35 47.0 4.9
692 *ISC 1991 4 19 6 39 16.82 -60.63 30.0 5.5
693 *ISC 1991 5 3 16 51 17.56 -60.99 50.0 4.7
694 *ISC 1991 5 28 19 43 14.88 -60.20 57.0 4.8
695 *ISC 1991 8 10 22 27 21.91 -67.88 10.0 4.7
696 *CMT 1991 8 17 12 39 10.54 -62.20 45.0 5.4
697 *ISC 1991 8 27 11 49 16.14 -60.70 56.0 5.1
698 *CMT 1991 9 21 19 1 11.18 -60.57 50.0 5.3
699 *ISC 1991 10 15 7 45 10.44 -62.68 45.0 5.0
700 *ISC 1991 10 21 19 46 18.02 -68.26 42.0 4.8
701 *CMT 1991 12 30 22 38 17.78 -61.14 15.0 5.3
702 *ISC 1992 1 9 9 7 10.60 -62.67 109.0 4.6
703 *ISC 1992 2 9 12 45 19.39 -67.63 14.0 4.5
704 *ISC 1992 3 19 6 16 17.56 -61.15 38.0 4.9
705 *ISC 1992 4 17 12 11 17.01 -61.66 81.0 4.5
706 *CMT 1992 8 3 2 53 16.08 -60.95 27.0 5.4
707 *ISC 1992 8 19 0 36 11.75 -60.83 71.0 4.5
708 *ISC 1992 9 8 13 53 20.15 -64.56 13.0 4.5
709 *ISC 1992 10 20 16 28 14.97 -60.55 62.0 4.6
710 *CMT 1992 11 23 6 31 18.51 -66.96 15.0 5.3
711 *ISC 1992 11 27 12 17 15.90 -61.30 107.0 4.6
712 *ISC 1992 12 28 19 52 19.45 -64.59 36.0 5.3
713 *ISC 1993 1 13 19 34 19.50 -64.51 17.0 4.8
714 *ISC 1993 1 13 19 53 19.48 -64.46 33.0 4.6
715 *ISC 1993 1 14 18 22 19.56 -64.49 33.0 4.5
716 *ISC 1993 4 20 3 38 18.05 -62.89 75.0 4.5
717 *ISC 1993 4 22 12 31 10.64 -62.55 107.0 4.5
718 *ISC 1993 8 1 3 18 18.25 -63.69 101.0 4.5
719 *CMT 1993 8 1 19 32 17.42 -65.71 15.0 5.3
720 *ISC 1993 8 10 7 37 19.38 -64.90 33.0 4.5
A154

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


721 *ISC 1993 9 11 18 56 10.87 -62.59 120.0 4.6
722 *ISC 1993 9 14 14 0 11.53 -60.35 40.0 4.5
723 *ISC 1993 11 5 2 7 19.03 -66.01 46.0 4.6
724 *ISC 1993 12 11 23 57 19.05 -64.73 56.0 4.5
725 *ISC 1993 12 15 10 20 17.16 -61.23 46.0 4.5
726 *ISC 1994 1 8 22 28 18.25 -64.31 102.0 4.6
727 *ISC 1994 2 25 17 34 19.14 -64.34 48.0 5.3
728 *ISC 1994 4 21 17 46 17.97 -62.91 78.0 4.6
729 *ISC 1994 5 3 16 36 10.34 -60.75 48.0 6.2
730 *ISC 1994 5 3 16 56 10.24 -60.65 42.0 4.8
731 *CMT 1994 6 1 3 13 11.75 -60.64 52.0 5.4
732 *ISC 1994 6 25 6 40 19.18 -66.75 17.0 5.0
733 *CMT 1994 7 23 7 22 16.76 -61.11 41.0 5.4
734 *CMT 1994 8 15 6 15 16.81 -60.76 18.0 5.5
735 *ISC 1994 9 23 2 52 18.46 -61.50 20.0 5.3
736 *ISC 1994 10 10 2 43 11.90 -60.31 15.0 4.5
737 *ISC 1994 11 17 14 59 18.53 -68.26 99.0 4.5
738 *ISC 1994 11 30 2 54 19.54 -64.61 16.0 5.1
739 *CMT 1995 3 8 3 46 16.59 -59.56 15.0 6.1
740 *ISC 1995 5 7 6 41 10.32 -62.41 43.0 4.7
741 *ISC 1995 6 28 23 28 17.67 -61.60 55.0 5.2
742 *CMT 1995 6 30 7 47 17.51 -61.39 20.0 5.4
743 *ISC 1995 7 9 18 50 19.53 -67.21 48.0 5.0
744 *ISC 1995 7 26 9 56 19.17 -64.66 55.0 4.9
745 *ISC 1995 8 11 4 7 17.68 -61.61 1.0 5.0
746 *ISC 1995 9 9 16 17 14.87 -60.63 110.0 4.5
747 *ISC 1995 10 8 4 53 19.00 -66.94 48.0 5.1
748 *ISC 1995 10 21 18 51 17.63 -60.93 9.0 5.2
749 *ISC 1995 12 7 10 28 17.65 -61.58 52.0 5.1
750 *CMT 1995 12 29 14 36 14.39 -59.31 49.0 5.2
751 *ISC 1996 1 1 9 38 11.40 -61.84 97.0 5.3
752 *ISC 1996 3 6 7 42 10.59 -64.31 13.0 5.2
753 *CMT 1996 5 11 2 18 19.54 -65.41 27.0 5.1
754 *ISC 1996 5 19 3 25 10.54 -62.81 5.0 5.2
755 *ISC 1996 7 14 22 40 10.34 -62.43 35.0 5.2
756 *ISC 1996 8 3 13 2 17.01 -60.71 19.0 4.7
757 *ISC 1996 8 16 6 0 16.87 -61.37 50.0 4.8
758 *CMT 1996 9 24 11 42 15.39 -61.22 139.0 5.6
759 *ISC 1996 11 6 2 0 18.75 -64.42 55.0 5.2
760 *ISC 1996 11 25 11 9 19.05 -66.23 57.0 5.1
761 ISC 1997 1 14 18 3 17.32 -61.66 60.0 5.1
762 ISC 1997 1 25 14 52 17.70 -61.63 51.8 5.1
763 ISC 1997 2 19 10 28 19.12 -64.42 51.0 4.5
764 ISC 1997 3 15 18 56 16.01 -60.99 54.6 4.5
765 ISC 1997 3 17 5 13 19.02 -62.79 26.8 5.2
766 ISC 1997 3 17 5 20 19.01 -62.83 29.7 5.2
767 ISC 1997 4 2 6 14 11.47 -61.19 73.4 6.1
768 ISC 1997 4 2 7 21 11.36 -60.83 70.4 4.5
769 ISC 1997 4 5 17 4 11.10 -61.00 55.8 5.0
770 ISC 1997 4 5 21 5 19.08 -63.10 33.0 4.5
771 ISC 1997 4 7 23 25 10.33 -62.22 51.1 4.6
772 ISC 1997 4 8 9 37 11.00 -60.69 7.7 4.6
773 ISC 1997 4 8 17 11 11.06 -61.02 45.2 5.5
774 ISC 1997 4 9 0 50 11.06 -61.06 46.3 5.0
775 ISC 1997 4 10 2 23 11.14 -60.92 1.4 5.1
776 ISC 1997 4 11 6 57 11.03 -60.79 14.1 5.0
777 ISC 1997 4 16 11 42 11.01 -60.67 9.9 4.5
778 ISC 1997 4 22 9 31 11.16 -61.09 47.1 6.7
779 ISC 1997 4 22 9 49 11.06 -61.10 67.1 4.7
780 ISC 1997 4 22 10 11 11.12 -60.99 50.1 6.1
A155

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


781 USGS-PDE 1997 4 22 10 18 11.06 -61.17 5.0 5.3
782 ISC 1997 4 22 10 19 11.21 -61.26 57.5 5.0
783 ISC 1997 4 22 10 22 11.07 -60.94 50.9 5.3
784 ISC 1997 4 22 11 0 11.08 -60.87 3.0 4.5
785 ISC 1997 4 22 11 46 11.12 -60.92 5.0 5.1
786 ISC 1997 4 22 12 45 11.08 -61.23 41.2 4.9
787 ISC 1997 4 23 22 25 11.13 -60.77 26.3 5.2
788 ISC 1997 4 24 12 44 11.05 -60.60 3.0 5.1
789 ISC 1997 4 25 16 4 11.14 -61.00 46.9 5.3
790 ISC 1997 5 3 22 39 11.02 -60.90 5.0 5.2
791 ISC 1997 5 4 1 44 11.08 -61.26 41.8 5.5
792 USGS-PDE 1997 5 4 22 39 11.07 -60.94 5.0 4.7
793 ISC 1997 5 16 21 17 11.06 -61.26 37.8 5.1
794 ISC 1997 6 15 5 5 10.78 -62.47 74.2 4.5
795 ISC 1997 7 9 19 24 10.51 -63.55 3.0 7.0
796 ISC 1997 7 9 19 26 10.64 -62.70 10.0 5.7
797 ISC 1997 7 9 20 6 10.56 -63.50 37.0 5.5
798 ISC 1997 7 9 20 53 10.53 -63.46 17.8 4.9
799 ISC 1997 7 9 22 21 10.66 -63.29 51.4 5.0
800 USGS-PDE 1997 7 10 1 1 11.00 -63.58 10.0 4.5
801 ISC 1997 7 12 0 47 10.70 -63.17 52.2 4.9
802 USGS-PDE 1997 7 13 0 42 10.40 -63.70 10.0 4.5
803 ISC 1997 7 30 14 6 10.72 -63.42 30.0 4.5
804 ISC 1997 8 6 0 42 15.75 -60.63 52.5 4.9
805 ISC 1997 8 12 3 53 10.48 -63.47 31.4 4.6
806 USGS-PDE 1997 8 19 0 29 11.52 -62.91 10.0 4.5
807 ISC 1997 9 22 8 59 10.35 -62.31 43.2 4.6
808 ISC 1997 10 16 8 27 10.78 -61.92 66.1 4.5
809 ISC 1997 11 2 11 24 19.35 -66.29 45.8 4.5
810 ISC 1997 11 10 2 7 10.52 -62.85 44.7 5.1
811 ISC 1997 11 15 2 34 10.49 -62.85 52.1 5.0
812 ISC 1997 12 14 7 6 10.38 -64.78 36.8 5.1
813 ISC 1997 12 20 16 48 10.68 -63.47 34.0 4.5
814 ISC 1998 2 16 15 2 10.54 -64.50 34.1 5.0
815 ISC 1998 2 16 15 9 10.51 -64.54 33.1 4.5
816 ISC 1998 3 2 21 37 10.44 -62.23 1.3 4.5
817 ISC 1998 3 5 7 8 20.01 -63.12 18.7 5.3
818 ISC 1998 3 25 8 27 19.28 -67.08 22.8 5.1
819 ISC 1998 5 3 14 59 15.02 -60.65 70.4 4.9
820 ISC 1998 6 15 8 28 19.44 -65.55 33.3 4.6
821 ISC 1998 6 25 21 3 17.74 -61.57 42.1 5.3
822 ISC 1998 7 19 9 27 18.30 -65.10 133.8 4.7
823 ISC 1998 7 22 16 21 12.11 -60.74 76.6 5.3
824 ISC 1998 7 23 6 36 11.15 -60.72 14.1 4.6
825 ISC 1998 9 13 19 0 17.19 -61.17 41.5 5.0
826 ISC 1998 9 18 15 24 14.28 -60.33 82.5 5.0
827 ISC 1998 10 24 1 1 18.87 -64.32 54.8 4.6
828 ISC 1998 11 21 23 19 15.86 -60.40 51.4 5.0
829 ISC 1998 12 8 2 32 18.80 -64.05 34.1 5.9
830 ISC 1998 12 17 3 25 16.90 -62.44 144.4 4.7
831 ISC 1998 12 26 11 29 10.55 -63.56 19.9 5.8
832 ISC 1999 1 15 20 29 10.54 -63.73 16.8 4.9
833 ISC 1999 1 18 19 41 18.84 -67.22 37.4 5.2
834 ISC 1999 1 24 9 37 15.19 -61.32 153.2 4.6
835 ISC 1999 1 25 10 36 16.87 -62.48 142.6 4.9
836 ISC 1999 1 25 17 18 19.50 -66.71 48.8 4.5
837 ISC 1999 2 19 11 57 14.99 -60.62 67.8 4.5
838 ISC 1999 2 28 3 35 19.33 -64.94 40.8 4.6
839 ISC 1999 3 7 18 1 19.40 -64.90 33.0 4.5
840 USGS-PDE 1999 3 21 12 3 11.23 -61.92 83.0 4.5
A156

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


841 ISC 1999 4 6 21 3 15.16 -60.53 64.5 4.9
842 ISC 1999 5 20 22 54 10.38 -62.97 67.1 4.9
843 CMT 1999 6 8 12 4 15.06 -60.24 37.0 5.8
844 ISC 1999 7 3 9 6 15.07 -60.40 61.7 4.7
845 ISC 1999 7 8 9 31 11.29 -60.62 59.4 5.0
846 ISC 1999 7 11 11 51 16.44 -59.84 0.0 5.0
847 ISC 1999 8 28 4 27 16.99 -61.49 62.5 5.3
848 ISC 1999 10 26 9 35 11.53 -61.08 71.9 4.5
849 ISC 1999 11 7 6 53 10.45 -62.59 49.5 5.0
850 ISC 1999 12 20 10 43 17.27 -61.73 62.5 5.7
851 ISC 2000 1 29 3 16 18.93 -64.29 42.2 4.5
852 ISC 2000 2 16 6 58 17.73 -60.85 37.3 4.7
853 ISC 2000 2 16 7 3 17.73 -60.89 43.2 5.5
854 ISC 2000 2 16 7 16 17.72 -60.96 52.5 4.6
855 ISC 2000 2 16 9 50 17.49 -60.81 41.4 4.5
856 ISC 2000 2 23 19 20 17.39 -60.75 29.3 5.5
857 ISC 2000 2 27 23 38 17.78 -60.95 39.1 5.3
858 ISC 2000 3 3 19 45 11.89 -59.13 10.0 4.7
859 ISC 2000 3 25 20 15 10.44 -62.57 35.2 5.1
860 ISC 2000 4 12 20 20 15.55 -60.63 52.5 4.7
861 ISC 2000 4 27 12 19 11.13 -62.30 117.5 4.5
862 ISC 2000 5 1 6 37 11.11 -62.36 111.3 4.5
863 USGS-PDE 2000 6 11 0 45 10.24 -62.37 10.0 4.7
864 ISC 2000 6 16 18 36 17.33 -59.92 14.3 4.7
865 USGS-PDE 2000 7 20 22 2 11.41 -62.11 143.0 4.6
866 ISC 2000 8 6 8 22 11.03 -62.15 75.6 5.0
867 ISC 2000 9 5 19 36 16.84 -61.12 56.3 5.1
868 ISC 2000 9 21 5 38 11.13 -62.48 123.5 4.9
869 ISC 2000 9 25 5 38 19.17 -62.58 44.2 4.7
870 ISC 2000 9 25 6 6 19.06 -62.40 0.0 5.2
871 CMT 2000 10 4 14 37 11.15 -62.48 110.4 6.1
872 ISC 2000 10 27 18 57 17.57 -61.18 50.0 5.0
873 ISC 2000 10 27 19 2 17.55 -61.21 45.9 5.6
874 ISC 2000 10 27 19 15 17.61 -61.23 44.8 5.5
875 ISC 2000 10 27 19 31 17.57 -61.11 44.9 5.2
876 ISC 2000 10 30 3 7 17.55 -61.21 45.9 5.8
877 ISC 2000 11 5 10 49 9.90 -62.07 27.9 4.7
878 ISC 2000 12 11 18 54 19.15 -67.09 44.1 5.5
879 USGS-PDE 2000 12 23 3 6 19.01 -65.58 69.0 4.5
880 ISC 2000 12 30 15 31 14.98 -60.60 69.1 5.1
881 ISC 2001 1 5 8 6 15.97 -61.02 54.9 5.7
882 ISC 2001 1 6 19 56 17.70 -62.08 55.3 5.0
883 ISC 2001 1 23 7 37 19.24 -64.41 28.4 5.2
884 ISC 2001 1 25 4 33 10.70 -62.57 85.5 4.6
885 ISC 2001 2 17 20 55 10.71 -61.45 64.7 5.3
886 ISC 2001 2 18 0 40 10.74 -60.91 27.6 4.7
887 ISC 2001 2 18 5 42 13.11 -60.57 44.9 4.9
888 ISC 2001 3 26 8 32 10.82 -62.53 98.2 4.6
889 USGS-PDE 2001 3 31 9 12 19.21 -65.06 49.0 4.5
890 ISC 2001 4 5 13 54 15.94 -60.87 46.8 5.5
891 ISC 2001 4 24 17 3 10.88 -62.45 89.0 4.9
892 ISC 2001 4 26 22 37 17.92 -63.69 130.3 4.6
893 ISC 2001 5 1 16 44 17.02 -60.22 41.0 5.1
894 ISC 2001 5 3 11 23 16.98 -60.18 38.3 4.9
895 USGS-PDE 2001 6 6 0 4 16.95 -60.30 46.0 4.5
896 ISC 2001 6 6 13 16 17.03 -60.18 8.8 4.5
897 ISC 2001 6 6 23 18 17.15 -60.17 9.2 4.9
898 USGS-PDE 2001 6 16 15 8 12.05 -61.85 147.0 4.5
899 ISC 2001 6 18 6 50 16.96 -60.23 46.1 4.5
900 ISC 2001 7 2 15 5 17.05 -60.24 40.3 4.7
A157

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


901 ISC 2001 7 3 2 8 17.01 -60.22 44.9 5.1
902 ISC 2001 7 24 19 0 19.29 -64.12 10.0 5.0
903 ISC 2001 8 11 15 23 12.13 -59.95 0.0 5.1
904 ISC 2001 9 25 23 16 17.08 -61.40 58.4 5.7
905 USGS-PDE 2001 10 13 16 15 17.11 -60.23 50.0 4.9
906 ISC 2001 10 16 15 27 19.37 -64.89 30.4 5.6
907 ISC 2001 10 17 11 29 19.34 -64.98 41.2 6.0
908 ISC 2001 10 17 20 45 18.47 -67.51 14.2 4.9
909 ISC 2001 10 18 19 15 19.34 -64.90 45.6 4.7
910 ISC 2001 10 18 21 36 19.32 -64.71 33.0 4.5
911 ISC 2001 10 22 14 42 19.25 -64.84 0.0 4.7
912 ISC 2001 10 22 15 29 19.41 -64.85 22.1 4.7
913 ISC 2001 10 22 15 36 19.30 -64.88 40.6 5.2
914 ISC 2001 10 22 15 39 19.36 -64.89 38.6 4.9
915 ISC 2001 10 22 16 12 19.32 -64.65 23.5 5.0
916 ISC 2001 10 24 9 51 19.40 -64.97 15.2 5.4
917 ISC 2001 10 24 14 14 19.36 -64.72 25.9 4.6
918 ISC 2001 11 17 1 45 10.57 -61.39 33.0 4.5
919 ISC 2001 11 27 4 14 17.92 -67.95 56.4 5.1
920 ISC 2001 12 4 17 10 19.66 -63.32 0.0 4.6
921 ISC 2001 12 7 15 59 16.74 -61.07 8.2 5.3
922 ISC 2001 12 28 7 39 10.85 -64.52 19.6 5.1
923 ISC 2002 1 10 18 58 19.56 -64.15 33.0 4.9
924 ISC 2002 1 15 2 18 19.28 -64.25 26.8 4.6
925 ISC 2002 1 15 9 13 19.39 -64.29 47.7 5.1
926 ISC 2002 1 30 14 50 15.41 -61.11 116.4 4.9
927 ISC 2002 3 4 1 30 17.01 -60.81 9.5 4.7
928 ISC 2002 3 11 14 24 19.53 -64.37 35.4 5.0
929 ISC 2002 3 11 22 45 15.81 -61.05 80.2 4.9
930 ISC 2002 3 18 18 36 17.10 -62.88 156.2 4.7
931 ISC 2002 3 29 20 22 10.10 -62.48 33.0 4.6
932 ISC 2002 3 30 19 26 10.25 -60.55 55.4 4.6
933 ISC 2002 4 7 1 51 18.64 -60.42 41.9 4.6
934 ISC 2002 4 11 2 19 19.24 -66.67 45.7 4.9
935 ISC 2002 5 13 9 54 18.05 -61.45 31.1 5.2
936 ISC 2002 5 15 16 11 19.43 -68.11 10.0 4.6
937 ISC 2002 5 28 4 10 19.26 -68.07 10.0 6.1
938 ISC 2002 5 29 19 49 14.70 -61.19 152.5 4.7
939 ISC 2002 6 12 12 48 18.08 -64.95 26.8 4.7
940 ISC 2002 6 14 4 5 19.50 -64.42 36.5 4.7
941 ISC 2002 6 21 19 35 11.23 -60.30 66.1 4.7
942 ISC 2002 7 16 15 15 15.07 -60.22 29.6 4.6
943 ISC 2002 7 31 12 32 19.30 -64.53 25.5 5.0
944 ISC 2002 9 15 7 14 16.18 -59.69 0.0 4.7
945 ISC 2002 9 17 22 35 12.22 -59.43 48.7 5.2
946 USGS-PDE 2002 9 27 16 26 10.73 -62.34 43.0 4.7
947 ISC 2002 10 1 7 0 18.83 -62.99 41.4 5.2
948 ISC 2002 10 4 15 43 10.38 -62.58 32.3 5.2
949 ISC 2002 10 15 23 12 10.76 -63.71 5.0 4.6
950 ISC 2002 10 18 12 4 19.51 -64.38 33.0 5.1
951 USGS-PDE 2002 10 23 8 49 19.55 -64.31 65.0 4.7
952 ISC 2002 10 25 18 56 19.57 -64.50 41.5 4.7
953 ISC 2002 11 9 5 24 19.32 -63.64 40.6 4.6
954 ISC 2002 11 9 6 15 14.63 -58.64 10.0 4.5
955 ISC 2002 11 10 2 47 19.36 -63.57 26.3 4.9
956 ISC 2002 11 13 20 26 18.89 -64.34 33.0 5.2
957 ISC 2002 11 23 9 15 10.90 -60.26 64.8 4.5
958 ISC 2002 11 23 9 21 10.94 -60.34 67.5 4.6
959 USGS-PDE 2002 11 24 15 9 10.78 -62.67 98.0 5.2
960 ISC 2002 11 25 4 35 10.62 -62.96 10.0 4.7
A158

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


961 ISC 2002 12 21 18 35 15.03 -60.61 77.6 5.1
962 ISC 2002 12 21 18 50 18.52 -62.32 32.1 5.1
963 USGS-PDE 2003 1 20 14 52 19.42 -63.31 33.0 4.6
964 ISC 2003 2 14 1 23 10.29 -62.30 17.6 5.1
965 ISC 2003 2 17 6 57 18.66 -64.51 57.1 4.6
966 ISC 2003 3 27 20 43 20.15 -64.70 10.0 4.7
967 ISC 2003 4 12 0 10 20.15 -64.61 10.0 4.9
968 ISC 2003 4 12 10 13 19.23 -62.71 40.3 4.6
969 ISC 2003 4 24 13 5 10.72 -62.90 121.7 4.7
970 ISC 2003 5 7 12 57 18.42 -65.01 120.4 4.5
971 ISC 2003 5 14 6 3 18.36 -58.63 20.3 6.6
972 ISC 2003 6 1 16 15 9.78 -62.79 57.8 5.1
973 ISC 2003 6 21 5 55 10.79 -59.27 10.0 5.3
974 ISC 2003 6 30 0 7 17.53 -61.10 33.0 5.9
975 USGS-PDE 2003 7 3 8 45 11.07 -62.35 129.0 4.6
976 ISC 2003 8 14 16 27 10.80 -62.52 120.7 4.7
977 USGS-PDE 2003 9 2 0 2 10.39 -62.53 51.0 4.7
978 ISC 2003 9 8 4 31 15.12 -61.41 160.6 4.9
979 ISC 2003 10 20 20 31 19.48 -63.16 41.6 5.2
980 ISC 2003 10 28 7 15 18.42 -62.76 53.2 4.6
981 ISC 2003 11 6 22 15 18.54 -63.33 54.4 5.2
982 ISC 2003 11 9 20 40 17.85 -61.47 30.4 4.7
983 ISC 2003 11 16 7 22 12.88 -67.95 33.0 5.1
984 ISC 2003 12 6 3 33 19.27 -67.33 32.9 5.4
985 ISC 2003 12 6 4 33 19.29 -67.30 18.2 5.0
986 ISC 2003 12 20 12 12 10.56 -63.22 6.8 5.0
987 ISC 2004 1 4 23 8 19.15 -64.71 41.0 4.9
988 ISC 2004 1 15 10 56 10.90 -62.38 97.0 5.3
989 ISC 2004 2 5 18 3 15.20 -61.23 133.4 4.9
990 ISC 2004 2 25 18 13 10.68 -62.64 111.8 4.7
991 ISC 2004 2 29 19 12 17.76 -61.65 52.5 5.0
992 ISC 2004 3 1 6 12 15.13 -60.66 48.5 5.3
993 ISC 2004 3 19 22 14 19.23 -64.65 36.0 5.0
994 ISC 2004 3 30 16 23 17.57 -62.18 71.0 5.0
995 ISC 2004 3 31 22 29 18.58 -66.88 119.6 4.6
996 USGS-PDE 2004 4 15 15 19 16.35 -62.58 189.0 4.5
997 USGS-PDE 2004 5 1 19 32 10.13 -62.95 10.0 4.7
998 ISC 2004 5 31 0 3 12.77 -60.92 93.0 4.6
999 ISC 2004 6 4 2 39 15.89 -60.67 24.6 4.6
1000 ISC 2004 6 10 3 18 13.67 -60.45 78.2 4.7
1001 ISC 2004 6 15 13 44 15.20 -60.19 37.0 4.6
1002 ISC 2004 6 19 4 3 19.17 -65.45 33.0 4.7
1003 ISC 2004 6 29 9 49 10.68 -62.62 114.0 4.6
1004 ISC 2004 7 23 12 6 10.41 -62.37 23.8 4.6
1005 ISC 2004 8 11 12 30 17.86 -61.43 29.5 4.9
1006 ISC 2004 8 14 6 40 17.79 -61.49 33.0 4.9
1007 ISC 2004 9 6 20 43 19.49 -65.30 21.6 5.1
1008 USGS-PDE 2004 9 15 13 13 19.56 -65.09 50.0 4.5
1009 ISC 2004 10 9 19 42 15.13 -61.12 138.1 4.7
1010 ISC 2004 11 4 21 28 10.89 -62.39 97.2 4.9
1011 ISC 2004 11 21 11 41 15.73 -61.68 21.2 6.3
1012 ISC 2004 11 21 11 47 15.83 -61.62 35.0 5.4
1013 ISC 2004 11 21 11 50 15.66 -61.67 36.8 4.5
1014 ISC 2004 11 21 11 51 15.76 -61.65 27.9 5.1
1015 ISC 2004 11 21 11 56 15.72 -61.61 27.8 5.3
1016 ISC 2004 11 21 11 58 15.85 -61.68 26.4 5.0
1017 ISC 2004 11 21 12 2 15.76 -61.60 22.7 5.1
1018 ISC 2004 11 21 12 8 15.77 -61.73 25.9 4.6
1019 ISC 2004 11 21 12 22 15.80 -61.73 19.4 4.7
1020 ISC 2004 11 21 12 45 15.75 -61.76 28.2 5.0
A159

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


1021 ISC 2004 11 21 12 58 15.81 -61.71 23.6 4.6
1022 ISC 2004 11 21 13 36 15.75 -61.57 26.2 5.4
1023 ISC 2004 11 21 13 49 15.76 -61.63 26.6 4.9
1024 ISC 2004 11 21 14 47 15.74 -61.75 29.8 4.6
1025 ISC 2004 11 21 18 53 15.76 -61.74 14.1 5.4
1026 ISC 2004 11 21 20 19 15.64 -61.69 35.6 4.6
1027 ISC 2004 11 21 22 56 15.73 -61.61 22.1 5.1
1028 ISC 2004 11 22 22 59 15.73 -61.77 33.0 4.6
1029 ISC 2004 11 22 2 1 15.73 -61.78 26.5 5.0
1030 ISC 2004 11 22 18 13 15.79 -61.69 25.4 4.5
1031 ISC 2004 11 22 21 23 15.77 -61.69 13.4 5.3
1032 ISC 2004 11 22 21 53 15.83 -61.65 17.8 4.5
1033 ISC 2004 11 22 22 56 15.76 -61.75 33.3 4.7
1034 ISC 2004 11 25 20 51 15.93 -61.57 13.7 4.5
1035 ISC 2004 11 25 20 59 15.82 -61.72 25.5 4.6
1036 ISC 2004 11 26 5 5 15.69 -61.54 17.7 5.5
1037 ISC 2004 11 26 22 49 15.81 -61.72 17.8 5.2
1038 ISC 2004 11 27 3 5 15.85 -60.39 36.4 4.5
1039 USGS-PDE 2004 11 27 16 2 15.67 -61.51 10.0 4.5
1040 ISC 2004 11 27 23 44 15.70 -61.62 14.7 4.9
1041 ISC 2004 11 29 16 36 15.70 -61.50 20.3 4.7
1042 ISC 2004 11 29 20 53 15.78 -61.57 25.5 4.9
1043 ISC 2004 12 2 14 47 15.83 -61.32 29.9 5.0
1044 ISC 2004 12 2 19 16 10.49 -61.45 48.2 5.8
1045 ISC 2004 12 3 3 46 10.54 -61.46 40.5 5.4
1046 ISC 2004 12 11 4 45 11.04 -62.48 116.7 4.7
1047 ISC 2004 12 11 19 45 18.69 -64.68 58.8 5.2
1048 ISC 2004 12 14 21 29 15.76 -61.50 24.1 4.9
1049 ISC 2004 12 18 0 35 19.19 -64.65 36.4 4.8
1050 ISC 2004 12 26 15 19 15.72 -61.66 27.2 4.7
1051 ISC 2004 12 27 20 58 15.80 -61.68 26.6 5.1
1052 ISC 2005 1 10 3 3 15.85 -61.53 27.2 4.5
1053 ISC 2005 1 12 13 52 8.71 -61.95 33.0 5.0
1054 ISC 2005 1 21 6 5 10.79 -62.61 108.0 4.5
1055 ISC 2005 1 25 18 37 11.39 -58.52 12.9 5.1
1056 ISC 2005 1 28 18 5 19.61 -65.42 37.1 4.6
1057 ISC 2005 1 29 6 36 19.71 -65.44 49.6 4.6
1058 ISC 2005 1 29 14 45 15.81 -61.58 27.5 5.1
1059 ISC 2005 2 8 2 18 15.48 -61.28 124.4 5.3
1060 ANSS 2005 2 8 2 18 15.48 -61.31 115.4 5.0
1061 ISC 2005 2 14 18 5 15.84 -61.73 17.1 5.8
1062 ISC 2005 2 14 18 8 15.73 -61.60 3.0 4.6
1063 ISC 2005 2 15 4 31 15.74 -61.70 19.0 4.5
1064 ISC 2005 2 16 23 24 15.84 -61.67 13.2 4.6
1065 ISC 2005 3 22 16 20 10.37 -62.47 11.9 4.7
1066 ISC 2005 3 23 18 13 10.61 -62.57 97.7 4.5
1067 ISC 2005 3 26 16 0 10.38 -62.47 5.0 4.7
1068 ISC 2005 3 30 16 17 15.02 -61.46 172.3 4.6
1069 ISC 2005 5 9 12 22 18.98 -64.32 34.8 4.9
1070 ISC 2005 6 6 1 20 15.76 -61.64 24.7 4.8
1071 ISC 2005 6 14 15 44 15.26 -61.37 147.8 5.2
1072 ISC 2005 7 4 4 4 10.44 -61.39 64.6 5.0
1073 ISC 2005 7 25 4 23 9.90 -63.05 33.0 4.5
1074 ISC 2005 8 4 4 35 10.97 -62.46 96.6 4.6
1075 ISC 2005 8 8 9 38 11.74 -66.21 38.4 4.7
1076 USGS-PDE 2005 8 17 6 42 9.40 -62.07 17.0 4.9
1077 ISC 2005 8 17 6 43 11.60 -66.12 17.0 4.6
1078 ISC 2005 8 30 14 2 15.02 -60.56 62.4 4.9
1079 ISC 2005 10 24 9 19 11.10 -62.39 137.7 5.1
1080 ISC 2005 10 27 6 54 17.81 -61.45 44.0 4.6
A160

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


1081 ISC 2005 10 28 22 30 11.11 -62.04 80.9 5.5
1082 ISC 2005 12 2 3 30 10.54 -61.68 17.4 4.5
1083 ISC 2005 12 22 19 8 15.79 -61.82 8.5 4.7
1084 ISC 2006 1 1 21 31 11.45 -60.18 64.7 4.6
1085 ISC 2006 1 18 12 42 17.57 -61.32 27.3 4.7
1086 ISC 2006 2 12 11 44 15.38 -60.79 41.8 4.9
1087 ISC 2006 3 2 23 35 19.16 -63.99 28.6 5.6
1088 ISC 2006 3 2 23 51 19.24 -63.96 9.3 4.6
1089 ISC 2006 3 3 13 9 19.17 -63.98 16.9 5.1
1090 ISC 2006 3 4 2 50 19.15 -64.01 20.6 4.7
1091 USGS-PDE 2006 3 9 9 16 16.22 -61.63 135.0 4.6
1092 ISC 2006 3 16 11 17 18.02 -61.57 25.4 4.5
1093 ISC 2006 3 26 2 14 19.24 -64.91 9.1 4.6
1094 ISC 2006 4 20 12 20 19.49 -62.61 30.1 4.5
1095 ISC 2006 4 28 15 46 10.13 -62.52 59.2 4.9
1096 ISC 2006 5 11 2 31 15.06 -60.31 35.0 4.9
1097 USGS-PDE 2006 5 14 21 2 17.84 -62.60 65.0 4.7
1098 ISC 2006 5 17 11 23 14.98 -60.43 47.6 4.5
1099 ISC 2006 6 11 0 24 14.19 -58.36 10.0 5.0
1100 ISC 2006 6 13 12 25 10.90 -62.36 87.3 4.6
1101 ISC 2006 6 17 10 42 19.23 -64.75 2.0 4.9
1102 ISC 2006 6 21 2 16 15.71 -61.59 13.5 4.9
1103 ISC 2006 6 21 12 34 15.73 -61.51 13.4 4.5
1104 ISC 2006 7 1 17 24 8.12 -64.26 0.3 4.5
1105 ISC 2006 7 5 22 22 15.74 -61.58 15.0 5.0
1106 ISC 2006 8 4 3 41 10.38 -62.33 11.8 5.2
1107 ISC 2006 8 6 9 23 18.46 -61.55 22.3 4.5
1108 ISC 2006 8 6 23 0 18.14 -68.05 87.2 4.5
1109 ISC 2006 8 14 13 9 19.13 -64.61 26.0 5.7
1110 ISC 2006 8 15 1 30 19.07 -64.69 21.3 5.1
1111 USGS-PDE 2006 8 25 12 47 10.82 -62.38 64.0 4.7
1112 ISC 2006 9 15 7 8 15.70 -61.28 100.5 5.0
1113 USGS-PDE 2006 9 24 0 50 10.87 -62.40 28.0 4.5
1114 ISC 2006 9 29 13 8 10.87 -61.75 55.4 6.3
1115 ISC 2006 9 29 18 23 10.79 -61.75 52.8 5.5
1116 ISC 2006 10 8 10 6 14.77 -59.24 8.3 5.1
1117 ISC 2006 10 19 16 4 17.79 -61.08 20.1 4.7
1118 ISC 2006 10 27 1 26 17.72 -61.01 35.0 4.7
1119 ISC 2006 11 15 20 25 10.78 -62.65 98.9 5.2
1120 ISC 2006 11 17 8 33 11.39 -62.24 135.8 4.9
1121 ISC 2006 11 27 17 52 14.39 -61.41 193.0 4.9
1122 ISC 2006 12 2 8 6 10.31 -59.14 4.8 5.2
1123 ISC 2006 12 15 12 59 16.00 -60.75 46.5 5.2
1124 USGS-PDE 2006 12 16 10 21 18.13 -61.41 30.0 4.6
1125 ISC 2007 1 7 6 15 19.31 -65.21 32.8 5.0
1126 USGS-PDE 2007 1 31 8 33 17.63 -60.99 10.0 4.6
1127 ISC 2007 2 13 5 29 18.68 -63.33 23.7 4.6
1128 ISC 2007 2 23 14 48 10.81 -61.72 52.3 5.2
1129 ISC 2007 2 27 9 42 17.07 -61.53 52.5 5.1
1130 ISC 2007 3 12 23 23 10.86 -62.21 71.8 4.5
1131 ISC 2007 3 29 5 24 13.63 -60.12 68.3 5.0
1132 USGS-PDE 2007 4 5 13 14 17.94 -60.71 67.0 4.7
1133 ISC 2007 4 8 16 8 19.02 -67.66 9.3 5.0
1134 ISC 2007 4 14 17 44 16.93 -60.77 14.1 4.6
1135 ISC 2007 4 16 14 45 19.15 -64.66 29.4 4.6
1136 ISC 2007 4 19 23 13 17.56 -61.91 38.1 4.6
1137 ISC 2007 4 21 9 47 10.69 -62.62 100.1 4.9
1138 USGS-PDE 2007 5 24 8 49 16.69 -60.81 35.0 4.5
1139 USGS-PDE 2007 5 28 18 24 16.20 -62.40 185.0 4.7
1140 USGS-PDE 2007 6 17 9 3 19.19 -64.83 19.0 5.1
A161

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


1141 USGS-PDE 2007 7 14 18 11 16.64 -61.64 49.0 4.9
1142 USGS-PDE 2007 7 16 1 32 10.63 -62.35 89.0 5.2
1143 USGS-PDE 2007 8 16 7 20 17.38 -60.31 47.0 5.2
1144 USGS-PDE 2007 8 16 8 28 15.27 -60.72 39.0 4.5
1145 USGS-PDE 2007 8 21 3 57 15.58 -61.24 86.0 4.7
1146 USGS-PDE 2007 9 13 6 37 14.24 -60.39 71.0 5.3
1147 USGS-PDE 2007 9 14 6 3 18.38 -64.45 94.0 5.1
1148 USGS-PDE 2007 9 16 23 46 19.14 -64.66 12.0 4.7
1149 USGS-PDE 2007 10 10 2 51 19.10 -65.87 111.0 4.6
1150 USGS-PDE 2007 11 1 17 44 10.08 -64.35 12.0 4.5
1151 USGS-PDE 2007 11 7 6 30 14.97 -60.61 63.0 5.1
1152 USGS-PDE 2007 11 22 7 1 18.17 -64.53 134.0 4.6
1153 USGS-PDE 2007 11 28 14 18 15.14 -60.53 80.0 5.1
1154 CMT 2007 11 29 19 0 14.97 -61.26 148.0 7.4
1155 USGS-PDE 2007 11 29 19 34 15.09 -61.38 146.0 5.4
1156 USGS-PDE 2007 11 29 20 11 15.04 -61.31 144.0 4.7
1157 USGS-PDE 2007 11 29 20 17 15.11 -61.22 150.0 4.7
1158 USGS-PDE 2007 11 30 4 0 15.03 -61.27 148.0 5.3
1159 USGS-PDE 2007 12 13 0 18 14.10 -61.05 126.0 4.9
1160 USGS-PDE 2007 12 19 2 48 19.19 -64.70 5.0 4.7
1161 USGS-PDE 2007 12 19 3 14 19.40 -64.57 39.0 4.9
1162 USGS-PDE 2007 12 24 2 32 14.99 -61.31 169.0 4.9
1163 USGS-PDE 2007 12 28 18 18 19.20 -64.52 65.0 4.9
1164 USGS-PDE 2008 2 17 1 50 14.82 -61.39 172.0 4.7
1165 USGS-PDE 2008 3 9 22 17 10.64 -62.57 83.0 4.7
1166 USGS-PDE 2008 3 20 15 28 10.44 -62.39 20.0 5.0
1167 USGS-PDE 2008 3 24 13 30 10.90 -62.46 103.0 4.9
1168 USGS-PDE 2008 4 24 22 20 20.05 -65.01 45.0 4.6
1169 USGS-PDE 2008 5 3 6 36 19.36 -64.17 49.0 4.7
1170 USGS-PDE 2008 5 4 11 15 20.06 -65.04 35.0 4.7
1171 USGS-PDE 2008 5 21 11 43 15.89 -60.97 69.0 4.9
1172 USGS-PDE 2008 6 16 7 43 19.59 -64.50 37.0 4.6
1173 USGS-PDE 2008 6 17 18 50 15.72 -61.76 11.0 4.5
1174 USGS-PDE 2008 6 28 8 3 15.05 -61.10 151.0 4.9
1175 USGS-PDE 2008 7 14 1 33 17.79 -60.85 82.0 4.6
1176 USGS-PDE 2008 7 14 17 31 10.44 -60.43 43.0 4.9
1177 USGS-PDE 2008 7 26 15 7 12.52 -59.44 49.0 4.6
1178 CMT 2008 8 11 7 19 10.51 -64.17 13.0 5.2
1179 USGS-PDE 2008 8 12 12 30 10.61 -60.51 30.0 4.6
1180 USGS-PDE 2008 9 2 6 18 13.86 -60.44 16.0 5.0
1181 USGS-PDE 2008 11 1 17 25 17.17 -62.14 78.0 4.7
1182 USGS-PDE 2008 11 1 18 14 16.93 -62.44 10.0 4.5
1183 USGS-PDE 2008 11 2 8 5 17.14 -62.38 10.0 5.1
1184 USGS-PDE 2008 11 2 21 1 19.65 -66.28 21.0 5.1
1185 USGS-PDE 2008 11 13 14 53 19.48 -66.35 15.0 4.9
1186 USGS-PDE 2008 11 23 9 5 19.49 -66.32 47.0 5.2
1187 USGS-PDE 2008 12 2 0 33 19.50 -66.36 35.0 4.9
1188 USGS-PDE 2008 12 10 21 33 15.99 -60.57 10.0 5.1
1189 USGS-PDE 2008 12 17 19 55 18.99 -65.19 28.0 4.9
1190 USGS-PDE 2008 12 21 12 55 10.82 -61.66 54.0 5.2
1191 USGS-PDE 2009 1 6 5 32 11.23 -60.34 47.0 4.5
1192 USGS-PDE 2009 2 5 6 8 16.09 -60.63 53.0 4.7
1193 USGS-PDE 2009 2 13 4 24 19.63 -66.30 35.0 4.7
1194 USGS-PDE 2009 2 13 4 39 19.47 -66.36 6.0 5.1
1195 USGS-PDE 2009 3 12 2 5 19.07 -66.37 15.0 5.4
1196 USGS-PDE 2009 4 1 2 3 10.51 -61.10 28.0 4.9
1197 USGS-PDE 2009 4 4 9 24 10.21 -61.23 45.0 5.2
1198 USGS-PDE 2009 4 5 17 37 19.33 -65.10 30.0 4.6
1199 USGS-PDE 2009 5 20 3 36 19.00 -65.50 18.0 5.3
1200 USGS-PDE 2009 5 24 5 13 13.33 -58.82 19.0 4.9
A162

Code Source Y M D h m Lat (°) Long (°) Dep (km) Mw


1201 USGS-PDE 2009 5 28 16 28 15.11 -60.39 52.0 4.9
1202 USGS-PDE 2009 6 1 4 33 19.48 -63.46 25.0 4.7
1203 USGS-PDE 2009 6 1 4 35 19.56 -63.47 39.0 5.0
1204 USGS-PDE 2009 6 13 16 12 18.96 -66.56 42.0 5.2

* Events from IPGH Catalogue.

** Corrected depth value.


SECTION B

Strong Motion Network:


feasibility study and data transmission

by

Paolo Augliera 1, Ezio D'Alema1, Simone Marzorati1, Marco Massa1,


Tony Gibbs2, Domalapally Rao2 and Walter Salazar2

1 Italian Institute of Geophysics and Volcanology (INGV), Milan, Italy.


2 The University of the West Indies (UWI), St Augustine, Trinidad.
INDEX

1. STRONG MOTION NETWORK ON TRINIDAD: FEASIBILITY STUDY .....................1

1.1 STATE OF THE ART: VELOCIMETRIC SENSORS ...........................................................................1

1.2 STATE OF THE ART: STRONG MOTION SENSORS ........................................................................4

1.3 INTEGRATION OF DIFFERENT SENSORS: THE INGV (ITALY) MULTIPARAMETRIC


NETWORK..............................................................................................................................................5

1.4 STRONG-MOTION NETWORK: STATIONS LOCATIONS AND TECHNICAL SPECIFICATIONS ...6

BIBLIOGRAPHY................................................................................................................................13
1.STRONG MOTION NETWORK ON TRINIDAD:
FEASIBILITY STUDY

The work defines the following “milestones” or reference indications:

¾ Definition of the financial resources.


Considering an order of magnitude, we can hypothesize costs relative to a
complete strong-motion recording system (considering supply and installation)
on the order of 20.000 Euro.

¾ All the choices rely on the available funds. We consider a total budget of the
order of 200.000 Euro (290.000 $ at 13/01/2009)

¾ Definition of the area of installation.

¾ Seismic station inventory for the study area.

¾ Definition of the standard for strong-motion sensor and recorders;

¾ Definition of the data transmission link;

¾ Planning for location of new stations.


For this task we consider, according to their economical importance, the
possibility of “reusable” velocimetric or strong-motion sites (or pre-existing
instrument housing).

For the planning of the new strong-motion stations we consider the inter-distances, the
near-fault location and the different soil conditions.

1.1 STATE OF THE ART: VELOCIMETRIC SENSORS


The Seismic Research Centre (SRC) of The University of the West Indies (UWI) is the
agency responsible for earthquakes and volcanoes monitoring for the English-speaking
islands of the Eastern Caribbean. In recent years, the SRC played a more active role in
promoting geologic hazard awareness through its education and outreach activities and it
B2

is also part of a regional effort to establish a tsunami warning system for the Caribbean
and adjacent areas.

The Seismic Research Unit of the University of West Indies maintains a seismograph
network which, when combined with other networks managed by cooperating agencies,
covers the entire Eastern Caribbean from Trinidad to Virgin Islands (Figure 1.1). In the
figure networks managed by the Institut de Physique du Globe de Paris (IPGP) in the
French Antilles are shown in blue boxes and those managed by the Fundacion
Venezolana de Investigaciones Sismologicas (FUNVISIS) and the Universidad de Oriente
(UDO) are shown in green boxes .

The map of Figure 1.1 shows the regional monitoring networks as they currently exist.

From 1952 to 1977 the seismograph network included at least one station in each of the
main islands of the Eastern Caribbean. The stations recorded data locally and the data
were passed to Unit headquarters in Trinidad by local station operators. In 1977 it was
started the process of linking all outstations by a combination of VHF/UHF radio
telemetry and leased telephone circuits directly to Trinidad ,where the data could be
studied in real time. This system came in to full operation in January 1980. Initially the
main recording medium was analogue magnetic tape but digital recording using first
minicomputers (PDP 11/30) and then microcomputers (PC's) was introduced in the mid
1980's. Increased efficiency both of the regional telephone and of internet
communications systems during the 1990's made it possible to subdivide the network into
a number of sub-networks, which are connected to a local microcomputer. These local
microcomputers transmit data to Trinidad through the internet at regular intervals.
Network calls can also be performed through the ordinary telephone system.
B3

Figure 1.1 Seismic networks in the eastern Caribbean area

Most stations are equipped by just vertical component short period sensors but each sub-
network includes at least one broadband three-component station. In particular, there are
different types of seismometers in use: 1 Hz, vertical component, L4C passive
seismometer; 3-component Lennartz, 1 Hz active seismometer and Guralp CMG-40T
Broad-Band seismometer (Figure 1.2). The signals from each seismometer, within a node,
are transmitted to the base on computer and digitized by a 16-bit analogue to digital
B4

converter at a sampling rate of 100 Hz (i.e. 100 samples/sec). A triggering algorithm


searches in continuous-mode the resulting time series from each station for possible
earthquakes and, once preset triggering criteria are met, stores the “event” preceded by a
selectable length of pre-triggering time record (Beckles et al., 1992). Since May 2002 all
the volcanic islands have the capacity to accurately record and locate local
microearthquakes. Data are routinely re-transmitted to Trinidad through a fast internet
link but can also be processed on the spot.
In this study we concentrate our attention only on Trinidad Island and strong-motion
data acquisition.

1.2 STATE OF THE ART: STRONG MOTION SENSORS


On the island of Trinidad 4 strong motion stations has been still operating (Figure 1.2
and Table 1.1). These stations have been recording data intermittently since the initial
installation (2000), especially in the initial phase there have been many teething problems
arising in the collection and analysis of the data that is obtained from the network. At
present, the major limitation of the network are the sparse density distribution of the
stations, the data transmission and the low dynamic range.

Table 1.1 Strong motion stations (from Douglas and Mohais, 2009)

Lat. Long. Elev. Installed


Name Code Site
(N) (W) (m) (DD/MM/YYYY)

Brigand Hill TBH 10.4840 61.0670 199 11/06/2000 Rock

Chaguaramas TCHG 10.6800 61.6600 21 20/12/2001 Rock

Point Fortin ALNG 10.1814 61.6883 0 25/01/2001 Rock

West Moorings TWMO 10.6700 61.5600 0 21/09/2000 Soil

The stations TBH and TWMO are located on ground floor of concrete structure,
whereas TCHG is located on ground floor of concreate building in military base and
ALNG is located on ground floor of small concreate shelter in natural gas processing
plant.

All strong-motion stations (Figure 1.2) are equipped with Kinemetric K2 3-D
accelerograph, with dynamic range of 114 db (19 bits).
B5

Figure 1.2 Seismic stations in the Trinidad area (from Lynch, 2008; the international seismic code for
CHAG station is TCHG)

1.3 INTEGRATION OF DIFFERENT SENSORS: THE INGV (ITALY)


MULTIPARAMETRIC NETWORK

As an example for Trinidad installation we can refer to the RING network (managed by
INGV-CNT) that integrate seismological and geodetic instruments. In particular, RING
integrates GPS receivers with broad band seismometers and accelerometers in real time
connection with 3 center of acquisitions. The co-location of different instruments allow
to detect all frequencies included in the earthquake process, from the inter-seismic strain
accumulation on faults to the radiative part during the rupture process. The seismometers
are Trillium broad-band (40s, 120s or 240s) produced by Nanometrics Seismological
Instruments Inc., as well as the Libra-Gui satellite transmission part. The accelerometers
are the Episensor from Kinemetrics Inc. The GPS receivers are the Leica GRX 1200PRO
B6

with AT504 choke-ring antenna. An example of the multi-parametric installation in the


northern Italy (SALO station) is presented in Figure 1.3.

Figure 1.3 An example of a multi-parametric station (SALO, Italy)

1.4 STRONG-MOTION NETWORK: STATIONS LOCATIONS AND TECHNICAL


SPECIFICATIONS

A strong motion network can be installed for different goals: monitoring; investigation of
site effects, source characteristics, attenuation; structural dynamics; specific dense
network (fault survey, early warning), seismic hazard assessment, shake maps and loss
estimates.
In this feasibility study we consider the network mainly devoted to monitoring and
seismic hazard assessment, with the aim of recording strong earthquakes (as well as weak
motions, depending on noise level values), for engineering and seismological applications.
Strong-motion recordings play an important role, in particular in high seismic risk areas,
and two sets of sensors will have to be installed so that the recordings never clips. Today
both types of sensors are frequently integrated into a single system: six-channel data
loggers with three weak and three strong-motion channels are the current state-of-the-art,
B7

so we can cover the whole dynamic range of seismic events, from the minimum level,
defined by the ambient noise, to the largest damaging events.

Some strong-motion sensors in the market have no DC response but a low-frequency,


high-pass corner at around 0.1 Hz. These sensors have an important drawback: their
records cannot be used for residual displacement determination. They are considered as
less appropriate for seismic applications where low-frequency signals can be important, as
in our case. The strong-motion sensor should be with high dynamic range (24 bits)
combined with low self-noise performances.
Three components high-dynamic range sensors with large full-scale recording ranges will
be installed in order to record earthquake even in near source, at near-fault locations. We
propose to enhance the dynamic range from 19 to 24 bits (> 144 db). A possible strong-
motion sensor with optimal characteristics is the Episensor FBA ES-T (produced by
Kinemetrics Inc.).

A preliminary location of the recording stations (Table 1.2) has be planned and will be
followed by field surveys and noise measurements in order to detect the most suitable
sites for the installations of the recording stations in the island of Trinidad, with attention
also to zones with high population density. We consider also the distribution of alluvial
formations in Trinidad and, as a preliminary step, we checked the availability of approach
roads for sites using satellite images.

Table 1.2 Sites for new strong motion stations

Locality Latitude Longitude Notes

Toco 10.8198 -60.9480 2 km S-SE from urban area (near sea)

Arima 10.6228 -61.2268 5 km E from urban area

Chaguanas 10.5161 -61.4122 Urban area (alluvial site)

Tabaquite 10.3824 -61.2831 2 km E from urban area

Rio Claro 10.2890 -61.1663 2 km S-SE from urban area (alluvial site)

Guayaguayare 10.1838 -61.0580 5 km S-SW from urban area (near sea)

Preau 10.1865 -61.3078 1.5 km S from urban area

At the beginning, the new network will be composed by 10 strong motion stations. To
reduce the economical budget of the project we consider 7 new sites needed to improve
the network geometry. We can use 3 of the actual 4 sites (but with upgraded
B8

instrumentation) to obtain the 10 strong-motion stations. The interdistance in this


configuration is of the order of 20 km in the center of the island (inner zone of the
network), although is becoming higher considering the stations in the seacoast area.

Figure 1.4 a) Existing situation for velocimetric (square) and strong-motion stations (star); b)
Proposed locations for new strong-motion stations. See text for explanations
B9

The next and crucial point is the selection of the data transmission system, in order to
optimize data transfer to the acquisition center in St. Augustine. The strong-motion data
transmission has to be guaranteed even in case of strong earthquakes.

A critical analysis of data transmission has be performed in order to optimize data


transfer to the acquisition centre. Particular attention has been devoted in order to assure
strong-motion data transmission even in case of strong earthquakes.

For this project, satellite communications (VSAT) have a number of distinct advantages
over terrestrial RF types. VSAT does not suffer from the line of sight problems of
terrestrial communications. Any location with a clear view of the satellite in the sky is a
suitable location. Furthermore, Seismic Research Center has extensive expertise in VSAT
system installation and management. Since June 2005, under the coordination of the
Intergovernmental Oceanographic Commission of UNESCO, the nations bordering the
Caribbean seas and adjacent waters have been working to establish a Tsunami Warning
Network in the region. A contributor in this initiative is the Seismic Research Unit of the
University of the West Indies, St. Augustine Campus, Trinidad & Tobago. In the Figure
1.5 we can observe the satellite antenna located at the acquisition center in St. Augustine.
B10

Figure 1.5 VSAT system at St. Augustine Campus

The nature of a VSAT system is such that there is no need for repeaters. Every system
uses the same basic elements making installation, commissioning and maintenance easy.
Networks can be installed and operated after a few days of training. Remote stations can
be placed virtually anywhere, even in isolated areas with hostile terrain. The remote site
equipment is housed in a single outdoor unit and requires no integration. The lightweight
package and small Ku-band antenna are easily transported and can be installed in a few
hours. Accordingly, a choice for the project can be to select the Nanometrics Inc. Libra
VSAT telemetry systems to establish a sustainable and robust seismic network using
velocimetric and strong-motion sensors.

Libra technology is used in over 90% of the world's VSAT based seismograph networks.

A VSAT solution was the best option for our purposes and also the SRU is analyzing
seismic data using Nanometrics Atlas data analysis and processing software.

For data transmission, in most countries, costs are as low as $40/station/month for
continuous high quality 24-bit data. As an example, “a Libra network can be operated for
as little as US $5800 per year, or only $40/station/month. This is based on a typical 16-
B11

stations transmitting continuous 3 ch data at 100 sps over an Intelsat 100 kHz lease, at a
ten-year lease rate of $5800. Larger networks are just as economical, due to Libra's use of
efficient QPSK (quadrature phase shift keying) modulation and TDMA (time domain
multiple access) protocol” (Nanometrics Inc, website).

A common equipment set for all stations combined with solar power operation reduces
spare parts requirements, and simplifies network maintenance. Satellite remote field
stations allow for a complete freedom in station placement, using the same essential
configuration for all applications.

In conclusion, satellite networks are economical to maintain, every system uses the same
basic elements making installation, commissioning and maintenance easy. These systems
have lower operating costs than similar networks using telephone lines, with cost savings
achieved and maintained within 12 to 14 months.
B12
BIBLIOGRAPHY

Abrahamson M and Silva W. (2008). “Summary of the Abrahamson and Sinva NGA ground-
motion relations”, Earthquake Spectra, Vol. 24, No. 1, pp. 67-97.

Beckles D. M., Shepherd J. B. and Aspinall W. P. (1992). The “Soufriere” system: PC-based
instrumentation for acquiring and processing data from seismograph networks.
Tectonophysics, 209: 47-49.

Douglas J. and R. Mohais (2009). Comparing predicted and observed ground motions from
subduction earthquakes in the Lesser Antilles, J. Seismol., 13: 577-587.

Lynch L. (2008). Retrofitting the Caribbean Earthquake Monitoring Networks to provide Tsunami
Surveillance: Implications on Earthquake Risk Reduction, Presentation at the project
meeting; “Evaluation and Mitigation of seismic risk in the Eastern Caribbean Region: a
Workshop convened to Explore Possible Forms of Scientific and Technical Cooperation and
Research”, Institute for Critical Thinking, The University of the West Indies, St. Augustine,
Trinidad - February 25-27, 2009

Seismological instrument web sites:

Kinemetrics Inc.
http://www.kinemetrics.com/

Nanometrics Inc.
http://www.nanometrics.ca/

Refraction Technology, Inc. (REF TEK)


http://www.reftek.com/
B14
SECTION C

Telecommunications and Remote Sensing

by

Fabio Dell'Acqua1, Diego Polli1 and Lloyd Lynch2

1 European Centre for Training and Research in Earthquake Engineering (EUCENTRE), Pavia,
Italy.
2 The University of the West Indies (UWI), St Augustine, Trinidad.
INDEX

1. REMOTE SENSING........................................................................................................................1

BIBLIOGRAPHY..................................................................................................................................7
1.REMOTE SENSING

The aim of this work is to illustrate, with the help of a real-world example, how to realize
the iniziative which is being carried forward by the Telecommunications and Remote
Sensing Section within the GEO (www.earthobservations.org) under the Task DI-09-01a
label, i.e. the use of Earth Observation (EO) data for large-scale evaluation of seismic
vulnerability.

In the case at hand, the work consists of a case study on selected urban areas in the
Caribbean islands.

Starting from the coordinates of some areas of interest in the Island of Trininad,
provided by the Caribbean partners, we have selected three urban areas. To perform this
work, Google Earth® was exploited as a tool, and three polygons were created in a
unique .kml file, highlighted in the next figure.
C2

Figure 1.1 The Island of Trinidad as seen on Google Earth ©. Highlighted in red are the three
polygons with the areas of interest selected with the cooperation of the Caribbean
partners from the University of West Indies

The northmost area is called Vaslayn and it has an extent of 35 km2; slightly southwards
we find the Las Lomas area covering 30 km2, while the southmost area, Golconda,
consists of 27 km2.

At this stage we have exploited the Earth Observation Data Service (EOWEB) of the
DLR (German Aerospace Center) to obtain some Synthetic Aperture Radar (SAR)
images of the island. Thanks to a formerly activated research project, our research unit
has the possibility to apply for new acquisitions on the German TerraSAR-X satellite,
featuring 1 m ground resolution in the Spotlight mode, which is the one of interest for
our work. We have thus submitted an order for a series of new acquisition as no high
resolution archive data was available on the area. The acquisitions took place on the next
few days, then we could download 5 images on two out of three areas via FTP. The third
area could not be acquired probably due to some planning conflicts with other satellite
C3

tasks, causing the cancellation of the request, which unfortunately often happens with
low-priority acquisitions such as those for scientific use. The images were acquired on
both ascending and descending orbits, thus providing two different wiewpoints on the
same area under investigation.

Table 1 .1 Data set

Image name Area Acquisition date Orbit


204455528_1 Vaslayn 11/11/2009 Ascending
204455528_2 Vaslayn 07/11/2009 Descending
204455618_1 Golconda 11/11/2009 Ascending
204455618_2 Golconda 07/11/2009 Descending
204457327_1 Vaslayn 18/11/2009 Descending

An example of radar image on the area is shown in Figure 1.2; it is clearly visible how the
interpretation of this sort of images is far from being intuitive. Anyway, these images
allow obtaining information not available at all from optical images.
C4

Figure 1.2 A small portion of SAR image on the area under examination. Even a non-expert can
notice the footprint shape of the building, however further details on these latter are not
obvious from the image

Starting from the same three .kml files from Google Earth © a further order was
submitted to the Eurimage company redistributing optical data. We have purchased three
standard, ortho-ready QuickBird images in bundle panchromatic+multispectral version at
0.61 and 2.44 m resolution respectively. All the images spring from the same acquisition
performer on the 21st December 2005. It was not possible to obtain more recent images
as the later acquisition were affected by relevant cloud cover, which prevented a thorough
analysis of the data.

Every optical+radar image pair on the area is potentially capable of providing all the
required information for an approximate evaluation of the seismic vulnerability on the
acquired buildings. More in detail:

¾ from a radar image one can extract the number of floors, in addition to the ratio
between the height of the ground floor and the height of the other floors; the
second radar image, on the opposite orbit, should provide a cross validation;

¾ from the corresponding optical image one may detect and extract footprint of
the buildings. In particular, the multispectral allows determining the spectral
C5

signature of the construction materials and thus to detect the presence of


buildings; the panchromatic image can instead track precisely their footprint.

So, exploiting the features of the different image types, one may perform a study on the
vulnerability of the buildings found in the urban areas. From the optical images, the
footprints may be extracted using a method already known in the literature [1], while
from the SAR images, thanks to their side looking geometry, it is possible to extract the
number of floors [2], and all these information can be fed into a suitable software
combining them together to provide an estimate of the building vulnerability as
represented in the diagram in Figure 1.3.

Figure 1.3 Block diagram of the method for EO-based vulnerability assessment

The relevant data, i.e. the footprint size and number of floors, are used as an input to the
SP-BELA model [3], capable of extracting the vulnerability curves at different damage
levels for a given building.

The final goal is to obtain a map, similar to the one reported in figure 4, where every
building examined is associated with a vulnerability level, represented with different
colours to represent different vulnerability classes and making the representation more
intuitive. A similar work is in progress on the urban site of Messina, and the reader is
referred to [4] for more details on the procedure.
C6

Unfortunately the limited amount of funding and the scarce cooperation from the local
researchers did not allow the work to be finished properly, nor the problem to be
investigated in deeper detail. We have limited ourselves to a few studies, which are
however expected to fruitfully guide further research in a possible follow-up of the work.

Figure 1.4 Example of vulnerability map. Different colours represent different classes of seismic
vulnerability
BIBLIOGRAPHY

Lisini G., Tison C., Tupin F., Gamba P. (2006). “Feature fusion to improve road network
extraction in high-resolution SAR images”, Geoscience and Remote Sensing Letters, IEEE,
vol.3,no.2,pp.217-221,
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1621082&isnumber=33954

Polli D., Dell'Acqua F., Gamba P. (2009). “First Steps Towards a Framework for Earth
Observation (EO)-Based Seismic Vulnerability Evaluation”. Environmental Semeiotics
(2009) 2(1), 16-30 . DOI 10.3383/es.2.1.2 © diaRnet® 2009

Borzi B., Crowley H., Pinho R. (2008a):.Simplified Pushover-Based Earthquake Loss Assessment
(SP-BELA) Method for Masonry Buildings. International Journal of Architectural Heritage,
2:4, 353-376.

Polli D., Dell’Acqua F., Gamba P.. “Seismic vulnerability assessment in the framework of GEO: a
case study on Messina, Italy”. Proc. of 2010 Gi4DM conference, 2-4 February 2010, Torino,
Italy. Proceedings on CD-ROM.