S43D-1901
Towards a global catalog of instrumental seismicity: current status
We report the current status of our efforts to create a global catalog of instrumental seismicity that can be used as an authoritative source of earthquake locations, magnitudes and other associated parameters (source mechanisms, rupture areas, casualties, references, etc) in a wide range of geoscience and social science investigations. These efforts can be summarized in three main tasks. First, we are systematically relocating all the instrumental seismicity before 1964 as reported in the bulletins of the International Seismological Summary using the same methodology as with more recent data (Engdahl et al., Bull. Seism. Soc. Am., 88, 722-743, 1998). The relocations for 1960-1963 were recently completed, and we present here our results for 1959. Second, we are compiling seismic moment estimates, and rupture zone geometries for the largest instrumentally recorded earthquakes. For earthquakes occurring after 1976 the main source of seismic moments is the catalog of the Global Centroid Moment Tensor project, while for earlier events we obtain them from modeling studies found in the scientific literature. Similarly, rupture zones for recent large earthquakes (Mw > 8) are obtained from finite fault inversions of the USGS, aftershock distributions, and other quantitative modeling results, while for earlier events we use information obtained from macroseismic investigations and/or field observations and paleoseismology. Finally, we are organizing all this information in a digital, searchable database that will be accessible through a web interface and/or client-server applications. The schema of this database is based on that adopted by IASPEI and the International Seismological Centre, with extensions for other information not currently included (e.g. rupture areas, surface rupture, slip distribution, detailed information on casualties and damage, etc).
S43D-1902
A Global Probability-based Magnitude of Completeness (PMC) Study
We use the Probability-based Magnitude of Completeness (PMC) method to compute the spatial variation of the detection capability for a global catalog. Unlike other completeness estimates which work only in the more seismically active regions, the PMC method allows for calculation of completeness based solely on properties of the network (not the seismicity). Using phase data, station information, and the network-specific attenuation relation, we produce estimates of recording completeness even for seismically less-active regions such as continental interiors. Many areas of research rely on estimates of detection completeness, particularly those involving statistical parameters of earthquake occurrence. We show for the first time detection probabilities for any given magnitude and completeness magnitudes for selected probability levels for a global network. The possibility of scenario computations with different sets of stations allow for design improvement of seismic networks. Simulations help estimating completeness drops due to random failures of stations and estimated increases in completeness resulting from new station installation.
S43D-1903
1997-2008: 11 years of European-Mediterranean Regional Centroid Moment Tensors and their dissemination
The European-Mediterranean Regional Centroid Moment Tensor (RCMT) Catalog collects seismic moment tensor solutions that have been routinely computed since 1997 for earthquakes with moderate magnitude (4.5 < M < 5.5) in the Mediterranean region. This database represents an extension to smaller magnitudes of the Global CMT catalog (http://www.globalcmt.org). RCMT computation is based on the analysis of seismograms recorded at regional distances, and modeling of intermediate period surface waves. The Catalog is regularly updated a few months behind real time, and reports are published in Phys. Earth Planet. Int. and on the worldwide web (http://www.bo.ingv.it/RCMT). However, moment tensors can also be computed on the basis of data available in quasi-real time. Such preliminary solutions are available within one or a few hours after an earthquake occurs, and published as a Quick RCMTs. The European- Mediterranean RCMT Catalog currently includes more than 850 seismic moment tensor solutions, mainly located in the most seismically active areas, such as northern Greece and the Hellenic trench. In the consideration of the size of the Catalog, we developed a web-search interface that allows to query the SQL database on geographical, time and magnitudes (mb, Ms and Mw) ranges. The results can then be formatted and exported in various formats, or mapped directly on the web page (beach balls). Two important flags were introduced, allowing us to include in the database new events still in the preliminary stages of analysis or in need of further refinements. The first flag distinguishes Definitive and Quick solutions, while the second one categorizes events in 4 Quality levels on the basis of changes of centroid coordinates or when the depth needs to be fixed or when the double couple component is too big (Pondrelli et al., PEPI, 2006).
S43D-1904
Quality Control on the IBERARRY broad-band seismic network
Systematic quality control of the seismic recordings acquired by the IBERARRAY broad-band seismic network is carried out. As part of the standard quality control procedure of the raw seismic data, the background noise power spectral density (PSD) is regularly estimated for all the stations of the IberArray portable seismic network and is statistically analyzed to compute probability density functions (PDFs) using the PQLX software package. These PDFs provide a useful tool for managing the network, as they allow to identify stations with unacceptable high noise levels in the frequency band of interest as well as temporal changes of the noise level that may indicate the convenience of changing the location of some sites. At long periods (20-120s), the vertical components usually lie 15db above the NLNM of Peterson (1993). The horizontal components are much noisier in this frequency range, often depassing the NHNM for the longest periods. At microseismic frequencies (0.05 – 0.3 Hz), the noise level is very similar between all the stations, while at high frequencies (> 1 Hz), the main contribution seems to arise from the cultural noise and therefore produces significant variations between the stations. Among the different features observed in the PDF curves, we can highlight the day/night differences in the mean noise level, specially significant for high frequencies, the importance of the local site effects, illustrated by two stations located less than 100 km away but lying in very different terrains and the observation of noise variations related to weather conditions in the microseismic band.
S43D-1905
Results from IMS Seismic Array in Niger
The International Monitoring System (IMS) of the Preparatory Commission of the Comprehensive Nuclear Test-Ban Treaty Organization (CTBTO) has built a sixteen element broadband seismic array in southwestern Niger near the town of Torodi. This is the first array to be built in West Africa. The array has a diameter of 6 km (three rings with a central element), contains twelve vertical and four 3-component broadband sensors, all with identical instrument responses (Guralp CMG-3TB broadband sensors, flat to velocity from 100 seconds to 40 Hz). Many elements have seismic noise characteristics at or below Peterson's Low Noise Model at greater than 0.5 Hz. All of the sensors are emplaced in 50 m boreholes in crystalline rock. The array is used to examine the North African Craton Structure and Seismicity. Receiver functions show that the crust is fairly simple with a Moho depth of about 38 km and Upper Mantle Discontinuities at 410 and 660 km. Because this is one of the few seismic arrays close to the Equator (array is centered at 13.2 N) and because its distance to Tonga-Fiji-New Zealand is 140-160 degrees, it could be used to look at temporal and spatial (north-south) variations in core phases. It is also used to look at Solid Earth Tides and long period diurnal signals. The array is capable of resolving and detecting events with magnitude (mb) less than 3.0 from events in Peru and Indonesia, showing that the array significantly adds to the IMS detection capabilities in the Africa region and world-wide.
S43D-1906
Peru Subduction Zone Seismic Experiment (PeruSZE): Preliminary Results From a Seismic Network Between Mollendo and Lake Titicaca, Peru.
This work describes preliminary results from a 50 station broadband seismic network recently installed from the coast to the high Andes in Peru. UCLA's Center for Embedded Network Sensing (CENS) and Caltech's Tectonic Observatory are collaborating with the IRD (French L'Institut de Recherche pour le Developpement) and the Institute of Geophysics, in Lima Peru in a broadband seismic experiment that will study the transition from steep to shallow slab subduction. The currently installed line has stations located above the steep subduction zone at a spacing of about 6 km. In 2009 we plan to install a line of 50 stations north from this line along the crest of the Andes, crossing the transition from steep to shallow subduction. A further line from the end of that line back to the coast, completing a U shaped array, is in the planning phase. The network is wirelessly linked using multi-hop network software designed by computer scientists in CENS in which data is transmitted from station to station, and collected at Internet drops, from where it is transmitted over the Internet to CENS each night. The instrument installation in Peru is almost finished and we have been receiving data daily from 10 stations (out of total 50) since June 2008. The rest are recording on-site while the RF network is being completed. The software system provides dynamic link quality based routing, reliable data delivery, and a disruption tolerant shell interface for managing the system from UCLA without the need to travel to Peru. The near real-time data delivery also allows immediate detection of any problems at the sites. We are building a seismic data and GPS quality control toolset that would greatly minimize the station's downtime by alerting the users of any possible problems.
S43D-1907
Recent Studies on the increase in Seismicity in the Antarctic Plate: Observations from BB Seismological Observatory (MAIT) at Maitri, Antarctica
The permanent Seismological Observatory was established in 1997 at Maitri in Central Dronning Maud Land, East Antarctica (70 °45' South 11 °43' East) primarily to monitor the seismicity in and around Antarctica, the space and time distribution of earthquake occurrences and obtain hypocentral parameters, magnitudes of earthquakes, velocity inversion for underground structure and earthquake source mechanism. The observatory has been upgraded during 25th Indian Silver Jubilee Scientific Expedition to Antarctica (December 2005 to February 2007) and 26th Indian Antarctic Expedition (IAE) with the new generation Geotech KS-2000M Seismometer and Smart 24R digitizer. During the 27th IAE the Seismic Observatory was further upgraded by adding Reftek 130 seismic system. Uninterrupted good quality digital Broad Band Seismic data is continuously being acquired. The SEISAN 8.1 software was used for final processing and analysis of about 300 earthquakes recorded. During the year 2006 the tele-seismic events, and quite a number of regional earthquakes of the order of 4 to 6.0 magnitude within Antarctic Plate, 23 in South Sandwich Islands, 7 in Scotia Sea, 2 in Macqurie Islands and 23 in Mid Oceanic Ridges in the Indian Ocean were recorded. 48 earthquakes of the magnitude above 4.5 from the nearby South Indian Ocean, South of South Africa, Chile, Argentina, Bolivia and about 40 earthquakes of the magnitude above 5.0 from the Indonesian Region were analysed. An earthquake of magnitude Ms=7.3 from the seismically active region of South Sandwich Islands Δ =16.5 °, Mb=7.8 earthquake from Tonga Islands and Mb=7.2 earthquake from Java were the large earthquakes that were recorded. Along with this the MOHO depth beneath MAIT was also estimated to be about 40km using receiver function analysis. All the analysed monthly data was reported to the I.S.C., U.K.,Global Data Centre for the final processing and inclusion in the yearly ISC Seismic Bulletin. The increasing seismic activity in and around Antarctic plate and along the oceanic ridges in the Indian Ocean confirms the emerging deforming zone between 75 ° East and 100 ° East longitude estimated by IERS in their plate characteristics in the Indian Ocean between India and Antarctica. All the results give an insight into the spreading rates of the ridges and reorganization of Plate Boundaries.
S43D-1908
New Developments Of The Geoscope Program
The GEOSCOPE observatory consists of a global seismic network and a data center. The observatory was
launched in 1982 by the French National Center of Scientific Research (CNRS/INSU) and progressively 30
stations have been installed across all continents and on islands throughout the oceans. The GEOSCOPE
stations are located on 18 countries and equipped with three component very broadband seismometers
(STS1 or STS2) and 24 or 26 bit digitizers, as required by the Federation of Seismic Digital Network (FDSN).
In most stations, a pressure gauge and a thermometer are also installed. In 2008, 17 stations send data in
real or near real time to GEOSCOPE Data Center.
In 2008, two stations have been upgraded and send real time data: UNM in Mexico and HDC in Costa Rica.
Two new stations have also been installed: CLF in France and FOMA south of Madagascar. In 2009, we plan
to upgrade stations PPT in French Polynesia, COYC in Chile, MPG in Guyana and to install a new station in
Rodrigues island. We also plan to progressively replace original STS1 electronics with Metrozet systems. In
the same time, we will reinstall all STS1 with warpless base plate systems in order to minimize the effects of
atmospheric pressure variations thus to improve signal noise ratio.
Continuous data of all stations are collected in real time or with a delay by the GEOSCOPE Data Center in
Paris where they are validated, stored and made accessible to the international scientific community.
Users have free and open access to:
- Real time data coming from 17 stations to Geoscope Data Center using the seedlink protocol developed by
GEOFON (GFZ, Germany). Seedlink also enables to make these data accessible in real time to Tsunami
Warning Centers and to other data centers. These data are also available to users through the GEOSCOPE
web interface.
- Validated continuous waveforms and meta data of all stations by using the NetDC system (Networked Data
Centers). Data can be requested from GEOSCOPE Data Center and from other networked centers
associated to the FDSN.
- A selection of seismograms corresponding to large earthquakes via a web interface
- The power spectrum estimates of the seismic noise averaged over sequences of 24 hours for each station.
The noise level of the last 10 years of continuous data has been computed and is accessible via the web.
The noise level of real time data is computed at day-8.
GEOSCOPE data center is now networked to the French virtual data center, FOSFORE, in order to give a
unique access to French seismological data. In Europe, GEOSCOPE data center participates in NERIES
project (NA3 activity) in order to create a distributed archive and database for all continuous digital waveform
recordings of the Euro-Med region.
http://geoscope.ipgp.fr
S43D-1909
FOSFORE: portal for distributions of French seismological data
FOSFORE (Fédération de l'Observation Sismologique Française) is a collaborative inter-university project aimed to distribute data of French permanent and temporary seismic networks and to make them freely available for the scientific community. FOSFORE is designed as a distributed data archive and is organized around four data centers located in universities at Grenoble, Nice, Paris, and Strasbourg. Currently available data include records of the GEOSCOPE global seismic network (Paris data center), of the French broadband seismic network (Nice and Strasbourg data centers), of the French accelerometric permanent network RAP (Grenoble data center), of numerous French temporary seismic deployments from different regions in the World (Grenoble data center), and of regional monitoring networks in France. The four data centers are connected through centralized Web portal (http://www.fosfore.ipgp.fr) where waveform data are available via NetDC data request protocol. In addition, the data are available directly in the data centers via NetDC or via AutoDRM request systems or via Web interfaces. In the near future, we plan to incorporate more temporary deployments as well as data from volcano-monitoring networks in La Reunion Island and French Antilles and data from the North Chile seismic experiment.
S43D-1910
IISEE's CMTs, Aftershock Distributions, Fault planes, and Rupture processes for large earthquakes in the world (1994-2004)
The IISEE started a project to determine the following earthquake information for large (Mw>=7.2) earthquakes in the world between 1994 and October, 2004 by the analytical techniques developed by the IISEE staff members and visiting researchers: centroid moment tensors, aftershock distributions, corresponding fault planes, and rupture processes. Centroid moment tensors are determined analyzing long period body wave data recorded at GSN (Global Seismological Network) stations by the grid search approach by Hara (Hara, 2004, Earth Planets Space, 56, 307-310; Hara, 2005, Earth Planets Space, 57, 179-183). Aftershocks including mainshock and foreshocks are relocated using P-wave arrivals from International Seismological Centre (ISC) CD-ROMs by the modified joint hypocenter determination method (MJHD. Hurukawa and Imoto, 1992, Geophys. J. Int., 109, 639-652; Hurukawa, 1995, Geophys. Res. Lett., 22, 3159- 3162; Hurukawa, Popa, and Radulian, 2008, Earth Planets Space, 60, 565-572). Then, the corresponding fault plane is determined based on the obtained aftershock distribution (nodal planes are taken from the Global CMT catalog). Earthquake rupture processes are determined analyzing broadband waveform data recorded at GSN (Global Seismological Network) stations following Yagi and Fukahata (2008, Geophys. J. Int. in press), who developed a waveform inversion technique considering covariance components in analyses of densely sampled observed data. In addition, we developed software to perform strong ground motion simulation for seismic bedrock using stochastic Greenfs function method, and performed simulations for some damaging earthquakes. We have made a database consisting of the above earthquake information and other information such as the Global CMT catalog, and have developed a web interface by which the above database can be searched, displayed and compared. We are going to make this interface open to public in September, 2008. We plan to continuously update this database. Also, we plan to perform further analyses. The one is multiple CMT analyses based on the technique of Hara (1997, Geophys. J. Int., 130, 251-256). The other is waveform inversion for rupture process using highly accurate synthetic seismograms computed by a newly developed finite difference method (Mizutani et al., 2008, General assembly of Asian Seismological Commission, abstract) with consideration of complex 3-D structure in source regions.
S43D-1911
Regional moment tensor determination in the Korean Peninsula
The number of broadband seismic stations in the Korean Peninsula has been continuously increased since 1995. This broadband seismic network provides high quality waveform data and makes it possible to determine source parameters of small to moderate sized earthquakes in and around the Korean Peninsula, which are too small to be studied using teleseismic waveforms. We analyzed 31 earthquakes ranging in size from ML=3.5 to 5.1 for the eight-year period 2001-2008. The Green's functions are computed using frequency-wavenumber integration method and 1D velocity structure is obtained from a previous study based on regional broadband waveforms and travel times. In order to determine moment tensor solutions and source depths, we used time domain moment tensor (TDMT) inversion technique, which is widely used in many regions (e.g., Northern California and Japan). Our preliminary result shows that strike-slip mechanism is dominant and it is consistent with several previous results based on P wave first motions and the principal strain rate measured by GPS in the same region.
S43D-1912
Receiver Functions Under the Provence, France: Resolution Ability of LSBB Low-Noise Underground Laboratory Seismic Network
We use the data of the LSBB laboratory (http://lsbb.unice.fr) that is located in the Provence region of France, to the north of Aix-en-Provence. The site was a launch control centre for the French strategic nuclear defense. LSBB is emplaced up to 500 meters underground. It includes more than 3.7 km of galleries. The underground seismic array is made up of six broad-band STS2 sensors that allow cross-comparison with measurements taken at the surface. This study makes use of the favorable measurements conditions (low noise) at the LSBB for determining receiver function images of the subsurface under the Provence. We also compare seismograms recorded near the free surface and those under 500 meters overburden, to demonstrate the usefulness of measurements at depth. Using frequency domain as well as iterative time- domain deconvolutions, we have obtained the following results: 1) depth and dip of the Moho are respectively 31 km and 10° toward the North-East. 2) We also detected converted PS waves from the transition zone interfaces. The P410S conversion is well-resolved whereas the P660S does not appear clearly. 3) Back-azimuthal variations and energy on the transverse component are observed for the P410S conversion. These variations are coherent with the measurements of shear-wave splitting of SKS waves realized in this region. We also used receiver functions as a tool to obtain the depth of an anisotropic layer.
S43D-1913
SPECTRAL CHARACTERISTICS OF NOISE IN BROADBAND STATIONS OF KANDILLI OBSERVATORY AND EARTHQUAKE RESEARCH INSTITUE
In this study, the seismic noise levels of the Broadband Stations operated by Bogazici University Kandilli Observatory and Earthquake Research Institue, in Turkey are investigated for periods ranging from 0.01 to 100 sec. The data are selected to reflect different conditions including seasonal and daily variations. The method which was applied consisted of first removing the instrument response to obtain raw ground noise data in the specified frequency range. The digital signal processing techniques were applied to obtain the power density spectra of the noise in various conditions. Namely data is segmented in large number of windows whose spectra are calculated individually and their average is found as final estimation of the spectrum of total data. The size of the windows are selected according to the trade off which exists between the resolution and the stability of the final result. Noise spectra are calculated for different stations and different components, during various stations. Identifying common behavior at all stations draws some general conclusions. In parallel to this common behavior, some stations displayed individual differences, which are related to the particularities of the physical conditions at that station. Obtained seismic noise level is affected from human activity and/or daily temperature variations in the 0.1 and 1 sec. period range. The seismic noise level was higher during the day than during the night at this period both in horizontal and vertical components. There is not observed daily variations for periods longer than 1 sec. There is no seasonal variations of short-period noise (<1 sec). In general the noise level at high periods (>10 sec) is less in vertical component compared to horizontal ones. This is clearly seen in full Broadband Stations.
S43D-1914
Seismic noise level variation in South Korea
The variations of seismic background noise in South Korea have been investigated by means of power spectral analysis. The Korea Institute of Geoscience and Mineral Resources (KIGAM) and the Korea Meteorological Administation (KMA) have national wide seismic networks in South Korea, and, in the end of 2007, there are 30 broadband stations which have been operating for more than a year. In this study, we have estimated the power spectral density of seismic noise for 30 broadband stations from 2005 to 2007. Since we estimate PSDs from a large dataset of continuous waveform in this study, a robust PSD estimate of McNamara and Buland (2004) is used. In the frequency range 1-5 Hz, the diurnal variations of noise are observed at most of stations, which are especially larger at coastal stations and at insular than at inland. Some stations shows daily difference of diurnal variations, which represents that cultural activities contribute to the noise level of a station. The variation of number of triggered stations, however, shows that cultural noise has little influence on the detection capability of seismic network in South Korea. Seasonal variations are observed well in the range 0.1-0.5 Hz, while much less found in the frequency range 1-5 Hz. We observed that strong peaks in the range 0.1-0.5 Hz occur at the summer when Pacific typhoons are close to the Korean Peninsula.
S43D-1915
Probabilistic Seismic Network Completeness Studies Around the World
An important characteristic of any seismic network is its detection completeness, which should be considered
a function of space and time. Many researchers rely on robust estimates of detection completeness,
especially when investigating statistical parameters of earthquake occurrence.
We present the Probability-based Magnitude of Completeness (PMC) method for computing the spatial
variation and temporal evolution of detection capability of seismic networks based on empirical data only:
phase data, station information, and the network specific attenuation relation.
We present studies of regional networks from California, Switzerland, Italy, Japan, and compare the result
with estimated completeness levels of other methods. We report on the time evolution of monitoring
completeness in these regions and show the depth dependence of detection probabilities. Scenario
computations show the impact of different possible network failures and offer estimates of possible network
optimization strategies. All presented results are published on the CompletenessWeb
(www.completenessweb.org) from which the user can download completeness data from all investigated
regions, software codes for reproducing the results, and publication-ready and customizable figures.
http://www.completenessweb.org
S43D-1916
Nostradamus: the Radar that wonted be a Seismometer
Surface waves emitted after large earthquakes are known to induce, by dynamic coupling, atmospheric infrasonic waves propagating upward through the neutral and ionized atmosphere. Those waves have been detected in the past at ionospheric heights using a variety of techniques, such as HF Doppler sounding or GPS receivers. The HF Doppler technique, particularly sensitive to the ionospheric signature of Rayleigh waves is used here to show ionospheric perturbations consistent with the propagation of Rayleigh wave phases R1 and R2 following the Sumatra Earthquakes on the 28 March 2005. This is in our knowledge the first time the the phase R2 is detected by ionospheric sounding. In addition, we prove here that the ionospheric signature of R2 is also observed by over-the-horizon (OTH) Radar. This latter was never used before to detect seismic signature in the ionosphere. Adding the OTH Radar to the list of the "ionospheric seismometers" we discuss and compare the performances of the different instruments mentioned above.
S43D-1917
Redesign of the Electronics and the Mechanical Sensor of the STS-1 Very Broadband Seismometer
The STS-1 VBB, widely viewed as the finest VBB sensor in the world, is currently the principal broad-band seismometer used by the Global Seismographic Network (GSN), GEOSCOPE, and several other global or regional seismic networks. Its continued operation is critical to future, fundamental research in a number of important disciplines within seismology. These include modal studies of the earth, the determination of source processes of very large earthquakes, and tsunami warning. We have recently completed the development of a replacement electronics module for the Streckeisen STS-1 Very Broad-Band (VBB) seismometer. This module maintains the outstanding analog performance of the original sensor electronics, while providing a number of unique performance enhancements that will improve the operability of STS-1 sensors within a modern, digital seismic network. We describe the attributes of this new electronics module, which has now reached production stage. As a second step in this collaboration between the Berkeley Seismological Laboratory and Metrozet (Inc), we have started the development of a commercially-viable replacement to the aging, but state-of-the-art mechanical sensor. We discuss the goals of this new project. The design of the new sensor aims at maintaining many of the unique features that have made the original sensor the world's finest instrument for low frequency seismic recording. It will, however, implement a number of enhancements that promise to improve operating performance and ease-of-use. These include the use of capacitive position sensing to reduce high frequency self-noise, an increase in the upper corner frequency of the sensor, much tighter tolerance (sensor-to-sensor) in scalar response values, and a new, highly-integrated package for deploying the sensors as a standard triaxial set. Importantly, the replacement sensor development is driven by principles of "design-for manufacturing". As such, the new sensor will include a number of features that will allow a modern, "deterministic manufacturing process" to replace what was historically an iterative, "workshop" approach to manufacture the STS-1. This is crucial to reducing the cost, and manufacturing lead time, of the new sensor.
S43D-1918
First step for mobile ocean bottom broadband seismic observation of the next generation
Since 1999, we have already developed the mobile broadband ocean bottom seismometer (BBOBS), and many practical observations in the northwestern Pacific Ocean and French Polynesian Sea have been conducted. But, through the evaluation of the seismic data, the noise level of horizontal components in long periods, those are important in data analyses, is rather high in average and its variation in time is also large. The reason of this high noise level is assumed as the small tilt variation of the large housing sphere due to a bottom current with time variation. To clear this problem, one idea of observation without the tilt variation is the use of small and low profile broadband sensor that enables to intrude into the sediment. Now, we are on the way to develop this next generation BBOBS (BBOBS-NX). Its final goal is the free fall deployment and self pop-up recovery system as same as the BBOBS in present. In June 2008, we made a first step for it during the cruise of the R/V Kairei (JAMSTEC) with the ROV, Kaiko 7000II. It is a practical test to know achievement of the noise reduction with the idea above under support of the ROV. The site of this observation was chosen as same position where we have conducted three-year BBOBS observation since 2005 for the stagnant slab project. Because, it makes us easy to compare the data with that of the BBOBS, and we can also evaluate the difference of noise level by different methods to deploy if they were deployed nearby and collect the data at same time. We put an ocean bottom doppler current meter (OBDC) to know relation of the noise level and strength of the bottom current, too. The prototype of BBOBS-NX was loaded on the ROV, and was located nearby the BBOBS that has already deployed in 2007. The sensor unit of BBOBS-NX was separated from the recording unit, and the sensor was pushed into the sediment. The OBDC had been dropped before the dive of the ROV, and was moved to nearby of them. The distance from the BBOBS to the BBOBS-NX is about 10m. We will recover these instruments several months later to get enough length of data to evaluate correctly.
S43D-1919
Connecting a Quanterra Data Logger Q330 on the GWR C021 Superconducting Gravimeter for low Frequency Seismology
Reference instrumentation such as a superconducting gravimeter (SG), barometer, and absolute traceability
of calibration and orientation are important components in modern networks. Array processing of seismic
data is not related only to program computers: somebody has to pay attention to the measurement physics,
operational accuracy and meaning of the acquired seismic data.
SGs, benefiting from a calibration at the 0.1% level in amplitude and 0.01 s in phase, can play an important
role for improved estimation of source parameters, in particular the magnitude of large earthquakes, as well
as for investigating Earth's gravest free oscillations.
Equally clearly, since the Q330 is a datalogger used not only by the largest open array in the world, the
USArray Transportable Array and by the largest aperture array, the GSN, and in numerous smaller-scale
deployments, the operation of core instrumentation with various types of sensors is a crucial design element
of future instrumentation.
In order to promote the SG data among the seismic community, a Q330 data logger was connected to the SG
GWR C021 at the Membach station, Belgium, for at least two reasons: (1) the Q330 provides the data in the
quite complex SEED (Standard for the Exchange of Earthquake Data) format directly and (2) the Q330 is a
standard data system used for acquisition of continuous seismological data. So, integration of SG data into
the global data distribution system using a Q330 and testing this data logger to monitor very low frequency
signal from an SG become straightforward.
This paper presents the solutions provided to optimize the Q330 data logger when connected to a SG to
ensure a reliable DC level and a stable calibration factor (at the 0.1% level). The noise contribution of the
data acquisition systems is below the noise affecting the SG in the 10-5 -0.01 Hz frequency band, which
includes the tidal and seismic free oscillation frequencies. The Q330 data logger is also a valuable tool to
determine easily the transfer function of the SG.
S43D-1920
An Optical Seismometer Without Feedback
We are developing a new seismometer, motivated by a desire to have an instrument whose performance rivals that of the best observatory sensors but which can operate within a borehole without electronics. This has led us to develop an optical seismometer consisting of a spring-suspended mass whose position is monitored interferometrically. We use a Michelson interferometer illuminated with a 1 mW laser that can be linked to the seismometer with optical fibers only. A digital signal processor (DSP) samples the fringe signal and produces a displacement record of the seismometer mass with a resolution of less than 1 pm (10-12 m). The maximum mass motion is limited by practical issues to about 10 mm, providing a dynamic range of 1010, equivalent to 33 bits. For a moderate size mass, readily achievable free period, and damping, the mass-spring sensor's fundamental thermal noise is less than ambient noise at the seismically quietest sites. Hence there is no need to abandon this basic, simple, mass-spring design. Elimination of electronics, however, means elimination of force feedback -- the paradigm in seismometry for the past several decades. We have tested our non-fedback, optical seismometer against a standard fedback observatory quality seismometer (STS-1) and found that it provides equivalent seismograms for signals ranging in frequency from tides to at least 10 Hz.
S43D-1921
The new Hamburg Ocean-Bottom-Tiltmeter (OBT): A first deployment at Columbo Seamount (Aegean Sea, Greece)
Assessing the state of volcanic activity of seamounts is quite more complicated than for onshore volcanoes, due to the difficulty of deploying instruments. At land, various techniques are applied (seismic networks, deformation studies, gas measurements etc.). At sea, mainly seismological observations are used. However, especially onshore deformation studies using INSAR have proven to be valuable in determining recharge of magmatic systems. We therefore developed a free fall, self leveling Ocean Bottom Tiltmeter (OBT) to observe deformation on the seafloor, using a two component high resolution tilt sensor with a resolution of about 1nrad (0.15μ°) and a maximum signal of about 0.045rad (0.5°). It is mounted inside a 17~inch glass sphere on a levelling stage, which relevels the instrument between ± 5° down to an accuracy of 0.006°. During the measurement this leveling stage is standing on the bottom of the glass sphere. For releveling, the instrument is pulled up by thin nylon strings and then locked to a gimbal system in order to compensate for tilt. This releveling procedure is done once every 48 hours. Data is recorded on an 18bit data logger at 50Hz sampling rate. Additionally to tilt and seismic signals (using a hydrophone), temperature, absolute pressure to measure uplift or subsidence, and orientation (electronic compass) are monitored. 4 OBT systems were deployed between June 2006 – March 2007 at Columbo seamount, a submarine volcano north-east of Santorini island, Aegean Sea, Greece, on a 3 km long profile perpendicular to the first principal stress axis σ_1 of the regional stressfield. Three of the instruments operated the whole time, one shut down due to a short circuit. First data processing indicates that small regional earthquakes as well as major tectonic earthquakes are properly recorded by the system. We find small, but permanent short- period deformations associated with local earthquakes and also observe long-period deformation processes occurring over a period of days. Additionally, subsidence of two stations relative to a third is observed with the absolute pressure gauges.
S43D-1922
An Introduction to SPEAR (Seismogram Picking Error from Analyst Review)
A grassroots initiative began in February of 2008 at the University of Texas at El Paso to understand how
seismologists measure earthquakes. The Seismogram Picking Error from Analyst Review (SPEAR) project is
designed to be a forum where seismologists can propose, discuss and experimentally test theories on proper
procedures to identify and measure seismic phases. We outline the history of seismogram analysis and
explore areas of seismogram analysis that still need to be defined. The main concern for SPEAR, at this time,
is the impact of picking errors produced by merging earthquake catalogs. Our initial effort has been to
establish a common data set for seismologists to pick. The preliminary studies from this data set have shown
that significant bias between authors of catalogs may exist. We provide techniques to ensure that these
biases can be identified and correctly managed to provide accurate mergers of earthquake measurements.
The overall goal of SPEAR is to provide a repository of information to aid seismologists in comparing and
sharing measurements. We want to document in the repository and explore all aspects of the picking
process, from the basics of learning how to read a seismogram to complex transformations and
enhancements of signals. Your participation in SPEAR will aid the seismological community to close the
knowledge gaps that exist in seismogram analysis.
http://www.geo.utep.edu/pub/SPEAR/SPEAR.html
S43D-1923
Correcting time offsets in seismic array data using noise correlation: Examples from a seismic array installed across Mexico
We are devising a new method to correct large offsets in time synchronization of seismic stations due to hardware faults using noise correlation. We focus on data collected during the Meso American Subduction Experiment: a 100 station 500 km linear seismic array deployed for 2 years across Central America. Up to 10% of the data is incorrectly timestamped with offsets ranging from a few seconds to a few thousand seconds. Our approach is to build a model of microseism propagation through the array using the noise correlation. The microseism energy coming from the Pacific travels north through the array with a dominant 6 second period while the energy coming from the Gulf of Mexico has a dominant 20 second period. To calculate a time correction for a given station we first calculate the group velocity between all stations using noise correlation analysis for long windows of data when the clocks are locked. We have observed that the lags for pairs of stations 50km or closer are shorter than expected resulting in a apparent increase in velocity. For pairs of station further than 50km, the velocities are as expected. We then correlate short time windows of data from a given station with that of distant stations which have the correct time, and compare the lag of the correlation peak with that predicted from the group velocity calculation. Unlocked clocks show up as differences between predicted and observed lags and can be corrected back to predicted values using the microseism propagation model.