190 research outputs found
Accelerated pacific plate subduction following interplate thrust earthquakes at the Japan trench
No full textInterplate thrust earthquakes are usually followed by afterslips, and they let the fore-arc move slowly trenchward. However, we do not know if the subducting oceanic plate is accelerated landward after such earthquakes. The westward velocity of Global Positioning System (GPS) stations in NE Japan show gradient decreasing from east to west reflecting the E-W contractional strain built up by the inter-plate coupling. Here we show that such coupling significantly enhanced (∼1.5 times) after the 2003 Tokachi-Oki earthquake (Mw 8.0), Hokkaido, in the segments adjacent to the ruptured fault. The coupling seems to be further enhanced (∼3 times) after the 2011 Tohoku-Oki earthquake (Mw 9.0). It is unlikely that interplate friction suddenly increased over such a large region, and relatively strong pre-2003 coupling there would not allow such enhancements even if full coupling is attained. Hence they are attributable to the temporary acceleration of the Pacific Plate subduction. We propose a simple 2-dimensional model in which down-dip acceleration of the slab let the force balance rapidly recover promoted by a thin low-viscosity layer on the slab surface. The accelerated subduction would account for temporary activations of regional interplate seismicity after megathrust earthquakes
間隙流体圧の時間発展を含む摩擦強度と長期断層挙動との関係に関する研究
No full text京都大学0048新制・課程博士博士(理学)甲第15743号理博第3583号新制||理||1523(附属図書館)28304京都大学大学院理学研究科地球惑星科学専攻(主査)教授 平原 和朗, 教授 飯尾 能久, 教授 Mori James Jiro学位規則第4条第1項該当Doctor of ScienceKyoto UniversityDA
Report on a characteristic oscillation about 38 mHz (26 s) in northeastern Japan following surface wave of the 2011 Tohoku megathrust earthquake
No full textWe try to detect an unidentified signal from the surface motion at northeastern Japan immediately after the 2011 Tohoku earthquake. A focused frequency range is 10-100 mHz (10-100 s). We find a peaky signal with frequency of about 38 mHz (26 s) based on the horizontal-to-vertical (H/V) spectral ratio using the high-rate GNSS data at 382 GEONET stations. We are not able to identify locality of the signal. The signal appears several minutes after the passing of surface wave fronts. The duration of the signal is about 2 min. Since the origin of the 38 mHz signal is unlikely to be local hydrologic tremors, tectonic tremors, or the tsunami, we speculate that the 38 mHz signal originates from a kind of a characteristic oscillation of Northeastern Japan triggered by the 2011 Tohoku earthquake. A normal-mode simulation implies that high-order radial overtones could create the signal with a spherically-layered velocity structure, however, the detailed mechanism of the signal still remains a mystery
Fault instability on a finite and planar fault related to early phase of nucleation
No full textWe numerically investigate the early phase of nucleation on a planar fault with the rate- and state-dependent friction law, loaded externally by steady slip, to clarify its relation to fault instability. We define R[n] as the invasion distance of the inward creep to characterize that phase. For a circular fault, the dependence of R[n] on the dimensionless parameters l[b], l[b−a], and l[RA] (all of these are proportional to the rigidity and the characteristic distance of the state evolution L and inversely proportional to the normal stress and the fault radius) can be compiled. We found that Rn is proportional to l[b] (both aging law and slip law of the state evolution) and l[b−a] (aging law). In the case of the aging law only, there are two regimes (ordinary events and slow events) separated by the value of l[RA]. The regimes have different trend lines, although we could not measure Rn for the case of l[RA] < 0.35 because of breaking of the mirror symmetry of instability along the loading direction. R[n] in the slow event regime is smaller. Moreover, we investigated the effect of fault shape and found that a model with a long radius along the mode 2 direction has similar parameter dependence to circular faults, but a model with a long radius along the mode 3 direction has different ones. Our results imply that we can qualitatively estimate the fault instability parameters from the early phase of nucleation, although further research is necessary to enable application to actual faults
Observation of Earth's free oscillation by dense GPS array: After the 2011 Tohoku megathrust earthquake
No full textDay-scale Earth’s free oscillation after large earthquakes has been detected by underground instruments such as strainmeters, gravimeters and seismometers, to investigate Earth’s internal structure, geodynamics, and source properties of earthquakes. Here we show that Global Positioning System (GPS) can also detect the signals of the Earth’s free oscillation. A dense GPS array in Japan (GEONET) recorded the surface deformation following the 2011 Tohoku megathrust earthquake. A simple array analysis over 300 stations reduces local noise in GPS time series. We find that the dense GPS array truly detected both spheroidal and toroidal fundamental modes in three-direction displacement. This new tool has a strong potential to investigate the free oscillations particularly in low-frequency bands
Megaquake cycle at the Tohoku subduction zone with thermal fluid pressurization near the surface
No full textFor the 2011 Tohoku earthquake, we propose a mechanical model to explain rare giant (M9-class) and frequent large (M7-class) earthquakes on a thrust fault in the subduction zone. Observations implied, in the M9 Tohoku earthquake, that extremely large slip on the order of tens of meters occurs in a shallower part to release a slip deficit, as well as substantial slip about ten meters or so in a deeper part including the source area of the M7-class earthquakes. Here, we present a model in which the extremely large slip is caused by hydrothermal weakening (dynamic thermal pressurization of pore fluid) on the fault plane, not by contrast of frictional properties in terms of rate- and state-dependent friction. The model explains that the Tohoku earthquake followed a M7-class earthquake in two days, but M7-class earthquakes are not always followed by a giant earthquake. In a giant event, large coseismic slip can occur over an area where quasistatic slip, namely, afterslip of M7-class earthquakes or spontaneous slow slip events, takes place. Slight differences of stress state in the shallow part can result in drastically different coseismic slips. We further perform numerical experiments varying hydraulic parameters and the length of effective hydrothermal weakening area. The experiments imply that observations for monitoring the effective hydrothermal weakening area need spatial resolution on the order of 10 km or finer
Coseismic thermal pressurization can notably prolong earthquake recurrence intervals on weak rate and state friction faults: Numerical experiments using different constitutive equations
No full textWe add a new perspective to component factors of earthquake cyclicity, namely coseismic thermal pressurization (TP) within fluid‐saturated fault zones, which is pore fluid pressurization caused by frictional heating. By using a single degree of freedom system with a rate‐ and state‐dependent friction law, we show that the short‐lived TP can prolong earthquake recurrence intervals. This lengthening effect can operate even without any notable shear heating in weak faults. Moreover, if the maximum increase in temperature is above a certain level, the permeability rather than the maximum temperature becomes important for the lengthening effect. Lower permeability causes longer recurrence intervals. By contrast, narrower slip zones (more pronounced heating) do not simply prolong recurrence intervals, although they entail higher dynamic undershoot and energy radiation. These features do not depend on whether the assumed evolution law is the Ruina law or the Dieterich law. However, our results indicate that if the degree of TP changes for each earthquake, the ideal time‐predictable model for earthquake cycles can be applicable only in the case of faults obeying the Ruina law. Furthermore, on the basis of the above‐mentioned dependence of the interval on the permeability, we point out that it is necessary to measure the permeability rather than the slip zone thickness (or the increase in temperature) in order to estimate the TP effect on long‐term earthquake cycles. Although it is currently difficult to measure the permeability under ground, measurements should be performed in the light of the importance of permeability in the prediction of future seismic hazards
- …
