Deformation and stress regimes in the Hellenic subduction zone from focal Mechanisms

Fault plane solutions for earthquakes in the central Hellenic arc are analysed to determine the deformation and stress regimes in the Hellenic subduction zone in the vicinity of Crete. Fault mechanisms for earthquakes recorded by various networks or contained in global catalogues are collected. In addition, 34 fault plane solutions are determined for events recorded by our own local temporary network on central Crete in 2000–2001. The entire data set of 264 source mechanisms is examined for types of faulting and spatial clustering of mechanisms. Eight regions with significantly varying characteristic types of faulting are identified of which the upper (Aegean) plate includes four. Three regions contain interplate seismicity along the Hellenic arc from west to east and all events below are identified to occur within the subducting African lithosphere. We perform stress tensor inversion to each of the subsets in order to determine the stress field. Results indicate a uniform N-NNE direction of relative plate motion between the Ionian Sea and Rhodes resulting in orthogonal convergence in the western forearc and oblique (40–50^∘) subduction in the eastern forearc. There, the plate boundary migrates towards the SE resulting in left-lateral strike-slip faulting that extends to onshore Eastern Crete. N110^∘E trending normal faulting in the Aegean plate at this part is in accordance with this model. Along-arc extension is observed on Western Crete. Fault plane solutions for earthquakes within the dipping African lithosphere indicate that slab pull is the dominant force within the subduction process and responsible for the roll-back of the Hellenic subduction zone.     

3-D velocity structure beneath the crust and upper mantle of Aegean Sea region

The region of the Aegean Sea and the surrounding areas in the Eastern Mediterranean lies on the boundary zone between the Eurasian and the African plates. It is a zone of widespread extensive deformation and, therefore, reveals a high level of seismicity.Three-dimensional velocity structure, beneath the crust and upper mantle of the region between 33.0°N–43.0°N and 18.0°E–30.6°E, is determined.The data used are arrival times ofP-waves from 166 earthquakes, recorded at 62 seismological stations. In total, 3973 residual data are inverted.The resultant structure reveals a remarkable contrast of velocity. In the top crustal layer, low velocities are dominant in Western Turkey and on the Greek mainland, while a high velocity zone is dominant in the Ionian Sea and in the southern Aegean Sea.In the upper mantle, high velocity zones dominate along the Hellenic arc, corresponding to the subducting African plate and in the northern part of the region, corresponding to the subducting African plate and in the northern part of the region, corresponding to the margin of Eurasian plate.A low velocity zone is dominant in the Aegean Sea region, where large-scale extension and volcanic activity are predominant, associated with the subduction of the African plate.     

Evidence for periodic seismicity in the inner Aegean seismic zone

The most complete and reliable data of strong (M _s6.5), shallow (h<70 km) earthquakes which occurred in the inner Aegean seismic zone have been utilized to describe its seismicity time variation during 1800–1986 by two independent statistical models. The first is a sequentially stationary model of seismicity rates which shows that intervals of low seismicity rate, lasting for some 37 years, alternate with high rate intervals of 8–12 years duration. The second model is a statistical model according which seismic energy released within 5-year time windows approximates a harmonic curve within a period of about 50 years. This model is in agreement with the notion that the time series of strong earthquake occurrences in the inner Aegean seismic zone consists of a random (shocks withM _s=6.5–6.8) and a nonrandom component (M _s6.9). Maxima and minima of the harmonic curve coincide with the high and low rate intervals, respectively. A model of regional stationary accumulation of thermal stresses along certain seismic belts and their cyclic relaxation may explain this periodicity.     

The Medina Wrench: a key to the kinematics of the central and eastern Mediterranean over the past 5 Ma

The Medina Wrenth in the central Mediterranean is a transform fault connecting the plate collision in northwest Africa and northern Sicily with that occurring at the Aegean plate boundary, south of Greece. The more than 800 km long crescent-shaped wrench zone is currently seismically quiet but exhibits major deformation since 5 Ma within a belt 30–100 km wide. It forms the southern boundary of two microplates moving eastward with respect to Africa and Europe. A simple plate rotation model constrained by recent paleomagnetic data indicates that a continental Iblean microplate and a hybrid continental/oceanic Ionian microplate, separated along the Malta Escarpment, have rotated anticlockwise by 11° and 12°, respectively, around poles in southern Italy. These rotations involved some 100 km of dextral eastward movement relative to Africa of the Ionian Basin north of the Medina Wrench since 5 Ma. Combining the published 26° clockwise rotation of the Peloponnesus and northwest half of the Aegean with the 12° anticlockwise rotation of the Ionian microplate results in (a) a 99% agreement between the length of the seismic Benioff Zone beneath Greece and the total convergence of the microplates, and (b) an average rate of convergence across the Aegean plate boundary southwest of the Peloponnesus of 6.6 ± 1cm a⁻¹ since the Miocene. Relative motion between microplates in a collision zone thus may be as much as 6 times faster than convergence between the major plates which spawned them, and they can be considered rigid to the first order over the time span involved.     

The geometry of the Burmese-Andaman subducting lithosphere

The gross seismotectonic features for the Burmese-Andaman arc system which defines the northeast margin of the Indian plate are rather well known but variations in the subduction zone geometry along and across the arc and fault pattern within the subducting Indian plate have not been studied. Present workaims to study these by using seismicity data whose results are presented in the form of: (a) Lithospheric across-the-arc sections at about every 100–120 km (approximately one degree latitude apart) covering the 3500 km longBurmese-Andaman arc system, (b) a structure contour map showing the depth tothe top surface of the seismically active lithosphere and (c) interpretationof focal mechanism solutions for 148 Benioff zone earthquakes. Both penetrationdepth and the dip of the Benioff zone vary considerably along the arc in correspondence to the curvature of the fold-thrust belt which varies from concave to convex in different sectors of the arc. Several extensive `Hinge faults' that abut at high angles to the arc orientation, are inferred from aninterpretation of the structure contour map. Active nature of the hinge faultsis established in several areas by their association with earthquakes andcorroborated through fault plane solutions. At shallow level of the Benioffzone along these faults, focal mechanism solutions display left lateral strikeslip movement while at deeper levels reverse fault solutions are common.     

Space-time migration of earthquakes along the North Anatolian fault zone and seismic gaps

The North Anatolian fault is a well-defined tectonic feature extending for 1400 km across Northern Turkey. The space-time distribution of seismicity and faulting of this zone has been examined with a particular emphasis on the identification of possible seismic gaps. Results suggest several conclusions with respect to the temporal and spatial distribution of seismicity. First, the earthquake activity appears not to be stationary over time. Periods of high activity in 1850–1900 and 1940 to the present bracket a period of relatively low activity in 1910–39. Second, there appears to have been a two-directional migration of earthquake epicenters away from a central region located at about 39°E longitude. The migration to the west has a higher velocity (>50 km/yr) than the migration to the east (10km/yr). The faulting associated with successive earthquakes generally abuts the previous rupture. Some existing gaps were filled by later earthquakes.At present there are two possible seismic gaps along the North Anatolian fault zone. One is at the western end of the fault, from about 29° to 30°E. Unless this is a region of ongoing aseismic creep, it could be the site of a magnitude 6 or greater earthquake. The other possible gap is at the eastern end, from about 42° to 43°E, to the west of the unexpected M=7.3 event of 24 November 1976.     

Regional stresses along the Eurasia-Africa plate boundary derived from focal mechanisms of large earthquakes

The focal mechanism solutions of 83 European earthquakes withM>6, selected from a total of 140, have been used to derive the directions of the principal axes of stress along the plate boundary between Eurasia and Africa from the Azores islands to the Caucasus mountains. Along most of the region, the horizontalP-axes are at an angle of 45° to 90° with the trend of the plate boundary. HorizontalT-axes are concentrated in central Italy and northern Greece in association with normal faulting. Large strike-slip motion of right-lateral character takes place at the center of the Azores-Gibraltar fault and the North Anatolian fault. From Gibraltar to the Caucasus the boundary is complicated by the presence of secondary blocks and zones of extended deformations with earthquakes spread over wide areas. Intermediate and deep earthquakes are present at four areas with arc-like structure, namely, Gibraltar, Sicily-Calabria, Hellenic arc and Carpathians.     

Historical intermediate-depth earthquakes in the southern Aegean Sea Benioff zone: modeling their anomalous macroseismic patterns with stochastic ground-motion simulations

We model the macroseismic damage distribution of four important intermediate-depth earthquakes of the southern Aegean Sea subduction zone, namely the destructive 1926 M?=?7.7 Rhodes and 1935 M?=?6.9 Crete earthquakes, the unique 1956 M?=?6.9 Amorgos aftershock (recently proposed to be triggered by a shallow event), and the more recent 2002 M?=?5.9 Milos earthquake, which all exhibit spatially anomalous macroseismic patterns. Macroseismic data for these events are collected from published macroseismic databases and compared with the spatial distribution of seismic motions obtained from stochastic simulation, converted to macroseismic intensity (Modified Mercalli scale, I_MM). For this conversion, we present an updated correlation between macroseismic intensities and peak measures of seismic motions (PGA and PGV) for the intermediate-depth earthquakes of the southern Aegean Sea. Input model parameters for the simulations, such as fault dimensions, stress parameters, and attenuation parameters (e.g. back-arc/along anelastic attenuation) are adopted from previous work performed in the area. Site-effects on the observed seismic motions are approximated using generic transfer functions proposed for the broader Aegean Sea area on the basis of V_S30 values from topographic slope proxies. The results are in very good agreement with the observed anomalous damage patterns, for which the largest intensities are often observed at distances >?100 km from the earthquake epicenters. We also consider two additional “prediction” but realistic intermediate-depth earthquake scenarios, and model their macroseismic distributions, to assess their expected damage impact in the broader southern Aegean area. The results suggest that intermediate-depth events, especially north of central Crete, have a prominent effect on a wide area of the outer Hellenic arc, with a very important impact on modern urban centers along northern Crete coasts (e.g. city of Heraklion), in excellent agreement with the available historical information.     

On the spatial distribution of seismicity and the 3D tectonic stress field in western Greece

We analyzed a large number of focal mechanisms and relocated earthquake hypocenters to investigate the geodynamics of western Greece, the most seismically active part of the Aegean plate-boundary zone. This region was seismically activated multiple times during the last decade, providing a large amount of enhanced quality new information that was obtained by the Hellenic Unified Seismological Network (HUSN). Relocated seismicity using a double-difference method appears to be concentrated above ∼35 km depth, exhibiting spatial continuity along the convergence boundary and being clustered elsewhere. Earthquakes are confined within the accreted sediments escarpment of the down-going African plate against the un-deformed Eurasian hinterland. The data arrangement shows that Pindos constitutes a seismic boundary along which large stress heterogeneities occur. In Cephalonia no seismicity is found to be related with the offshore Cephalonia Transform Fault (CTF). Onshore, NS crustal extension dominates, while in central and south Peloponnesus the stress field appears rotated by 90°. Shearing-stress obliquity by 30° is indicated along the major strike-slip faults, consistent with clockwise crustal rotation. Within the lower crust, the stress field appears affected by plate kinematics and distributed deformation of the lower crust and upper mantle, which guide the regional geodynamics.     

Seismic Sources on the Iberia-African Plate Boundary and their Tectonic Implications

—The plate boundary between Iberia and Africa has been studied using data on seismicity and focal mechanisms. The region has been divided into three areas: A; the Gulf of Cadiz; B, the Betics, Alboran Sea and northern Morocco; and C, Algeria. Seismicity shows a complex behavior, large shallow earthquakes (h < 30 km) occur in areas A and C and moderate shocks in area B; intermediate-depth activity (30 < h < 150 km) is located in area B; the depth earthquakes (h 650 km) are located to the south of Granada. Moment rate, slip velocity and b values have been estimated for shallow shocks, and show similar characteristics for the Gulf of Cadiz and Algeria, and quite different ones for the central region. Focal mechanisms of 80 selected shallow earthquakes (8 m_b 4) show thrust faulting in the Gulf of Cadiz and Algeria with horizontal NNW-SSE compression, and normal faulting in the Alboran Sea with E-W extension. Focal mechanisms of 26 intermediate-depth earthquakes in the Alboran Sea display vertical motions, with a predominant plane trending E-W. Solutions for very deep shocks correspond to vertical dip-slip along N-S trends. Frohlich diagrams and seismic moment tensors show different behavior in the Gulf of Cadiz, Betic-Alboran Sea and northern Morocco, and northern Algeria for shallow events. The stress pattern of intermediate-depth and very deep earthquakes has different directions: vertical extension in the NW-SE direction for intermediate depth earthquakes, and tension and pressure axes dipping about 45 ° for very deep earthquakes. Regional stress pattern may result from the collision between the African plate and Iberia, with extension and subduction of lithospheric material in the Alboran Sea at intermediate depth. The very deep seismicity may be correlated with older subduction processes.     

Focal mechanisms of intraplate earthquakes in Bolivia,South America

Intraplate seismic activity in Bolivia is mainly located in the central region (16°–19°S, 63°–67°W) which includes the East Andean Cordillera and the Sub-Andean Sierras. At this region there is a bend in the trend of the main geological structures from NW-SE in the north to N-S in the south. Focal mechanisms have been calculated for 10 earthquakes of magnitudes 4.9–5.6, using first motionP-waves from long period instruments. Their solutions correspond to reverse faulting, some with a large component of strike-slip motion. Their solutions can be grouped into two types; one with pure reverse faulting on planes with azimuth NW-SE and the other with a large strike-slip component on planes with azimuths nearly N-S or WNW-ESE. The maximum stress axis (P-axis) is practically horizontal (dipping less than 5°) oriented in a mean N56°E direction. This orientation may be related with the direction of compression resulting from the collision of the Nazca plate against the western margin of the South American continent. Wave-form analysis of long-periodP-waves for one event restricts the focal depth to 8 km in the Sub-Andean region. Seismic moments and source dimensions determined from spectra of Rayleigh waves are in the range of 10¹⁶–10¹⁷Nm and 17–24 km, respectively. The Central Bolivia region can be considered as a zone of intraplate deformation situated between the Bolivian Altiplano and the Brazil shield.     

Relocation of shallow earthquakes in southern Alaska using Joint Hypocenter Determination method

A subset of 2660 shallow earthquakes (0–50 km) that occurred from 1988 to 1996 in south central Alaska between 155°W and 145°W and 59°N and 63°N was relocated using the joint hypocenter determination (JHD) method. Both P- and S-wave observations recorded by the regional seismic network were used. Events were relocated in twenty different groups based on their geographic location and depth using two velocity models. As a result of the relocation, the majority of the hypocenters shifted downward, while the epicenter locations did not change significantly. The distribution of the shallow subduction zone earthquakes indicates the existence of two seismically independent blocks, with one block occupying the northeastern part and the other occupying the central and western parts of the study area. The boundary between the blocks is marked by a 15 to 20 km wide seismicity gap to the southeast of 149.5°W and 62°N. The analysis of the fault plane solutions for shallow subduction zone earthquakes shows that an overwhelming majority of the solutions represent normal, oblique-normal or strike-slip faulting with predominant WNW-ESE orientation of T-axes. This indicates a down-dip extensional regime for the subducting slab at shallow depths. Very few earthquakes yielded fault plane solutions consistent with thrusting on a contact zone between the overriding and subducting plates. This result may be an indication that currently either the strain energy is not released at the contact zone or it is associated with aseismic motion.     

Seismicity and Seismic Hazard in Alexandria (Egypt) and its Surroundings

— Alexandria City has suffered great damage due to earthquakes from near and distant sources, both in historical and recent times. Sometimes the source of such damages is not well known. Seismogenic zones such as the Red Sea, Gulf of Aqaba-Dead Sea Hellenic Arc, Suez-Cairo-Alexandria, Eastern-Mediterranean-Cairo-Faiyoum and the Egyptian costal area are located in the vicinity of this city. The Egyptian coastal zone has the lowest seismicity, and therefore, its tectonic setting is not well known. The 1998 Egyptian costal zone earthquake is a moderate complex source. It is composed of two subevents separated by 4 sec. The first subevent initiated at a depth of 28 km and caused a rupture of strike (347°), dip (29°) and slip (125°). The second subevent occurred at a shallower depth (24 km) and has a relatively different focal parameter (strike 334°, dip 60° and slip 60°). The available focal mechanisms strongly support the manifestation of a complex stress regime from the Hellenic Arc into the Alexandria offshore area. In the present study a numerical modeling technique is applied to estimate quantitative seismic hazard in Alexandria. In terms of seismic hazard, both local and remote earthquakes have a tremendous affect on this city. A local earthquake with magnitude M_s = 6.7 at the offshore area gives peak ground acceleration up to 300 cm/sec². The total duration of shaking expected from such an earthquake is about three seconds. The Fourier amplitude spectra of the ground acceleration reveals that the maximum energy is carried by the low frequency (1–3 Hz), part of the seismic waves. The largest response spectra at Alexandria city is within this frequency band. The computed ground accelerations due to strong earthquakes in the Hellenic Arc, Red Sea and Gulf of Aqaba are very small (less than 10 cm/sec²) although with long duration (up to 3 minutes).     

Mediterranean island arcs and origin of high potash volcanoes

Active volcanoes of the Mediterranean Sea are distributed along two arc structures: the Hellenic arc in the Aegean Sea and the Calabrian arc in the Tyrrhenian Sea. The active volcanoes in both arcs lie above earthquakes with focal depth greater than 100 km. The depth of these earthquakes increases generally northward reaching a maximum depth of about 200 km in the Aegean Sea and more than 300 km in the Tyrrhenian Sea.     

Seismic quiescence precursory to a past and a future Kurile island earthquake

A systematic search was made for seismicity rate changes in the segment of the Kurile island arc from 45°N to 53°N by studying the cumulative seismicity of shallow (h100 km) earthquakes within 11 overlapping volumes of radius 100 km for the time period 1960 through beginning of 1978. We found that in most parts of this island arc and most of the time the seismicity rate as obtained from the NOAA catalogue and not excluding any events is fairly constant except for increased seismicity in the mid 1960s in the southern portion due to the great 1963 mainshock there, and for seismicity quiescence during part of the time period studied within two well defined sections of the arc. The first of these is a volume of 100 km radius around a 1973 (M _s =7.3) mainshock within which the seismicity rate was demonstrated at the 99% confidence level to have been lower by 50% during 2100 days (5.75 years) before this mainshock. The second volume of seismic quiescence coincides with the 400 km long north Kuriles gap. In this gap the seismicity rate is shown (at the 99% confidence level) to be lower by 50% from 1967 to present (1978), in comparison with the rate within the gap befor 1967, as well as with the rate surrounding the gap. We propose that the anomalously low seismicity rate within the Kuriles gap is a precursor to a great earthquake, the occurrence time of which was estimated by the following preliminary relation between precursory quiescence time and source dimensionT=190L ^0.545. We predict that an earthquake with source length of 200–400 km (M>8) will occur along the north Kurile island arc between latitude 45.5°N and 49.2°N at a time between now and 1994.     

Historical seismicity of the southeastern Caribbean and tectonic implications

We have relocated the twenty-eight largest magnitude (4.3M _s 7.3) historical (1922–1963) earthquakes of the southeastern Caribbean. We also present new focal mechanisms for seven of these events. The relocations are based on reported ISSP andS arrival times that we analyzed using generalized linear inversion techniques. The new focal mechanisms were constrained by first motionP polarities as reported by the ISS and as picked by us where records were available, and by the polarities and ratios ofSH andsSH, andSV andsSV arrivals that we determined from seismograms. The results of the relocations are commensurate with the distribution of seismicity observed in the recent era: hypocenters are shallow and intermediate in depth (0–200 km), and the events occur almost exclusively in areas known to be currently seismic. The frequent seismic activity in the vicinity of the Paria Peninsula, Venezuela, is clearly a persistent feature of the regional earthquake pattern; intermediate depth earthquakes indicative of subduction beneath the Caribbean plate occur here and along the Lesser Antilles arc. The Grenadines seismic gap is confirmed as an area of low seismic moment release throughout the historical era. Trinidad and the eastern Gulf of Paria were also largely quiescent.The new focal mechanisms, despite being a sparse data set, give significant insight into both subduction processes along the Lesser Antilles arc and into the shallow deformation of the Caribbean-South America plate boundary zone. The largest earthquake to have occurred in this region, the 19 March 1953 event (M _m =7.01), is a Lesser Antilles slab deformation event, and another earthquake in this region of the Lesser Antilles is probably a rarely-observed interplate thrust event. Shallow deformation in the plate boundary zone is complex and, near the Paria Penninsula, involves mixed southeastward thrusting and dextral strike-slip on east-striking faults, and secondarily, normal faulting. Bending of the subducting Atlantic-South American plate also seems to generate seisms. The rather high ratio of intraplate deformation to interplate deformation observed along the Lesser Antilles subduction zone in the more recent era seems to have been operative in the historical era as well.     

Anelasticity of the crust and upper mantle beneath north and central India from the inversion of observed Love and Rayleigh wave attenuation data

The fundamental mode Love and Rayleigh waves generated by earthquakes occurring in Kashmir, Nepal Himalaya, northeast India and Burma and recorded at Hyderabad, New Delhi and Kodaikanal seismic stations are analysed. Love and Rayleigh wave attenuation coefficients are obtained at time periods of 15–100 seconds, using the spectral amplitude of these waves for 23 different paths along northern (across Burma to New Delhi) and central (across Kashmir, Nepal Himalaya and northeast India to Hyderabad and Kodaikanal) India. Love wave attenuation coefficients are found to vary from 0.0003 to 0.0022 km^–1 for northern India and 0.00003 km^–1 to 0.00016 km^–1 for central India. Similarly, Rayleigh wave attenuation coefficients vary from 0.0002 km^–1 to 0.0016 km^–1 for northern India and 0.00001 km^–1 to 0.0009 km^–1 for central India. Backus and Gilbert inversion theory is applied to these surface wave attenuation data to obtainQ ^–1 models for the crust and uppermost mantle beneath northern and central India. Inversion of Love and Rayleigh wave attenuation data shows a highly attenuating zone centred at a depth of 20–80 km with lowQ for northern India. Similarly, inversion of Love and Rayleigh wave attenuation data shows a high attenuation zone below a depth of 100 km. The inferred lowQ value at mid-crustal depth (high attenuating zone) in the model for northern India can be by underthrusting of the Indian plate beneath the Eurasian plate which has caused a low velocity zone at this shallow depth. The gradual increase ofQ ^–1 from shallow to deeper depth shows that the lithosphere-asthenosphere boundary is not sharply defined beneath central India, but rather it represents a gradual transformation, which starts beneath the uppermost mantle. The lithospheric thickness is 100 km beneath central India and below that the asthenosphere shows higher attenuation, a factor of about two greater than that in the lithosphere. The very lowQ can be explained by changes in the chemical constitution taking place in the uppermost mantle.     

Bayesian probabilities of earthquake occurrences in Greece and surrounding areas

The Bayesian extreme-value distribution of earthquake occurrences has been adopted to estimate the seismic hazard in some seismogenic zones in Greece and surrounding regions. Seismic moment, slip rate, earthquake recurrence rate and magnitude were considered as basic parameters for computing the prior estimates of the seismicity. These estimates are then updated in terms of Bayes' theorem and historical estimates of seismicity associated with each zone.High probabilities for earthquakes withM6.0 have been obtained for the northwestern part of Greece as well as for the southwestern part of the Hellenic arc.     

Probabilities of occurrence of large earthquakes in the Aegean and surrounding area during the period 1986–2006

Repeat times of large shocks are obtained for 17 seismic fracture zones of the Aegean and surrounding area, from times of historic and present century earthquakes. The mean standard deviation of the repeat times is approximately 50% of any one observation.A probabilistic approach is then used to forecast the likelihood of large future earthquakes in each fracture zone, using as input the time of the last large shock, the average repeat time and its standard deviation. Shallow and intermediate depth earthquakes are examined separately. The calculated probabilities are high for the entire Hellenic arc, both for shallow and intermediate depth seismicity, for the area of Leucas island (Ionian), of Lesbos island (Aegean), for Patraikos-west Corinthiakos Gulfs, for Evoikos Gulf as well as for southern Bulgaria.The probability estimates based on the most recent large earthquakes, involve a number of basic physical assumptions and we would think that they provide a semi-stochastic approach to the problem of earthquake prediction in Greece.     

Source properties of the 1997–98 Central Italy earthquake sequence from inversion of long-period and broad-band seismograms

We study source properties of the main earthquakes of the 1997–98 Umbria-Marche (central Italy) sequence by analysis of regional-distanceand teleseismic long period and broadband seismograms recorded by MedNet and IRIS/GSN stations. We use a modified Harvardcentroid-moment tensor (CMT) algorithm to allow inversion of long period waveforms, primarily Rayleigh and Love waves, for small earthquakes (4.2 M_W 5.5) at local to regional distances (<15^°). For the seven largest earthquakes (M_W>5.2) moment tensors derived from local and regional data agree well with those determined using teleseismic waveforms and standard methods of analysis. We also determine moment tensors for a foreshock and 12 other aftershocks, that were too small for global analysis. Focal depth and rupture propagation are analyzed for three largest shocks by inversion of teleseismic broadband body waves. The earthquakes are generally located at shallow depth (5 km or shallower) and are characterized by normal faulting mechanisms, with a NE-SW tension axis. The presumed principal fault plane dips at a shallow angle towards the SW. Only one of the events analyzed has an entirely different faulting geometry, indicating instead right-lateral strike-slip motion on a plane approximately E-W, or left-lateral faulting on a N-S plane. The other significant exception to the regular pattern of mechanisms is represented by the March 26, 1998, event, located at 51 km depth. Its connection with the shallow earthquake sequence is unclear and intriguing. The time evolution of the seismic sequence is unusual,with the mainshock accounting for only approximately 50% of the total moment release. The broadband teleseismic waveforms of the main, September 26, 09:40, earthquake are very complicated for the size of the event and suggest a complex rupture. In our favored source model, rupture initiated at 5 km depth, propagated updip and was followed, 3 seconds later, by a shallower subevent with a slightly rotated mechanism.