Marine Geology in the Region of the Messina Straits, and a Puzzling Tale of Faults, Earthquakes and Tsunamis

Transcription

1 Marine Geology in the Region of the Messina Straits, and a Puzzling Tale of Faults, Earthquakes and Tsunamis A. Argnani Institute of Marine Sciences, CNR, Bologna, Italy andrea.argnani@ismar.cnr.it Abstract The Straits of Messina is one of the most tectonically active areas of the Mediterranean, and has been the site of the destructive 1908 Messina earthquake. In spite of the hazard potential of this marine area, studies based on modern geophysical data are still lacking. A marine multichannel seismic survey was purposely carried out with the aim to outline the fault pattern in the area of the Messina 1908 earthquake, and to better understand its significance within the tectonic frame of the region. Within the Messina Straits, faults have been imaged on the Calabrian side, with a 30 km long NW-trending fault, located at the SW tip of Calabria, that is affecting the sea floor, whereas we did not image any extensional fault plane attributable to the Taormina Fault, on the Sicilian side of the straits. The geodynamic implication is that extension in south-eastern Sicily, on the Ionian side of the Hyblean Plateau, and extension in southern Calabria and Messina Straits belong to two different tectonic systems and, therefore, cannot be mechanically linked. Finally, the damages produced by the1908 ground shaking were aggravated by the effects of a remarkable tsunami, with up to 11 m of run-up height, that followed the earthquake. The origin of the tsunami associated with the 1908 earthquake is not yet fully understood, but geological and geophysical evidences substantially undermine a recent proposal that the 1908 tsunami originated by a large landslide offshore Giardini Naxos. 1 Introduction and Geological Setting The Messina Straits and its surroundings are one of the most tectonically active areas of the Mediterranean, as indicated by several lines of geological and geophysical evidence. Relatively large earthquakes struck the area in historical times (Figure 1), although the recurrence time for 1908-type events (Mw=7.1) seems to be about 1500 yr [4]. The Sicilian and Calabrian side of the straits are characterised by uplifted marine terraces. The flight of emergent marine terraces along the coast of Sicily, from Taormina to Briga, were uplifted since 125 ka with rates of 1.07 mm yr 1 [5]. The occurrence of a large and active extensional fault named Taormina Fault, running offshore, has been inferred on the basis of coastal geomorphology (Figure 1 [6]). According to some investigators (e.g., [7]), the systems of faults of southern Calabria and south-eastern Sicily, on the Ionian side of the Hyblean Plateau (e.g., [8]), both associated to large and destructive earthquakes, can be linked through the Taormina

2 Marine Geology Figure 1: Map of the Messina Straits with location of relevant historical earthquakes (after [1]; [2]). Dashed boxes represent poorly constrained events. Fault. However, along the belt corresponding to the inferred Taormina Fault a lack of seismicity is indicated by historical data and recent instrumental records [9] (Figure 1). The assumption that the hypothesized Taormina Fault is part of a single rift system, connecting Calabria to south-eastern Sicily, implies that it represents one of the most hazardous seismic gaps in Italy, a potential site for large future earthquakes. A flight of uplifted marine terraces characterizes also the Calabrian coast of the Messina Straits. Twelve to fourteen orders of terraces have been identified, with the highest terraces, dated Middle Pleistocene, having an elevation of about 1350 m above sea level [10], with an stimated uplift rate of about mm yr 1 in the last ka. GPS velocities [11], indicate a NW-SEdirected extension at a rate between 1.7 to 3 mm yr 1 between the Sicilian and Calabrian sides of the Messina Straits. In particular, GPS-derived interseismic strain appears to fit aseismic slip along a 30, SEdipping normal fault, locked above about 8 km (Serpelloni, p.c. 2009). Onshore structural studies (e.g., [3]) have shown the occurrence of a fault system that runs along the Sicilian coast, but the faults that have been most active during the Pleistocene are those located on the Calabrian side of the Messina Straits [3]; these faults belongs to two main sets trending NW-SE and NE- SW and are not obviously related to the N- 846

3 Marine research at CNR Figure 2: Map showing the traces of fault planes proposed for the Messina 1908 earthquake, with the authors annotated. S trend of the faults inferred to be responsible for 1908 Messina earthquake (Figure 2). Most authors agree that the extensional faults bounding the Reggio Calabria basin have been active through Late Pliocene and Early Pleistocene. However, subsequent fault activity is debated, and some authors suggest that the faults were not active since Middle Pleistocene (e.g., [10]). The large 1908 Messina earthquake (Mw = 7.1 [2]), for which extensional focal mechanisms were obtained, occurred within this tectonic frame. The December 28, 1908 Messina Earthquake has been ranked as one of the most destructive events of the last centuries, and costed the highest toll in human life in Italy s history of seismicity, with over 80,000 people dieing in the cities of Messina and Reggio Calabria and the surrounding area. Besides buildings collapse and fires, the damages produced by ground shaking were aggravated by the effects of a remarkable tsunami, with up to 11 m of run-up height, that followed the earthquake [12]. Despite such catastrophic effects the location of the causative fault is not fully assessed (e.g., [13]). Inverse modelling of seismograms and geodetic levelling, and geological studies have produced a variety of results in terms of position, direction, length and dip of the fault, with the most recent solutions proposing 847

4 Marine Geology Figure 3: Map with traces of multichannel seismic profiles in the study area. The trace of the supposed Taormina Fault is in red. Onshore faults after Ghisetti [3]. Seismic profiles shown in Figures 4-6 are in thick green lines. long, E-dipping faults trending about N-S (e.g., [14, 2], and references therein Figure 2). At present, the most accepted seismogenic source for the 1908 earthquake is a 40 km-long, blind fault dipping 30 to the ESE which is thought to accounts for the topography of the Messina Straits (DISS, This fault plane, which trends NNE-SSW, has a minimum depth of 3 km, and would crop out along the coast of Sicily (Figure 2). As for the seismogenic fault, the source of the tsunami related to the earthquake is still a matter of debate [15], and contribution from a so far unidentified submarine slide is called upon. In spite of the hazard potential of this marine region, geophysical surveys purposely devised to investigate the neotectonic features are lacking, with the notable exception of the early work of Selli [16] which, however, did not employ modern geophysical techniques. In order to bridge this gap a multichannel seismic survey, aimed at defining the structural pattern of the Messina Straits, was carried out within the frame of INGV-DPC seismolog- 848

5 Marine research at CNR Figure 4: Seismic profile TAO 09 across the central part of the supposed Taormina fault. Arrows mark the South Calabrian fault. See Figure 3 for location. ical projects (Figure 3 [13]). 2 Discussion Three key aspects related to the tectonics of the Messina Straits will be briefly discussed below, based on the results obtained from multichannel seismic data. A more detailed presentation of the data and interpretation procedures is given elsewhere [13], with references therein), together with the correlation of seismic units with onshore outcrops and short offshore wells that allows to date the seismic units. 2.1 The elusive Taormina Fault The hypothesized occurrence of a large extensional fault parallel to the coastline and located offshore, between Taormina and Briga (Taormina Fault) can be rejected by seismic data [13]. That stretch of coastline is actively deforming, as suggested by uplifted marine terraces and tilting of marine strata along the offshore slope (Figure 4), but such deformation cannot be related to footwall uplift of a normal fault. The geodynamic implication is that extension in south-eastern Sicily, on the Ionian side of the Hyblean Plateau, and extension in southern Calabria and Messina Straits, belong to two different tectonic systems and cannot be mechanically linked. The seismological implication is that the lack of earthquakes is not indicating the occurrence of a seismicity gap. 2.2 The faults of the Messina Straits and the 1908 Earthquake Within the northern part of the Messina Straits, the imaged faults are located on the Calabrian side and dip to the west. These faults appear connected to the fault system reported onshore near Reggio Calabria (Figure 3 [3]). A fault trending NW-SE has been mapped between the towns of Reggio Calabria and Messina. The fault 849

6 Marine Geology Figure 5: Seismic profile TAO 08 within the Messina Straits. A listric fault (arrows) is flattening at shallow depth. Location in Figure 5. plane dips to the west and displays a listric shape, with growth strata in the lower part of the half graben sedimentary fill, that can be dated as Late Pliocene-Early Pleistocene (Figure 5). Towards the coast of Calabria the fault is sealed by a sedimentary wedge, which can be dated as Middle Pleistocene-Holocene. The profiles approaching SW Calabria show a 30 kmlong west-dipping fault that is affecting the sea floor with a NW-SE trend, which represents the longest lineament observed within the Messina Straits (Figure 6). The W-dipping southern Calabrian fault has a low-angle plane when depth-migrated therefore, assuming that the epicentre is located above the deeper, seismogenic portion of the fault plane, it does not satisfy the 1908 macroseismic intensity observed along the Calabrian coast. However, this fault might be responsibile for the March 28, 1780 and May 22, 1932 earthquakes (Figure 7), two events for which a seismogenic fault is difficult to be identified [1]. The lack of evident extensional faults within the narrower part of the Messina Straits might support the interpretation of a seismogenic fault located to the south of this area or, alternatively, of a blind fault located within the northern Messina Straits. Hints suggesting that the occurrence of a blind fault, possibly east-dipping as indicated by seismological studies, are not obvious on seismic profiles in the northern part of the straits. Moreover, given the large magnitude of the 1908 event, it cannot be ruled out that more than one fault were activated at the same time, as already proposed (Figure 2). In this event, the long fault observed offshore southern Calabria could be activated together with a blind east-dipping fault located further to the north, possibly contributing to the tsunami related to the 1908 earthquake the tsunami waves observed all along the east coast of Sicily, in fact, are hardly compatible with a blind fault located within the northern part of the Messina Straits [15]. 850

7 Marine research at CNR Figure 6: Seismic profile TAO 17. Arrows mark the South Calabrian fault. See Figure 3 for location. 2.3 The 1908 Messina Earthquake and its related Tsunami Because of the uncertainty in the location of the causative fault, the modelling of the associated tsunami is not satisfactory [15], and the possibility to have a contribution from a submarine slide has been explored [19, 20, 21]. A. Billi et al. [22] have recently proposed that the tsunami that stroke the coast of the Messina Strait in December 1908 originated from a very large submarine landslide (20 km 3 ) located offshore Giardini-Naxos. However, geophysical data and work on tsunami modelling cast doubt on the proposed hypothesis that a large submarine landslide that was caused by the 1908 earthquake is located offshore Giardini-Naxos [23]. The morphological expression of a 100-yr old landslide, of its headscarp and its deposit, should be little affected by subsequent erosion and reworking. High-resolution morphobathymetry and seismic profiles clearly show that the morphology of the area is the product of long-lasting erosion without any trace of a 100 yr-old large-scale landslide. 3 Conclusions The data collected during the Taormina survey open some interesting questions concerning the position, direction and extent of active faults within the Messina Straits, where it is difficult to find a single fault that is long enough to account for the Mw 7.1, 1908 Messina earthquake. Moreover, the trend of the observed faults, though consistent with faults onshore Calabria, is different from the trend of many of the faults proposed on the basis of inversion of seismological and geodetic data (Figure 2). Seismic data show that there is no N-S fault cutting the sedimentary successions south of the epicenter of Schick, 1977 (Figure 7). In fact, the only long 851

8 Marine Geology Figure 7: Summary map with main faults in the study area. For the 1908 Messina earthquake the followings are indicated: a) the relocation of [17] (ellipse with green star), b) the epicenter of [18] (yellow star), c) the isoseismal X [12], and the breaks in the telegraph cables. The DISS fault plane is indicated in green. fault that is affecting the sea floor is located at the SW tip of Calabria, trends NW- SE and dips to the west with a low angle. Such fault parameters are not compatible with the 1908 earthquake, as the hypocentral depth would be located too far to the west. However, this fault could have contributed to the tsunami wave that came together with the 1908 earthquake better than any faults located within the northern part of the Messina Straits. As a working hypothesis, it seems that a system of interconnected faults (e.g.,[3]), partly exploiting preexisting fault planes, can better describe the observed geological (i.e., long term) deformation. This interpretation leaves the possibility to have more than a fault active at the same time an event that has been inferred for the 1908 earthquake (e.g., Figure 2). At present it is difficult to say whether the observed active faults are just the surface expression of a single, deeper and blind seismogenic fault, as suggested by seismological arguments (e.g., DISS), or if they truly represent the complex response of an area that suffered a great deal of geological deformation and where more than one fault can be active at the same time. The large magnitude of the 1908 earthquake makes the first hypothesis 852

9 Marine research at CNR perhaps more appealing. As far as the 1908 tsunami is concerned, an alternative to a fault that extends to the south of the Straits consists in introducing a submarine slide that adds to the earthquake in contributing to the tsunami. In this regards, the recent proposal that the 1908 tsunami was due to a very large landslide of about 20 km 3, the deposit of which would be located offshore Giardini-Naxos [22], is not supported by evidence [23]. Moreover, modelling of the 1908 tsunami indicates that a much smaller slide, less than 2 km 3, is enough to produce the effects observed along the shores [20, 21]. The location of this potentially tsunamigenic slide, however, is difficult to constrain from the inversion of poor quality tsunami runup observations, and no landslide deposit that can be related to the 1908 earthquake has been so far identified on swath bathymetry. The 1908 earthquake has been a complex event. At present, the combined action of a blind fault and a submarine slide is thought to be able to account for the observed ground shaking and tsunami. However, given the structural complexity of the area, other solutions, where more than one slide, and perhaps more than one fault, are active at the same time, cannot a priori be disregarded. References [1] R. Azzaro, F. Bernardini, R. Camassi, and V. Castelli. The 1780 seismic sequence in NE Sicily (Italy): shifting an underestimated and mislocated earthquake to a seismically low rate zone. Natural Hazards, 42: , [2] N.A. Pino, A. Piatanesi, G. Valensise, and E. Boschi. The 28 December 1908 Messina Straits Earthquake (Mw 7.1): A Great Earthquake throughout a Century of Seismology. Seismological Research Letters, 80: , [3] F. Ghisetti. Fault parameters in the Messina Strait (southern Italy) and relations with the seismogenic source. Tectonophysics, 210: , [4] G. Valensise and D. Pantosti. A 125 Kyr-long geological record of seismic source repeatability: the Messina Straits (southern Italy) and the 1908 earthquake (Ms 7 1/2). Terra Nova, 4: , [5] F. Antonioli, L. Ferranti, K. Lambeck, S. Kershaw, V. Verrubi, and G. Dai Pra. Late Pleistocene to Holocene record of changing uplift rates in southern Calabria and northeastern Sicily (southern Italy, Central Mediterranean Sea). Tectonophysics, 422:23 40, [6] S. Catalano and G. De Guidi. Late Quaternary uplift of northeatern Sicily: relation with the active normal faulting deformation. J. Geodynamics, 36: , [7] C. Monaco and L. Tortorici. Active faulting in the Calabrian arc and eastern Sicily. J. Geodynamics, 29: ,

10 Marine Geology [8] A. Argnani and C. Bonazzi. The Malta Escarpment fault zone offshore eastern Sicily: Pliocene-Quaternary tectonic evolution based on new multichannel seismic data. Tectonics, 24(TC4009):1 12, [9] G. Neri, G. Barberi, G. Oliva, B. Orecchio, and D. Presti. A Possible Seismic Gap within a Highly Seismogenic Belt Crossing Calabria and Eastern Sicily. Bull. Seismological Society of America, 96: , [10] B. Dumas, P. Gueremy, and J. Raffy. Evidence for sea-level oscillations by the characteristic thickness of marine deposits from raised terraces of Southern Calabria (Italy). Quaternary Sci. review, 244: , [11] E. Serpelloni, G. Vannucci, S. Pondrelli, A. Argnani, G. Casula, M. Anzidei, P. Baldi, and P. Gasperini. Kinematics of the Western Africa-Eurasia plate boundary from focal mechanisms and GPS data. Geoph. J. International, 169: , [12] M. Baratta. La catastrofe sismica calabro-messinese. 28 Dicembre pages 1 426, [13] A. Argnani, G. Brancolini, C. Bonazzi, M. Rovere, F. Accaino, F. Zgur,, and E. Lodolo. The results of the Taormina 2006 seismic survey: Possible implications for active tectonics in the Messina Straits. Tectonophysics, 476: , [14] A. Amoruso, L. Crescentini, and R. Scarpa. Source parameters of the 1908 Messina Straits, Italy, earthquake from geodetic and seismic data. J. Geoph. Res., 107(B4):1 11, [15] E. Boschi, D. Pantosti, and G. Valensise. Modello di sorgente per il terremoto di Messina del Atti VIII Convegno GNGTS, pages , [16] R. Selli. Geologia e simotettonica dello Stretto di Messina. Atti Accademia Nazionale dei Lincei, 43: , [17] A. Michelini, A.J. Lomax, A. Bono, A. Nardi, B. Palombo, A. Rossi, and the INGV SISMOS Group. Relocation of instrumentally recorded, historical earthqukes in the Italian region. Geophysical Research Abstracts, 6(07642), [18] R. Schick. Eine seismotektonische Bearbeitung des Erdbebens von Messina im Jahre Geol. Jahrb., 11:3 74, [19] A. Piatanesi, S. Lorito,, and F. Romano. Il Grande maremoto del 1908: analisi e modellazione [20] S. Tinti, A. Armigliato, F. Zaniboni, R. Tonini, G. Pagnoni, S. Gallazzi, A. Manucci, and P. Pontrelli. Quale sorgente per il maremoto del 28 Dicembre 1908 nello Stretto di Messina? Terremoto, frana sottomarina o entrambe? GNGTS, 27 Convegno Nazionale, Extended Abstract, pages ,

11 Marine research at CNR [21] M. Favalli, E. Boschi, F. Mazzarini, and M.T. Pareschi. Seismic and landslide source of the 1908 Straits of Messina tsunami (Sicily, Italy). Geoph. Res. Lett, 36(L16304):1 6, [22] A. Billi, R. Funiciello, L. Minelli, C. Faccenna, G. Neri, B. Orecchio,, and D. Presti. On the cause of the 1908 Messina tsunami, southern Italy. Geophys. Res. Lett., 35(L06301):1 5, [23] A. Argnani, F.L. Chiocci, S. Tinti, A. Bosman, M.V. Lodi, G. Pagnoni, F., and Zaniboni. Comment on On the cause of the 1908 Messina tsunami, southern Italy by Andrea Billi et al. Geoph. Res. Lett., 36(L13307):1 2,