The
motivation of this work originates from the difficulties that any
investigator,
interested in seismotectonics and seismic hazard analysis of Italy and
surrounding regions, encounters when collecting fault plane solutions
from the
literature. Italy is in fact a region with a relatively moderate
seismicity,
where the earthquakes above the magnitude threshold of the Harvard CMT
catalog
(about Mw>5.5) are only few dozens since 1976, hence the
contribution given
by first pulse mechanisms published in the literature is also necessary
to well
characterize the tectonic styles of the various seismic source
areas.
The difficulties in using such data
mainly come from the fact the they were very dispersed over a large
number of
papers as well as that, due to the papery format, they were not easily
available
for analyses by computer codes. Moreover, a number of mistakes can be
occasionally found on some
of them, going from the
incorrect use of terms (i.e. strike instead of
dip-direction) to
inconsistencies
of the reported data (i.e. non-orthogonality of
fault planes and
deformation
axes, inconsistency between plane and axes, etc.).
Therefore,
taking advantage of a call for proposals within the Framework
Project 2000/2002 of the Italian Gruppo
Nazionale Difesa dai Terremoti (GNDT), our group submitted a where one of
the tasks was Coordinate Project ("Revision
of the theoretical and observational grounds of seismic hazard
estimates at a
national scale")concerned to the
development of a
reliable catalog of focal mechanisms to be used in new hazard
estimates. We
thus started to collect focal solutions, from the national and
international
literature. Initially our interest was limited to Italy and surrounding
region,
but later the area has been extended to include the whole Mediterranean
Sea and
surrounding regions, from the Atlantic Ocean to Iran. The first version
of the
database, published more than one year ago (Vannucci and Gasperini,
2003),
was containing about 5000 mechanisms.
Such version was recently used to constrain the tectonic styles of the
seismogenic source areas of the new seismic zonation ZS9 (Working
Group, 2004)
which has been developed within the framework of the INGV
initiative
following
the Italian Prime Minister Ordinance of March 20, 2003 for the
assessment of
the new seismic hazard map of Italy.
The database is embedded in a
MS-Access application
allowing the visual comparison between original and recomputed focal
mechanisms
data, the importing of the data from the on-line Global CMT
Harvard
catalog and
from two regional RCMT catalogs (at INGV
and ETH), the
selection of
mechanisms on
the basis of various criteria as well as the exporting onto ASCII files
of the
data to be used in further computations.
An added value of the EMMA database
is the checking for
consistency of all the mechanism parameters and, in case of problems,
their
correction so that the mechanisms are immediately usable for drawing
maps or
making further computations like the ones shown in the previous article
of this
issue. We checked the angles between planes and between axes to not
differ from
90 degrees by more than three degrees. Also, we checked the consistency
between fault
planes and axes and between moment tensor components and planes and/or
axes. To
make all of these checks as well as to homogeneously re-compute all
relevant
parameters (fault planes and deformation axes angles, moment tensor
components)
for each mechanism, we have developed a structured package of Fortran
77
subroutines (FPSPACK, Gasperini and Vannucci, 2003) performing the most
common
computations and checks on focal mechanism data. This package is freely
available
from the ftp server of Computers & Geosciences journal: ftp://ftp.iamg.org/VOL29/v29-07-08.zip
and includes, among the others, routines to compute nodal planes from P
and T
axes and vice versa as well as to compute moment tensor components from
planes
or axes or best double couple parameters from moment tensor components.
All
definitions conventions and formulas we have used are reported in
Appendix A.
To make reliable selections on the
basis of the earthquake
size as well as to compute the seismic moment tensor, we uniformly
computed the
scalar seismic moment, using empirical regressions with available
magnitude
estimates, for all of the mechanisms for which this parameter is not
reported
on the original paper.
Another feature of the EMMA
database is the choice of the best mechanism among the available
duplicates. We
weighted each solution on the basis of a series of objective criteria
based on:
i) the correctness of the solution (presence or absence of errors in
the
published FPS parameters), ii) the originality of the source (original
sources
are preferred with respect to indirect ones), iii) the
"authoritativeness" of
the source (chosen roughly proportional to the impact factor of the
journal
where the mechanism is published), iv) the recentness of the
publication (most
recent papers override previous ones). We did not consider any kind of
quality
estimators (solution quality factor, number of stations, etc) as these
are only
given by a minority of the published solutions. The "best solution" is
the one
with higher weight and is marked in the database by a specific flag
that can be
included in selection.
References and related papers
Gasperini,
P., Vannucci, G., 2003. FPSPACK: A package of simple FORTRAN
subroutines to manage earthquake focal mechanism data, Computers &
Geosciences, 29, 893-901.
Vannucci, G., Gasperini, P., 2003. A database of
revised
fault
plane solutions for Italy and surrounding regions, Computers &
Geosciences, 29, 903-909.
Vannucci, G. Gasperini,
P. 2004, The new release of the database of Earthquake Mechanisms of
the Mediterranean Area (EMMA Version 2), Annals of Geophysics,
Supplement to V. 47, N.1,
307-334.