Yahya Djamour; S. Tabibi; M. M. Hossainali
Abstract
The accuracy of GPS derived positions in short term measurements largely depends on the better modeling of residual errors which is normally reduced in long term measurements by averaging the obtained results. To increase the accuracy of the movements obtained through the analysis of high rate data in ...
Read More
The accuracy of GPS derived positions in short term measurements largely depends on the better modeling of residual errors which is normally reduced in long term measurements by averaging the obtained results. To increase the accuracy of the movements obtained through the analysis of high rate data in geophysics applications, systematic errors in the existing measurements in the corresponding frequency range should be reduced. Calibration techniques and the error reduction are based on the repeatability of the system constellation. For example, those errors affecting the resulting accuracy of the high rate positions in the time scales of 10-600s depend much on the constellation geometry of the GPS satellites and GPS stations. Since the satellite orbits are fixed, those errors are highly repeatable in time. This characteristic is the base for the development of sidereal filtering techniques for reducing this kind of errors. It is assumed that the repeatability occurs based on the nominal repetition period of the satellites; however, the true time of this repeatability varies even for each satellite. In this paper the mean period is estimated as 23h, 55m, and 55s and used for the noise reduction of time series of the relative displacement. By using this period, noises with frequencies less than 0.01Hz of high rate GPS positions have been reduced significantly. Accessible co-seismic displacements are reduced to less than 4mm in horizontal components and less than 10mm in the vertical components.
S. Tabibi; M. M. Hossainali; Yahya Djamour
Abstract
Better understanding of earthquakes primarily requires more accurate dynamic and kinematic models for fault rupture. There are several methods for ground motion detection; each of them has its own advantages and limitations. The processes, needed for the estimation of displacements by the seismic data, ...
Read More
Better understanding of earthquakes primarily requires more accurate dynamic and kinematic models for fault rupture. There are several methods for ground motion detection; each of them has its own advantages and limitations. The processes, needed for the estimation of displacements by the seismic data, generally increase the noise. Accelerometers, for example, record the details of strong ground motion close to the earthquake source; however it is difficult to transform the measured accelerations into displacement. Broadband seismometers are more sensitive and more accurate than accelerometers but even those may be saturated or clipped in far distances from a large earthquake. InSAR observations can provide good spatial images of some of the surface displacement components in the rupture area. It has, however, drawbacks in some regions, as the InSAR has no sufficient temporal resolution for the analysis of dynamic short period changes during an earthquake. Most of the GPS monitoring systems process the daily or hourly data in order to achieve the station coordinate with millimeter accuracy. But in warning systems, the temporal delay between the natural event and the act of warning must be the least as much as possible. Increasingly more continuous GPS receivers, established primarily for geophysical studies, are now running in seismic frequencies such as 1-Hz. GPS seismology is the unexpected result of the geodetic networks which at first were established to measure the deformation of plates and tectonic plate boundaries. A GPS receiver can accurately measure the movements in the geological time scales (i.e. 1 mm/yr) and that of seismology (i.e. 500 mm/sec). In this paper, the shape of the seismic waves, obtained from thirteen GPS stations, being in 36 to 74 kms of epicentral distances of San-Simeon Earthquake, 2003 are determined. The efficiency of the relative methods of positioning using high rate data has been analyzed; estimated co-seismic displacements have been validated using similar results obtained from the integration of seismic records.