Content uploaded by Eugen Oros
Author content
All content in this area was uploaded by Eugen Oros
Content may be subject to copyright.
Rom. Journ. Phys., Vol. 53, Nos. 7–8, P. 955–964, Bucharest, 2008
SEISMOLOGICAL DATABASE
FOR BANAT SEISMIC REGION (ROMANIA) – PART 1:
THE PARAMETRIC EARTHQUAKE CATALOGUE
E. OROS, M. POPA, I. A. MOLDOVAN
National Institute of Research and Development for Earth Physics, Bucharest, Romania;
eoros@mail.com, mihaela@infp.ro, iren@infp.ro
Received June 28, 2007
The most comprehensive Seismological Database for Banat Seismic Region
has been achieved. This paper refers to the essential characteristics of the first
component of this database, namely the Parametric Earthquakes Catalogue for the
Banat Seismic Region (PECBSR). PECBSR comprises 7783 crustal earthquakes
(3 ≤ h ≤ 25 km) with 0.4 ≤ M
i
≥ 5.6 (M
i
is M
L
, M
D
, M
S
, M
W
, Mm and/or mb from
compiled sources) occurred in the Banat region and its surroundings between 1443
and 2006. Different magnitude scales were converted into moment magnitude scale,
Mw. The completeness of PECBSR strongly depends on the time.
Key words: earthquakes, hypocenter parameters, macroseismic data, databases,
Banat Seismic Region.
1. INTRODUCTION
The western and southwestern territory of Romania, called, in this paper,
Banat Seismic Region (BSR), is the most important region of the country as
concerns the seismic hazard determined by crustal earthquakes sources. The
seismic risk in the region is also very high due to local risk factors and
vulnerabilities: weak dwellings, old and unprotected buildings in the large cities,
dams and chemical factories, high density of localities, great towns, and so on.
The studies of seismicity and seismotectonic sources used for a realistic local
hazard assessment and the reliable inputs for seismic hazard computations need
high quality data and information. Consequently a high quality seismological
database has been elaborated focused on BSR (Seismological Database for
Banat Seismic Region, SDBSR).
SDBSR has been designed in a comprehensive format because BSR is a
transfrontier region of Romania and thus the local seismic hazard is also
determined by the earthquakes occurred in the neighboring areas of Hungary,
Serbia, Bulgaria and also in nearby seismic provinces of Romania (e.g. those
defined by [2], namely Crisana, Transylvania and Western Muntenia). On the
other hand, the seismotectonic peculiarities identified inside BSR pass beyond its
boundaries. SDBSR has conceived as a relational database and it has two main
956 E. Oros, M. Popa, I. A. Moldovan 2
components: the Parametric Earthquakes Catalogue for BSR (PECBSR) and the
Catalogue of Focal Mechanism Solutions for BSR (CFMSBSR). The two
catalogues are linked together and to other components of SDBSR (e.g.
macroseismic data, digital archives, etc.).
PECBSR is the result of an elaborated and well-documented research work
started in 2003 as doctoral studies of the first author. Partially these studies have
performed into the framework of several projects of the National Institute for
Earth Physics (NIEP) Bucharest. These are the projects from the Program
„Researches about seismic hazard on national and local scale”, 2003–2005 and
the Project “Assessment of seismic hazard due to earthquakes occurred in
southwestern part of Romania and neighboring areas: its implications on
environment security and territory development”, Hasver/Ceres 2004–2006. The
work also benefited by the results of the first author’s partnership into an
International Project of European Seismological Commission, known as
EuroSeismos (“Saving and Studying the Seismograms of the strongest Euro-
Mediterranean Earthquakes”) that started in 2002 and is on development.
The final catalogue has a high level of homogeneity and is up-to-date until
the end of 2006. Its elaboration was possible especially because good scientific
and informational conditions fulfilled. Firstly, a long tradition of seismological
research focused on local seismicity and seismotetconics exists at Timisoara [7,
15]. Secondly, the region was instrumentally monitored for a long time by a
local seismic network with good performance, called Banat Seismic Micro-
network [8]. The first seismograph was installed at Timisoara in 1942 [7, 8, 15].
The EuroSeismos Project (www.storing.ingv.it/es_web) provides us with many
unique instrumental data for early historical events (since 1900).
There are many sources representative for PECRSB compilation. Other para-
metric and/or descriptive catalogues are very important. To our knowledge, only
few regional catalogues focused on BSR earthquakes exist, such as [9, 12–14].
A comprehensive catalogue covering Carpatho-Pannonian Realm was published by
[21]. Enlarged catalogues were compiled for a European/North Balkan area and
have special significance for our work, e.g. [16, 17]. Additional several catalogues
of international agencies, such as [22], were published for a long time period.
The main aim of this paper is to prezent PECBSR, the core of SDBSR. The
another catalogue, CFMSBSR, is to be published as a separate paper due to
particularities of its content and structure [10].
2. THE PARAMETRIC EARTHQUAKES CATALOGUE, PECBSR
2.1. ELABORATION PROCEDURE
The users of Parametric Earthquakes Catalogues consider them as samples
of primary data, which is unfortunately false. Generally, these ones are
3 Parametric earthquake catalogue for Banat Seismic Region 957
compilations and it is important to know all about their content and structure. In
other words, they have to be transparent as much as possible. PECBSR has been
elaborated into a comprehensive format and it is based on compiled and/or
reinterpreted earthquakes data and new computed focal parameters. The links
with other information fields of SDBSR assure a high level of transparency of
the catalogue.
The work out of PECBSR followed the next stages: settling the
geographical limits of the catalogue; identifying earthquake catalogues needful
for selecting all seismic events occurred between previous limits; collecting all
data and primary information (historical, macroseismic and instrumental ones);
analyzing and compiling the final parameters. Then all collected data were
processed using uniform criteria and principles. Additional investigations
established when they was necessary. A special attention was paid to the
historical and macroseismic available primary data that was analyzed taking into
account some recommendations of [18, 19].
The homogeneous cataloguing of earthquakes is the master principle of
PECBSR elaboration. Thus, each earthquake has a uniform standard set of
parameters, like origin time (GMT time), geographical coordinate of epicenters
(latitude and longitude), focal depth, magnitude, and epicentral intensity (degree
on EMS98 scale). Errors are also associated with each above-mentioned
parameter. Except the date and origin-time, the other parameters mutually
correlated according the conformity principle in Seismology used by [13].
2.2. SPACE, TIME AND MAGNITUDE LIMITS
To establish the space limits of PECBSR we define two areas: BSR senso
stricto (BSRss) and BSR senso largo, (BSRsl) respectively. BSRss is the area of
interest for the study with a surface of about 25000 km
2
and is bounded by 46.5–
44.5 N latitude, 21.3–22.8 E longitude as well as the national frontiers with
Hungary and Serbia. BSRss is defined mainly using the limits of historical
provinces [4], seismic risk factors and the conditions of instrumental monitoring
[7]. BSRsl is an enlarged area of BSRss and it has critical significance for
seismic hazard assessment. Its borders have been drawn at a distance defined
conservatively. Firstly, we chose an earthquake scenario with the M
max
observed
in the southern Europe, Ms = 6.6 [17] and the maximum focal depth in the
region h
max
= 25 km [10]. Then we established I = VI
O
EMS as the minimum
intensity that can be produced by this maximum earthquake into at least one
locality situated inside BSRss, nearby its borders. Finally, using the intensities
attenuation relationships from [17] we computed an average critical distance of
80 km, resulting BSRsl area of about 100000 km
2
. The epicenters located
outside BSRsl, up to 10 km distance from its borders are also catalogued.
958 E. Oros, M. Popa, I. A. Moldovan 4
PECBSR covers a time interval of 563 years (1443–2006). No pre-
established limits for magnitudes and intensities exist.
2.3. DATA SOURCES AND PROCESSING OF INFORMATION
All available sources of data and information used to elaborate PECBSR
have been classified by their structure, nature and quality of useful information
and the level of reliability. Thus, three groups of sources were identified, namely
i) main or core sources for compilation with earthquakes data (parametric and
descriptive earthquake catalogues [e.g. 1, 2, 6, 15, 16, 20], internal reports of
National Institute for Earth Physics [e.g. 11–13], earthquake catalogues
elaborated in the framework of several national projects [e.g. 9], international
bulletins [21]; ii) sources of primary information about historical earthquakes
(studies of particular earthquakes (e.g. [19]), macroseismic chestionaires,
manuscripts, documents, corespondences and notices from different archives,
historical documents like monographies, cronicles, documents from national
archives, newspapers reports, original historical seismograms obtained by direct
partenership in EuroSeismos Project (www.storing.ingv.it/es_web), analogic and
digital seismograms, seismic bulletins, etc.); iii) auxiliary sources with non-
seismic information (documents and papers about history, education, religion,
and so on, needed to interpret correctly the historical data); iv) analog and
digital seismograms, obtained especially since 1980 by National Seismic
Network; v) seismograms recorded with mobile stations network which were
installed on July and December 1991.
Three categories of investigations were applied on the datasets: i) historical
investigations for being able to interpret correctly all available information;
ii) macroseismic investigations for obtaining all intensity data points and for
mapping the macroseismic field based on new criteria; iii) instrumental
investigations for refining and/or computing new hypocentral parameters.
Historical investigations have been applied especially for become
acquainted with the conditions of recording, preservation and recovering of any
useful information for the region. Thus, the culture of the inhabitants,
demography data and political conditions, religions, economical and
administrative development, the name of localities, migrations, wars, etc. are
very important information. As an example, it is very important to know that a
terrible war between 1718 and 1789 destroyed completely numerous localities
from southeastern zone of the region and it has been repopulated after many
years only. These historical circumstances tell us a lot about the chances to find
out some seismic information. This information tell us that only some official
war reports or letters of soldiers probable preserved into an archive, a library or
somewhere else could be a chance to recover new seismic data about the region.
5 Parametric earthquake catalogue for Banat Seismic Region 959
By studying historical documents, papers, books, newspapers and letters, it was
also possible to improve the parameters of some earthquakes.
The parameters of all historical earthquakes have been revised using
calibrated macroseismic data on well-documented earthquakes basis. We obtain
some new equations of the macroseismic field, describing the mutual
relationships between magnitude, intensities and distance from the source.
Instrumental data concerning arrival times of P and S, L, R waves,
amplitudes and duration of seismic signal were obtained directly by us from
original seismograms or seismic bulletins. The earthquake occurred on October
19, 1915 is the first local event from PECBSR that has been studied using
instrumental data. We used Seisan software package [3], a simple velocity model
and station corrections to (re)locate the catalogued earthquakes. Magnitude M
D
is routinely determined using local stations recordings. M
L
and M
W
were also
computed when digital waveforms were available. Macroseismic magnitude,
Mm, is (re)determined using the new equations. Finally, PECBSR comprise all
types of magnitudes determined by us or collected from other sources. The
homogenized M
W
magnitude was obtained using a hierarchically conversion
scheme similar with [6] for conformity with national catalogue, but we also
applied some new conversion relationships obtained in this work.
2.4. STRUCTURE, FORMAT AND COMPOSITION
PECBSR has a complex structure due to the characteristics of processed
data and information and to the compiled results. There are several informational
levels interconnected and linked with the other components of the database. The
core of PECBSR is presented in a standardized form with the following
parameters: Date (year, month and day/yyyymmdd), Origin time GMT (hour,
minutes and seconds/hhmmss.s), latitude N (degree, F5.3), longitude E (degree,
F5.3), depth (km, F4.1), magnitudes (F2.1), intensity (degree, F3.1). There are
many events without any locations, too. In this case, the parameters are the
origin time, magnitude/intensity, P polarities, S-P differences if instrumental data
exist, and geographical coordinates of the locality with maximum intensity
and/or corresponding to the azimuth and to the epicentral distances computed
using 3 components polarities and arrival time differences. These events are
marked with distinct flags being eventually connected with a main shock of a
seismic sequence, with an earthquake swarm, a locality or a map or with any
other useful data.
Macroseismic maps, intensity data points (IDP), instrumental data and so
on constitute information included into PECBSR but as complementary data
being accessible either separately as independent information or as a link with
the standard main catalogue.
960 E. Oros, M. Popa, I. A. Moldovan 6
The catalogue finally comprises 7783 earthquakes. Magnitudes and
intensities range between 0.2 ≤ Mi ≥ 5.6 (Mi are magnitudes collected from
original sources and may be M
D
, M
L
, M
S
, M
W
, mb) and 2.0 ≤ Ii ≥ 9.0,
respectively (Ii is maximum observed or epicentral intensity). Fig. 1 displays a
map of earthquake epicenters (locations obtained using minimum three stations
arrival times, explosions have been excluded as far as possible). Their
distribution show many clusters of epicenters related with areas of high seismic
activity and with destructive potential historicaly confirmed (I
O
= VIII
O
EMS)
[10]. Two groups of clusters could be separated from this distribution: one in the
North-West of BSR, called Banat Seismogenic Zone and the other in South-East,
or Danube Seismogenic Zone, respectively [1, 10].
Fig. 1 – Map of epicenters for PECBSR (1443–2006). Magnitudes are in M
D
. Dashed line bounds
Banat Seismic Region. The continuous line limits the critical area around BSR (details in text).
Magnitude distribution prezented in Fig. 2 has a multimodal character. Two
main maxima with several secondary picks are highlited. These ones may be
7 Parametric earthquake catalogue for Banat Seismic Region 961
related to seismic sequences occurred frequently in the region, both as seismic
swarms and aftershocks series.
Fig. 2 – The frequency-magnitude (dM = 0.1) distribution for PECBSR (1443–2006).
Focal depths distribution is presented in Fig. 3. As it can be seen, the focal
depths vary in the region between h
min
= 3.0 km and h
max
= 25 km (these
Fig. 3 – The distributions of focal depths for PECBSR (after [10]). Statistics has computed only
for earthquakes with high quality of focal depths: for earthquakes with macroseismic focal
depths errors are σ < 0.25 h (for h > 10 km) and σ < 0.5 h for h < 10 km; for instrumental
computed focal depths σ < 5 km. The two diagrams display the distributions for BSRss (left) and
BSRsl (right) (see the text for explanations).
962 E. Oros, M. Popa, I. A. Moldovan 8
extreme values correspond to the statistics of only high quality locations from
the catalogue). This distribution emphasizes a relative concentration of the
hypocenters at different levels into the crust. The average depths computed at
BSRss and BSRsl scale, h = 11.4 and h = 12.4 km respectively, are similar to the
average depth found out by [21] for all Carpato-Pannonian Basin, without
Vrancea epicentral area.
The time distribution of threshold magnitudes displayed in Fig. 4 shows a
common trend of earthquakes catalogues in Seismology: lower limits of
magnitudes strongly depend on the different historical circumstances being
higher and higher as we go back in time. The clustering of strongest earthquakes
in the catalogue (M = 5.1–6.0) could reflect a specific time behaviour of seismic
activity with significance for hazard and prediction studies.
Fig. 4 – Time distribution for earthquakes with different magnitude classes
(1600–2006).
3. CONCLUSIONS
A high quality seismological database for Banat Seismic Region (SDBSR)
composed by two main components, Parametric Earthquakes Catalogue
(PECBSR) and the Catalogue of Focal Mechanism Solutions (CFMSBSR)
respectively, was elaborated recently in the framework of several projects of
NIEP and the Minister for Education and Research. The work also used the
results obtained by the first author as partner into the ESC EuroSeismos Project.
The paper presented the main characteristics of PECBSR. PECBSR
comprises 7783 events. It fulfills all criteria and principles required by this kind
of informational samples. Thus, it is i) homogenous (all entries were analyzed
and processed using the same philosophy, methods, algorithms and software);
ii) complete (depending on time), iii) accurate (errors of hundred meters to tens
of km depending of the data); iv) up-to-date (until December 2006). PECBSR,
by its structure and format, assures an easy access to many types of data, as:
standardized parameters of earthquakes sources, the primary data and information,
9 Parametric earthquake catalogue for Banat Seismic Region 963
complementary datasets, maps, other catalogues and so on (e.g. focal mechanism
solutions, macroseismic maps, historical/digitized/analog and digital seismo-
grams, etc.).
Fig. 5 – The distribution location errors for the earthquakes occurred since
1900. 72% of location errors are smaller than 15 km in latitude and longi-
tude, from which 91% are smaller than 10 km.
Acknowledgement. This research has been partially carried out in the framework of the
National Research and Development Programme, CERES/HASVER project, of the Minister for
Education and Research, Romania, under Contract no. 4-15/4.11.2004. The instrumental study of
some historical events has been possible by the partnership of the first author within the
EuroSeismos Project and by the useful colaboration of Dr. Graziano Ferrari and Dr. Marco
Caciagli from SGA Bologna, Italy. We are grateful especially to Ms Edith Toro from Timisoara.
Without her valuable information and kindness, most of this work could not be accomplished. We
thank to Ms. Lucia Nitoiu for her useful contribution to read and interpret many seismograms.
REFERENCES
1. I. Atanasiu, The earthquakes of Romania (in Romanian). Romanian Acad. Press, Bucharest,
(1961).
2. L. Constantinescu, V. I. Marza, A computer-compiled and computer-oriented Cataloque of
Romania’s Earthquakes during a Millenium (984–1979). Rev. Roum. Geologie,
Geophysique et Geographie, Geophysique, 24, 2, pp. 193–234 (1980).
964 E. Oros, M. Popa, I. A. Moldovan 10
3. J. Havskov, L. Ottemoller (eds.), Seisan: Earthquake analysis software. Version 7.2 ºi 8.0.
Inst. of Solid Earth Physics, Univ. of Bergen, Norway (www.ifif.uib.no), (2004).
4. V. Karnik, Seismicity of the European Area. Part 1. Catalogue of earthquakes (1901–1955).
ESC Project. Publishing House of the Czechoslocak Academy of Sciences, Praga, (1968).
5. M. Mare, Banatul între secolele IV–IX, Excelsior Art Press, Timiºoara, (2004)
.
6. M. C. Oncescu, V. Marza, M. Rizescu, M. Popa, The Romanian earthquakes catalogue
between 984–1999, In Vrancea Earthquakes: Tectonics, Hazard and Risk Mitigation;
F. Wenzel, D. Lungu (eds.), 43–47, Kluwer Academic Publishers, (1999). Updated version:
www.infp.ro.
7. E. Oros, L. Nitoiu, Timisoara (Romania) Seismological Observatory-sixty years of existence,
Abstracts, XXVIII General Assembly of ESC, 1–6 Septembrie 2002, Genoa, Italia,
(www.dipteris.unige.it/ geofisica/esc2002), SCB-5_06-P, (2002).
8. E. Oros, Banat Seismic Network (Romania). Evolution and performances. Studii si Cercetari
de Geofizica, 41, 111–125, (2003).
9. E. Oros, The Catalogue of the earthquakes occurred in Banat in the time interval 1990–2004
(in Romanian), Report of Research for the Project Ceres/Hasver, Contr. 4–15/4.11.2004,
January 2005, Bucharest, (2005).
10. E. Oros, Macroseismic and instrumental seismicity of the Banat Region and its significance on
the local seismic hazard and risk. Proc. and CD-Rom of the “Thirty Years from the
Romania Earthquake of March 4, 1977” Symposium, 1–3 March 2007, Bucharest, (2007).
11. E. Oros, M. Popa, I. A. Moldovan, E. Popescu, Seismological Database for Banat Seismic
Region of Romania. Part 2: The Catalogue of Focal Mechanism Solution, Roumanian J. of
Physics (this issue), (2007).
12. C. Radu, E. Toro, L. Niþoiu, The Catalogue of banatic earthquakes occured before 1901 (in
Romanian), Internal Report CFPS/CSEN, I, A2, pag. 13–43, Bucureºti, (1986a).
13. C. Radu, E. Toro, L. Niþoiu, The Catalogue of banatic earthquakes occured in the time
interval 1901–1950 (in Romanian), Internal Report CFPS/CSEN, III, A2, pag. 13–28,
Bucureºti, (1986b).
14. C. Radu, E. Toro, L. Nitoiu, On the seismic activity from Timisoara area (1950–1980), (in
Romanian), Internal Report CFPS/CSEN, 30-81-8/1981, Bucuresti, 112–126, (1981).
15. C. Radu, E. Toro, Development of instrumental seismology at Timisoara Station, Proc. WG
Historical Seismic Instruments and Documnents: a Heritage of Great Scientific and Cultural
Value, ESC, 16–18 Mai, 1994, Luxemburg, (1997).
16. N. V. Shebalin, V. Karnik, D. Hadzievski (eds), Catalogue of Earthquakes and Atlas of
isoseismal map, UNESCO, Skopje, (1974).
17. N. V. Shebalin, V. Leydecker, G. Mokrushina, N. G. Tatevossian, R. E. Erteleva, O. O. Yu.
Vassiliev (1998): Earthquake Catalogue for Central and Southeastern Europe 342 BC –
1990 AD. European Commission, Final Report to contract ETNU CT93-0087, Brussels,
(1998).
18. M. Stucchi, Recomandation for compilation of a European parametric earthquake catalogue,
with special reference to historical records. In Materials of CEC Project „Review of
Historical Seismicity in Europe”, (Albini P., Moroni A., eds), CNR, Milano, 2, 254 pp,
(1994).
19. M. Stucchi, et al. (1998). A Basic European Earthquake Catalogue and a database (BEECD).
An EC „Environment and Climate” Project. Contr. EVV-CT94-0497 (1995–1998), (1998).
20. Zsiros T. (1983a). The Ermellek earthquake of 1834. Acar. Geodaet., Goephzs., et Montanist.
Hunh., 18 (1–2), pp. 129–134.
21.
T. Zs
iros, A Karpat-medence szeizmicitasa es foldrenges veszelyessege: Magyar foldrenges
katalogus (456–1995), MTA FK GGKI, Budapest, ISBN 9638381159, (2000).
22. *** International Seismological Center, On-Line Bulletin, http://www.isc.ac.uk, Thatcham, U.K.
