A time-term method is a simple travel time inversion for seismic refraction crustal studies. Although this method is based on simple travel time calculation, it is robust in determining a basement velocity under the complicated surface structure. In the present paper, this method is extended to detect lateral velocity variation in the crystalline crust. We consider a model where a basement is composed of segments with a different velocity. Defining a likelihood function for unknown time-terms and refractor velocities, their best estimates are obtained by minimizing Akaike's Bayesian Information Criterion (ABIC). To realize accurate velocity estimation, we also incorporate an effect of vertical velocity gradient within the refractor, which was ignored in the conventional algorithms. Numerical tests undertaken for realistic crustal structures show that our method successfully detects lateral velocity variation in a basement even for a case with undulated surface layers. It is particularly stable and reliable if the surface layers are not thick and their velocities are more than 30-40% smaller than the refractor velocities. Then, the refractor velocities are almost decoupled to the time-terms in the process of inversion. The incorporation of the effect of the velocity gradient is inevitably important when travel time data at far distances are included in the inversion.

抄録全体を表示

We present a magnetostratigraphic record from the western Philippine Sea that is tied to a marine δ¹⁸O record for the past 2.14 million years. The ages of geomagnetic reversals were astronomically calibrated by tuning the oxygen isotopic stratigraphy, yielding a chronology for the following subchrons: Matuyama/Brunhes boundary, 781±3 ka (slightly above δ¹⁸O Stage 19.3); top of the Santa Rosa polarity interval, 920 ± 2 ka (Stage 23/24); base of the Santa Rosa polarity interval, 925 ± 1 ka (Stage 24); top of the Jaramillo subchron, 988 ± 3 ka (Stage 27); base of the Jaramillo subchron, 1072 ± 2 ka (Stage 31); top of the Cobb Mountain subchron, 1173 ± 4 ka (Stage 35/36); base of the Cobb Mountain subchron, 1185 ± 5 ka (Stage 36); top of the Olduvai subchron, 1778 ± 3 ka (Stage 63/64); base of the Olduvai subchron, 1945 ± 4 ka (Stage 71/72); top of the Réunion II subchron, 2118 ± 3 ka (Stage 80/81); and base of the Réunion II subchron, 2133 ± 5 ka (Stage 81). This astronomically calibrated chronology independently confirms the ages of major reversals in recently published astronomically calibrated polarity timescales for the Matuyama chron. It also provides the first astronomically calibrated dates for the lower and upper reversals associated with the Cobb Mountain and Santa Rosa polarity intervals, respectively.

抄録全体を表示

The EISCAT VHF radar (224 MHz) and the EISCAT dynasonde (1-10 MHz) were operated simultaneously during a Polar Mesosphere Summer Echoes (PMSE) event on 22 June 1994. We investigate the echo characteristics revealed by these two instruments. The antenna of the VHF radar was set 10° off-zenith to the north for both transmission and receiving, while the dynasonde was operated with vertical beam to record ionograms every 3 minutes. The Height-Time-Intensity (HTI) variation of the VHF radar PMSE is studied, which is characterized layered structures. The intensity profiles are quite intermittent which show similarity to those data recorded by the dynasonde. To study the ionospheric condition with the dynasonde, the quantities of signal-to-noise ratio, echo frequency, echo polarization, echolocation, and Doppler velocity are investigated. Variations in the quantities demonstrate that during the PMSE conditions, echoes of the dynasonde are mainly attributed by reflection of many small irregularities within the observed ionospheric volume.

抄録全体を表示

Four years of auroral latitude radar echo signal fading time data from the Tromsø MF radar compared with geomagnetic and solar indices yield strong evidence to suggest, although not prove, that upper mesospheric turbulence may be enhanced during periods of geomagnetic activity. We present a critique of hypotheses to explain such effects.

抄録全体を表示

The possible existence of a positron radiation belt in the inner magnetosphere of the Earth, its space location, flux, energy distribution and the ratio of e⁺/e^- fluxes is considered. The source of the positrons/electrons is assumed to be the decay of charged pions (π - μ - e decay chain) produced, in the nuclear interactions between protons and the neutral constituents of the atmosphere. The production of excess positron fluxes over electron ones through this process is examined for two different proton populations and their atmospheric interactions in the altitude range of 80 to 1000 km. Monte Carlo simulations of intra-nuclear cascade process through SHIELD code was utilized to simulate these interactions. Considering the trapped proton fluxes in the inner magnetosphere as a source, the simulations for interactions in rarified atmosphere shows excess of positrons over electrons with ratios above 2 for proton energies below 2.5 GeV. Although, protons above this energy do not produce excess of positrons compared to electrons in atmospheric interactions, we assumed the primary cosmic rays with the energy greater than 8-10 GeV also as a source for these interactions, and utilizing the east-west asymmetry in the arrival directions of these primary cosmic rays combined with the exponential nature of the atmosphere density we also obtained an excess of positrons over electrons escaping from the atmosphere to the altitudes of satellite orbits. A comparison is attempted with the recent data of AMS experiment on board the space shuttle.

抄録全体を表示

We study two important aspects of the waveform moment tensor inversion for the shallow earthquakes in the subduction region: the effect of the intense lateral inhomogeneity in the structure, and the strategy to invert the waveform data for the focal mechanisms. For the first aspect, using a “forward” finite difference modeling, we demonstrate that the effect of the inhomogeneity is quite large on the surface waves with a period of about 20s, and the current knowledge on the subduction region structure is practically effective in reproducing the characteristics in the observed waveforms. For the second aspect, we develop a “reciprocal” moment tensor inversion method that can generate the Green's functions for a large quantity of source locations (19, 200 in this study) in a realistic inhomogeneous structure by only three finite difference calculations per a single station. The inversion with a grid search scheme result in a reasonable source location, moment tensor and fit of the waveforms using data from only two stations. The constraint on the epicenter in the “transverse” direction is found to be somewhat weak in the case of single-station inversions, but the two-station inversion improves the constraint.

抄録全体を表示

To solve for the induced electromagnetic fields in a conductive medium the quasi-static Maxwell's equations may be reduced to a second order elliptic system, formulated in terms of either the electric or magnetic vector fields. We show with 1-D and 3-D numerical experiments that solutions obtained from equations formulated in terms of the electric fields are less sensitive to grid resolution than those obtained from the magnetic formulation. On a fine enough mesh, solutions from both approaches are nearly identical, while on coarser meshes solutions form the electric field formulation tend to be closer to exact solutions (where available), or fully converged fine mesh solutions.

抄録全体を表示

The radially dependent mean free path of solar energetic particles was calculated by considering the radially dependent magnetic field fluctuation, its correlation length and the variation of magnetic field strength along an Archimedean interplanetary magnetic field. A longer mean free path can be deduced as radial distance approaches the sun. That is, the mean free path at 0.1 AU is increased by 10 times for some typical values of radial dependence of magnetic field fluctuation (e.g., power index of radial dependence = -2), compared to the value at 1 AU. The focused pitch angle transport equation without adiabatic deceleration was solved for the obtained mean free path. The calculated time profiles are compared with observations of solar energetic particles. In some cases, very good matches to observed data for longer periods are obtained with shorter mean free paths than the original authors mentioned.

抄録全体を表示