Figure - available from: Journal of Geophysical Research: Solid Earth
This content is subject to copyright. Terms and conditions apply.
Hypocenter distribution of the r‐LFEs (large circles) and other earthquakes (blue dots) whose FI values are calculated for the (a) 2008 Iwate‐Miyagi and (b) 2011 Iwaki Earthquakes. The r‐LFEs are colored with respect to the elapsed time since each mainshock. The A‐B vertical cross sections are shown in the bottom panels. The mainshock hypocenters are shown by the large green stars. The hypocenters of two moderate‐size foreshocks (M6.0 and M6.1) prior to the 2011 Iwaki Earthquake are shown by the small green stars in (b). The black and gray triangles denote active and Quaternary volcanoes, respectively. FI, frequency index.
Source publication
Low‐frequency earthquakes (LFEs) that predominantly occur at depths of 20–30 km are categorized as a particular class of earthquakes whose spectral power is concentrated at 1–4 Hz. While the tectonic LFEs along megathrust boundaries occur as shear failure, the genesis of LFEs in the continental plate is poorly understood due to the diversity of foc...
Similar publications
We conduct a P‐wave receiver function analysis of the Bransfield Strait (West Antarctica) to determine the lithospheric structure of this back‐arc basin, thanks to 31 temporary and permanent stations. Our main finding is a 15 km tear of the Phoenix slab, coinciding with the location of the 2020–2021 Orca earthquake swarm's epicenters. Teleseismic w...
Subduction zone architecture and properties are thought to control megathrust slip behavior, but few constraints on crustal structure and megathrust properties are available at sufficient resolution and depth, hindering understanding of linkages between structure and behavior. Here we present a P‐wave seismic velocity model based on wide‐angle seis...
Repeating earthquakes, or repeaters, which repeatedly rupture the same location and release the strain energy, are interpreted as repeated ruptures of an isolated asperity patch surrounded by a stable sliding regime. While repeaters are frequently observed along plate‐boundary faults, the occurrence of repeater sequences are reported in various tec...
Tectonic tremor has been explained as a swarm of low‐frequency earthquakes (LFEs), which are located on a narrow fault at the plate boundary. However, due to the lack of clear impulsive phases in the tremor signal, it is difficult to determine the depth of the tremor source with great precision. The thickness of the tremor region is also not well c...
The subduction of continental lithosphere is a complex process because the buoyancy of the crust is higher than the oceanic and should resist sinking into the mantle. Anyway, studies on the Alpine-Himalayan collision system indicate that a large portion of the continental crust is subducted, while some material is accreted in the orogens. The Apenn...
Citations
... Plot of fp versus FI for regular earthquakes (gray) and LFEs (red) identified by the Japan Meteorological Agency (JMA). The green rectangle shows the domain of LFEs re-defined byNakajima and Hasegawa (2021), containing mainly earthquakes identified as LFEs by the JMA. Modified fromNakajima and Hasegawa (2021). ...
... The green rectangle shows the domain of LFEs re-defined byNakajima and Hasegawa (2021), containing mainly earthquakes identified as LFEs by the JMA. Modified fromNakajima and Hasegawa (2021). ...
... Distributions of (a) shallow (depth ≤ 15 km) and (b) deep (depth >15 km) LFEs defined byNakajima and Hasegawa (2021). Gray triangles denote active volcanoes. ...
Low-frequency earthquakes (LFEs) are anomalous earthquakes with a lower predominant frequency than that expected from the earthquake magnitude. LFEs are also unique with respect to their anomalously deep focal depths, where ordinary earthquakes do not occur. Due to these characteristics, the generation mechanisms of LFEs have attracted seismological, lithological, and geochemical attention. Here, we review the observation and their interpretation of LFEs. Most LFEs worldwide occur near active volcanoes, although some LFEs in Japan are detected far from such volcanoes on account of the high sensitivity of the seismic network. The focal mechanisms of LFEs include double-couple that represent fault slip, isotropic and compensated linear vector dipole (CLVD) that suggest a volume change in the source. Tomographic studies have shown that the source areas of LFEs are characterized by low velocity and high VP/VS ratios, suggesting the contribution of geofluids to LFE occurrence. Recent observations in Japan have revealed the occurrence of LFEs at shallow depths, even in the upper crust, sometimes located close to ordinary earthquakes. These observations and the variation in focal mechanisms support an LFEs’ source model of tensile-shear crack. LFEs can occur in the upper crust if the pore fluids are close to H2O rather than magma. The focal mechanism is double-couple if shear motion is dominant but is characterized by isotropic or CLVD components if crack opening is dominant, with the different focal mechanisms probably reflecting the pressure of pore fluids.
... This area is 124 surrounded by many borehole seismic stations, and recent studies have identified relocated 125 hypocenters for small earthquakes (e.g., Okada et al., 2012;Yoshida et al., 2014a) in this 126 region, making it suitable for studying the source properties of earthquakes (Figure 1 and 127 Figure 3). Nakajima and Hasegawa (2021) reported that shallow LFEs (~ 10 km) also occur in 128 and around this area ( Figure 1 and Figure 2). The nature of shallow LFEs is also not yet clear, 129 ...
... Standard deviations calculated for each factor were 382 used as measures of uncertainty. We calculated the radiated energy ( ) and seismic moment ( 0 ) using the derived 394 source spectra and obtained the scaled energy ( ) for 1464 regular earthquakes ( ≥ 395 2.0), 169 deep LFEs, and 52 shallow LFEs detected by Nakajima and Hasegawa (2021). Figure 396 8 shows the relationship between the moment magnitudes estimated in this study and 397 those listed on the F-net moment tensor catalog ( ) based on broadband data. ...
... Standard deviations calculated for each factor were 382 used as measures of uncertainty. 383 We calculated the radiated energy ( ) and seismic moment ( 0 ) using the derived 394 source spectra and obtained the scaled energy ( ) for 1464 regular earthquakes ( ≥ 395 2.0), 169 deep LFEs, and 52 shallow LFEs detected by Nakajima and Hasegawa (2021). Figure 396 8 shows the relationship between the moment magnitudes estimated in this study and 397 those listed on the F-net moment tensor catalog ( ) based on broadband data. ...
Many unknowns exist regarding the energy radiation processes of the inland low-frequency earthquakes (LFEs) often observed beneath volcanoes. To evaluate their energy radiation characteristics, we estimated the scaled energy for LFEs and regular earthquakes in and around the focal area of the 2008 Mw 6.9 Iwate-Miyagi earthquake. We computed the source spectra for regular earthquakes, deep LFEs, and shallow LFEs by correcting for the site and path effects from direct S-waves. We computed the radiated energy and seismic moments, and obtained the scaled energy (eR) for 1464 regular earthquakes, 169 deep LFEs, and 52 shallow LFEs. The eR for regular earthquakes is in the order of 10-5 to 10-4, typical for crustal earthquakes, and tends to become smaller near volcanoes and shallow LFEs. In contrast, eR is in the order of 10-7 and 10-6 for deep and shallow LFEs, respectively, one to three orders of magnitude smaller than that for regular earthquakes. This result suggests that LFEs are associated with a much lower stress drop and/or slower rupture and deformation rates than regular earthquakes. Although the energy magnitudes derived from radiated energy generally show good agreement with the local magnitudes for the three types of earthquakes, the moment and local magnitudes show a large discrepancy for the LFEs. This suggests that the local magnitude based only on the maximum amplitude of the observed seismic records may not provide good information on the static sizes of LFEs whose eR values are substantially different from those of regular earthquakes.
... They are often considered as building blocks of tremors Figure 1d) and of the broadband slow earthquake phenomena (i.e., Brownian walk model) (Ide, 2008;Ide & Yabe, 2018). There are also reports on LFEs that are not associated with tremor signals (Arai et al., 2016;Aso et al., 2013;Nakajima & Hasegawa, 2021). Meanwhile, to date, LFEs have not been reported in association with shallow (<15 km) tremor signals. ...
... Some of these differences can be explained by the different source and receiver geometries (deep or shallow). Nevertheless, the LFEs that are not associated with tremor signals (e.g., Nakajima & Hasegawa, 2021) have been detected also at relatively shallow depths (<15 km), making it difficult to consider short-duration tremors as LFEs at shallow depths. Furthermore, Figure 3a shows that the short-duration tremors are likely end-members of tremors with varying source durations. ...
Plain Language Summary
At least two types of earthquakes occur in shallow subduction zones: ordinary earthquakes and tremors. Tremors are known to exhibit long signal duration compared to ordinary earthquakes. To date, tremors' long‐duration signal has been solely interpreted by their source process. Here, we discovered tremors that exhibit short duration signals when recorded close from the source which we referred to as "short‐duration tremors". They suggest that tremors' source process is not always long and structural effects may partially form the typical long‐duration signals. We performed numerical simulations on elastic wave propagation and demonstrated that the observations can be qualitatively reproduced by assuming a strongly scattering material surrounding the seismic source. On the other hand, it has been reported based on ocean bottom drilling project that tremor source region may consist of patchily distributed aquifers. The inclusions in our model may be the seismic expressions of the geologically detected aquifers. Further, such a structure could be embedded along the slow‐earthquake fault zone and play a key role in their source process.
... We defined afterslip-type repeaters in the continental crust that occurred in aftershock areas of 10 M ≥ 6.5 earthquakes (mainshocks) in the analyzed period of 2003-2017 (areas outlined in blue in Figure 6a) because the repeaters in these areas are likely to be affected by afterslip of the mainshock (see Table 1 in Nakajima & Hasegawa, 2021, for details of the 10 mainshocks). We classified all other repeaters outside the 10 aftershock areas as burst-type repeaters, where no marked aseismic slip was geodetically observed. ...
... The temporal correlations of the observed high repeater rates and short T r with the inferred period of high pore-fluid pressures suggest that repeater activity is enhanced by the elevated pore-fluid pressures. We compared frequency index (FI) of repeaters and non-repeaters (1.5 ≤ M ≤ 2.5) in the Kitakata swarm, where FI is a metric characterizing the frequency content of waveforms with smaller FI enriched in low-frequency energy (Nakajima & Hasegawa, 2021). Figure 8h shows that repeaters tend to possess low FI compared to non-repeaters for the same magnitude range of 1.5 ≤ M ≤ 2.5, suggesting a relatively enriched low-frequency energy in repeater waveforms. ...
... Previous studies have suggested that enhanced pore-fluid pressures weaken the fault strength, thereby playing a crucial role on the triggering of earthquakes (e.g., Sibson, 2013). This hypothesis is consistent with observations of hypocenter migrations (e.g., Yoshida & Hasegawa, 2018), seismicity rate change with time (e.g., Nakagomi et al., 2021), and reduced stress drop of earthquakes (e.g., Nakajima & Hasegawa, 2021;Yoshida & Hasegawa, 2018). In this framework, the elevated pore-fluid pressures are a direct control on earthquake generation. ...
Repeating earthquakes, or repeaters, which repeatedly rupture the same location and release the strain energy, are interpreted as repeated ruptures of an isolated asperity patch surrounded by a stable sliding regime. While repeaters are frequently observed along plate‐boundary faults, the occurrence of repeater sequences are reported in various tectonic settings. Here we systematically investigate whether repeaters occur in the continental crust and subducting slabs beneath the Japanese Islands and show a prevalence of repeaters in both locations. The crustal and intraslab repeaters show the power‐law decay of seismicity rates that is identical to that along plate‐boundary faults and occur on well‐defined fault planes coincidentally with non‐repeating earthquakes. These observations suggest that repeaters share a common generation process independent of the tectonics regime and are not distinct from non‐repeaters in their source locations and occurrence times. We thus infer that repeaters and non‐repeaters are both generated by ruptures of locked asperities in response to stress loading by aseismic slip in the surrounding stable regime. Repeater sequences can be observed when aseismic slip is large enough to rupture an isolated asperity more than twice, while no repeaters are generated when the amount of aseismic slip is not sufficient to cause multiple ruptures at an isolated asperity. Since an earthquake cannot be triggered unless the amount of the aseismic stress loading exceeds the strength of asperities at the fault, hidden aseismic slip probably occurs repeatedly and frequently during an interseismic period.
... However, the relocated hypocenters with the 3D velocity model do not have enough resolution to discuss the detailed temporal and spatial evolution of seismicity. Furthermore, this study has no observations to suggest what broke the low-permeability seal, even though the candidates would probably include either episodic aseismic deformation (Nakajima and Uchida 2018), or stress loading by external forces such as strong motion (Nakajima and Hasegawa 2021), or fast increases in porefluid pressure itself to some threshold value (Shapiro et al. 2018). ...
We carried out seismic tomography study to reveal three-dimensional (3D) seismic velocity structures in the Noto peninsula, Japan, where swarm-like seismic activity started in December 2020. The obtained results reveal a highly heterogeneous structure in the crust. The most striking feature is the existence of a low-velocity anomaly in the lower crust beneath the Noto earthquake swarm. Although the data set used in this study cannot resolve the upper mantle structure, previous regional tomographic studies suggest that a low-velocity anomaly exists at depths of 50–150 km around the Noto peninsula that is probably interpreted as a fluid-rich region. We infer that fluids have been supplied from the uppermost mantle to the lower crust over a geological time scale and a large volume of fluids have accumulated below the seismogenic zone beneath the Noto peninsula. A further upward migration of fluids to the upper crust, which may have suddenly started in December 2020, probably triggers numerous earthquakes at depths of 10–15 km. Since major active faults exist at shallower extensions of the hypocenters of the Noto earthquake swarm, we consider that the earthquake swarm occurs along pre-existing and weak fault planes. Dense temporary seismic observations will highlight a smaller-scale (5–10 km) 3D seismic velocity model and finer hypocenter distribution, which provide additional information for better understanding of the generation mechanisms of the Noto earthquake swarm.
Graphical Abstract
... Taking advantage of a dense local network, Yoshida et al. (2020) found intermingled regular earthquakes and LFEs in the upper crust beneath Hakodate, Hokkaido, extending a pipe-like cluster of LFEs that reaches down to the base of the crust. In a systematic analysis across Japan, Nakajima and Hasegawa (2021) identified shallow nonvolcanic LFEs that are dominantly observed within aftershock sequences following ten large (M w > 6:5) inland earthquakes, including the 2016 M L 7.0 Kumamoto earthquake. These studies propose that the shallow tectonic LFEs may be ascribed to an aqueous fluid movement stemming from the underlying subduction zone (Yoshida et al., 2020), crustal fluids redistributed by the recent large mainshock ruptures (Nakajima and Hasegawa, 2021), or heterogeneous lithologies with variable mechanical and frictional properties (Barnes et al., 2020). ...
... In a systematic analysis across Japan, Nakajima and Hasegawa (2021) identified shallow nonvolcanic LFEs that are dominantly observed within aftershock sequences following ten large (M w > 6:5) inland earthquakes, including the 2016 M L 7.0 Kumamoto earthquake. These studies propose that the shallow tectonic LFEs may be ascribed to an aqueous fluid movement stemming from the underlying subduction zone (Yoshida et al., 2020), crustal fluids redistributed by the recent large mainshock ruptures (Nakajima and Hasegawa, 2021), or heterogeneous lithologies with variable mechanical and frictional properties (Barnes et al., 2020). However, the general mechanism of shallow LFEs is still enigmatic. ...
... We aim to confirm the existence of shallow LFEs and then identify their generation mechanism, considering the possible role of enhanced local attenuation in the immediate vicinity of a large rupture in the upper crust, fluid-related processes, and other previously proposed mechanisms of LFEs. We apply the frequency index (FI) method (Buurman and West, 2010;Matoza et al., 2014), which was used for detecting volcanic LFEs during the 2006 eruption of Augustine Volcano, Alaska, (Buurman and West, 2010) and at Mammoth Mountain, California (Hotovec-Ellis et al., 2018), as well as for the selection of shallow tectonic LFEs in a regular earthquake catalog (Nakajima and Hasegawa, 2021). ...
Low-frequency earthquakes (LFEs) generally have relatively stronger spectral components in the lower frequency range compared with what is expected for regular earthquakes based on their magnitude. LFEs generally occur in volcanic systems or deep (>∼15 km) in plate boundary fault zones; however, LFEs have also been observed in nonvolcanic, upper crustal settings. Because there are few studies that explore the spatiotemporal behaviors of LFEs in the shallow crust, it remains unclear whether the shallow-crustal LFEs reflect local attenuation in their immediate vicinity or differences in their source mechanism. Therefore, it is important to identify shallow-crustal LFEs and to characterize their spatiotemporal activity, which may also improve our understanding of LFEs. In this study, we focus on detecting shallow-crustal LFEs and explore the possible generation mechanisms. We analyze 29,646 aftershocks in the 2019 Ridgecrest, California, earthquake sequence, by measuring the frequency index (FI) to identify candidate low-frequency aftershocks (LFAs), while accounting for the magnitude dependency of the FI. Using small earthquakes (ML 1–3) recorded in the borehole stations to minimize the attenuation effects in near-surface layers, we identify 68 clear LFAs in total. Based on their distribution and comparisons with other seismic parameters measured by Trugman (2020), the LFAs possess distinct features from regular events in the same depths range, including low corner frequencies and low stress drops. Events in the close vicinity of LFAs exhibit lower average FI values than regular aftershocks, particularly if the hypocentral distance between an LFA and its neighbors is less than 1 km. Our results suggest that LFAs are related to local heterogeneity or a highly fractured fault zone correlated with an abundance of cross faults induced by the aftershock sequence at shallow depths. Zones of high pore-fluid pressure in intensely fractured fault zones could cause the bandlimited nature of LFAs and LFEs in general.
Low-frequency earthquakes (LFEs) are seismic phenomena with the shortest timescale among various slow earthquakes observed on broadband time scales. To understand the nature of such a broadband slow phenomenon, it is important to investigate the rupture evolution process of individual slow events, such as LFEs. Here, we investigated the moment-duration relationship of LFEs at plate interfaces and volcanic regions, and showed that the moment-duration relationship of both tectonic and volcanic LFEs is characterised by a moment proportional to the cubic duration, similar to that in ordinary earthquakes. The difference between our obtained moment-duration relationship and the broadband scaling suggests that the evolution process of LFEs may not be controlled, but only triggered by the slow earthquakes with longer durations, such as slow slip events driven by aseismic diffusion. The seismic moments of the LFEs are approximately three orders of magnitude smaller than those of ordinary earthquakes with similar durations. This result indicates that LFEs have rupture growth similar to that of ordinary earthquakes, although the rupture velocity and/or stress drop are much smaller. Considering the hypocentre spread of LFEs, the estimated rupture velocity and stress drop were approximately 100 m/s–1 km/s and 2 kPa–1 MPa, respectively. Additionally, the estimated moment magnitudes are much larger than the local magnitudes determined based on the maximum amplitudes, which is due to the longer durations and resultant smaller amplitudes of LFEs than those of ordinary earthquakes.
How to estimate a strain-rate field from spatially discrete geodetic data has been a longstanding issue. In this paper, a method based on basis function expansion with Akaike's Bayesian information criterion (ABIC) is introduced, by which strain-rate fields can be obtained objectively and stably. By applying the method to GNSS data in Japan, strain-rate fields are obtained for three periods: 1997-1999, 2006-2009, and 2017-2020. Except for deformation related to volcanic activity and large earthquakes, the obtained strain-rate fields are roughly stationary in time, while showing large variations in space. In order to interpret such spatially heterogeneous deformation, a framework of inter-arc and intra-arc deformation is used, considering Japanese Islands to be composed of five island arcs (Kuril, northeast Japan, west Japan, Izu–Bonin, and Ryukyu) and that these island arcs are defined with little ambiguity, though the northeast- and west-Japan arcs are collectively treated as the Honshu arc in this study. Inter-arc deformation between the Kuril and Honshu arcs is characterized by EW contraction, the Izu–Bonin and Honshu arcs by NS to NW–SE contraction, and the Ryukyu and Honshu arcs by NS extension with EW contraction. Regarding intra-arc deformation, the Kuril arc shows high strain rates from the Pacific coast to the back of the volcanic arc, the northernmost part of the Izu–Bonin arc shows significant EW to NE–SW extension, and the Ryukyu arc shows NS extension with EW contraction similar to the inter-arc deformation with the Honshu arc, although the EW contraction is weaker to the south. The Honshu arc shows zones of high strain rates along the eastern margin of the Japan Sea via the Niigata–Kobe tectonic zone (NKTZ) to the Median Tectonic Line and along the Ou-backbone Range, while it also shows low strain rates in the Chugoku district and in the zone from northern Ibaraki prefecture via the northern Kanto district to northern Aichi prefecture, which is named the Hitachi–Mikawa forearc low strain-rate zone (HMLSZ).
Low-frequency earthquakes (LFEs) occurring in the continental plate are reviewed. Most LFEs in the continental plate occur at depths of ~15-45 km in the uppermost mantle to the lower crust beneath volcanoes, but they also occur within the same depth range beneath non-volcanic areas. Because they occur at greater depths than the typical depth limit for shallow regular earthquakes, they are called “deep low-frequency earthquakes (deep LFE).” However, a recent study reveals that LFEs also occur at depths shallower than 15 km in the upper crust where many regular earthquakes occur. This indicates that LFEs occur over the entire depth range from the uppermost mantle to the upper crust. In the upper crust, LFEs and regular earthquakes coexist and occur in close proximity. Focal mechanisms and activity patterns of LFEs show that tensile-shear crack is the dominant mechanism generating LFEs. In addition, the long duration of waveforms is probably caused by resonance in the fluid-filled crack. Distributions of peak frequency (fp) and frequency index (FI) values of waveforms, both of which are expected to be significantly small for LFEs and large for regular earthquakes, show that there is no clear boundary for fp and FI values between LFEs and regular earthquakes; rather, they are distributed continuously. It is presumed that the distribution of high and low pore fluid pressures in source faults creates such distributions of small and large fp and FI values, respectively, and a LFE occurs when the pore pressure is extremely high. This indicates that pore pressure is directly related also to the genesis of regular earthquakes. In source areas of recent large inland earthquakes, LFEs are activated by the mainshock, and FI and fp values and stress drop synchronously decrease immediately after the mainshock, gradually recovering thereafter. The activation of LFEs by the mainshock and such temporal changes of FI, fp, and stress drop after the mainshock can be explained within the framework of fault-valve behavior. Furthermore, fp and FI values tend to be small along prominent tectonic lines, such as the Itoigawa-Shizuoka tectonic line. It is inferred that pore fluid pressure is locally high along those tectonic lines, thereby facilitating the current crustal deformation.
