Inter-event time statistics of earthquakes as a gauge of volcano activity

Anca Opris; Sumanta Kundu; Takahiro Hatano; Yosuke Aoki

REVIEW 4 major objections 4 minor 29 references

Inter-event time statistics of earthquakes as a gauge of volcano activity

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper claims that the inter-event-time power-law exponent of volcanic earthquakes takes distinct phase values — 1.0 pre-eruption, 0.6–0.7 steady, 1.3 during bursts — and proposes α≈1.0 as an eruption index.

desk verdict A plausible and interesting empirical staging of IET exponents across volcanoes, but the preburst alpha about 1.0 indicator rests on retrospective window choices and untested stationarity, so the forecasting claim is work in progress. read the letter →

arxiv 2506.09203 v1 pith:GNZPFYKA submitted 2025-06-10 physics.geo-ph

Sumanta Kundu , Anca Opris , Yosuke Aoki , Takahiro Hatano This is my paper · ORCID

classification physics.geo-ph

keywords inter-eventtimepower-lawexponentvolcanicearthquakeseruptionprecursorseismicswarmvolcanoactivityphasesearthquakestatistics

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper asks whether the statistics of waiting times between consecutive volcanic earthquakes can reveal the state of a volcano. It claims that the power-law exponent α of the inter-event time distribution takes characteristic values that depend on the activity phase: 0.6–0.7 during steady seismicity, close to 1.0 in the preburst phase leading to a major eruption, about 1.3 during the burst itself, and near 0.8 afterward. Because the preburst value is distinct from the other two, the paper proposes α≈1.0 as a potential index for an imminent eruption accompanied by a burst of volcanic earthquakes. The claim rests on seven volcanoes from Japan and Hawaii, with the caveat that Kilauea does not show the preburst value while still matching the burst and post-burst values.

What carries the argument

The central object is the inter-event time (IET) — the duration between two consecutive earthquakes in a fixed region — and its probability density $p(\tau)$, which the paper fits as a power law $A\,\tau^{-\alpha}$ on a log-log scale in the intermediate time range. Because the IET is defined without reference to mainshocks or aftershocks, it applies to volcanic swarms that violate the Omori-law framework. The exponent $\alpha$ is the carrier of the signal: it is estimated by least-squares regression over logarithmically binned data, with robustness to catalog completeness checked by varying the cutoff magnitude from $M_{\min}$ to $M_c$. The paper's phase classifications (preburst, burst, post-burst) come from cumulative event-count curves, and the distinguishing numbers $\alpha\approx0.6$–$0.7$, $\approx1.0$, $\approx1.3$, and $\approx0.8$ are what make the IET exponent a candidate gauge of volcanic activity.

What would settle it

A prospective test is to compute the IET exponent in sliding windows on a volcano before any eruption is known and check whether $\alpha\approx1.0$ reliably precedes bursts and rarely appears without one; if it also appears during non-eruptive swarms, it is a swarm signature rather than an eruption precursor.

Watch

Extended reading notes

Core claim

The central discovery is that a single scalar — the exponent $\alpha$ in $p(\tau)\sim \tau^{-\alpha}$ fitted to the intermediate range of inter-event times — organizes volcanic seismicity by phase. For steady volcanic activity the exponent is 0.65(5) at Iwate and 0.61(3) at Nikko; for volcanoes with repeated minor bursts it is 1.21(7) at Yakedake and 1.24(6) at Hakone. For the Miyakejima 2000 eruption, the preburst phase gives $\alpha=1.03(6)$, the burst phase $\alpha=1.32(3)$, and the post-burst phase $\alpha=0.84(3)$; Sakurajima repeats the pair $\alpha=0.96(3)$ preburst and $1.31(6)$ burst. Kilauea matches the burst (1.33(6)) and post-burst (0.83(2)) values but shows $\alpha=0.67(3)$ in its preburst phase, which the authors attribute to low magma viscosity. The paper states that the preburst value $\alpha\simeq 1.0$ "could be interpreted as an index for imminent eruption."

Load-bearing premise

The load-bearing premise is that the phases of volcanic activity can be identified by looking at the cumulative event count after the eruption date is known, and that the time ranges used for the power-law fits are the right ones; if those choices are changed, the claimed exponent values and the proposed eruption index could shift or overlap.

Editorial extensions

If this is right

If the preburst exponent $\alpha\approx1.0$ is reproducible, monitoring the IET exponent in near-real time could provide a measurable precursor window before an eruption-driven burst.
The exponent distinguishes steady, pre-eruptive, and bursty volcanic seismicity without requiring a mainshock–aftershock model, making it applicable to swarm-dominated catalogs.
The common burst-phase value $\alpha\approx1.3$ across subduction-zone and hotspot volcanoes suggests a shared statistical feature of magma-driven fracture.
The post-burst value $\alpha\approx0.8$ differs from the steady value 0.6–0.7, indicating that a volcano's temporal statistics do not return to its pre-eruption state immediately.
The Kilauea exception, which the paper attributes to low magma viscosity, means the preburst index may be specific to higher-viscosity magma systems.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Beyond the paper: a sliding-window version of the IET exponent could turn the retrospective classification into a prospective eruption alarm; the paper itself only demonstrates the correlation after the fact.
Beyond the paper: because $\alpha\approx1.0$ sits between the steady and burst values, it may reflect an intermediate degree of temporal clustering; if so, the same statistic could quantify precursory deformation or pressurization in other geophysical time series.
Beyond the paper: a natural extension is to apply the same phase–exponent analysis to other well-documented explosive eruptions and to check whether $\alpha\approx1.0$ appears in non-eruptive swarms; the latter would show it is a swarm signature rather than an eruption precursor.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

This paper makes a genuinely interesting empirical observation: the inter-event time power-law exponent for volcanic seismicity clusters by activity phase across several volcanoes. Prior work was inconclusive (Lehr et al. at Villarrica), so the systematic staging is a real advance. The authors analyze seven public catalogs, show a consistent pattern—alpha around 0.6–0.7 for steady activity, around 1.3 during bursts, around 0.8 post-burst, and around 1.0 in the preburst phase for the two subduction-zone single-burst cases—and they are transparent about the Kilauea exception, where preburst alpha is 0.67. The fitting method is straightforward and the cutoff-magnitude robustness checks are reasonable. Credit where due: the burst and post-burst values are consistent across all volcanoes, and those phase classifications are unambiguous because the bursts are obvious in the cumulative counts.

The soft spots sit exactly on the claimed indicator. The preburst phase is defined retrospectively—you know an eruption is coming because you are looking at the catalog after the fact. The power-law fit ranges are chosen per phase without a prespecified rule, and the preburst windows are long (8–16 years). If the rate is slowly accelerating, pooling IETs over that window yields a mixture of exponentials that can look like tau^{-1}—alpha around 1.0—without any distinct precursor process. The authors assert stationarity by eye but do not run a formal test, and they do not show a sliding-window analysis demonstrating that alpha rises only shortly before the burst. Given that Kilauea gives 0.67 for its preburst phase, the 'imminent eruption index' claim already has a known false negative, and the viscosity explanation is post hoc and untested.

That said, the core empirical staging is likely reproducible, and the paper deserves a serious referee. A referee should push for a prospective test, a formal phase-detection procedure, and a stationarity test before alpha around 1.0 is treated as operational. I would send this to peer review rather than desk-reject it; it is the kind of paper that benefits from careful scrutiny and could become a solid, properly caveated empirical contribution after revision.

Referee Report

4 major / 4 minor

Summary. The paper analyzes inter-event time (IET) distributions of volcanic earthquakes from seven volcanoes in Japan and Hawaii. It reports that the IET distribution follows a power law, p(τ)∼τ^{-α}, in an intermediate time range, and that the exponent α depends on the stage of volcanic activity: approximately 0.6–0.7 for stationary seismicity (Iwate, Nikko), about 1.2 for volcanoes with many bursts (Yakedake, Hakone), and phase-dependent values for single-major-burst volcanoes. For Miyakejima the authors report α=1.03(6) for the preburst phase (1992–2000), α=1.32(3) for the burst phase (2000–2003), and α=0.84(3) for the post-burst phase; Sakurajima gives α=0.96(3) preburst and α=1.31(6) burst; Kilauea gives α=0.67(3) preburst, α=1.33(6) burst, and α=0.83(2) post-burst. The central claim is that α≈1.0 in the preburst phase may serve as an indicator of imminent eruption.

Significance. If the exponent values are robust and phase-specific, the result would provide a simple statistical gauge of volcano activity that is complementary to magnitude–frequency analyses. The paper has notable strengths: it draws on publicly available JMA and USGS catalogs, it checks robustness against cutoff magnitude in several cases, and it identifies a systematic difference between stationary and non-stationary volcanic seismicity. However, the empirical support for the preburst indicator is currently limited by the retrospective definition of phases, the absence of a sliding-window test, and one acknowledged counterexample (Kilauea). The claim of universality is therefore not yet established, but it is a testable hypothesis.

major comments (4)

[§3, 'Non-stationary time series - Single Major Burst: Miyakejima'; Fig. 3] The preburst phase (April 1992–May 2000) is identified by visual inspection of the cumulative event count after the date of the 2000 eruption is known, and the exponent α=1.03(6) is measured over this entire retrospective window; the paper does not provide a sliding-window analysis showing that α departs from the stationary value of about 0.65 only when an eruption is imminent, so the conclusion that α≈1.0 can be 'interpreted as an index for imminent eruption' is a retrospective fit rather than a prospective indicator.
[Methods, 'IET Distribution'; Fig. 3b] The preburst phase is an 8–16 year window (Miyakejima 1992–2000, Sakurajima 2000–2016), and the statement that the seismicity 'appears to be stationary' is not backed by any statistical test; for a point process with a slowly varying rate, the pooled IET density is a mixture of exponentials, and a sufficiently broad rate distribution inside the window can produce an apparent power law with exponent near 1 without any change in the underlying event-time law, so the alpha≈1.0 signature could be a rate-trend artifact rather than a distinct precursor state.
[Fig. 5b; Conclusions] Kilauea is the only single-burst case with a well-defined preburst phase that yields α=0.67(3), close to the stationary value and not the claimed ≈1.0; the viscosity-based explanation in the Conclusions is post hoc and untested, so as written the preburst indicator is supported by two positive cases (Miyakejima, Sakurajima) and one counterexample (Kilauea).
[Fig. 3b; Figs. 2b and 2d] The power-law fit ranges are selected per phase and differ between phases (preburst 10^2<τ<2×10^5 s versus burst and post-burst 10^2<τ<5×10^4 s for Miyakejima), and for Yakedake and Hakone the whole-period exponent of about 1.2 is fitted to a time series that mixes many bursts with quiet intervals, so it is not clear whether these exponents reflect phase-specific states or are artifacts of the chosen fitting windows and rate mixing; a prespecified fitting rule or a systematic sensitivity scan over ranges is needed.

minor comments (4)

[References, Ref. [15]] Reference [15] lists the authors as 'L. Johanna, B. Stefan, R. Wolfgang, T. Martin, and F. Luis', while the text calls them 'Lehr et al.'; the reference should be completed with the full author list or a standard journal citation format.
[References, Ref. [23]] There is a LaTeX encoding artifact in Ref. [23] ('K ¯ ılauea' should be 'Kīlauea' or 'Kilauea').
[Figs. 4 and 5] The captions state that data are 'arbitrarily shifted in the vertical direction for visual clarity'; the paper should state explicitly that the shifts do not affect the fitted exponents and that unshifted curves are available in the supplementary material.
[Data availability] Although catalog sources are indicated in Table S1, the paper does not include a data availability statement specifying the exact version or download date of the JMA and USGS catalogs; adding this would improve reproducibility.

Circularity Check

1 steps flagged · score 6.0 of 10

Preburst α≈1.0 'index' is an in-sample fit on retrospectively labeled pre-eruption windows, not a forward prediction.

fitted input called prediction [Conclusions; Results, 'Non-stationary time series - Single Major Burst: Miyakejima' (Fig. 3b)]
"The time series of seismic activity in Miyakejima is thus classified into three different phases: preburst steady activity (April 1992 - May 2000), a burst of volcano-tectonic earthquakes (June 2000 - March 2003), and the post-burst steady activity (April 2003 - October 2020). ... preburst seismicity may be characterized by a certain value of the exponent, α≃1.0. Since this is significantly different from the stationary phase (0.6 to 0.7) and the burst phase (≈1.3), this could be interpreted as an index for imminent eruption with a burst of volcanic earthquakes."

The 'preburst' windows are selected with knowledge that a burst follows, and the IET exponent is then fitted inside those windows (Miyakejima α=1.03, Sakurajima α=0.96). Calling this fitted value an 'index for imminent eruption' is an in-sample restatement: the period is labeled preburst precisely because an eruption/burst is known to occur later, so the fitted α is not a forward prediction. No sliding-window or out-of-sample test demonstrates that α≈1.0 precedes a burst when the burst is not already used to define the window, and a control set of non-eruptive α≈1.0 intervals is not examined. The predictive claim therefore reduces to the fitted value on a hindsight-selected subset.

full rationale

Most of the paper is a self-contained empirical analysis of seven catalogs: there are no load-bearing self-citations, no imported uniqueness theorem, and no ansatz smuggled in by citation. The stationary-phase exponents (Iwate 0.65, Nikko 0.61) and the burst exponents (≈1.2–1.33) are measured directly and are not circular. The central circularity lies in the claim that 'preburst seismicity may be characterized by a certain value of the exponent, α≃1.0' and that this could serve 'as an index for imminent eruption.' The preburst windows are defined retrospectively using the eruption as the labeling event: Miyakejima's 'preburst steady activity (April 1992 - May 2000)' is selected because the June 2000–March 2003 burst is known, and Sakurajima's preburst interval is selected because the 2017–2022 burst is known. The IET exponent is then fitted inside those outcome-defined windows and promoted to a precursor indicator without an out-of-sample or sliding-window test. The fitted exponent is thus renamed as a prediction, which is the 'prediction reduces to the fit' pattern. The paper even reports a known false negative (Kilauea preburst α=0.67) and offers an untested viscosity explanation, underscoring that the claimed universality is not established by independent prediction. Because this is a partial circularity in the central prospective claim rather than an equation-for-equation equivalence, the score is 6 rather than higher.

Assumptions & free parameters 22 free parameters · 6 assumptions · 0 invented entities

The central claim is empirical: exponents fitted to selected time windows. The main costs are the hand-chosen phase boundaries, the completeness magnitudes used to truncate catalogs, and the assumed power-law form over the selected fit range.

free parameters (22)

α_Iwate = 0.65 ± 0.05
Least-squares slope of log-binned IET distribution for Iwate, stationary phase (2000-2020).
α_Nikko = 0.61 ± 0.03
Same for Nikko (2004-2018), stationary.
α_Yakedake = 1.21 ± 0.07
Same for Yakedake, multi-burst catalog.
α_Hakone = 1.24 ± 0.06
Same for Hakone, multi-burst catalog.
α_Miyakejima_preburst = 1.03 ± 0.06
Same for Miyakejima preburst window (Apr 1992 - May 2000).
α_Miyakejima_burst = 1.32 ± 0.03
Same for Miyakejima burst window (Jun 2000 - Mar 2003).
α_Miyakejima_postburst = 0.84 ± 0.03
Same for Miyakejima post-burst window (Apr 2003 - Oct 2020).
α_Sakurajima_preburst = 0.96 ± 0.03
Same for Sakurajima preburst (2000-2016).
α_Sakurajima_burst = 1.31 ± 0.06
Same for Sakurajima burst (2017-2022).
α_Kilauea_preburst = 0.67 ± 0.03
Same for Kilauea preburst (2012-2017).
α_Kilauea_burst = 1.33 ± 0.06
Same for Kilauea burst (2018-2019).
α_Kilauea_postburst = 0.83 ± 0.02
Same for Kilauea post-burst (2020-2024).
M_c_Iwate = 0.4
Completeness magnitude estimated from GR law fit.
M_c_Nikko = 0.4
Completeness magnitude estimated from GR law fit.
M_c_Yakedake = 0.6
Completeness magnitude estimated from GR law fit.
M_c_Hakone = 0.1
Completeness magnitude estimated from GR law fit.
M_c_Miyakejima = 1.8
Completeness magnitude estimated from GR law fit.
M_c_Sakurajima = 0.5
Completeness magnitude estimated from GR law fit.
M_c_Kilauea = 1.8
Completeness magnitude estimated from GR law fit.
phase boundaries (Miyakejima, Sakurajima, Kilauea) = MYJ: 1992-04/2000-05, 2000-06/2003-03, 2003-04/2020-10; SAK: 2000-2016, 2017-2022; KIL: 2012-2017, 2018-2019, 2020-2024
Time windows chosen by visual inspection of cumulative curves (Figs. 3a, 4a, 5a); the exponents and the preburst indicator are defined relative to these windows.
Fit range endpoints = e.g., 10^2 to 2*10^5 s for Miyakejima preburst; varies per catalog
The intermediate-time-scale window over which the power law is fitted is selected per dataset without a stated criterion.
IET minimum threshold = 10 s
Inter-event times shorter than 10 s are discarded due to waveform overlap; no sensitivity test is shown in main text.

assumptions (6)

domain assumption The Gutenberg-Richter law holds for volcanic earthquake catalogs above the completeness magnitude.
Used to estimate M_c for each catalog in Table S1 and to argue the catalogs are complete for M>=M_c.
domain assumption The IET distribution is adequately described by a pure power law A*τ^{-α} on an intermediate time scale.
The fits in Figs. 1-5 assume this form, following the Gamma-distribution framework of Corral (2004), but no model comparison is performed.
domain assumption The selected spatial windows and maximum depths isolate volcanic earthquakes from tectonic events.
Region vertices and d_max in Table S1 are chosen to contain shallow volcanic seismicity; deeper events are excluded or assumed to be a separate class.
ad hoc to paper The phases (preburst, burst, post-burst) inferred from cumulative count curves correspond to physically distinct activity states.
Phase boundaries are set visually after the eruption is known (Figs. 3a, 4a, 5a), not by a changepoint algorithm, and the preburst indicator is defined on these same windows.
domain assumption Inter-event times shorter than 10 seconds can be disregarded without biasing the exponent.
The SI states this cutoff to avoid overlapping seismic waves, but no sensitivity check for the 10 s threshold is reported.
domain assumption The catalogs are complete above M_c within the selected region and time window.
Standard seismological assumption; robustness of exponents to cutoff magnitude is tested only for Yakedake, Hakone, and Miyakejima.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Inter-event time statistics of earthquakes as a gauge of volcano activity." pith.science (2026). https://pith.science/paper/GNZPFYKA

@misc{pith2026250609203,
  author       = {Pith},
  title        = {Pith review of: Inter-event time statistics of earthquakes as a gauge of volcano activity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GNZPFYKA}},
  note         = {Machine review of arXiv:2506.09203}
}

read the original abstract

The probability distribution of inter-event time (IET) between two consecutive earthquakes is a measure for the uncertainty in the occurrence time of earthquakes in a region of interest. It is well known that the IET distribution for regular earthquakes is commonly characterized by a power law with the exponent of 0.3. However, less is known about other classes of earthquakes, such as volcanic earthquakes, which do not manifest mainshock-aftershocks sequences. Since volcanic earthquakes are caused by the movement of magmas, their IET distribution may be closely related to the volcanic activities and therefore of particular interest. Nevertheless, the general form of IET distribution for volcanic earthquakes and its dependence on volcanic activity are still unknown. Here we show that the power-law exponent characterizing the IET distribution exhibits a few common values depending on the stage of volcanic activity. Volcanoes with steady seismicity exhibit the lowest exponent ranging from 0.6 to 0.7. During the burst period, when the earthquake rate is highest, the exponent reaches its peak at approximately 1.3. In the preburst phase, the exponent takes on the intermediate value of 1.0. These values are common to several different volcanoes. Since the preburst phase is characterized by the distinct exponent value, it may serve as an indicator of imminent volcanic activity that is accompanied by a surge in seismic events.

Figures

Figures reproduced from arXiv: 2506.09203 by the authors.

**Figure 3.** FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

**Figure 5.** FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 29 canonical work pages

[1]

Ogata, Statistics of earthquake activity: Models and methods for earthquake predictability studies, Annual Review of Earth and Planetary Sciences45, 497 (2017)

Y. Ogata, Statistics of earthquake activity: Models and methods for earthquake predictability studies, Annual Review of Earth and Planetary Sciences45, 497 (2017)

work page 2017
[2]

Hor´ alek, T

J. Hor´ alek, T. Fischer, P. Einarsson, and S. S. Jakobsd´ otir, Earthquake swarms, inEncyclopedia of Earthquake Engineering, edited by M. Beer, I. Kou- gioumtzoglou, E. Patelli, and I. K. Au (Springer, Berlin, Heidelberg, 2021)

work page 2021
[3]

Hor´ alek and T

J. Hor´ alek and T. Fischer, Role of crustal fluids in trigger- ing the west bohemia/vogtland earthquake swarms: Just what we know (a review), Studia Geophysica et Geodaet- ica52, 455 (2008)

work page 2008
[4]

D. R. Shelly, D. P. Hill, F. Massin, J. Farrell, R. B. Smith, and T. Taira, A fluid-driven earthquake swarm on the margin of the yellowstone caldera, Journal of Geophysical Research: Solid Earth118, 4872 (2013)

work page 2013
[5]

The heavy volcanic activity continued until March

work page
[6]

Traversa and J.-R

P. Traversa and J.-R. Grasso, How is volcano seismicity different from tectonic seismicity?, Bulletin of the Seis- mological Society of America100, 1755 (2010)

work page 2010
[7]

Chouet, Volcano seismology, Pure and Applied Geo- physics160, 739 (2003)

B. Chouet, Volcano seismology, Pure and Applied Geo- physics160, 739 (2003)

work page 2003
[8]

Br¨ auer, H

K. Br¨ auer, H. K¨ ampf, and G. Strauch, Earthquake swarms in non-volcanic regions: What fluids have to say, Geophysical Research Letters36, L17309 (2009)

work page 2009

Show all 29 references

[9]

D. C. Roman and K. V. Cashman, The origin of volcano- tectonic earthquake swarms, Geology34, 457 (2006)

2006
[10]

Nishimura and M

T. Nishimura and M. Iguchi,Volcanic Earthquakes and Tremor in Japan(Kyoto University Press, 2011)

2011
[11]

S. R. McNutt, Volcanic seismology, Annual Review of Earth and Planetary Sciences33, 461 (2005)

2005
[12]

Corral, Long-term clustering, scaling, and universality in the temporal occurrence of earthquakes, Phys

A. Corral, Long-term clustering, scaling, and universality in the temporal occurrence of earthquakes, Phys. Rev. Lett.92, 108501 (2004)

2004
[13]

´Alvaro Corral, Dependence of earthquake recurrence times and independence of magnitudes on seismicity his- tory, Tectonophysics424, 177 (2006), dynamics of Seis- micity Patterns and Earthquake Triggering

2006
[14]

N. S. Roberts, A. F. Bell, and I. G. Main, Are volcanic seismic b-values high, and if so when?, Journal of Vol- canology and Geothermal Research308, 127 (2015)

2015
[15]

Johanna, B

L. Johanna, B. Stefan, R. Wolfgang, T. Martin, and F. Luis, Comparative interevent time statistics of de- gassing and seismic activity at villarrica volcano (chile), ESS Open Archive 10.1002/essoar.10503030.1 (2020)

2020 doi
[16]

Japan Meteorological Agency, Recent seismic activity in the miyakejima and niijima-kozushima region, japan — the largest earthquake swarm ever recorded—, Earth, Planets and Space52, i (2000)

2000
[17]

Davidsen and G

J. Davidsen and G. Kwiatek, Earthquake interevent time distribution for induced micro-, nano-, and picoseismic- ity, Phys. Rev. Lett.110, 068501 (2013)

2013
[18]

Furuya, S

M. Furuya, S. Okubo, W. Sun, Y. Tanaka, J. Oikawa, H. Watanabe, and T. Maekawa, Spatiotemporal gravity changes at miyakejima volcano, japan: Caldera collapse, explosive eruptions and magma movement, Journal of Geophysical Research: Solid Earth108, 2219 (2003)

2003
[19]

Nakada, M

S. Nakada, M. Nagai, T. Kaneko, A. Nozawa, and K. Suzuki-Kamata, Chronology and products of the 2000 eruption of miyakejima volcano, japan, Bulletin of Vol- canology67, 205 (2005)

2005
[20]

Ukawa, E

M. Ukawa, E. Fujita, E. Yamamoto, Y. Okada, and M. Kikuchi, The 2000 Miyakejima eruption: Crustal deformation and earthquakes observed by the NIED Miyakejima observation network, Earth, Planets and Space52, xix (2000)

2000
[21]

Iguchi, T

M. Iguchi, T. Tamaguri, J.-I. Hirabayashi, and H. Nakamichi, Forecasting volcanic eruption of sakura- jima volcano based on magma intrusion rate, Bulletin of the volcanological society of Japan64, 33 (2019)

2019
[22]

Iguchi, T

M. Iguchi, T. Yamada, and T. Tameguri, Sequence of vol- canic activity of sakurajima volcano, japan, as revealed by non-eruptive deflation, Frontiers in Earth Science10, 10.3389/feart.2022.727909 (2022)

2022
[23]

Koike and H

M. Koike and H. Nakamichi, Dike inflation process be- neath sakurajima volcano, japan, during the earthquake swarm of august 15, 2015, Frontiers in Earth Science8, 10.3389/feart.2020.600223 (2021)

2021
[24]

jp/svd/eqev/data/bulletin/hypo_e.html

The JMA unified catalog,https://www.data.jma.go. jp/svd/eqev/data/bulletin/hypo_e.html
[25]

The USGS catalog,https://earthquake.usgs.gov/ earthquakes/search/
[26]

D. C. Roman, A. Soldati, D. B. Dingwell, B. F. Houghton, and B. R. Shiro, Earthquakes indicated magma viscosity during k ¯ ılauea’s 2018 eruption, Nature 592, 237 (2021)

2021
[29]

Yukutake, R

Y. Yukutake, R. Honda, M. Harada, R. Arai, and M. Matsubara, A magma-hydrothermal system beneath hakone volcano, central japan, revealed by highly re- solved velocity structures, Journal of Geophysical Re- search: Solid Earth120, 3293 (2015). S1 Supplementary information Inter...

2015
[2003]

The change in the seismicity and the signature of the strong non-stationary behavior during this period is clearly visible in the time series shown in Fig. 3(a). The time series of seismic activity in Miyakejima is thus classified into three different phases: preburst steady a...

1992
[2024]

Earth- quake and Volcano Hazards Observation and Research Program

phases withN= 3440,38750,and 3070 events, respec- tively. For visual clarity, the data for post-burst phase has been arbitrarily shifted in the vertical direction. smoother and therefore rather similar to the stationary phase. The common exponent values in burst and post- burs...

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Ogata, Statistics of earthquake activity: Models and methods for earthquake predictability studies, Annual Review of Earth and Planetary Sciences45, 497 (2017)

Y. Ogata, Statistics of earthquake activity: Models and methods for earthquake predictability studies, Annual Review of Earth and Planetary Sciences45, 497 (2017)

work page 2017

[2] [2]

Hor´ alek, T

J. Hor´ alek, T. Fischer, P. Einarsson, and S. S. Jakobsd´ otir, Earthquake swarms, inEncyclopedia of Earthquake Engineering, edited by M. Beer, I. Kou- gioumtzoglou, E. Patelli, and I. K. Au (Springer, Berlin, Heidelberg, 2021)

work page 2021

[3] [3]

Hor´ alek and T

J. Hor´ alek and T. Fischer, Role of crustal fluids in trigger- ing the west bohemia/vogtland earthquake swarms: Just what we know (a review), Studia Geophysica et Geodaet- ica52, 455 (2008)

work page 2008

[4] [4]

D. R. Shelly, D. P. Hill, F. Massin, J. Farrell, R. B. Smith, and T. Taira, A fluid-driven earthquake swarm on the margin of the yellowstone caldera, Journal of Geophysical Research: Solid Earth118, 4872 (2013)

work page 2013

[5] [5]

The heavy volcanic activity continued until March

work page

[6] [6]

Traversa and J.-R

P. Traversa and J.-R. Grasso, How is volcano seismicity different from tectonic seismicity?, Bulletin of the Seis- mological Society of America100, 1755 (2010)

work page 2010

[7] [7]

Chouet, Volcano seismology, Pure and Applied Geo- physics160, 739 (2003)

B. Chouet, Volcano seismology, Pure and Applied Geo- physics160, 739 (2003)

work page 2003

[8] [8]

Br¨ auer, H

K. Br¨ auer, H. K¨ ampf, and G. Strauch, Earthquake swarms in non-volcanic regions: What fluids have to say, Geophysical Research Letters36, L17309 (2009)

work page 2009