Investigation on tsunami effects in the central Adriatic Sea during the last century ' a contribution

Nat. Hazards Earth Syst. Sci., 7, 15–19, 2007
Natural Hazards
www.nat-hazards-earth-syst-sci.net/7/15/2007/
© Author(s) 2007. This work is licensed
and Earth
under a Creative Commons License.
System Sciences
Investigation on tsunami effects in the central Adriatic Sea during
the last century – a contribution
A. Maramai1, L. Graziani1, and S. Tinti2
1Istituto Nazionale di Geo?sica e Vulcanologia, Sezione Roma 2, Rome, Italy
2University of Bologna, Dipartimento di Fisica, Settore Geo?sica, Bologna, Italy
Received: 28 September 2006 – Revised: 13 December 2006 – Accepted: 13 December 2006 – Published: 10 January 2007
Abstract. In this work we present the result of a study aimed
Thouvenot and Fr´echet, 2006; Muc¸o, 2006). They put in evi-
at examining the Italian earthquake sequences that occurred
dence that the Adriatic sea presents a quite low seismic activ-
in the area of the central Adriatic sea with the purpose of
ity, mainly with intraplate earthquakes characterised by mod-
understanding whether some of them were accompanied by
erate magnitudes. Earthquakes catalogues show large uncer-
tsunami effects. The motivation for this research was the up-
tainties in the hypocentre locations and in the magnitude es-
date and enrichment of the Italian Tsunami Catalogue. The
timates: the historical seismicity is poorly documented and,
result was that evidence was found for two new cases of
in particular, one of the major dif?culties is to establish if an
earthquake-induced tsunamis: these are the August 1916 Ri-
earthquake occurred in the Adriatic sea or inland. Similarly,
mini and the October 1930 Ancona events.
instrumental seismicity shows mislocations probably due to
The bulk of the present research consisted in collecting
badly spaced recording stations. In the last decades, signif-
all the available data on the earthquakes that affected the se-
icant seismicity has been noticed in this area, with impor-
lected area in the past century and in identifying those po-
tant seismic sequences having main shocks of relevant mag-
tentially capable of generating tsunamis. During the study
nitude.
all the available material was gathered, which includes spe-
The most recent Italian Tsunami Catalogue (Tinti et al.,
ci?c monographs and scienti?c papers, articles available in
2004), spanning from 79 AD to 2004, contains a number of
contemporary chronicles and in local and national newspa-
tsunamis observed along the Adriatic coasts which can be
pers. The ?nal result of this research will improve our knowl-
put in relation to the earthquake activity. In the catalogue
edge of the tsunamigenic activity of the central Adriatic sea
each event is characterised by a “reliability” value, a very
and contribute to the assessment of the tsunami hazard and
important parameter which expresses the quality of the data,
risk along these coasts, that especially in the peak season
and by a tsunami “intensity” value. The reliability is graded
form one of the most densely populated areas of the Italian
according to the 5-degree modi?ed Iida scale (Iida, 1956)
peninsula with ?at and large beaches and water front resorts
and ranges from degree 0, equivalent to a “very improbable
crowded of tourists.
tsunami”, to degree 4 corresponding to a “de?nite tsunami”.
The tsunami intensity value is attributed on the basis of the
effects of a tsunami on structures and environment and it is
1
Introduction
based on the Sieberg-Ambraseys scale (Ambraseys, 1962).
The total number of tsunami events reported in the Ital-
In the frame of the geophysical knowledge of the Mediter-
ian Tsunami Catalogue for the central Adriatic region is 8
ranean region the geodynamic evolution of the Adriatic mi-
(Fig. 1), most of them of low intensity (intensity 2 “light” on
croplate, very complex and still debated, plays a key role.
the Ambraseys-Sieberg scale) and one reaching intensity 5
Many international studies have been carried out with the aim
(“very strong”) that is the 30 July 1627 Gargano event. The
of better assessing the seismicity of this region (Bada et al.,
catalogue divided the Adriatic region into four subregions,
1996; Babbucci et al., 2004; Battaglia et al., 2004; Venisti et
namely North Adriatic, in which one tsunami occurred, cen-
al., 2005; Pinter and Grenerczy, 2006; Altiner et al., 2006;
tral Adriatic, with 3 events and Gargano and Apulia, both af-
Correspondence to: A. Maramai
fected by 2 tsunamis. All of them, except one, are of seismic
(maramai@ingv.it)
origin.
Published by Copernicus GmbH on behalf of the European Geosciences Union.

---

16
A. Maramai et al.: Tsunami effects in central Adriatic
46.000 N
44.900 N
19.800 E
15.100 E
N
NORTH
N
ADRIATIC
Cervia
ADRIATIC
Cesenatico
A
Rimini
SEA
D R I
1672
Pesaro
I T
1875
CENTRAL A T I C S E
A
ADRIATIC
Ancona
L Y
A
43.100 N
10.700 E
GARGANO
1627
Fig. 2. Localities affected by the 1672 and 1875 Adriatic tsunamis.
APULIA
Stars represent the epicentres of the two tsunamigenic earthquakes.
40.300 N
10.600 E
agitation (Boschi et al., 1997; Tinti et al., 2004). The tsunami
Fig. 1. Geographic distribution of tsunamis in the Adriatic Sea (af-
intensity attributed in the catalogue is 2.
ter the Italian tsunami catalogue, Tinti et al., 2004). Stars repre-
On 17 March 1875 another tsunami hit the central Adri-
sent tsunami sources. Note that the 1978 tsunami source north of
atic area after a strong earthquake (Me=5.8 (CFTI, 2004),
Gargano is still uncertain and is taken here from Bedosti (1980).
I=VIII MCS scale) with epicenter in the sea close to Rimini,
The four subregions are shown.
involving central Italy and part of northern Italy. The shock
was very violent at Cesenatico, Rimini and Cervia (Fig. 2),
where many chimneys ruined and some houses suffered dam-
In this work we focus on the central Adriatic subregion,
age. At Rimini some ?shermen in boats felt a violent hori-
in order to perform a deep investigation of the area, also ?-
zontal shock and noted some big waves. In the harbour the
nalised to update the Italian Tsunami Catalogue. The most
sea level was at a high tide, passed suddenly to a low tide
relevant earthquakes that occurred in the last century have
returning to high tide again 2 min later. At Cervia the sea
been analysed, with the aim of establishing whether they
violently ?ooded quite a large part of the beach and some
could have been accompanied by tsunamis. To accomplish
sailors felt the shock on the boats. At Cesenatico some big
this goal, we gathered all suitable documents available in
waves ?ooded the beach. At Pesaro eye-witnesses on the
contemporary chronicles, in local and national newspapers
beach were reached by two big waves up to the waistband.
as well as in speci?c monographs and scienti?c articles.
At Ancona eye-witness report of some big waves (Tinti et
al., 2004). The intensity attributed to this event by the Italian
catalogue is 3 (“rather strong”).
2
Italian Tsunami Catalogue: tsunami events in central
A third tsunami event occurred on 21 June 1978: anoma-
Adriatic
lous coastal oscillations of the sea, with a period of about
10–15 min, were observed along a 300 km stretch of coasts
As mentioned before, the Italian Tsunami Catalogue (Tinti
in the central Adriatic, between Giulianova and Bari (Fig. 3).
et al., 2004) reports that the central Adriatic subregion expe-
No earthquake was recorded in that day to justify the anoma-
rienced three tsunami events, two of them of seismic origin
lous event. Bedosti (1980) hypothesises that a possible cause
and the third one of uncertain cause, probably due to a sub-
was a submarine landslide beneath the 200 m isobath more
marine landslide. Here a short description of the events is
or less between Termoli and Vasto (see star in Fig. 3). We
given.
remark that in the Italian Tsunami Catalogue this event was
The ?rst tsunami of the area occurred on 14 April 1672
attributed an unknown cause, and we still persist in this po-
when a strong earthquake caused severe damage and many
sition since new investigation is needed to unveil the real
fatalities (some sources report 200, some others more than
cause. We remark, however, that if the tsunami was induced
500). The earthquake, with equivalent magnitude Me=5.7
by a submarine landslide, the source can very unlikely be lo-
(Me is deduced from macroseismic data. For the de?nition
cated in the place suggested by Bedosti (1980) so close to the
of Me, refer to CFTI, 2004) and I=VIII Mercalli Cancani
coast. Recent investigations show that this area is not suscep-
Sieberg (MCS) scale, was located in land a few kilometres far
tible to large offshore mass failures (Trincardi et al., 2004),
from Rimini (Fig. 2). According to the coeval descriptions,
while evidence for widespread failures has been found in the
at Rimini the sea withdrew and then ?ooded the coast by
south-western Adriatic margin (Minisini et al., 2006) more
about 20 paces (15 m) and some ?shermen, that were in the
than 100 km offshore south of Gargano, where steep slopes
sea close to the Rimini coast, observed an unusual strong sea
link the shallow Adriatic platform to the deep Ionian sea.
Nat. Hazards Earth Syst. Sci., 7, 15–19, 2007
www.nat-hazards-earth-syst-sci.net/7/15/2007/

---

A. Maramai et al.: Tsunami effects in central Adriatic
17
Split
43.700 N
19.700 E
A
Vis
44.300 N
D R I
N
A
Vela Luka
T I C S E
Dubrovnik
Giulianova
Cattaro
1916
Ortona
Bar
Fossacesia
A
1930
Vasto
CENTRAL
Termoli
Vieste
Campomarino
ADRIATIC
Trani Bisceglie
Molfetta
Bari
42.300 N
40.300 N
12.900 E
Fig. 3. Localities where tsunami effects were observed during the
Fig. 4. Geographic distribution of the studied earthquakes in the
1978 Adriatic event. The star represents the position of the subma-
central Adriatic subregion.
rine landslide hypothesised by Bedosti (1980), which is still under
debate.
and the sea water ?ooded the cost by 650 m inland. Also at
Cattaro (Montenegro), as well as at Dubrovnik and at the Vis
At Giulianova the sea withdrew by about 50 m and it came
island, anomalous waves were observed.
back 6 min later surpassing the normal level by about 60 cm.
Oscillations repeated many times with period of 15 min and
We notice further that, in spite of the fact that the event
amplitude of 60 cm in the harbour. At Ortona only one oscil-
took place in modern instrumental era and affected a vast
lation was observed, with a withdrawal of 15 m. At Bisceglie
area, and that the Adriatic tides were monitored by a number
the sea withdrew by 2–3 m and then ?ooded the beach by
of tide-gauge stations at the time, we were not able to ?nd
200 m. The maximum height was 50 cm. Two ?shing boats
any instrumental records of this tsunami.
broke their mooring. At Molfetta a 10 m withdrawal was no-
ticed. The period was 5 min. Anomalous oscillations also
were observed at Fossacesia, Vasto, Termoli, Campomarino
3
Further tsunami events in central Adriatic
and Trani at different times.
Anomalous disturbances were also seen on the east-
In order to have a more complete picture of the tsunamigenic
ern coast of the Adriatic Sea: at Bar (30 cm), Dubrovnik
potential of the studied area, we performed accurate investi-
(40 cm) and Split (45 cm) (Zoe-Armanda, 1988; Papadopou-
gations on the earthquakes that occurred in the last century
los, 1993).
with the aim of identifying those potentially capable of gen-
Some newspapers report detailed description of the phe-
erating tsunamis, also minor, that are not included in the Ital-
nomenon, as follows. At Termoli the sea withdrew by tens
ian Tsunami Catalogue. To this purpose, all the available
of metres leaving the harbour dried and then violently came
material was gathered, with a scrupulous and careful search
back; some swimmers referred that the sea water suddenly
for the sources, including speci?c monographs and scienti?c
became cold and immediately after became also turbolent
papers, articles available in contemporary chronicles and in
and strongly agitated (Il Messaggero, 1978) At Bari the sea
national and local newspapers.
water suddenly withdrew many times leaving the sea?oor
The ?rst step of the analysis consisted in examining re-
dried; some moored ships in the harbour went aground and
cent Italian seismic catalogues (Boschi et al., 1995, 1997;
some ?sh was found on the beach. At Bari and Vieste the
CPTI2, 2004) and selecting the most relevant earthquakes,
sea suddenly withdrew by about 50–100 m leaving boats
with magnitude M?5.0, in the geographical area included
aground and after a few minutes the sea violently inundated
between latitude 42.300 N to 44.300 N. We examined events
the shore with very high waves, causing damage to the boats
occurred since 1900, with epicentre located in the sea or in
and some injured people. The sea water swept away the
land near the coast, namely not more than 25 km far from the
beach-umbrellas (Il Messaggero, 1978; Il Corriere Adriatico,
shore. Thirteen signi?cant earthquakes have been identi?ed
1978).
(Table 1 and Fig. 4.), and only for two of them tsunami ev-
In many islands and places along the Yugoslavian coasts
idences have been found, respectively the 16 August 1916
the sea water suddenly rose; at Vela Luka (Korcula island)
and the 1930 events.
the waves were 8 m high, inundating the ?rst ?oors of all the
The tsunami events have been examined by using the same
buildings on the shore. Many ships suffered severe damage
evaluation parameters adopted during the Italian Tsunami
www.nat-hazards-earth-syst-sci.net/7/15/2007/
Nat. Hazards Earth Syst. Sci., 7, 15–19, 2007

---

18
A. Maramai et al.: Tsunami effects in central Adriatic
Table 1. List of the studied earthquakes taken from the Italian catalogue CPTI2 (2004). In bold the two events for which tsunami evidences
were found. For a de?nition of Me see CPTI2 (2004).
Date
Latitude (N?)
Longitude(E?)
Magnitude (Me)
Intensity (MCS)
16 Nov 1908
43.155
13.596
5.4
IV
19 Feb 1911
44.120
12.080
5.4
VII
26 March 1911
44.061
12.507
5.2
VII
17 May 1916
44.000
12.630
5.8
VIII
16 Aug 1916
43.970
12.670
5.9
VIII
5 Nov 1917
43.506
13.586
5.4
VI
2 Jan 1924
43.736
13.141
5.6
VII–VIII
30 May 1928
43.706
13.122
5.0
V–VI
30 Oct 1930
43.659
13.331
5.9
VIII–IX
30 Nov 1934
44.100
14.000
5.8
V
23 Jan 1962
43.921
12.806
5.0
VI
4 Feb 1972
43.633
13.550
5.2
VIII
14 June 1972
43.650
13.600
5.4
VIII
44.900 N
On the basis of the found information we attribute to this
15.100 E
tsunami the highest reliability value 4, that means “de?-
N
nite tsunami”, and the tsunami intensity 2 “light” on the
Ambraseys-Sieberg scale.
ADRIATIC
Rimini
SEA
Riccione
1916
Pesaro
Senigallia
3.2
30 October 1930 Senigallia
1930
Ancona
43.100 N
10.700 E
This earthquake called “the Senigallia earthquake”, with epi-
center located on the coast, hit the coastal area in the cen-
tral Adriatic, with maximum effects at Senigallia (Mw=5.9
Fig. 5. Localities affected by the 1916 and 1930 tsunamis. Stars
(CPTI2, 2004), I=VIII-IX MCS scale) (Fig. 5). Here the
indicate the epicentres of the two shocks.
shock caused severe damage with the total destruction of
many buildings and 18 victims.
Catalogue compilation. Therefore each event is characterised
Boschi et al. (1995) mention a strong tsunami in the har-
by a reliability and a tsunami intensity value.
bour of Ancona. The analysis of the contemporary newspa-
pers gives a more detailed picture of the effects. They re-
3.1
16 August 1916 Rimini
port that at Ancona a sudden high tide was noted and the
sea boiled. An American steamship, moored in the “Molo
A long seismic sequence, that started on 17 May and lasted
della Sanit`a”, broke its mooring because of the waves and it
till December, caused disastrous damage at Rimini and in the
was slammed against the dock, suffering heavy damage. The
neighboring villages (Fig. 5). The main shock occurred on
railing of the harbour of?ce was almost completely pulled
17 May, but the shock of 16 August (Mw=5.9 (CPTI2, 2004)
off. The docks in the harbour were broken with large cracks.
I=VIII MCS scale) made worse the pre-existing damage and
Many ?shermen, that were on their ships in the harbour,
it was particularly strong at Rimini and Riccione and in the
quickly came back to the beach, frightened by the extraor-
localities along the coast. At Rimini 4 victims and more than
dinary movement of the water (Il Messaggero 31 October
60 injured people were counted. Descriptions of the tsunami
1930; La Nazione 31 October 1930; Corriere Padano 31 Oc-
effects are found in some newspapers where it was reported
tober 1930).
that during the shock of 16 August at Tavollo, close to Pe-
saro, small tsunami waves were observed (Corriere della Sera
Also to this event the highest reliability value 4 has been
17 August 1916; Gazzetta di Parma, 17 August 1916). Alfani
attributed and, on the basis of the description of the effects,
(1916) and Boschi et al. (1997) also report the same descrip-
the tsunami intensity is 4 “strong” on the Ambraseys-Sieberg
tion.
scale.
Nat. Hazards Earth Syst. Sci., 7, 15–19, 2007
www.nat-hazards-earth-syst-sci.net/7/15/2007/

---

A. Maramai et al.: Tsunami effects in central Adriatic
19
4
Conclusions
Battaglia, M., Murray, M. H., Serpelloni, E., and B¨urgmann, R.:
The Adriatic region: An independent microplate within the
In this study we investigate all the potential tsunamigenic
Africa-Eurasia collision zone, Geophys. Res. Lett., 31, L09605,
earthquakes occurred in the central Adriatic Sea since 1900.
doi:10.1029/2004GL019723, 2004.
Among the thirteen examined earthquakes only two were fol-
Bedosti, B.: Considerazioni sul maremoto adriatico (tsunami) del
lowed by a tsunami. These events, that were not included in
21.6.1978, Supplemento Bollettini Sismici Provv., 12-14-20,
the Italian Tsunami Catalogue, are the 16 August 1916 Ri-
giugno, luglio, ottobre 1980, Osservatorio Meteorico-Sismico
Valerio, Pesaro, 1980.
mini event and the 30 October 1930 Senigallia event. These
Boschi, E., Ferrari, G., Gasperini, P., Guidoboni, E., Smriglio, G.,
tsunamis were examined by following the catalogue guide-
and Valensise, G.: Catalogo dei forti terremoti in Italia dal 461
line, therefore the present research is both a contribution for
a.C. al 1980, Istituto Nazionale di Geo?sica – SGA Storia Ge-
the updating of the catalogue itself and an improvement on
o?sica Ambiente, Bologna, 1995.
the knowledge of the tsunamigenic potential of the central
Boschi, E., Guidoboni, E., Ferrari, G., Valensise, G., and Gasperini,
Adriatic region. As a result of this study, the number of
P.: Catalogo dei forti terremoti in Italia dal 461 a.C. al 1990, Is-
tsunamis involving central Adriatic passes from three to ?ve,
tituto Nazionale di Geo?sica – SGA Storia Geo?sica Ambiente,
and this increasing puts in evidence that a non-negligible haz-
Bologna, 1997.
ard and risk related to tsunamis exist along the central Adri-
CPTI2 – Catalogo Parametrico dei Terremoti Italiani, versione
atic coasts. These coasts are characterised by ?at and large
2 (CPTI04): INGV, Bologna, http://emidius.mi.ingv.it/CPTI,
beaches and the risk is especially high in the summer season,
2004.
CFTI – Catalogo dei Forti Terremoti Italiani: dal 461 a.C al 1997 –
when the beaches and the water front resorts are crowded
Versione on-line, release 3,1. INGV-SGA, 2004.
with tourists. We can also state that this region has the high-
Iida, K.: Earthquakes accompanied by tsunamis occurring under the
est tsunamigenic potential among the whole Adriatic subre-
sea off the islands of Japan, J. Earth Sci., Nagoya Univ., 4, 1–43,
gions, particularly for the number of occurred events, also
1956.
taking into account that the analysis was limited only to the
Minisini, D., Trincardi, F., and Asioli, A.: Evidence of Slope Insta-
last century. Nevertheless, the results we obtained represent
bility in the Southwestern Adriatic Margin, Nat. Hazards Earth
a valuable base for future tsunami modelling and hazard/risk
Syst. Sci., 6, 1–20, 2006,
assessment.
http://www.nat-hazards-earth-syst-sci.net/6/1/2006/.
The next steps of this work will be the study of potential
Muc¸o, B.: Seismicity of the Adriatic Microplate and a Possi-
tsunamigenic earthquake occurred before 1900 and, in order
ble Triggering: Geodynamic Implication, edited by: Pinter, N.,
Springer Verlag, ISBN 1402042345, 351–368, January 2006.
to get a more complete knowledge of the tsunami activity of
Papadopoulos, G. A.: Seismic faulting and nonseismic tsunami gen-
the whole Adriatic Sea, the events occurred on the Croatian,
eration in Greece, Proc. IUGG/IOC International Tsunami Sym-
Albanian and Dalmatian coasts will be studied.
posium, 23–27 August, Wakayama, Japan, 1993, 115–122, 1993.
Pinter, N. and Grenerczy, G.: Recent advances in peri-Adriatic geo-
Edited by: E. Pelinovsky
dynamics and future research directions, edited by: Pinter, N.,
Reviewed by: B. Ranguelov and another referee
Springer Verlag, ISBN 1402042345, 1–20, January 2006.
Tinti, S., Maramai, A., and Graziani, L.: The new catalogue of Ital-
ian tsunamis, Nat. Hazards, 33, 439–465, 2004.
References
Thouvenot, F. and Fr´echet, J.: Seismicity along the Northwestern
Edge of the Adria Microplate, edited by: Pinter, N., Springer
Alfani, G.:
Note ed Osservazioni sui terremoti della re-
Verlag, ISBN 1402042345, 335–350, January 2006.
gione littoranea Adriatica nell’agosto 1916, Pubblicazioni
Trincardi, F., Cattaneo, A., Correggiati, A., and Ridente, D.: Evi-
dell’Osservatorio Ximeniano dei PP. Scolopi, n. 122, Firenze,
dence of soft sediment deformation, ?uid escape, sediment fail-
1916.
ure and regional weak layers within the late Quaternary mud de-
Altiner, Y., Marianovic, M., Medved, M., and Rasic, L.: Active
posits of the Adriatic Sea, Mar. Geol., 213, 91–119, 2004.
Deformation of the Northern Adriatic Region: Results from
Venisti, N., Calcagnile, G., Pontevivo, A., and Panza, G. F.: Tomo-
the CRODYN Geodynamical Experiment, edited by: Pinter, N.,
graphic Study of the Adriatic Plate, Pure Appl. Geophys., 162,
Springer Verlag, ISBN 1402042345, 257–268, January 2006.
311–329, doi:10.1007/s00024-004-2602-6, 2005.
Ambraseys, N. N.: Data for investigation of seismic sea waves in
Zore-Armanda, M.:
Tsunamis in the Adriatic, Pomorskog
the Eastern Mediterranean, Bull. Seism. Soc. Am., 52, 895–913,
Zbornika, 26, 657–668, 1988.
1962.
Newspapers
Babbucci, D., Tamburelli, C., Viti, M., Mantovani, E., Albarello,
D., D’Onza, F., Cenni, N., and Mugnaioli, E.: Relative motion
Il Corriere Adriatico, Ancona, Year 1978, June 23.
of the Adriatic with respect to the con?ning plates: seismolog-
Il Corriere della Sera, Milano, Year 1916, August 17.
ical and geodetic constraints, Geophys. J. Intern., 159, p.765,
Il Corriere Padano, Ferrara, Year 1930, October 31.
doi:10.1111/j.1365-246X.2004.02403, 2004.
La Gazzetta di Parma, Parma, Year 1916, August 17.
Bada, G., Horvath, F., Gerner, P., and Fejes, I.: Review of the
Il Messaggero, Roma, Year 1930, October 31.
present-day geodynamics of the Pannonian basin: progress and
Il Messaggero, Roma, Year 1978, June 23.
problems, J. Geodynamics, 27(4), 501–527, 1999.
La Nazione, Firenze, Year 1930, October 31.
www.nat-hazards-earth-syst-sci.net/7/15/2007/
Nat. Hazards Earth Syst. Sci., 7, 15–19, 2007

---