S52A-01 10:20h
The MAFI Project: Mapping Active Faults in Italy by Using Microseismicity Data.
In past years, earthquake forecasting and seismic hazard in Italy have been approached by using geological and geophysical data yielding only a partial definition of seismic release for the main active structures. In this project, we collect seismological and geodetic data to yield deterministic constraints for seismic hazard studies in areas where large earthquakes are expected to occur in a near future, called lacunae. The basic idea is to massively deploy arrays of instruments in the lacunae areas to acquire seismic and geodetic data with the goals of defining location, geometry and kinematics of the active faults and possibly constraining their strain rate. We selected three target regions: two along the Apennines (Northern Umbria and Abruzzo) and one in the Southern Alps (Alpago-Cansiglio). These areas are characterized by different tectonics and different historical seismic release. We present results for the areas located along the Apennines: the Umbria 2000-2001 and the Abruzzo 2003-2004 experiments while for the Alpago-Cansiglio we are still collecting and processing data. Preliminary results for the Umbria lacuna shows that the collected microearthquakes allow us to clearly recognize the fault system geometry and the deep structure (P- and S-wave velocity and attenuation).
S52A-02 10:35h
Does Tectonic Knowledge Help Identifying Accelerating Moment Release Prior to Large Earthquakes? Examples From Italy
Before large events, moderate magnitude seismicity has often increased in a broad area around the future epicenter. Apparently, the seismic moment release accelerates (AMR) according to a time-to-failure power law. Several models explain such a phenomenon based on regional damage mechanics or the physics of criticality, where the (critical) region and the (critical) time involved are statistically determined. This work exploits the feasibility of the accelerated moment release concept for Italy, testing if the region showing the AMR pattern can be set a-priori. Specifically, we select areas based only on tectonics by including the major and antithetic faults. In such predetermined areas, we search the historical (1600-1980) and instrumental (1987-2003) earthquake catalogues for constant or accelerated MR patterns. In addition, we consider all earthquakes above a magnitude threshold (M 6.0 historical, M 5.3 instrumental) to define successes, failures, and false alarms. Results show that: (a) all earthquakes occurring in the Apennines fold-and-thrust belt were correctly anticipated; (b) the large shock is not required to occur at the centre of the region. On the contrary, (c) AMR patterns did not precede earthquakes in the less-faulted and slowly-deforming Apulia platform, although (d) identification of false alarms needs additional work. We conclude that tectonics strongly influences the detection of AMR patterns and provides useful a-priori information.
S52A-03 10:50h
Recent Seismic Activity of the Messina Straits Area, Italy, and the Magnitude 7, 1908 Messina Earthquake
The 1908 Messina earthquake is one of the strongest historical seismic events that ever occurred in Italy, with more than 60,000 casualties and extensive damage. Precision double-run levelings were carried out a few years before the 1908 earthquake, and some lines were resurveyed just after the seismic event. Such data were used to model the source fault (Amoruso et al. 2002). Recent seismicity level in the Messina Straits was low, and only one earthquake exceeded magnitude 5.0 (M=5.6, 1978). Earthquakes occurred between January 1978 and March 2001 at crustal depth in the Messina Straits area were located using an up-to-date 3D velocity model (Barberi et al. 2004). Best-quality earthquake fault-plane solutions were used to obtain stress tensor. Fault plane was identified for the majority of the events (Neri et al. 2004). A preliminary comparison among the results of the above-mentioned studies has proved encouraging (Amoruso et al. 2004). This presentation deals with further improvements and final results. We have re-analyzed coseismic geodetic data of the 1908 Messina earthquake and tested robustness of retrieved fault parameters and slip distribution, by taking into account the effects of crustal layering and combining random measurement errors and uncorrelated uplift residuals. The data set of recent seismicity has been extended to 2004, and a more detailed 3D velocity model with an inversion grid spacing of 10 km has been obtained. We have checked for possible correlation between the 1908 earthquake slip distribution and the spatial distribution of recent earthquakes located in the fault volume and the related seismic strain. Results have been compared with the geology and the geomorphology of the Messina Straits area.
S52A-04 11:05h
Seismic Structure and Activity of the North Anatolian Fault in the Sea of Marmara From the SEISMARMARA Leg1 Seismic Survey
The Sea of Marmara is the continuation to the West of the North Anatolian Fault (NAF) that produced the two destructive earthquakes of Izmit and Duzce in 1999 just to its East. It is prone to future major earthquakes as it has experienced in the past and is a present seismic gap. In 2001, the SEISMARMARA project was carried out in the whole North Marmara Trough with the aim to improve the knowledge of the regional tectonics and the evolution at crustal scale by imaging faults, basins and deep geological features by their structure. This experiment was a multi-method approach combining regional scale multichannel seismic profiles (MCS), refraction and wide-angle reflection towards OBS and land stations as well as earthquake recording by the same receivers. The traveltime modelling from coincident vertical reflection and refraction observations has been implemented for the two longest regional lines striking EW the Sea of Marmara : one line on the southern rim of the Trough and an other across the succession of basins and highs. The structural constraints for the shallow part is given by the MCS coincident data whereas the basement topography and the structure of the lower crust is sampled and modelled by the OBS and land stations data. This methodology is being extended to the transects that cross these two linesto sense the structure in space and with the help of new structural constraints obtained thanks to Pre-Stack Depth Migration (PSDM). The PSDM has been performed on 3 transects : 2 cross lines in the zone of the Central Basin and an other line in the region connecting the Central Basin to the Cinarcik Basin. The PSDM allowed us to image in details the fine heterogeneity related to the transtensionnal environment. The depth section across the deepest part of the Central Basin gives us information about the geometry and the detailed velocity field for the 5 km thick sedimentary layers when the depth section on the eastern side of this major basin reveals us clearly a new fault, distinct from the one commonly considered. This fault was simply suggested by a conventional seismic processing and which can be interpreted as the south boundary fault which could have caused the rhomboid Central Basin as a pull-apart. The depth section corresponding to the line across the small Kumburgaz Basin and the Central High brings us new constraints on the architecture of the NAF, especially on the northern rim of the Kumburgaz basin. Improving constraints on structure in this latter zone is of critical interest because this is the segment of the NAF which appears silent in current seismicity. This could be either because there is no active fault or because it is aseismic and creeping or in contrary because it is seismogenic and locked. It will be supplemented by detailed earthquake analysis on the OBS data of the survey.
S52A-05 11:20h
The 22 June 2002 Changureh (Avaj) Earthquake in Qazvin Province, NW Iran: Epicentral Re-location, Source Parameters, Surface Deformation and Geomorphology.
The Mw 6.4 Changureh (Avaj) earthquake occurred on the 22 June, 2002 in Qazvin province, NW Iran. We use observations from seismology, field investigation and analysis of satellite imagery and digital topography to suggest that slip on a previously unrecognised thrust fault (herein named the Abdarreh fault) was responsible for the earthquake. Inversion of long-period P and SH body-wave seismograms shows rupture on a thrust fault dipping 49 degrees to the southwest and with a centroid depth of about 10 km. Multiple-event relocation of the main-shock and aftershock epicentres, and discontinuous surface ruptures observed after the earthquake are compatible with a NW propagating rupture on a SW-dipping thrust, but maximum recorded displacements are much less than expected from seismology, suggesting that much of the slip failed to reach the surface and was accommodated as folding at the surface instead. Long-term folding is difficult to see in the topography of the epicentral region as the Abdarreh fold is growing through a relict Neogene topography. Anticlinal uplift can however be inferred from drainage disruption and stream incision. The 22 June, 2002 Changureh earthquake shows the importance of being able to interpret diagnostic features of active faulting in the landscape.
S52A-06 11:35h
Fossil fluid reservoir beneath a duplex fault structure within the Central Range of Taiwan: implication for earthquake rupturing by leakage and volatilization of confined metamorphic fluid at depth
We investigated the central part of the Taiwan slate belt and discovered potential fossil fluid reservoirs beneath a duplex fault structure. We described the occurrence of the fossil fluid reservoirs, examined microstructures of the highly silicified zones, and microprobed the veining minerals within the reservoirs. The investigation was driven by better understanding the mechanisms which likely facilitated the speedy rupturing process of the 1999 Mw 7.6 Taiwan Chi-Chi earthquake. We examined exposed rocks in regions, where the pressure and temperature conditions mostly resemble the hypocenter of the Chi-Chi earthquake, i.e., about sub-greenschist facies. Field observations and composition analyses of the silicified vein-rich zones beneath the duplex structure suggest that impermeable slate layers may serve as cap rocks for confining deep-seated fluids. These fluids most likely come from the Taiwan metamorphic complex at depths by the dehydration and decarbonation reactions (or partial melting). The presence of albite, K-feldspar, calcite and traces of REE minerals (monazite, and rutile) within the analyzed veins indicates a high-temperature metamorphic origin of the fluids. In addition, the gouge zone of a link fault above the detachment also indicates the presence of high-pressure fluids during faulting. To explain these observations, it is probable that episodic leakage of the confined fluid pockets may provide essential fluids for fault lubrication during earthquake rupturing. After leakage and subsequent volatilization of gases, the volatile fluids injected into earthquake fault planes and significantly ease the earthquake rupturing process
S52A-07 11:50h
The Relation between Seismicity and Active Faults in Northwestern Taiwan
Northwestern Taiwan is a major economical, transportation and cultural center in our country. Geologically, this region is covered by red soil terraces. These terraces are displaced by several active faults. Seismicity reveals potential of strong ground motion in this region. We deploy a microearthquake network to locate the active faults in the region. The aim of our study is to find the characteristics of seismicity in the Taoyuan, Hsinchu, and Miaoli region, and the relation between seiemicity and active faults. The Central Weather Bureau (CWB) in Taiwan has deployed a short period seismic network in whole Taiwan. Many earthquakes were located by the CWB seismic network, but many small mircoearthquakes occurred in the local area can not be detected. Thus we deploy a temporary seismic network in the area since 2001. Ten stations were deployed. Each station includes an accelerograph with GPS and three component velocity-type sensors. The sampling rate is 200 points/sec. There are two ways to increase the accuracy of the earthquake location. One is to increase the seismic stations in the study area. The other way is to improve the location method. We locate the earthquakes by using the data of both CWB seismic network and our seismic network. The results show most earthquakes occurred in the mountain area in the southeast part of the seismic network. After we relocate the earthquakes, the focal depth changes significantly. Most focal depths are less than fifteen kilometers and become densely clustered.