We have investigated how the automated moment tensor determination (AMTD) system using the FREESIA/KIBAN broadband network is likely to behave during a future large earthquake. Because we do not have enough experience with a large (M>8) nearby earthquake, we computed syntheticwaveforms for such an event by assuming the geometrical configuration of the anticipated Tokai earthquake and several fault rupture scenarios. Using this synthetic data set, we examined the behavior of the AMTD system to learn how to prepare for such an event. For our synthetic Tokai event data we assume its focal mechanism, fault dimension, and scalar seismic moment. We also assume a circular rupture propagation with constant rupture velocity and dislocation rise time. Both uniform and heterogeneous slip models are tested. The results show that performance depends on both the hypocentral location (i.e. unilateral vs. bilateral) and the degree of heterogeneity of slip. In the tests that we have performed the rupture directivity appears to be more important than slip heterogeneity. We find that for such large earthquakes it is necessary to use stations at distances greater than 600 km and frequencies between 0.005 to 0.02 Hz to maintain a point-source assumption and to recover the full scalar seismic moment and radiation pattern. In order to confirm the result of the synthetic test, we have analyzed the 1993 Hokkaido Nansei-oki (M_J7.8) and the 1995 Kobe (M_J7.2) earthquakes by using observed broadband waveforms. For the Kobe earthquake we successfully recovered the moment tensor by using the routinely used frequency band (0.01-0.05 Hz displacements). However, we failed to estimate a correct solution for the Hokkaido Nansei-oki earthquake by using the same routine frequency band. In this case, we had to use the frequencies between 0.005 to 0.02 Hz to recover the moment tensor, confirming the validity of the synthetic test result for the Tokai earthquake.

View full abstract

The dispersion characteristics of group velocity of Love waves are measured across a dent in the continental crust that has a maximum thickness of about 50 km; the numerical modeling analyses are performed using the finite difference method. Measurement of the group velocity over the entire width of the mountain root displays several distinct dispersions. The group velocities for the crustal dent structure are lower than those for the stratified medium with a maximum crustal thickness of the dent at short periods of 20-30 s. The period range becomes longer as the sloping angle of the dent increases or as the crustal thickness of the dent increases. The period indicating a group velocity minimum for the crustal dent structure is shorter than that for the stratified medium with a maximum crustal thickness of the dent. An example of the interpretation of observational data is shown. In the example, the group velocities for the propagation path over the Tibetan Plateau have properties concordant with the above dispersion characteristics.

View full abstract

The 21 September 1999 Chi-Chi, Taiwan, earthquake of magnitude MW = 7.6 (M_L = 7.3) severely deformed the Earth's crust in the central Taiwan region. The earthquake created an 85-km-long surface rupture along the Chelungpu fault. The epicenter was located at 23.85°N, 120.81°E, near the southern end of the rupture zone. Three-dimensional displacements of 285 geodetic control stations were determined in this study from Global Positioning System (GPS) observations collected before and after the earthquake. The detailed surface displacement field shows that individual stations are vertically uplifted by up to 4 m and displaced horizontally by up to 9 m, with the largest displacement occurring near the northern end of the ruptured thrust fault. The azimuth of the surface displacement field is approximately parallel to the direction of tectonic convergence of the Eurasian and Philippine Sea plates. The maximum three-dimensional displacement of 9.9 m is among the largest fault movements ever measured for modern earthquakes.

View full abstract

Many active plate boundaries, such as mid-ocean ridges and trenches, are under the sea. Seafloor crustal deformation data will contribute significantly to understanding the nature of the tectonic processes at these plate boundaries. We have developed a seafloor positioning system with a GPS-acoustic link. This system consists of two main components; (1) the surface positioning by differential GPS to on-land reference and (2) the precise acoustic ranging using the M-sequence between the surface and seafloor references. The position and attitude of the surface GPS-acoustic link unit are determined from the GPS observations. Simultaneously, the acoustic ranging between the surface unit and seafloor references are carried out. The positions of the seafloor references are determined from these observations and a sound-speed structure model of the seawater. We performed preliminary seafloor positioning experiments. In these experiments, simple 1-D structure models are assumed for the sound-speed in the sea. The results show that the positions of the seafloor references are estimated with an accuracy on order of 10 cm. The residuals for acoustic ranging imply that there are systematic differences between the assumed and real sound-speed structure. It is necessary to reduce the uncertainties of sound-speed structures for more accurate positioning.

View full abstract

The Utror Volcanic Formation forms a NE-SW belt with in Kohistan island arc, which lies between the Indian and Eurasian continents in the western Himalayas of northern Pakistan. The Utror Volcanic Formation formed during Late Paleocene, when Kohistan existed as an Andean-type arc on the southern margin of Eurasia. Five to ten block samples were collected from 17 sites of the formation for paleomagnetic studies. Magnetic minerals that serve as remanent carriers are maghemite, magnetite, hematite and titanohematite. Magnetite, hematite, and titanohematite carry the characteristic remanent magnetization (ChRM). The declination values of ChRM are highly discordant before and after structural correction. While inclination values show uniformity in geographic coordinates. The ChRM carried by magnetite yields downward inclinations, whereas the ChRM carried by hematite or titanohematite have upward inclinations in geographic coordinates. The correlation between polarity state and magnetic mineralogy suggests that the acquisition of magnetization occurred during two distinct time intervals. An inclination only fold test of ChRM of these two mineral assemblages indicates that they are post-folding magnetizations. The calculated paleolatitudes for ChRM carried by magnetite and hematite or titanohematite are 9±4°N and 13±4°N respectively. A comparison of these paleolatitudes with the Indian apparent polar wander path (APWP) shows that the remagnetization likely happened between 55 to 45 Ma. Discordant declinations indicate that these volcanics suffered local rotations after remagnetization event.

View full abstract

Paleomagnetic investigation was performed on 26 Potassium-Argon dated Pliocene to Pleistocene lava flows, collected at 34 sites from northern Hyogo, Southwest Japan. Most magnetic polarities were found to be coincident with the geomagnetic polarity timescale (GPTS), although two lava flows do not agree with the timescale. One of 2 lava flows incompatible with the GPTS, the Oliveine Andesite lava flow from the Oginosen volcano group dated at 0.92+/-0.05Ma, shows normal polarity. This finding is a possible evidence confirming the existence of the Santa Rosa Event proposed recently. Twenty-two paleomagnetic data sets were used to investigate paleosecular variation (PSV) having passed the following criteria: 1) characteristic components are obtained from four specimens or more for a lava flow, 2) α₉₅ of the lava flow mean direction is < 10° and 3) the virtual geomagnetic pole (VGP) latitude calculated is > 50° or < -50°. The mean VGP was calculated to be latitude = 85.5°N and longitude = 285.3°E (k = 34.8 and α₉₅ = 5.3°). The resultant angular standard deviation (ASD) with respect to the geographic pole was calculated to be 14.5° with the upper limit = 18.3° and the lower limit = 12.1°. The ASD value is almost equal to most existing results from the Japan Islands. Brunhes Chron PSV in Japan was estimated using the identical criteria for selection; 66 of 87 available data sets passed. The mean VGP was calculated to be latitude = 89.3°N and longitude = 14.4°E (k = 33.1 and α₉₅ = 3.1.). The resultant ASD with respect to the geographic pole was calculated to be 14.2° with the upper limit = 16.1° and the lower limit = 12.7°. The ASD values for the past 3 million years in northern Hyogo and for the Brunhes Chron in Japan are compatible with the global trend for the past 5 million years fitted by physical PSV Model F and phenomenological PSV Model G. This suggests that the PSV in Japan is not low compared with anomalously low PSV area in the Pacific Region.

View full abstract

The sensing of absolute precipitable water vapor (PW) by the Global Positioning System (GPS) and a Water Vapor Radiometer (WVR) is presented. The GPS approach requires a priori knowledge of the relationship between the weighted mean temperature of the atmosphere and surface temperature whose regression relationship is derived based on ten-year climatological data observed by radiosonde and surface meteorological instruments. Similarly, the WVR scheme needs a priori information of the relationship between sky brightness temperature and PW whose regression relationship is characterized based on the same set of climatological data. GPS-derived PW are compared with those observed by WVR and radiosondes. The GPS and WVR data were collected at the Taipei weather station of Taiwan Central Weather Bureau (CWB) from March 18 to 24, 1998. To obtain the estimates of absolute PW at the Taipei site, GPS data acquired from Tsukuba, Japan, at a distance of 2155 km from Taipei were utilized. It is found that GPS-derived PW agrees reasonably well with observations by the WVR and radiosondes. The average of GPS-derived PW is 3.38 cm with a standard deviation of 0.39 cm. The difference between the average GPS-derived and WVR-observed PW is 0.27 cm with a bias of -0.04 cm, while the difference between the average GPS-derived and radiosonde-observed PW is somewhat larger, 0.36 cm with a bias of -0.42 cm. These differences are larger than differences reported at higher latitudes in regions with lower average humidity.

View full abstract

Using a world-wide set of stations for 15 years, quantitative estimates of changes to equatorial spread-F (ESF) occurrence rates obtained from ionogram scalings, have been determined for a range of geomagnetic activity (GA) levels, as well as for four different levels of solar activity. Average occurrence rates were used as a reference. The percentage changes vary significantly depending on these subdivisions. For example for very high GA the inverse association is recorded by a change of -33% for R_z ≥ 150, and -10% for R_z < 50. Using data for 9 years for the equatorial station, Huancayo, these measurements of ESF, which indicate the presence of TIDs, have also been investigated by somewhat similar analyses. Additional parameters were used which involved the local times of GA, with the ESF being examined separately for occurrence pre-midnight (PM) and after-midnight (AM). Again the negative changes were most pronounced for high GA in R_z-max years (-21%). This result is for PM ESF for GA at a local time of 1700. There were increased ESF levels (+31%) for AM ESF in R_z-min years for high GA around 2300 LT. This additional knowledge of the influence of GA on ESF occurrence involving not only percentage changes, but these values for a range of parameter levels, may be useful if ever short-term forecasts are needed. There is some discussion on comparisons which can be made between ESF results obtained by coherent scatter from incoherent-scatter equipment and those obtained by ionosondes.

View full abstract