25
UPDATING AND REVISION OF THE EUROPEAN TSUNAMI CATALOGUE
A.MARAMAI 1, L. GRAZIANI 1, S.TINTI 2
1
Istituto Nazionale di Geofisica e Vulcanologia
Via di Vigna Murata 605, 00143 Rome, Italy
2
Università di Bologna, Dip.to di Fisica, Settore di Geofisica,
Viale Carlo Berti Pichat, 8 40127, Bologna, Italy
Abstract
The first catalogue of the European tsunamis was built in the frame of the EU Projects
called GITEC and GITEC-TWO. The catalogue was implemented as a FoxPro 2.5
database, and it can be fully used on PC with Windows 3.1 or with the first versions of
Windows 95. In the present work, we describe the new version of the database that has
been totally rebuilt within the Visual FoxPro 6.0 DBMS environment in order to make it
suitable for the operating systems currently on the market (i.e. Windows 98, Windows
2000, Windows NT, etc.). The general structure of the previous data base version is
preserved. The catalogue is accessible through a main screen containing a) functional
buttons to perform the basic inquiry and browser actions, and b) parametric and textual data
concerning a specific event that can be selected by the user. The data base has been
enriched by including new tsunami entries, as the result of revision of historical sources,
and by adding a new category of data, namely graphical data, such as digitised tide-gauge
records, photos, relevant maps, etc., that form a special section accessible from the main
screen. Moreover, the auxiliary data base of the references has been updated by introducing
all the contributions, such as papers and scientific studies that have been published in the
last years.
1. The European Tsunami Catalogue
Tsunami catalogues are indispensable means to assess the tsunami potential of a given
region and to evaluate vulnerability and risk exposure of coastal areas and environments.
The first European tsunami catalogue was the fruit of the work of various research
groups co-operating in the frame of the projects finance by the European Union named
GITEC and GITEC-TWO and was delivered in 1998. It was not built by simply gathering
and putting together the available national catalogues, but it was the result of efforts aimed
at establishing a standard format and structure for the data base and applying uniform
criteria to evaluate the informational sources and to parameterise the data. The catalogue
was implemented as a digital data base through the DBMS FoxPro 2.5, running under
Windows 3.1 and the first version of Windows 95, but not usable correctly in PCs with the
operating systems distributed today. The problem of migrating the catalogue to the current
A. C. Yalçıner, E. Pelinovsky, E. Okal, C. E. Synolakis (eds.),
Submarine Landslides and Tsunamis 25-32.
@2003 Kluwer Academic Publishers. Printed in Netherlands
26
Figure 1. The Catalogue Interface Screen (CIS) is the main user interface of the data base: this example
displays the data of the Messina Strait, December 1908 tsunami.
PC systems (Windows 98, Windows 2000, Windows NT and so on) has been solved by
implementing a new electronic version of the catalogue under Visual FoxPro 6.0. Since this
latter is a DBMS substantially different from FoxPro 2.5, the new catalogue could not be
obtained by automatically converting the old data base, but all the basic data entry had to
be done again.
The general architecture of the catalogue is the same as the old version, to favour the
user approach. Each event is presented through the main Catalogue Interface Screen, CIS,
(displayed here below in Figure 1), showing all the event data available in the database
either directly or indirectly through the activation of further linked windows.
Examples of event data that are immediately displayed on the screen are: identification
code, date, reliability, source region and subregion, tsunami information (runup, intensity
and magnitude), and source information (cause, coordinates, etc). A short one-line verbal
text provides a description of the tsunami. But further details can be learned through the
button “Description”: by pressing it, a new temporary window, containing a brief note on
the cause, the account of the tsunami and the list of the references, opens on the CIS, and
the user can later close it by clicking the mouse on any points of the screen.
In addition to the “Description” button, the CIS includes other functional buttons that
enable the user to perform operations on the event data base or to have access to other
ancillary data bases of the catalogue, namely the data bases of the references and of the
images. This latter database is a new addition of the present version of the catalogue,
consisting in a set of graphical data such as maps, photos and mareograms of the most
relevant tsunamis that can be selected and examined by the user. In Fig.2 an example of
one of the tide-gauge figures inserted in the images section is given.
27
Figure 2. Picture included in the images database of the catalogue. It shows the mareograms with the signal
of the Messina Strait 1908 tsunami, recorded by the tide-gauge stations of Palermo, Napoli and
Civitavecchia, Italy.
2. Updating: state of the art
If we compare a conventional catalogue published in the form of a printed book, booklet or
paper, to a digital data base, we see that the main difference existing between them is that
the former is static and cannot be changed, whilst the second is subject by its nature to a
continual process of modification and updating, that in a broad sense is part of the daily
maintenance activity. The expression “state of the art” we include in the title of this section
underlines intentionally the concept that a snapshot of the catalogue as it is “today” will be
provided in this section. Up to now, updating of the catalogue resulted in addition of new
events, in the definite decision that some cases cannot be included in the catalogue and in
inclusion of additional entries in the database of the references.
We started from the re-examination of the Italian events and, in particular from 34 cases
of the XIX and XX century that were reported in the Caputo and Faita [1] catalogue of the
Italian tsunamis, but that were not inserted in the Italian section of the GITEC catalogue
[2], basically because they were supported by extremely insufficient sets of data. With the
chief purpose of acquiring new data, a very minute, careful and exhausting search has been
undertaken for each event, by consulting the most important national libraries and archives,
examining with special attention the newspapers collections (about 50 different
newspapers). Despite our efforts, we were not able to find any additional data for 32 cases.
On considering the number and importance of the data archives where no mention exists of
them, we definitely conclude that these cases are not tsunamis and cannot be entries in the
European tsunami catalogue. On the other hand, we discovered valuable data on 2 cases
that in our view have to be considered tsunami events: their occurrence dates are July 3,
1809 (Ligurian Sea) and January 15, 1940 (Northern Sicily).
Case history studies and collaborative relationships with European tsunami specialists
have been advantageous to extend the catalogue with inclusion of new well supported
events, one that occurred in Italy in 1988 and six that took place in Norway between 1867
28
and 1998. The following two subsections will be used to provide more details on the new
tsunami additions.
A further step of catalogue updating was made by going through the recent
contributions in scientific and technical literature on European events. This led us to
include several more entries in the ancillary database of the references, such as papers,
books and studies that have been published in the last 3 years.
2.1. THE ITALIAN EVENTS
The 34 Italian cases taken from Caputo and Faita catalogue [1] of which we have tried to
find evidential support through an extensive search are listed here below in chronological
order, with the further indication of the subregion and the location of supposed occurrence.
Underlined are the only 2 events that we consider worth of inclusion as new entries in the
catalogue.
January 4
1802
Dalmatian Coasts
Dalmazia Istria
July 31
1804
Campania
Gulf of Naples
August 26
1806
Latium
Ardea
July 3
1809
Liguria Côte d’Azur
La Spezia
July 27
1809
Campania
Gulf of Naples
June 28
1812
Liguria Côte d’Azur
Marseille Harbour
April 7
1813
Central Adriatic
Ancona
Nov. 12-15
1816
Liguria Côte d’Azur
Genoa
January 8
1819
Liguria Côte d’Azur
Genoa
July 23
1820
Liguria Côte d’Azur
Genoa
March 20
1822
Northern Sicily
Marsala
April 10
1822
Eastern Sicily
Catania
1824
Campania
Gulf of Naples
March 18
1826
Central Adriatic
Pesaro Senigallia
July 20
1828
Liguria Côte d’Azur
Genoa
May 26
1831
Liguria Côte d’Azur
Ventimiglia
July 2
1831
Sicily Channel
Sciacca
August 10
1838
Dalmatian Coasts
Dalmatia
October 11
1843
Dalmatian Coasts
Ragusa
June 18
1845
Messina Strait
Messina
December 29
1854
Liguria Côte d’Azur
Genoa
November 24
1862
Liguria Côte d’Azur
Genoa
Oct. –Nov.
1870
Mediterranean Sea
January 17
1871
Western Liguria
December 23
1876
Liguria Côte d’Azur
Nice
April 25
1880
Western Liguria
Nov. 15-16
1892
Campania
Ponza
December 27
1894
Northern Sicily
Filicudi
October 16
1896
Liguria Côte d’Azur
Sanremo
May 13-14
1903
Northern Sicily
Palermo
January 27
1939
Northern Sicily
Filicudi
January 15
1940
Northern Sicily
Palermo
March 16
1941
Northern Sicily
Palermo
June 22
1978
Central Adriatic
Marche
29
The 1809 event occurred in the La Spezia gulf, Ligurian Sea. We cannot attribute a
certain cause for the tsunami generation, but on the basis of the sea bottom morphology, we
can suggest the occurrence of a submarine landslide in the gulf. On the grounds of the
gathered information we assigned it reliability 2 (meaning “questionable tsunami” in the
reliability catalogue scale, i.e. to say that doubts on the tsunami occurrence can be
legitimately raised) and tsunami intensity 2, referred to the Ambraseys-Sieberg scale [3],
that means “light tsunami” with waves generally noticed on very flat shores.
A short piece of the original description quoted in the “Gazette Nationale” [4] is
reported here: “…the inhabitants of La Spezia and those living in the whole gulf observed
an extraordinary tide on July 4. …At about 8 a.m. the sea, that till that time was absolutely
calm, suddenly rose about 1 m above its usual limit. This extraordinary tide lasted for about
15-20 min. rising and falling. No apparent cause was observed…The tide was so strong and
quick that the sea water flew up to the city of La Spezia through a small canal that crosses
the city itself. Some merchants that were settled in the embankments ran away. The
seafloor in the gulf was submerged and immediately afterwards large parts of the shallow
beach were left dried and some big fish were drag by the water and trapped in the dried
beach…The first flux of the sea water was followed by 4 or 5 others that gradually
diminished their strength. …We can suppose that the effects of this extraordinary tide were
due to some seismic shock or submarine or in land.” Observe that the newspaper reporter
makes the hypothesis of an event with seismic origin, on which we cannot agree, being
rather in favour of a generation mechanism related to a submarine mass failure.
The January 15, 1940 event, in Palermo, northern Sicily, was due to a strong shock
(VIII MCS), with epicentre located in land about 20 km away from the coast, which caused
one victim and severe damage. Many localities were involved. According to “L’Ora” [5]
and the “Ufficio Centrale di Meteorologia e Geodinamica” [6], in Palermo some people
noted a strong agitation of the sea just a few seconds before the shock and immediately
after some sudden sea waves were seen in the gulf. This event has been attributed tsunami
reliability 2 and tsunami intensity 2.
One more Italian event, that occurred in 1988 at Vulcano (Aeolian Islands) and that was
never inserted in previous catalogues, has been studied and finally introduced in the data
base. A specific study on this case has been published by Tinti et al. [7]. On April 20, 1988
a large landslide occurred on the north-eastern flank of volcano La Fossa on the island of
Vulcano, during a period of increasing volcanic and seismic activity. The landslide entered
the sea generating a tsunami. Instrumental data do not exist since no tide gauges were
present in the area.
Observational data are from eyewitnesses, who upon specific interviews gave detailed
reports. A fisherman, who at the time of the landslide was in his boat just inside the
generation area, reports that at about 5:30 am local time he heard a big noise and, in
looking at the coast, he had the perception that the mountain was running towards him. A
positive, approximately 1-2 m high wave was excited that raised the boat without damaging
it. Then he felt the boat to go down, and other smaller oscillations followed the first wave.
Another boat in the same bay but farther from the source was impacted by the tsunami with
no damage. In the harbour of Porto di Levante many people observed sudden waves
entering the harbour, similar to those produced in the sea by a storm, but the weather was
fine with no wind. Wave amplitude was estimated to be approximately 0.5 m. Information
about observed waves with 0.5m amplitude in some places of Lipari Island is further
available. The reliability attributed to this event is 4, which means “definite tsunami” with
100 percent certainty on its occurrence.
30
2. 2. THE NORWEGIAN EVENTS
In the Norwegian section of the European tsunami catalogue 6 new events have been
added, chiefly thanks to the revision carried out by Dr. C. Harbitz from Norwegian
Geotechnical Institute, who made his results available to us.
On May 7, 1867 after a strong earthquake, an unusual withdrawal of the sea, followed
by oscillations for about half an hour, was observed at several locations between
Haugesund and Lindesnes, south-west Norway. The wave came from SSW, and caused
runup heights of 2-6 feet. Boats were turned around. Eyewitnesses also report roaring and
shaking of the ground [8]. This event has been attributed reliability 4 and tsunami intensity
3 (“rather strong, generally noticed, with some shipwreck”).
The second studied tsunami occurred on August 31, 1940 in northern Norway, and has
a two phase occurrence. According to eyewitnesses a landslide fell into a fjord, causing a
railway embankment to disappear in the sea, and determining a tsunami. About 5 minutes
later the sediments in the inner part of the fjord were involved in a tremendous subaqueous
slide and disappeared together with a landing pier and a small harbour, causing a second
impressive wave that was observed at the failure point of the railway embankment a couple
of minutes later [9]. The reliability attributed to this event is 4 and the tsunami intensity is
2.
The January 9, 1952 tsunami was engendered by a gravitative landslide in Mid-Norway.
A 1-2 m high wave swept past a dregder within the area of sliding in Follafjorden, NordTrøndelag [9]. From the gathered information, this event has reliability 4 and tsunami
intensity 2.
On October 6, 1979 a rockslide of 5000 cubic meters dropped vertically into the fjord
Bindalsfjorden at Hildringen near Terråk, Nordland (northern Norway). The upper part was
released from 110 m a.s.l. The runup height at the island Øksninga (1.5 km to the north)
was more than 2 m at high tide, and caused damage to boat houses and harbours [10, 11].
The detailed description suggests event reliability 4 and tsunami intensity 3.
The August 18, 1983 western Norway tsunami was generated by a 150,000 cubic
meters of rocks and by about the same amount of scree that fell from Kleppura, in the
mountain Middagshaugen, into the fjord Årdalsfjorden. A surface elevation of about 2.53.5 m was measured in the fjord, and runup heights of 5-7 m in the harbour area of
Årdalstangen, 1-1.5 km to the north-east. Small sailing vessels and light structures suffered
damage, while parts of quay structures were lifted or damaged [12, 10]. On the basis of the
above data, the event deserves reliability 4 and tsunami intensity 4.
The last event examined occurred on March, 19, 199. A gravitative rock-slide with
mass of 20-30,000 cubic meters in Sørefjorden, Hyllestad, Sogn og Fjordane (western
Norway) produced runup heights exceeding 6 m on the opposite side of the fjord. The
sources referred extensive damage on quays and boat-houses, and no victims were
registered [13].
31
3. Conclusions
The 1998 FoxPro 2.5 European catalogue of tsunamis has been replaced by a new data base
implemented in the environment of Visual FoxPro version 6.0, in order to provide a
catalogue suitable for the operating systems distributed today. The new version has been
enriched with the insertion of new events and with the creation of a specific data section
containing digitized tide-gauge records, photos and maps for the most interesting tsunamis.
The Italian and Norwegian sections of the European tsunami catalogue have been revised
and updated.
As regards the Italian part, 34 cases from 1800 to 1978 that were not included in the
1998 catalogue due to scarcity of information have been re-examined; 2 of them have been
inserted as new entries in the data base, whilst the others have been definitely eliminated.
Further, one more Italian event that took place in 1988 at the Vulcano Island has been
added to the tsunami catalogue.
As regards the Norwegian section, the data base has been enriched with 6 new tsunami
events, from 1867 to 1998.
Moreover, the European tsunami catalogue updating also involved the references data
base, with inclusion of papers, books and studies published in the last years.
At present two more events are under investigation, both causing tsunami effects along
the Italian coast but probably originated in the southern Mediterranean area.
The next steps will be in the first place the revision of all events in the data base with
reliability 0 and 1, and then the extension of the updating to all European events that were
not revised in implementing the GITEC version of the catalogue.
Acknowledgements
The authors are indebted to Dr Carl Harbitz from the Norwegian Geotechnical Institute for
information on Norwegian tsunamis. This research was carried out on funds from the
Gruppo Nazionale di Difesa dai Terremoti (GNDT) of the Istituto Nazionale di Geofisica e
Vulcanologia (INGV) and from the Ministero dell’Università e della Ricerca Scientifica e
Tecnologica (MURST).
References
1.
2.
3.
4.
5.
6.
7.
Caputo, M. and Faita, G. (1984) Primo catalogo dei maremoti delle coste italiane, Atti Accademia Nazionale
dei Lincei, Memorie Classe Scienze Fisiche, Matematiche, Naturali, Roma, serie VIII, 17, 213-356.
Tinti, S. and Maramai, A. (1996) Catalogue of tsunamis generated in Italy and in Côte d’Azur, France: a step
towards a unified catalogue of tsunamis in Europe, Annali di Geofisica, vol.XXXIX, n.6, 1253-1299.
Ambraseys, N.N. (1962) Data for investigation on seismic sea waves in the Eastern Mediterranean, Bull
Seism.Soc.Am., 52, 895-913.
Gazette Nationale, (1809) July 18, Paris.
L’Ora, (1940) January 16, Palermo.
Ufficio Centrale di Meteorologia e Geodinamica, (1940) Notizie sismiche pervenute all’UCMG per l’anno
1940, Roma.
Tinti, S., Bortolucci, E. and Armigliato, A. (1999) Numerical simulation of the landslide-induced tsunami of
1988 on Vulcano Island, Italy, Bull.Vulcanol.,61, 121-137.
32
8.
9.
10.
11.
12.
13.
Mhon, H (1867) Meddelt i Videnskabsselskapet i Christiania den 24 mai 1867, Meddelelse angaaende en
usædvanling Bevægelse af havet paa Norges Vestkyst.
Bjerrum, l (1971) Subaqueous Slope Failures in Norwegian Fjords, Norwegian Geotechnical Institute, 88, 28.
Larsen, J.O. (1999) Tension cracks and landslides in steep hard rock mountains in the Norwegian fjord
districts, Dept. of Geotech. Engn., NTNU, Trondheim, Norway.
Larsen, J.O. (1979) Hildringsflauget i Bindal kommune-befaring etter skred, Norwegian Geotechnical
Institute, Report n° 79923-1, April 23, pp. 11.
Lied, K., Larsen, J.O., Schieldrop, B., Gjevik, B. and Pedersen, G. (1984) Årdalstangen, Årdal kommune Vurdering av fare for flodbølge ved skred i Kleppura, Norwegian Geotechnical Institute, Report n° 83483-1,
Oct. 12.
Harbitz, C. and Domaas, U. (1999) Hyllestad kommune-Vurdering av skredfare og bølgehøyder i Åfjorden,
Norwegian Geotechnical Institute, report n° 981014-1.