Crustal and upper mantle structure of the Kozani-Grevena area obtained by non-linear inversion of P and S travel times

The present study focuses on the P and S crustal and uppermost mantle velocity structure in the broader Kozani-Grevena area. The velocity structure is derived from the inversion of travel times of local events. The main data source is the travel times from the aftershock sequence of the large event of 13 of May 1995 (M_w = 6.6) which occurred in the study area. An appropriate preconditioning of the final linearized system is used to reduce ray density effects on the results. An attempt is made to interpret the features and details of the crustal structure in terms of the geotectonic setting of the area. The observed features of the deeper crustal and uppermost mantle structure are in very good agreement with previous results. Specifically, a crustal thickening is observed along a ENE-WSW direction, perpendicular to the well-known Dinaric trend (NNW-SSE) of the geological formations of the area, in accordance with the theoretical expectation of a thicker crust under the accretion prism which starts at the SW edge of the study area.     

Seismic hazard assessment and design spectra for the Kozani-Grevena region (Greece) after the earthquake of May 13, 1995

The Kozani-Grevena (Greece) destructive earthquake occurred in a region of low seismicity. A considerable amount of strong-motion data was acquired from the permanent strong motion network of the Institute of Engineering Scismology and Earthquake Engineering (ITSAK) as well as from a temporary one installed after the earthquake. On the basis of this data set as well as on the observed macroseismic intensities, local attenuation relations for peak ground acceleration and velocity are proposed. A posteriori seismic hazard analysis is attempted for the affected and surrounding areas in terms of peak ground acceleration, velocity, bracketed duration and spectral acceleration. The analysis shows that the event of May 13, 1995 can be characterized as one with a mean return period of 500 to 1000 years. Relying on the observed spectral-acceleration amplification factors and the expected peak ground acceleration for mean return period of 500 years, region-specific elastic design spectra for the buildings of the Kozani and Grevena prefectures are proposed.     

On the reliability of different methods of seismic hazard assessment in Greece

Historical and present century instrumental data have been used to determine seismic hazard in 35 sites of Greece by the application of Cornell's method (Cornell, 1968) and the mean value method. The macroseismic intensity has been considered as a measure of seismic hazard. Comparison of the results of the two methods showed that, in general, the mean value method gives higher values, particularly for low probabilities of exceedance. In addition, for some sites, the differences of the expected intensities resulting from the two methods, indicate that finer tuning of the seismogenic souce model is required, or suggest time dependence. Although each one of these methods has its own merits, the method based on seismic zonation (Cornell's method) has several advantages and must be preferred when an accurate zonation is possible by the use of macroseismic and instrumental seismic data, together with geological and geomorphological information. However, reliable estimates of seismic hazard at a particular site require work on a microzoning scale, incorporating historical, archaeological, and recent geological data.Paper presented at the 21st General Assembly of the European Seismological Commission held in Sofia, 1988.     

Local habitat drivers of macrobenthos in the northern,central and southern KwaZulu-Natal Bight,South Africa

The relatively wide KwaZulu-Natal Bight between St Lucia and Durban on the north-east shelf of South Africa is characterised by several circulation features driven by the Agulhas Current, wind and coastal inputs. A large multidisciplinary programme investigated the sources and relative influences of nutrients on the shelf. Within this, and to address a critical knowledge gap, this study describes macrobenthic (<1 mm) composition and frequency from 16 stations, assigned amongst four oceanographic focus areas. The areas were predetermined across the disciplines to represent upwelling, outwelling and a semi-persistent eddy, with nutrients and primary productivity being measured at each. Environmental variables such as sediment distribution, sediment TOC and bottom water physico-chemistry were determined at a significantly larger spatial scale. Our study postulated that oceanographic focus areas support significantly different macrobenthic assemblages, and that composition and relative distribution is due to measurable habitat attributes at each. Macrofauna were relatively abundant and particularly rich at >1 000 taxa. Annelida, Arthropoda, Mollusca, Echinodermata, Sipuncula and Cnidaria (>50 taxa each) were the dominant macrobenthic groups in the bight. Annelida were dominated by the polychaete families Spionidae, Terrebelidae and Cirratullidae, which were generally associated with outwelling and a mud depocentre off the Thukela River. Two unique and distinctive assemblages were found, one in the Thukela Mouth focus area and another on the midshelf between Thukela and Durban. The latter is influenced by poorly sorted, coarse sand and with probable influences from the Durban Eddy. There assemblages were abundant, rich and specific to this habitat. Correlation, PERMANOVA and CAP analyses showed assemblage fidelity to the focus areas. Medium sand, fine sand, mud and the variance of overall sediment type were the habitat drivers underlying macrofaunal abundance distributions.     

Applications of hybrid genetic algorithms in seismic tomography

Almost all earth sciences inverse problems are nonlinear and involve a large number of unknown parameters, making the application of analytical inversion methods quite restrictive. In practice, most analytical methods are local in nature and rely on a linearized form of the problem equations, adopting an iterative procedure which typically employs partial derivatives in order to optimize the starting (initial) model by minimizing a misfit (penalty) function. Unfortunately, especially for highly non-linear cases, the final model strongly depends on the initial model, hence it is prone to solution-entrapment in local minima of the misfit function, while the derivative calculation is often computationally inefficient and creates instabilities when numerical approximations are used. An alternative is to employ global techniques which do not rely on partial derivatives, are independent of the misfit form and are computationally robust. Such methods employ pseudo-randomly generated models (sampling an appropriately selected section of the model space) which are assessed in terms of their data-fit. A typical example is the class of methods known as genetic algorithms (GA), which achieves the aforementioned approximation through model representation and manipulations, and has attracted the attention of the earth sciences community during the last decade, with several applications already presented for several geophysical problems.In this paper, we examine the efficiency of the combination of the typical regularized least-squares and genetic methods for a typical seismic tomography problem. The proposed approach combines a local (LOM) and a global (GOM) optimization method, in an attempt to overcome the limitations of each individual approach, such as local minima and slow convergence, respectively. The potential of both optimization methods is tested and compared, both independently and jointly, using the several test models and synthetic refraction travel-time date sets that employ the same experimental geometry, wavelength and geometrical characteristics of the model anomalies. Moreover, real data from a crosswell tomographic project for the subsurface mapping of an ancient wall foundation are used for testing the efficiency of the proposed algorithm. The results show that the combined use of both methods can exploit the benefits of each approach, leading to improved final models and producing realistic velocity models, without significantly increasing the required computation time.     

Recent reliable observations and improved tests on synthetic catalogs with spatiotemporal clustering verify precursory decelerating–accelerating seismicity

We examined the seismic activity which preceded six strong mainshocks that occurred in the Aegean (M?=?6.4–6.9, 33–43° N, 19–28° E) and two strong mainshocks that occurred in California (M?=?6.5–7.1, 32–41° N, 115–125° W) during 1995–2010. We find that each of these eight mainshocks has been preceded by a pronounced decelerating and an equally easily identifiable accelerating seismic sequence with the time to the mainshock. The two preshock sequences of each mainshock occurred in separate space, time, and magnitude windows. In all eight cases, very low decelerating seismicity, as well as very low accelerating seismicity, is observed around the actual epicenter of the ensuing mainshock. Statistical tests on the observed measures of decelerating, q _d, and accelerating, q _a, seismicity against similar measures calculated using synthetic catalogs with spatiotemporal clustering based on the ETAS model show that there is an almost zero probability for each one of the two preshock sequences which preceded each of the eight mainshocks to be random. These results support the notion that every strong shallow mainshock is preceded by a decelerating and an accelerating seismic sequence with predictive properties for the ensuing mainshock.     

Present patterns of decelerating–accelerating seismic strain in South Japan

Decelerating generation of preshocks in a narrow (seismogenic) region and accelerating generation of other preshocks in a broader (critical) region, called decelerating–accelerating seismic strain (D-AS) model has been proposed as appropriate for intermediate-term earthquake prediction. An attempt is made in the present work to identify such seismic strain patterns and estimate the corresponding probably ensuing large mainshocks (M ≥ 7.0) in south Japan (30–38° N, 130–138° E). Two such patterns have been identified and the origin time, magnitude, and epicenter coordinates for each of the two corresponding probably ensuing mainshocks have been estimated. Model uncertainties of predicted quantities are also given to allow an objective forward testing of the efficiency of the model for intermediate-term earthquake prediction.     

Fault-plane Solutions Determined by Waveform Modeling Confirm Tectonic Collision in the Eastern Adriatic

—?Source parameters for thirteen earthquakes in the SE Adriatic region have been determined using P and SH body-waveform inversion. The results of this modeling are combined with eleven other earthquakes with M?≥?5 whose focal mechanisms have been determined mainly by waveform modeling. The results confirm that movement on mainly low-angle reverse faults causes the deformation in coastal southern Yugoslavia through Albania up to the Lefkada Island in NW Greece. This zone of thrusting has a NW–SE trend (N34°W), follows the coastline, and dips towards the continent. The slip vectors of these events trend at ～N229° along the Dalmatian coasts, to ～N247° along Albania and NW Greece. The deformation is attributed to the continental collision between the Adriatic block to the west and Eurasia to the east. Along the mountain line in eastern Albania (Albanides Mts.) and in NW Greece (Hellenides Mts.), E–W extension is occurring. The E–W extension associated with the orogenic belt could be attributed to a variety of models such as: gravity, internal deformation of the thrust wedge, a probable down bulge of the dense lithosphere of the Adriatic block beneath the Eurasian lithospheric plate in combination with the compressional stresses applied along the collision belt.     

Current accelerating seismic excitation along the northern boundary of the Aegean microplate

According to previous observations [Geophys. Res. Lett. 27 (2000) 3957], the generation of large (M≥7.0) earthquakes in the western part of the north Anatolian fault system (Marmara Sea) is followed by strong earthquakes along the Northern Boundary of the Aegean microplate (NAB: northwestermost Anatolia–northern Aegean–central Greece–Ionian islands). Therefore, it can be hypothesized that a seismic excitation along this boundary should be expected after the occurrence of the Izmit 1999 earthquake (M=7.6). We have applied the method of accelerating seismic crustal deformation, which is based on concepts of critical point dynamics in an attempt to locate more precisely those regions along the NAB where seismic excitation is more likely to occur. For this reason, a detailed parametric grid search of the broader NAB area was performed for the identification of accelerating energy release behavior.Three such elliptical critical regions have been identified with centers along this boundary. The first region, (A), is centered in the eastern part of this boundary (40.2°N, 27.2°E: southwest of Marmara), the second region, (B), has a center in the middle part of the boundary (38.8°N, 23.4°E: East Central Greece) and the third region, (C), in the westernmost part of the boundary (38.2°N, 20.9°E: Ionian Islands). The study of the time variation of the cumulative Benioff strain in two of the three identified regions (A and B) revealed that intense accelerating seismicity is observed especially after the occurrence of the 1999 Izmit mainshock. Therefore, it can be suggested that the seismic excitation, at least in these two regions, has been triggered by the Izmit mainshock.Estimations of the magnitudes and origin times of the expected mainshocks in these three critical regions have also been performed, assuming that the accelerating seismicity in these regions will lead to a critical point, that is, to the generation of mainshocks.