Master Event Research Articles

On the aftershock sequence of a 4.6 mb earthquake of the Garhwal Himalaya

Locally recorded data for eighteen aftershocks of a magnitude(mb) 4.6 earthquake occurring near Ukhimath in the Garhwal Himalaya were analysed. A master event technique was adopted to locate seventeen individual aftershock hypocentres relative to the hypocentre of the eighteenth aftershock chosen as the master event. The aftershock epicentres define an approximately 30 km2 rupture zone commensurate with the magnitude of the earthquake. The distribution of epicentres within this zone and the limited amount of first motion data support the view that a group of parallel, sub-vertical, sinistral strike-slip faults oriented N46°, transverse to the regional NW-SE trend of the Garhwal Himalaya, was involved in this seismic episode. Since the estimated focal depth range for aftershocks of this sequence is 3–14 km, we infer that this transverse fault zone extends through the upper crustal layer to a depth of 14 km at least.

Source characterization and fault plane determination for MbLg = 1.2 to 4.4 earthquakes in the Charlevoix Seismic Zone, Quebec, Canada

Abstract Two earthquake doublets and two multiplets recorded by the Charlevoix Telemetered Network (CLTN) in the Charlevoix Seismic Zone (CSZ) of southern Quebec, Canada, have been analyzed using an empirical Green's function (EGF) method to derive the relative source time functions (RSTF's) of seven master events with MbLg = 1.2 to 4.4. We identified the doublets and multiplets using a waveform cross-correlation and relative event location technique to verify that each earthquake pair had similar focal mechanisms and hypocentral locations. Three-component S waveforms recorded by the high dynamic range (126 dB) instrumentation of the CLTN were used to extract the RSTF's. The RSTF's reveal that six of the seven events are simple with single-source pulses having durations of 0.05 to 0.2 sec. Another earthquake (920310-0545, M 3.3) appears to be a double event with two episodes of rupturing. Azimuthal variations of the RSTF pulse amplitudes and widths provide strong evidence for the rupture directivities of five of the earthquakes (M = 1.2 to 4.4). The azimuthal variations in the RSTF pulse amplitudes were used to estimate the rupture directions and rupture velocities. Lower-bound estimates of the rupture velocity range from 0.5 to 0.7 Vs. Estimates of the rupture direction were combined with P-wave focal mechanisms for the four largest events (M 3.3 to 4.4) to identify the fault plane for these earthquakes. Source parameters were measured for the RSTF's of the master events, including seismic moments of 3.5 × 1018 to 5.3 × 1021 dyne-cm, fault radii of 100 to 330 m, and static stress drops of 2 to 90 bars. The fault radii and stress-drop estimates for M > 3 events agree well with estimates obtained by other researchers for M ∼ 3 to 4.5 earthquakes in the CSZ. We also observed apparent scaling between the stress drop and the earthquake size, which has been reported in other studies of stress drop in northeastern North America.

The 1986 Crested Butte earthquake swarm and its implications for seismogenesis in Colorado

In August and September 1986, an earthquake swarm of possibly several hundred earthquakes occurred near Crested Butte, Colorado. The epicentral area is located within the Ruby Range in a region of extensive middle Tertiary volcanic and intrusive activity. The recording of this sequence has provided the best data to date to evaluate the source characteristics of an earthquake sequence in Colorado and the associated tectonic stresses. At least 200 events were recorded at regional distances; 30 events were Richter magnitude ( ML ) 1.6 and greater, and 16 were reported felt. The largest event, ML 3.5, occurred on 3 September. In addition to the regional recordings, a portable seismographic network was deployed from 19 to 26 August. Based on these data, 78 events were relocated using a master event technique. The earthquakes define a 6-km-long, northwest-striking planar zone dipping steeply to the northeast, between the depths of 2 and 11 km. Focal mechanisms indicate predominantly normal faulting with a minor left-lateral component on an approximately northwest-striking, northeast-dipping plane. These observations are all consistent with possible slip on the Treasure Mountain fault, a late Tertiary structure within the Ruby Range. The northeast-oriented T axes exhibited by the Crested Butte focal mechanisms are consistent with the regional extensional stress direction characteristic of the southern Rocky Mountains as indicated by other earthquake focal mechanisms. Within this extensional stress regime, earthquakes in western Colorado appear to be the result of normal slip on reactivated preexisting faults that are favorably oriented to the contemporary stress field.

Rupture history of the Great Bolivian Earthquake: Slab interaction with the 660‐km discontinuity?

Teleseismic body waves of the great Bolivian earthquake of June 9, 1994 are analyzed to determine the focal parameters and rupture history. Broadband seismograms reveal a complex rupture process: A small initial event (Mw 7.2) was followed about 10 s later by a large moment release pulse of about 40 s duration. Focal mechanisms determined for the mainshock indicate normal faulting with one very shallow NE dipping plane. Azimuthal variation in body‐wave displacement pulse widths suggest northward rupture. From master event and body‐wave inversion, the main moment release is located 25–50 km NE of the initiation point at about 650‐km depth with only small depth variations between the initial and main event. This suggests that rupture was N directed on the near‐horizontal plane. Because the slab along other parts of the Andean arc at about 600‐km depth dips steeply, a sub‐horizontal plane may imply shearing perpendicular to slab dip. Downdip compression on a sub‐horizontal plane would then imply that the slab does not penetrate the 660‐km seismic discontinuity, but rather, is being sheared out to the NE. This interpretation is not unique as other scenarios are also possible. Such an event immediately above the 660‐km discontinuity suggests massive deformation above the discontinuity with no smooth slab penetration into the lower mantle.

The 1990 Rudbar‐Tarom Iranian earthquake sequence: Evidence for slip partitioning

The Rudbar‐Tarom Iran earthquake (June 20, 1990, MS= 7.7) occurred in the high Alborz mountains south of the Caspian Sea in northwest Iran. Although the structural style of the region is dominated by reverse faulting and folding, the focal mechanisms of the events indicate pure right‐lateral strike slip motion. We have studied the source parameters of the 11 largest aftershocks (MS > 4.6) using three different methodologies: (1) joint inversion of teleseismic and regional distance body waves; (2) single‐station moment tensor inversion (SSMT); and (3) comparison of entire wave trains to those of master events. The aftershocks show a clear bipolar distribution: thrusting earthquakes on NW‐SE trending planes and strike slip mechanisms with fault geometry similar to that of the mainshock. We interpret the faulting behavior of the 1990 earthquake sequence to indicate slip partitioning. In northwest Iran the Arabian and Eurasian plates are converging in a NE direction and the Alborz mountains are undergoing shortening and left‐lateral shear. This oblique motion results in dominantly thrusting earthquakes and infrequent strike slip earthquakes.

Main shock and aftershocks of the December 13, 1990, Eastern Sicily earthquake

n this paper we describe the location and the fault plane solution of the December 13, 1990, Eastern Sicily earthquake (ML = 5.4), and of its aftershock sequence. Because the main shock location is not well constrained due to the geometry of the permanent National Seismic Network in this area, we used a "master event" algorithm to locate it in relation to a well located aftershock. The revised location is slightly offshore Eastern Sicily, 4.8 km north of the largest aftershock (ML = 4.6) that occurred on December 16, 1990. The main shock has a strike-slip mechanism, indicating SE-NW compression with either left lateral motion on a NS plane, or right lateral on an EW plane. Two days after the main event we deployed a local network of eight digital stations, that provided accurate locations of the aftershocks, and the estimate of source parameters for the strongest earthquake. We observed an unusual quiescence after the ML = 5.4 event, that lasted until December 16, when a ML = 4.6 earthquake occurred. The fault plane solution of this aftershock shows normal faulting on E-W trending planes. Between December 16 and January 6, 1991, a sequence of at least 300 aftershock" was recorded by the local network. The well located earthquakes define a small source region of approximately 5 x 2 x 5 km3, with hypocentral depths ranging between 15 and 20 km. The paucity of large aftershocks, the time gap between the main shock occurrence and the beginning of the aftershock sequence (3.5 days), their different focal mechanisms (strike-slip vs. normal), and the different stress drop between main shock and after- shock suggest that the ML = 5.4 earthquake is an isolated event. The sequence of aftershocks began with the ML = 4.6 event, which is probably linked to the main shock with a complex mechanism of stress redistribution after the main faulting episode.

A new look back at the 1969 Santa Rosa, California, earthquakes

On 1 October 1969, two damaging earthquakes of Richter magnitude ( ML ) 5.6 and 5.7 occurred near the city of Santa Rosa followed by at least 200 aftershocks. These are the two largest events to occur in the northern San Francisco Bay area since the great 1906 San Francisco earthquake and its aftershocks. The epicentral area coincides with a 1-km right step-over from the northwest-striking Rodgers Creek fault to the Healdsburg fault zone and a 6-km right step-over to the Maacama fault. The 1969 sequence and its relationship to the Healdsburg and Rodgers Creek faults has taken on greater significance because of the potential for a moment magnitude ( Mw ) 7 earthquake occurring on the latter, possibly in the next several decades. We have relocated the five largest events in the 1969 sequence using the largest aftershock recorded by a temporary network operated by the U.S. Geological Survey (USGS) as the master event. The locations and focal mechanisms of the principal events, in addition to many of the aftershocks, are consistent with right-lateral strike-slip faulting on the northwest-striking Healdsburg fault. An interesting feature of the aftershocks is that they exhibit a cruciform pattern suggesting possible activity on a northeast-striking cross-fault located within the step-over. The significance of such a cross structure is that it may act as a barrier to northward rupture from a large earthquake on the Rodgers Creek fault. Seismicity since 1969 continues to image an active southern Healdsburg fault surrounding a quiescent rupture zone of the 1969 earthquakes and the possible cross-fault. The Rodgers Creek fault is generally aseismic during this period.

Ray synthetic seismograms: a useful tool in the International Data Center environment

In this paper some of the results of a feasibility study on ray synthetic seismograms usage are reported. A computational method, ways of composing synthetic traces, an application of the source wavelet and the radiation pattern and integration of such an approach into the Center for Seismic Studies (CSS) revision 3.0 structure are outlined. Further on, results obtained for model examples, nuclear explosions, and earthquakes are presented. Conclusions of the undertaken feasibility study help to understand that ray synthetic seismograms represent a very fast tool (results in “no time”) and simultaneously represent a complex tool with all needed features. The ray synthetic seismograms can be implemented in various ways: to be computed automatically and used within the Intelligent Monitoring System, to be computed automatically and provided to the analyst, to provide a database of master events, to be computed interactively by an analyst during routine daily analysis.

Seismicity, active faults and stress field of the North Muya Region, Baikal Rift: New insights on the rheology of extended continental lithosphere

The eastern Baikal rift is characterized by a succession of young en echelon half‐grabens distributed in a 300‐km‐wide zone of high seismic activity. In this study we present the overall seismotectonic setting of the north Muya region, which is an along‐strike transfer zone between two en echelon dip‐slip fault systems. Using both regional and local seismological networks, we refine the hypocenters and the single‐event focal mechanisms of 704 earthquakes recorded during 3 years, including the main period of activity of a very dense cluster, the Angarakan swarm. Our final hypocenter locations, obtained after determining a region‐specific velocity model and applying a master event technique, allow us to observe a widespread and unusual seismic deformation throughout the upper 30 km of the continental crust. Deep earthquakes (15–30 km) define a planar surface related to a large north‐dipping basement fault (Upper Muya), and also occur in a small cluster at 23 km depth at the intersection of two major structural trends. Along more than 150 km of strike length from west to east, maximum focal depths continuously increase from 20 to 30 km, whereas fault dip concomitantly increases from 30° to 55°. On the other hand, most activity of the Angarakan swarm is confined in the “classical” depth range of 0–15 km and depicts a narrow and steeply south‐dipping seismic band related to a very recent fault scarp at the surface (eastern Kovokta). This fault geometry, which is observed at depth and in the field, and the opposite tilting of regularly 30‐km‐spaced horsts and grabens, reveal an asymmetrical rift system. The Agreement of fault lengths estimated from field studies with rupture lengths estimated from large historical earthquakes strengthens the inference that large earthquakes may occur near the base of the seismogenic layer of the crust, i.e., 30 km. Most of the 39 focal mechanisms determined in the study region show a dominant normal dip‐slip displacement on nodal planes striking in the main inherited structural direction WSW‐ENE, whereas some limited pure strike‐slip faulting mainly occurred in the highly fractured Angarakan zone. From inversion of 37 focal mechanisms, we deduce a tensional stress tensor with a σ3 axis striking N160° and a σ2 axis slightly compressional (shape factor R of 0.4). Comparing these results with stress tensors computed along the eastern Baikal rift and with the focal mechanisms of three strong earthquakes that occurred in the Muya region during this century shows that small‐magnitude earthquake data give consistent and accurate constraints on the regional stress regime. Whether this present stress tensor remained stable since the beginning of opening of the Baikal rift and during Quaternary is unknown. Our results confirm that continental lithosphere submitted to rapid extension in an early stage of rifting may retain significant rigidity.

Fault plane determinations for three small earthquakes along the San Jacinto Fault, California: Search for cross faults

Using rupture directivity and aftershock locations, three earthquakes (ML 4.2 – 4.3) were studied to determine which of the two nodal planes of the focal mechanism was the rupture plane. All three events occurred near the San Jacinto fault zone during December 1989 and February 1990 and had a large component of strike‐slip motion with nodal planes oriented in the northeast and northwest directions. Smaller earthquakes were used as empirical Green functions to deconvolve the path and site complications from the short‐period P waveforms. The resultant far‐field source time functions were inverted for slip on a finite fault and tested to see which nodal plane provided a better fit to the data. The results of the waveform inversion showed that the earthquake near the Anza Gap (December 2, 1989) and the earthquake near San Bernardino (December 28, 1989) ruptured northwest trending planes. The earthquake near Buck Ridge (February 18, 1990) appeared to rupture a northeast trending plane, but the waveform results for this event were not considered as reliable. Available aftershock data for the three events were also examined using master event relocations. The aftershock sequence following the Buck Ridge event showed a clear lineation trending toward the northeast. The results of this study indicate that the Anza and San Bernardino earthquakes ruptured areas on or parallel to the San Jacinto fault, while the Buck Ridge event ruptured a northeast trending cross fault. In addition to determining the fault plane, the waveform analysis resolved details of the slip distribution for these relatively small earthquakes.

ML:M0 as a regional seismic discriminant

AbstractThe mb:MS ratio determined by teleseismic observations has proven to be an effective discriminant, for explosive sources tend to be significantly richer in short-period energy than are earthquakes. Unfortunately, this method is limited by the detection threshold of teleseismic surface waves. However, recent advances in instrumentation allowing low amplitude surface wave measurements coupled with new analytical techniques make it feasible to use regional waveform data to determine the long-period source excitation level of low magnitude events. We propose using the ratio of ML (local magnitude) to M0 (scalar seismic moment) as an analogous regional discriminant. We applied this criterion to a data set of 299 earthquakes and 178 explosions and found that this ratio seems to be diagnostic of source type. For a given M0, the ML of an explosion is more than 0.5 magnitude units larger than that of an earthquake. This separation of populations with respect to source type can be attributed to the fact that ML is a short-period (1 Hz) energy measurement, whereas seismic moment is determined from long-period body wave phases (period > 4 s) and surface waves (10 to 40 sec). Using regional stations with sources 200 to 600 km away, the effective threshold for magnitude measurements for this discriminant is found to be ML = 3.1 for earthquakes and ML = 3.6 for explosions. This method does require the determination of regional crustal models and path calibrations from master events or by other means.

Master-event correlations of weak local earthquakes by dynamic waveform matching

SUMMARY Dynamic waveform matching (DWM) performs a non-linear correlation between two seismograms that are similar in shape but may be squeezed or stretched relative to each other. It extends the application of master-event comparisons to seismograms of greater spatial distance and retains the high-timing resolution of correlation techniques that act on the original time series. The DWM approach is applied to data recorded by a small array being part of the BOCHUM UNIVERSITY GERMANY (BUG) network which monitors the mining-induced seismicity in the Ruhr basin of NW Germany. The observed epicentres occur in clusters and therefore display only a limited number of seismogram waveform types. In one application an automatized cluster association with DWM obtains a resolution of about 100 metres at an epicentral distance of 20 to 40 km, using 10-20 defined master events for each region. These results are confirmed both by seismograms from a near-site station for mininginduced events from the Hamm region and by blast reports for a quarry region near Wuppertal. In another application of DWM, array traces from the BUG array are correlated to yield azimuth and slowness for epicentre location. As for the master event application, this approach is tuned for high performance on weak local events using a priori information about the approximate epicentral region. The implemented processes are shown to be capable of locating events with a rate of success equal to the performance of an experienced seismologist when processing all seismograms of four years BUG registration.

Clustering and location of mining induced seismicity in the Ruhr Basin by automated master event comparison based on Dynamic Waveform Matching (DWM)

Abstract Most of the local seismicity in the Ruhr Basin can be separated into characteristic clusters of similar, mining induced earthquakes. Each cluster can be represented by a strong master event. Therefore, it is possible to associate weak events to the corresponding clusters by master event comparison. The seismic signal matching is performed by a nonlinear correlation termed DWM for the entire seismogram length. DWM permits stretchings and shortenings between the two signals and overcomes the ambiguities in phase correlation by a consistent matching path. The automatic cluster association searches for the best DWM-correlation between the actual event and all master events of the appropriate epicenter region. Knowing the P- and S-onsets of the master event, they can be transposed to the actual event by the correlation path with one sample accuracy. The method has been applied to all BUG small array recordings 1987–1990 of local events from the Hamm -region to investigate spatial and temporal clustering. Within the clusters, a high percentage of weak events could be located relative to its master event. The temporal clustering resolved seismic activities that typically last a few months per cluster, but single aftershocks occur in the following years.

A comparison of epicenter relocation techniques: a case study for the Juan de Fuca plate

The ‘master event relocation’ and ‘hypocentroid’ techniques are compared in an effort to determine their relative performance in evaluating the location of epicenters in a region hampered with recording station location bias and a poorly understood seismic velocity structure. The study areas chosen are the Explorer and Gorda plates, two small lithospheric subplates of the oceanic Juan de Fuca plate, which show a high degree of apparent intraplate seismicity. Previous studies have indicated systematic eastward errors in the location of epicenters due to the lack of recording stations to the west. Both techniques produce similar results in defining plate boundary vs. intraplate seismicity; however, noteworthy differences exist. The selection of a well-constrained master event has a clear effect on the alignment of relocated epicenters with plate boundaries. An event on the Gorda plate located by ocean bottom seismometer allows for a relocation of epicenters about the Gorda Rise. Using a more widely recorded master event on the Explorer plate produces parallelism of the events with the plate boundaries but an additional uniform shift in the relocated epicenters is required to align seismicity with the boundaries. Location of the master event, with respect to tectonic features of the plate, appears to affect the results, probably owing to differences in lithospheric conditions and their effect on wave travel paths through the plate. A comparison of results of the two techniques on intraplate seismicity on the Gorda plate shows opposing direction vectors for the relocations. The hypocentroid technique is less affected by changing conditions within the plate as events are clustered according to location. The relocations are still affected by the station location bias, as the events are relocated in relation to the weighted center of the event cluster. The technique produces greater definition of the trends of the relocated epicenters; however, in this situation an additional shift of the epicenters is required for alignment with the boundaries.

A waveform correlation method for identifying quarry explosions

Abstract A waveform correlation method is presented for identifying quarry explosions by attributing them to known mines characterized by multiple master events. The objective is to provide a reliable automatic procedure for screening the large number of quarry explosions likely to be detected by networks of in-country stations monitoring compliance with test-ban treaties. The method generalizes existing correlation techniques to compare waveforms from an unlocated event recorded at an array of sensors with a linear combination of master event waveforms recorded at the same array. The use of a linear combination reduces the chance of a missed location caused by some variation in mechanism or spectral excitation between the events being compared. The weights in the linear combination are filters, offering some compensation for variations in source time functions and errors of waveform alignment. The use of array data reduces the likelihood of false attribution by reducing bias and variance in the correlation measurement. In a test conducted with P-wave data segments recorded at a 13-element array, the method successfully resolves two source regions separated by 4 km at a range of 150 km. Resolution with single-station waveform correlations is marginal due to the limited amount of data. The statistics of the sample waveform correlation coefficient are developed and demonstrate that single-station waveform correlations are unreliable unless estimated with large signal durations T or bandwidths B. A time-bandwidth TB product exceeding 100 (or smaller TB with more stations) is necessary for reliable event attribution. The related problem of separating superimposed waveforms from two events in different source regions may be solved by cancellation. The waveforms of one event are again approximated by a linear combination of waveforms from master events in the same mine. The residual signals, obtained by subtracting the approximation from the superimposed waveforms, estimate the waveforms from the second event. This method achieves significant separation of waveforms from events 6 km apart at a range of 150 km, using data from the 13-element array. Its resolution exceeds that of conventional beam-forming methods.

Comparison of the aftershock distribution and the fault plane size of the M 4.9 Kameoka earthquake on May 28, 1987, Japan

It has been pointed out in several investigations that an aftershock gap exists in the neighborhood of the main shock, where the main shock slip is large for large earthquakes. Since the source process of small earthquakes is thought to be less complex than that of large earthquakes, it is expected that by comparing the size of the fault plane and the pattern of aftershock distribution for small earthquakes the physical meaning for large earthquakes may be clarified. The 28 May 1987 Kameoka earthquake with M 4.9 occurred in the midst of a telemetered seismic network and the data, therefore, were recorded with high accuracy. The aftershocks were relocated by means of the master event technique with the aftershock gap found in the neighborhood of the hypocenter of the main shock. P wave forms radiated by the Kameoka earthquake were observed with low-gain velocity seismographs having a natural period of 10 s without saturation. The fault plane size was estimated by comparing observed seismograms having small hypocentral distances with synthetic seismograms obtained by forward modeling. It was shown that the gap of aftershocks was comparable to the fault plane size of the small earthquake such as the M 4.9 Kameoka earthquake.

Slowness power spectrum analysis of the coda composition of two microearthquake clusters in northern Switzerland

The S-wave coda composition of two microearthquake clusters in northern Switzerland has been analysed using the slowness power spectrum analysis proposed by Spudich and Bostwick. The purpose of this paper is to estimate the directions in which individual coda wavelets leave the source region for lapse times more than six times the S-wave travel time. The data set consists of 12 and 16 events in the magnitude range of 1.9–2.6 and 1.4–3.4, respectively. The depths range from 7 to 8 km and from 6 to 7 km. Digital data from five vertical-component stations (ENB, RBF, GEF, GIF and TSB) and two three-component stations (ACB and CHE) were available. The hypocentral distances to the recording sites vary between 29 and 63 km for the first cluster (Günsberg) and 10 and 24 km for the second cluster (Läufelfingen), respectively. All events have been relocated using a master event technique with inter-event arrival times estimated by a cross-correlation method. The slowness power spectra for the Läufelfingen cluster at all stations show essentially two different patterns. In agreement with the results of Spudich and Bostwick, we found that the early coda following the direct S wave immediately was composed of wavelets leaving the source region with the same slowness vector as the direct wave. We interpret this coda part as being very probably produced close to the recording site. However, with increasing lapse times we observe a different coda type which receives its energy from wavelets leaving the source region in a variety of directions. The transition between the two coda ‘types’ takes place often at 1.5–2 times and always before three times the S-wave arrival time.

The 1967 Caracas Earthquake: Fault geometry, direction of rupture propagation and seismotectonic implications

The fault plane orientation of the July 30, 1967, Caracas earthquake (Mw = 6.6) has been a source of controversy for several years. This strike‐slip event was originally thought to have occurred on an east‐west oriented fault plane, reflecting the relative motion between the Caribbean and South American plates. More recently, however, the complex seismic radiation from this event was interpreted as being indicative of a north‐south striking fault that ruptured along three en echelon segments. In this study we synthesize evidence based on the intensity and damage reports, the distribution of aftershocks, and the results of a joint formal inversion of the P and SH waves and show that these data clearly indicate that the rupture of the 1967 earthquake occurred on an east‐west trending fault system. Using a master event technique, the largest aftershock, which occurred 40 min after the main event, is shown to lie 50 km east of the epicenter of the mainshock. The epicentral distances of small aftershocks registered in Caracas, based on the S–P arrival time differences and the polarizations of the P waves, are also consistent with these events occurring on an east‐west oriented fault system north of Caracas. A joint inversion of the teleseismic P and SH waves, recorded on long‐period seismographs of the World‐Wide Standardized Seismographic Network, shows that in a time frame of 65 s, four distinct bursts of seismic moment release (subevents) occurred, with a total seismic moment of 8.6×1018 N m. The first three subevents triggered sequentially from west to east, in a direction that is almost identical to the east‐west trending nodal planes of the source mechanisms. The average depth of these three subevents is 14 km. The fourth, and last identifiable, subevent of the sequence shows a reverse faulting mechanism with the nodal planes oriented roughly east‐west. It occurred at a 21‐km depth, about 50 km to the north of the fault zone defined by the strike‐slip subevents. This fourth subevent appears to reflect compressional deformation of the southern Caribbean, possibly related to underthrusting along the proposed Curaçao trench. The complexity of the fault system causing the 1967 earthquake suggests that the relative motion along the Caribbean‐South America plate boundary in central Venezuela is taken up over a broad, highly faulted, and highly stressed zone of deformation and not by a simple, major throughgoing fault.

Master event relocation of Gorda Block earthquakes: Implications for deformation

We have used the “master‐event” technique to relocate 73 earthquakes in the northern half of the Gorda block. The revised epicenters indicate a systematic eastward bias of ∼30 km in epicenters as published in the ISC Bulletin. The new seismicity pattern reflects a more symmetric distribution of earthquakes about the Gorda Rise, and a reduced degree of seismicity within the Gorda block. Although the level of intraplate activity has apparently been reduced, a significant number of events (10–20) still appears to be related to intraplate deformation. Further relocation studies, coupled with focal mechanism determinations, are necessary to better constrain deformation models for this region.

Lower Adriatic Sea seismic sequence (January 1986): spatial definition of the seismogenic structure

In January 1986 a swarm-like seismic sequence (main shock M L = 4.8) took place in the lower Adriatic Sea. This sequence was recorded by the Italian Telemetered Seismological Network using a digital acquisition system. Using additional data collected by other Italian, Yugoslavian and Albanian institutions, 15 shocks of the sequence were located by means of the “master event” technique, and for two of them a focal mechanism solution was obtained. We attempted to frame our results within the knowledge available in the literature. The Adriatic microplate is regarded by many authors as a whole block; nevertheless it exhibits moderate seismicity. The question has arisen as to whether it can be considered an African promontory or whether it behaves independently. The area actived by the sequence is rather small (less than 10 km in diameter) owing to the low level of seismic energy release and does not show any particular trends. The focal mechanisms are consistent with the recent stress field in the Adriatic area inferred from previous studies, which indicate a compression axis oriented NE-SW.