Sbarra, P., Tosi, P., & De Rubeis, V. (2017). Role of African–Eurasian plate setting in the felt areas of intermediate‐depth earthquakes: an investigation using crowdsourced data. Terra Nova, 29(1), 36-43. Accepted version. The final publication is available at http://onlinelibrary.wiley.com/doi/10.1111/ter.12245/abstract
Role of African-Eurasian plate setting in the felt areas of intermediate-depth earthquakes: an
investigation using crowdsourced data.
Sbarra Paola1*, Tosi Patrizia1, and De Rubeis Valerio1
1 Istituto Nazionale di Geofisica e Vulcanologia, Via di Vigna Murata 605, Roma, Italy [email protected]
ABSTRACT
Greek intermediate-depth earthquakes, occurring in the subducted plate of the Hellenic Arc are felt at greater distances than expected, reaching Italy in some cases. We study in detail macroseismic intensity data from intermediate-depth Italian and Greek earthquakes collected from Internet users who felt the shaking in Italy. The huge amount of data allowed to outline the felt/not-felt limit and to find a correspondence between attenuation areas and presence of asthenospheric material at shallow depths. We show that plate boundaries, known to produce the majority of earthquakes, are, in some specific cases, the boundaries of areas in which earthquakes are felt. The Ionian subducted lithosphere propagates seismic waves with low attenuation over large distances, whereas high- attenuation zones in Italy, linked to asthenospheric upwelling, limit the propagation as evidenced also by PGA values. We identify a typical pattern that can be used to recognise intermediate-depth earthquakes, and to properly locate historical events.
KEYWORDS
Intermediate-depth earthquakes, plate boundary, macroseismic data, macroseismic field, subduction.
1
INTRODUCTION
The interplay among the African and Eurasian plates produces a broad zone of deformation in
Mediterranean region (Fig. 1). The Ionian oceanic lithosphere, of African plate, is characterised by a divergent double subduction (Fig. 1a, c): the low-angle eastward-directed Hellenic subduction, where earthquakes occur at depths < 180 km (Papazachos et al., 2000), and the westward-directed
Calabrian subduction, where earthquake depths reach more than 400 km (Selvaggi and Chiarabba,
1995). The Adria continental lithosphere, of the African plate (Altiner et al. 2006), dips beneath the
Eurasian continental plate in northern Sicily (Hyblean subducted lithosphere, Fig. 1a, b; Doglioni et al., 2001) and in Apulia region sinks beneath the Eurasian plate with a steep slab (Fig. 1a, d;
Carminati and Doglioni, 2005). The Calabrian slab seems to be the only active subduction in Italy
(Selvaggi and Chiarabba, 1995; Faccenna et al., 2005), because in the other portions of the
Apennine Range there is an absence of deep seismicity. On the other side the presence of contractional seismicity in the accretionary wedge all around the Apennines suggests that the subduction is active also outside the Calabrian arc (e.g., Cuffaro et al., 2010; Carminati et al.,
2012); while the explanation of the paucity of deep seismicity can be searched in the continental nature (Fig. 1) of the subducting lithosphere (Carminati et al., 2002).
Some Greek events of the past showed an anisotropic macroseismic fields. For example in the 1926
M 7.4 earthquake, near the Rhodes Island, intensities were attenuated in the back-arc region of the
Hellenic arc, while the felt area reached Lower Egypt and southern Italy with maxima intensities of
VI Mercalli-Cancani-Sieberg (MCS) and in few cases of VII (Ambraseys and Adams, 1998, and references therein). This pattern was later confirmed by seismological observations recorded during
Greek intermediate-depth earthquakes (Konstantinou and Melis, 2008). Utsu et al. (1966) and
Oliver and Isacks (1967) observed the same phenomenon respectively in Japan and in the Tonga
Arc, and hypothesised that seismic waves propagate with low-attenuation (high-Q) through the subducting lithosphere with a large contrast to the surrounding asthenosphere characterised by high
2 attenuation (low-Q). Then the low-attenuation and the small-scale heterogeneities in the slab allow the seismic waves to travel longer distances with lesser attenuation than those propagating outside the slab (Chen et al., 2013; Garth and Rietbrock, 2014), generating large amplitude and high- frequency signals, which produce strong seismic intensity anomalies (Furumura and Kennett, 2005;
Kennett and Furumura, 2008; Chen et al., 2013; Sun et al., 2014). Here, we show how the six intermediate-depth Greek earthquakes that occurred between February 2007 and August 2014 with moment magnitudes (Mw) > 5.4, inside the rectangle showed in figure 1 (Table 1), were felt in Italy and were recorded by instruments, with the aim to investigate the relation between the anomalous macroseismic pattern and the plate tectonic setting. We also analysed the cases of two intermediate- depth Italian earthquakes to confirm the found relation.
DATA
Data were collected through the web-based macroseismic questionnaire, about effects produced by earthquakes, of Istituto Nazionale di Geofisica e Vulcanologia (Sbarra et al., 2010; Tosi et al.,
2015). The questionnaire is compiled by volunteers after the occurrence of any earthquake potentially felt in the Italian territory. Since 2009 a volunteer can become a registered user (now approximately 24,500, homogeneously distributed on Italian territory). A registered user receives an email, immediately after the occurrence of an earthquake within a specific distance depending on earthquake magnitude and depth, with an invitation to fill up the questionnaire also in the case of
“not felt” earthquake.
Crowdsourcing supplies a huge amount of macroseismic data, enabling analysis that reveals new details, even for small magnitude earthquakes (M < 3). Questionnaire data were used to determine municipal macroseismic intensity values on the Mercalli–Cancani–Sieberg (MCS) scale (Hai
Sentito Il Terremoto, “HSIT” database, available at http://www.haisentitoilterremoto.it), following the method of Tosi et al. (2015). Considering that a reliable distinction between the macroseismic degrees I (not felt) and II (felt by less than 5% of people) should require a large sample of
3 population, we do not differentiate these two degrees because we assumed on-line questionnaire data to be not suitable for such discrimination (Tosi et al. 2015).
A comparison between macroseismic and peak ground accelerations (PGA) data was made using waveforms from European Integrated Data Archive (EIDA) infrastructure within ORFEUS. PGA values derived from the largest of either horizontal components were converted to MCS degrees
(experimental law by Piromallo, personal communication, obtained using Northern Italy 2012 sequence data). PGA values greater than 0.0339m/s2 correspond to IV MCS, 0.007≤PGA<0.0339 to
III MCS. Values less than 0.007m/s2, corresponding to I-II MCS, were further divided above and below 0.003m/s2 in order to give more detail.
ANALYSIS OF GREEK EARTHQUAKE CASES
The macroseismic fields of the six intermediate-depth Greek earthquakes (Table 1 and Fig. 1 red stars) show that the events were felt in southern Italy (Fig. 2 a-f), far from their sources, with intensities of III, IV, and V MCS with an irregular pattern. To evaluate the presence of an anomalous extension of the felt area we compared these macroseismic fields with the boundaries of the expected III degree intensity areas (Fig. 2 a-f, insets) calculated with the intensity prediction equation (IPE) provided by Atkinson and Wald (2007), derived using online questionnaires data for magnitudes comparable with the studied Greek earthquakes. Following Atkinson and Wald (2007)
IPE we highlight that the six earthquakes should not have been felt in Italy. On the other hand, as expected, shallow crustal earthquakes (Fig. 1a, green stars) occurred in Greece are generally not felt in Italy; in fact six of eight shallow events occurring in the same period and spatial window (Table
1 and Fig. 3c) were not felt (based on non-response or ‘not felt’ questionnaire responses), whereas the other two events (Fig. 1a stars number 14 and 17) generated only a few ‘felt’ responses (Fig. 3 a, b). Anyhow, these two last events had equal or larger magnitudes (Mw ≥ 6.4) and shorter epicentral distances than the intermediate-depth earthquakes.
4 For the events illustrated in Fig. 2d and f, we were able to have also ‘not felt’ data, as we manually enlarged the area in which registered users were solicited immediately after the events for information in these two cases. The areas characterised by I-II MCS values for these events generally correspond to those lacking data for other events (Fig. 2a–c, e) thus it is reasonable to assume that areas with lack of answers are mostly representative of not felt areas. This hypothesis is confirmed by PGA data (Fig. 2 g-l, 3 d-f), which showed a general agreement with macroseismic intensity fields. All six intermediate-depth investigated earthquakes were felt mainly in the southernmost part of Sicily and in Apulia, whereas they were not felt or only scarcely felt in
Calabria, northern Sicily, and west of Apulia (Fig. 2). This characteristic pattern of seismic intensity is not observed for shallow earthquakes (Fig. 3).
An explanation for such unexpected high intensity of these events in southern Italy could be, as said before, the efficient propagation of seismic waves through the subducted lithosphere, which allowed them to reach greater distances than seismic waves originating from shallow earthquakes.
Under this hypothesis seismic waves propagated efficiently from the hypocentre through the Ionian lithosphere, reaching the Hyblean and the Apulian platforms belonging to the African plate (Fig. 1b and 1d). The efficiency of propagation seems to depend on hypocentre depth; for example, the deepest event (111 km; Fig. 2e, k) was felt intensely although its magnitude was only 5.6.
Moreover, the Italian macroseismic fields are irregular: the majority of felt data are located to the
African plate, whereas the majority of ‘not felt’ and ‘scarcely felt’ data are located to the west of the
African–Eurasian plate boundary on the Eurasian continental plate (Fig. 1a). This abrupt attenuation, beyond the boundary, can be explained by considering the anomalous upwelling at shallow depths of asthenospheric material under the Eurasian plate, related to the mantle wedge typically present in the hanging wall of W-directed subduction zones (Doglioni, 1991; Doglioni et al., 2009). The asthenospheric material reaches the plate boundary with African plate (Gvirtzman and Nur, 1999) as evidenced by low Vp anomalies, which attenuates seismic waves (Chiarabba et al., 2008; Di Stefano et al., 2009). In other subduction zones, wave propagation is usually efficient
5 to the volcanic arcs (Furumura and Kennett, 2005; Konstantinou and Melis, 2008), behind which waves are attenuated by the asthenosphere. Conversely, in the Italian cases, asthenospheric material reaches the plate boundary due to rollback (Doglioni et al., 2001; Di Stefano et al., 2009; Schellart,
2010) or detachment (Spakman et al., 1993; Piromallo and Morelli, 1997; Gvirtzman and Nur,
1999; Chiarabba et al., 2008) of the slabs, which explains the areas in which the intermediate-depth
Greek earthquakes were not felt or scarcely felt (Fig. 2).
The different macroseismic field of shallow and intermediate-depth earthquake can be a marker of the event depth. The European-Mediterranean Seismological Centre (EMSC) estimated that the 12
October 2013 event (Fig. 2d) occurred at a depth of 47 km, whereas the Centroid Moment Tensor catalogue lists a depth of 15 km for the same event. The macroseismic field of this event is similar to those of the intermediate-depth earthquakes analysed, thus supporting the EMSC location.
ANALYSIS OF ITALIAN EARTHQUAKE CASES
Also some intermediate-depth earthquakes occurred in southern Italy had anomalous macroseismic fields with the same pattern of the Greek ones. One of these earthquakes, occurred on 7 November
2009 with ML=4.1 (Fig. 4a), was generally scarcely or not felt (few data points with intensities > I-
II MCS) near the epicentre, located in the forearc region on the Eurasian plate, whereas it was felt in southern Sicily, on the Hyblean platform belonging to African plate (Fig. 1a, b). Another event with an anisotropic macroseismic field (Fig. 4b), MW=4.4, occurred in the Calabrian slab on 24
March 2013 (Fig. 1a, c); shaking was scarcely or not felt in Calabria (Eurasian plate), near the epicentre, whereas it was felt mainly in southern Sicily (African plate). The lack of ‘felt’ data near the epicentre, highlighted even by intensities derived from PGA values (Fig. 4c, d), could be connected also in these cases, with the presence of asthenospheric material at shallow depths, under the Eurasian plate, (Fig. 1a, b, c) in central Sicily and Calabria (Chiarabba et al., 2008; Di Stefano et al., 2009). The hypocentres of both these earthquakes have depth shallower than the Greek ones
(Fig. 1a, 1b, 4). Considering the felt data, both the fields of figure 4 had intensities greater than
6 expected (insets of Fig. 4), according to the IPE by Tosi et al. (2015) that was elaborated with HSIT data, and it is more appropriate for Italian small magnitude events. However even using the IPE by
Atkinson and Wald (2007) this result is confirmed. This behaviour shows that both the continental
Hyblean subducted lithosphere (Fig. 4) and the oceanic Ionian lithosphere of the Calabrian slab efficiently propagate the seismic waves despite the fact that the hypocentres were probably located at the tectonic contact on the top of the African plate. In particular the second case (Fig. 4b) suggests the continuity of the Hyblean and Calabrian slabs at shallow depths, whereas they are probably laterally discontinuous and independent at depths > 150 km as highlighted by the tomographic analysis of Chiarabba et al. (2008).
IMPLICATION FOR HISTORICAL EARTHQUAKE LOCATION
These results emphasise the importance of considering deep structures when locating historical earthquakes. For example, an earthquake felt mainly in the Hyblean platform may be misinterpreted as a shallow earthquake that occurred near the macroseismic epicentre, instead of an intermediate-depth event with a hypocentre located in the Hyblean, Calabrian, or even Hellenic slab. Furthermore, historical catalogues may contain two entries, one with an epicentre located in
Greece and the other located in Italy, that are actually referred to the same intermediate-depth
Greek earthquake. One such event may be that reported in the inscriptions dating the reconstruction of churches on Rhodes Island (Stiros et al., 2006). It occurred in north-western Crete around 1717 and had an estimated magnitude > 7.0 (Fig. 1a, grey star) (Papazachos and Papazachou, 2003).
With a hypothesised intermediate-depth hypocentre, shaking would have been felt in southern
Sicily and Apulia. The Database Macrosismico Italiano 11 (Locati et. al., 2011) records four events felt in Sicily around 1717, but if our interpretation is correct, it is possible to exclude three of them felt in northern Sicily (Eurasian plate). Thus the most probable duplicate event occurred on 4 April and caused slight or moderate damage in Vittoria, southern Sicily (Adria lithosphere). This correspondence provides a constraint for the date of earthquake occurrence. The presence of
7 damages in Vittoria places emphasis on the consequences that a strong intermediate-depth Greek earthquake could cause in Italian territory, as happened in the case of the 26 June 1926 Crete earthquake (Critikos, 1928), which reached VI MCS (corresponding to slight damage) in Italy.
DISCUSSION AND CONCLUSIONS
The irregular intensity distribution in the far field, observed when an intermediate-depth earthquake occurs in subducted lithosphere, is a phenomenon currently not fully understood. The studies made in the regions of Vrancea (Sørensen et al. 2010) and in Taiwan (Chen et al., 2013), show that the main factor for intensity amplification is the presence of a subducting lithosphere while the radiation pattern and to the local site effect played a secondary role. In Mediterranean region, the presence of the Ionian divergent double subduction, on which seismic waves travel for many kilometres without following common attenuation laws, allows a Greek intermediate-depth earthquake to be felt in Italy and, in theory, vice versa.
Several authors examined the macroseismic fields of two intermediate-depth Greek earthquakes abnormally felt in Italy: 26 June 1926, Ms = 7.4 (Ambraseys and Adams, 1998 and referenced therein) and 11 August 1903, Ms = 7.3 (Sieberg 1932) but the scarcity of data and the grouping of low intensity degrees did not allow the definition of a clear pattern. The greater amount of data and the availability of not-felt reports provided this work with a more detailed macroseismic field allowing a better understanding of wave propagation in this region. We have shown that intermediate-depth Greek earthquakes of Mw > 5.4 are generally felt in Italy on the African plate and that these events are generally not felt on the Eurasian plate showing a sharp change in macroseismic intensity in correspondence of the plate boundary as evidenced also by PGA data that showed a quite good agreement with macroseismic intensities. This study gives the first evidence for a characteristic macroseismic pattern of intermediate-depth earthquakes that can be used as an indication of the deep or shallow nature of an earthquake occurring within the studied area. In this way we demonstrate that non-instrumental data, such as macroseismic data obtained by
8 crowdsourcing, can highlight the presence of deep structures related to the African–Eurasian plate boundary zone, revealing high-attenuation areas, associated with the presence of asthenospheric material at low depth, that in this particular case reaches the plate boundary (Gvirtzman and Nur,
1999) as evidenced by tomographic studies (Chiarabba et al., 2008, Di Stefano et al., 2009) and amplification areas connected to the efficient propagation of seismic waves in the subducted lithosphere.
Our results also indicate that deep structures should not be neglected to properly locate historical events, to predict macroseismic intensity and to realise seismic hazard maps.
ACKNOWLEDGMENT
M. Di Bona, F. Mele, C. Piromallo, A. Rovelli and R. Vallone are acknowledged for helpful advice and D. Sorrentino for information technology architecture of http://www.haisentitoilterremoto.it.
REFERENCES
Altıner, Y., Bačić, E., Bašić, T., Coticchia, A., Medved, M., Mulić, M.I. and Nurçe, B., 2006.
Present-day tectonics in and around the Adria plate inferred from GPS measurements: Special
Paper 409. Postcollisional Tectonics and Magmatism in the Mediterranean Region and Asia,
409, 43–55, doi: 10.1130/2006.2409(03).
Ambraseys, N. N., and Adams, R. D., 1998. The Rhodes earthquake of 26 June 1926. Journal of
Seismology, 2, 267–292.
Atkinson, G. M., and Wald, D. J., 2007. "Did You Feel It?" Intensity Data. A Surprisingly Good
Measure of Earthquake Ground Motion. Seismological Research Letters, 78, 362–368.
Carminati, E., and Doglioni, C., 2005. EUROPE | Mediterranean Tectonics. Encyclopedia of
Geology, 135–146, doi: 10.1016/b0-12-369396-9/00135-0.
9 Carminati, E., Giardina, F., and Doglioni, C., 2002. Rheological control of subcrustal seismicity in
the Apennines subduction (Italy). Geophysical Research Letters, 29,
doi:10.1029/2001GL014084.
Carminati, E., Lustrino, M., and Doglioni, C., 2012. Geodynamic evolution of the central and
western Mediterranean: Tectonics vs. igneous petrology constraints. Tectonophysics, 579,
173-192, doi:10.1016/j.tecto.2012.01.026.
Chen, K. H., Kennett, B. L., and Furumura, T.,2013. High frequency waves guided by the
subducted plates underneath Taiwan and their association with seismic intensity anomalies.
Journal of Geophysical Research: Solid Earth, 118, 665-680, doi: 10.1002/jgrb.50071.
Chiarabba, C., Gori, P.D., and Speranza, F., 2008. The southern Tyrrhenian subduction zone: Deep
geometry, magmatism and Plio-Pleistocene evolution. Earth and Planetary Science Letters,
268, 408–423, doi: 10.1016/j.epsl.2008.01.036.
Critikos, N. A., 1928. Le tremblement de terre de la mer du Crete du 26 juin 1926. Annales de
l'Observatoire national d'Athènes, 10, 39-47.
Cuffaro, M., Riguzzi, F., Scrocca, D., Antonioli, F., Carminati, E., Livani, M., & Doglioni, C.,
2010. On the geodynamics of the northern Adriatic plate. Rendiconti Lincei, 21, 253-279,
DOI 10.1007/s12210-010-0098-9.
Di Stefano, R., Kissling, E., Chiarabba, C., Amato, A., and Giardini, D., 2009. Shallow subduction
beneath Italy: Three-dimensional images of the Adriatic-European-Tyrrhenian lithosphere
system based on high-quality P wave arrival times. Journal of Geophysical Research, 114,
B5, doi: 10.1029/2008jb005641.
Doglioni, C., 1991. A proposal for the kinematic modelling of W‐dipping subductions‐possible
applications to the Tyrrhenian‐Apennines system. Terra Nova, 3, 423-434.
Doglioni, C., Innocenti, F., and Mariotti, G., 2001. Why Mt Etna? Terra Nova, 13, 25–31, doi:
10.1046/j.1365-3121.2001.00301.x.
10 Doglioni, C., Tonarini, S. and Innocenti, F., 2009. Mantle wedge asymmetries and geochemical
signatures along W-and E–NE-directed subduction zones. Lithos, 113, 179-189,
doi:10.1016/j.lithos.2009.01.012.
Faccenna, C., 2005. Constraints on mantle circulation around the deforming Calabrian slab.
Geophysical Research Letters, 32, doi: 10.1029/2004gl021874.
Furumura, T., and Kennett B. L. N., 2005. Subduction zone guided waves and the heterogeneity
structure of the subducted plate: Intensity anomalies in northern Japan. Journal of
Geophysical Research, 110, B10, doi: 10.1029/2004jb003486.
Garth, T., and Rietbrock, A., 2014. Downdip velocity changes in subducted oceanic crust beneath
Northern Japan insights from guided waves. Geophysical Journal International, 198, 1342–
1358, doi: 10.1093/gji/ggu206.
Gvirtzman, Z., and Nur, A., 1999. The formation of Mount Etna as the consequence of slab
rollback. Nature, 401, 782–785, doi: 10.1038/44555.
Kennett, B. L. N., and Furumura, T., 2008. Stochastic waveguide in the lithosphere: Indonesian
subduction zone to Australian craton. Geophysical Journal International, 172, 363-382, doi
10.1016/S0065-2687(08)00007-1.
Konstantinou, K.I., and Melis, N.S., 2008. High-Frequency Shear-Wave Propagation across the
Hellenic Subduction Zone. Bulletin of the Seismological Society of America, 98, 797–803,
doi: 10.1785/0120060238.
Locati, M., Camassi R., & Stucchi, M., eds.,2011: DBMI11, the 2011 version of the Italian
Macroseismic Database. Milano, Bologna, http://emidius.mi.ingv.it/DBMI11, DOI:
10.6092/INGV.IT-DBMI11
Oliver, J., and Isacks, B., 1967. Deep earthquake zones, anomalous structures in the upper mantle,
and the lithosphere. Journal of Geophysical Research, 72, 4259–4275.
11 Papazachos, B., Karakostas, V., Papazachos, C., and Scordilis, E., 2000. The geometry of the
Wadati–Benioff zone and lithospheric kinematics in the Hellenic arc. Tectonophysics, 319,
275–300, doi: 10.1016/s0040-1951(99)00299-1.
Papazachos B., and Papazachou, C., 2003. The earthquakes of Greece: Editions Ziti, Thessaloniki.
Piromallo, C., and Morelli, A., 1997, Imaging the Mediterranean upper mantle by P-wave travel
time tomography. Annali di Geofisica, 11, 963-979.
Sbarra, P., Tosi, P., and Rubeis, V.D., 2010. Web-based macroseismic survey in Italy: method
validation and results. Natural Hazards, 54, 563–581, doi: 10.1007/s11069-009-9488-7.
Schellart, W.P., 2010. Mount Etna-Iblean volcanism caused by rollback-induced upper mantle
upwelling around the Ionian slab edge: An alternative to the plume model. Geology, 38, 691–
694, doi: 10.1130/g31037.1.
Selvaggi, G., and Chiarabba, C., 1995. Seismicity and P-wave velocity image of the Southern
Tyrrhenian subduction zone. Geophysical Journal International, 121, 818–826, doi:
10.1111/j.1365-246x.1995.tb06441.x.
Sieberg, A., 1932. Die Erdbeben. In: Gutenberg, B., Handbuch der Geophysik, 4, 93–94, Berlin.
Sørensen, M. B., Stromeyer D., and Grünthal G., 2010. A macroseismic intensity prediction
equation for intermediate depth earthquakes in the Vrancea region, Romania. Soil Dynamics
and Earthquake Engineering, 30,.1268-1278, doi:10.1016/j.soildyn.2010.05.009.
Spakman, W., Lee, S.V.D., and Hilst, R.V.D., 1993. Travel-time tomography of the European-
Mediterranean mantle down to 1400 km. Physics of the Earth and Planetary Interiors, 79, 3–
74, doi: 10.1016/0031-9201(93)90142-v.
Stiros, S., Papageorgiou, S., Kontogianni, V., and Psimoulis, P., 2006. Church repair swarms and
earthquakes in Rhodes Island, Greece. Journal of Seismology, 10, 527–537, doi:
10.1007/s10950-006-9035-x.
12 Sun, D., Miller, M. S., Agostinetti, N. P., Asimow, P. D., & Li, D. (2014). High frequency seismic
waves and slab structures beneath Italy. Earth and Planetary Science Letters, 391, 212-223,
doi: 10.1016/j.epsl.2014.01.034.
Tosi, P., Sbarra, P., Rubeis, V.D., and Ferrari, C., 2015. Macroseismic Intensity Assessment
Method for Web Questionnaires. Seismological Research Letters, 86, 985–990, doi:
10.1785/0220140229.
Utsu, T., 1966. Regional difference in absorption of seismic waves in the upper mantle as inferred
from abnormal distribution of seismic intensities. Journal Faculty Sciences Hokkaido
University, 2, 359–374.
FIGURE CAPTIONS
Figure 1. Geographic and tectonic setting of the study area. a, Schematic map of the African–
Eurasian plate boundary (demarcated in red) with tracks of schematic sections, epicentres of intermediate-depth earthquakes analysed in this study (red stars), epicentre of 1717 event (gray star) and epicentres of shallow earthquakes (green stars). The black rectangle bounds the analysed Greek spatial window (21.2–27° east longitude, 35.5–40.5° north latitude); b, schematic section A–B–C; c, schematic section D–E–C; d, schematic section F–G.
Figure 2. Italian macroseismic fields (a-f) and macroseismic intensities obtained through PGA values (g-l) of intermediate-depth Greek earthquakes. Epicentre location, magnitude, and depth were obtained from the EMSC catalogue. Blue insets areas depict expected intensities ≥ III MCS
(a-f). Geographic coordinates for the maps are between 12° and 20° east longitude and between
36°and 43° north latitude. a and g, 3 February 2007 (Fig. 1a, star no. 4); b and h, 6 January 2008
(Fig. 1a, star no. 1); c and i, 1 April 2011 (Fig. 1a, star no. 6); d and j, 12 October 2013 (Fig. 1a, star no. 5); e and k, 4 April 2014 (Fig. 1a, star no. 3); f and l, 29 August 2014 (Fig. 1a, star no. 2).
13 Figure 3. Italian macroseismic fields (a-c) and macroseismic intensities obtained through PGA values (d-f) of shallow Greek earthquakes examined in this study. Epicentre location, magnitude, and depth were obtained from the EMSC catalogue. Blue insets areas depict expected intensities ≥
III MCS (a-c). Geographic coordinates for the maps are between 12° and 20° east longitude and between 36°and 43° north latitude. a and d, 14 February 2008 (Fig. 1, star no. 17); b and e, 8 June
2008 (Fig. 1, star no. 14); c and f, 24 May 2014 (Fig. 1, star no. 10).
Figure 4. Macroseismic fields (a-b) and macroseismic intensities obtained through PGA values (c- d) of Italian earthquakes examined in this study. Epicentre location, magnitude, and depth were obtained from the Italian Seismological Instrumental and parametric Database catalogue. Blue insets areas depict expected intensities ≥ III MCS (a-b). Geographic coordinates for the maps are between 12.3° and 18.5° east longitude and between 36°and 39.5° north latitude. a and c, 7
November 2009 (Fig. 1a, star no. 7); b and d, 24 March 2013 (Fig. 1a, star no. 8).
14
Star ID Date Time UTC Lat. Lon. Depth (km) Mw Responses Fig. [MM/DD/YY]
1 01/06/08 05:14:18 37.16 22.64 72 6.2 317 in 148 municipalities 2b
2 08/29/14 03:45:06 37.26 23.61 95 5.8 1175 in 2f 518 municipalities
3 04/04/14 20:08:07 37.19 23.74 111 5.6 374 in 115 municipalities 2e
4 02/03/07 13:43:20 35.78 22.57 60 5.5 34 in 19 municipalities 2a
5 10/12/13 13:11:54 35.56 23.31 47 6.4 1115 in 431 municipalities 2d
6 04/01/11 13:29:11 35.54 26.63 60 6 40 in 30 municipalities 2c
7 11/07/09 07:10:05 37.69 13.85 34 4.1 146 in 42 municipalities 4a (ML)
8 03/24/13 15:47:21 37.67 16.46 38 4.4 285 in 101 municipalities 4b
9 1717 ? 35.90 22.90 ? 7.0 / /
10 5/24/14 09:25:02 40.29 25.40 27 6.9 1112 in 493 municipalities 3c
11 01/08/13 14:16:08 39.66 25.54 10 5.7 0 /
12 4/16/12 11:23:44 36.78 21.65 35 5.5 0 /
13 6/21/08 11:36:25 36.15 21.89 18 5.6 0 /
14 06/08/08 12:25:28 37.97 21.48 5 6.4 47 in 31 municipalities 3b
15 2/20/08 18:27:07 36.47 21.71 26 6.2 0 /
16 2/14/08 12:08:54 36.34 21.86 20 6.2 0 /
17 2/14/08 10:09:22 36.57 21.75 30 6.9 32 in 21 municipalities 3a
Table 1. Localization and numbers of responses for each analysed earthquake.
15
Figure 1
16
Figure 2
17
Figure 3
Figure 4
18