Skip to content
Back to the question

Evidence record

Can an earthquake be predicted?

Every sentence on the exploration belongs to at least one claim below. Each says what kind of statement it is, what it rests on, what the evidence does not establish, and how far it was checked. Nothing here is a score, and none of it is a verdict on a belief or a theory.

Last checked 24 September 2026.

  • Source checked: 42
  • A judgement, not a fact: 3
  • Presented as open: 2

If something here is wrong, misquoted, out of date or unfair, write to support@unknownoceans.com and give the claim number, which does not change.

What the question takes for granted

A scientific question with a settled half and an open one. That the time of a particular large earthquake cannot now be predicted is what the field and its agencies hold; whether it could ever be is argued, and the page reports the argument rather than settling it.

  • A prediction names time, place and magnitude, and a forecast gives a probability

    Established

    The distinction the whole page rests on is the one the ICEF report defines formally and the US Geological Survey uses.

  • Earthquakes could, in principle, be predicted

    Contested

    Geller and colleagues argued in 1997 that they cannot be; the ICEF called predictability poorly understood in 2011; a 2023 GPS study reported a precursory signal whose reality is argued over.

How the question is labelled

Claim: The forecast and the prediction

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Earthquakes

Kind
Framing. Signposting. It asserts nothing a source could check.

The question

Claim: Where and roughly how often, yes, well enough to build against.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: Long-term models are currently the most important forecasting tools for civil protection against earthquake damage, because they guide earthquake safety provisions of building codes

  • Qualifies

    Why earthquake hazard maps often fail and what to do about itAbstract

    Cited · Link answered, 19 October 2026 · The cited passage was read, 24 September 2026 · Earth and its record · Review of the field

    The source reads: of highly destructive earthquakes that occurred in areas predicted by earthquake hazard maps to be relatively safe

What this does not establish

The ICEF also records large uncertainty in long-term recurrence rates, and some destructive earthquakes have struck where the maps showed low hazard (claim 19).

Later work looked for

Crossref, 2012 on, "seismic hazard maps fail": Stein, Geller and Liu 2012, now on the page as a qualification; the maps remain the basis of building codes.

Corrected

24 September 2026. Previously: “Where and roughly how often, yes, and with real precision.” The ICEF report stresses large uncertainty in long-term recurrence rates, and Stein, Geller and Liu document destructive earthquakes in areas the maps rated relatively safe. "Real precision" overstated it; the use the report endorses is building codes.

Claim 3 · checked 24 September 2026 · AI-assisted audit

Claim: When, no,

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Can you predict earthquakes?FAQ answer, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: Neither the USGS nor any other scientists have ever predicted a major earthquake. We do not know how, and we do not expect to know how any time in the foreseeable future

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: The search for diagnostic precursors has not yet produced a successful short-term prediction scheme

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Claim 4 · checked 24 September 2026 · AI-assisted audit

Claim: and there is a serious argument that no is permanent.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Earthquakes Cannot Be Predictedp. 1616, second column

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Argument

    The source reads: Thus, the consensus of the meeting was that individual earthquakes are probably inherently unpredictable.

  • Context

    Earthquakes Cannot Be PredictedJournal summary of the Perspective

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Argument

    The source reads: whether or not this happens depends on unmeasurably fine details of conditions in Earth’s interior. Earthquakes are therefore inherently unpredictable

What this does not establish

The authors say individual earthquakes are "probably" inherently unpredictable. The argument is one side of a debate the page reports in claims 23 to 28.

Claim 5 · checked 24 September 2026 · AI-assisted audit

Claim: Seismic hazard maps are quantitative and useful

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Crossref, 2012 on, "seismic hazard maps fail": Stein, Geller and Liu 2012 question how well the maps perform, which the page now reports; it does not dispute that they are quantitative or used.

Claim 6 · checked 24 September 2026 · AI-assisted audit

Claim: and short-term prediction of a specific earthquake has never been demonstrated.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

The Haicheng earthquake of 1975 is the claimed exception; the ICEF notes that the prediction’s formality has been questioned and that nothing like it has succeeded since.

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Claim 7 · checked 24 September 2026 · AI-assisted audit

Claim: The distinction between forecasting a rate and predicting an event is the whole subject

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own framing. It follows from the ICEF’s formal definitions of prediction and forecast, which the report is organised around.

  • Context

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Review of the field

    The source reads: a deterministic statement that a future earthquake will or will not occur in a particular geographic region, time window, and magnitude range ... a probability (greater than zero but less than one) that such an event will occur

Claim 8 · checked 24 September 2026 · AI-assisted audit

Claim: and where it is not made clearly, informal predictions fill the gap, as they did in L’Aquila before the earthquake of 2009.

Kind
Historical claim. A statement about the past that historical evidence can test.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

The ICEF found no evidence that a valid prediction of the mainshock was passed to the authorities, and at least two of the informal predictions were false alarms.

Corrected

24 September 2026. Previously: “and the confusion between them has cost lives and reputations.” No source read carries "cost lives": that was the contested charge in the L’Aquila trial, which no cited source covers. The ICEF report does document the informal predictions and the public concern they caused.

Claim 9 · checked 24 September 2026 · AI-assisted audit

Claim: Every claimed precursor, radon in wells, animal behaviour, changes in electrical resistivity, has been reported before some earthquakes and absent before others.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Resistivity is the inverse of the conductivity the ICEF report discusses. For animal behaviour the report finds the evidence anecdotal rather than systematically recorded.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, radon emission

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: The preseismic anomalies account for only about 10% of the total observed anomalies

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, electrical conductivity changes

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: Changes in electrical conductivity prior to earthquakes have been reported since the early 1970s ... No precursory changes have been observed for any of the M > 4.0 earthquakes near Parkfield since 1989

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, anomalous animal behaviour

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: there is no credible scientific evidence that animals display behaviors indicative of earthquake-related environmental disturbances

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, Search for Diagnostic Earthquake Precursors

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026

    The source reads: none of the proposed precursors considered by the IASPEI Sub-Commission or otherwise published in the scientific literature had been demonstrated to be diagnostic

    The ICEF’s summary of the critical reviews, which carries the page’s "every": no proposed precursor has been shown to be reliable. The three named are examples it discusses.

  • Context

    The precursory phase of large earthquakesAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Observational study: associations, not causes

    The source reads: these observations do not directly precede earthquakes, are not seen before most events, and are also commonly observed without being followed by earthquakes

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Claim 10 · checked 24 September 2026 · AI-assisted audit

Claim: A signal that appears sometimes and is followed by nothing at other times is not a warning.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s reading of the precursor record in claim 10; the USGS draws the same conclusion.

  • Context

    Can you predict earthquakes?FAQ answer, fourth paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: Unfortunately, most such precursors frequently occur without being followed by an earthquake, so a real prediction is not possible

Claim 11 · checked 24 September 2026 · AI-assisted audit

Claim: Turning it into one is the part nobody has done.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Claim 12 · checked 24 September 2026 · AI-assisted audit

What can be done

Claim: What can be done

Kind
Framing. Signposting. It asserts nothing a source could check.

Hazard is forecastable and is forecast

Claim: Hazard is forecastable and is forecast

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

How the method works; not a finding later work revises.

Claim 14 · checked 24 September 2026 · AI-assisted audit

Claim: Long-term seismic hazard maps state the probability of ground shaking of a given intensity over a given period, and they are built from mapped faults, measured strain and the statistics of past events.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

How the method works; not a finding later work revises.

Claim 15 · checked 24 September 2026 · AI-assisted audit

Claim: They are good enough to write building codes against, which is the use that saves the most lives.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Crossref, 2012 on, "building codes seismic hazard mitigation": nothing that changes it.

Claim 16 · checked 24 September 2026 · AI-assisted audit

Claim: In January 2010 a magnitude 7.0 earthquake caused immense loss of life in Haiti, where large earthquakes were anticipated and buildings were not made to withstand them.

Kind
Historical claim. A statement about the past that historical evidence can test.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.E, value of forecasts

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: The M W 7.0 Haiti earthquake of 12 January 2010 caused immense destruction and loss of life in a region where large earthquake were anticipated but the built environment was not constructed to withstand intense seismic shaking

Corrected

24 September 2026. Previously: “A country that builds to its hazard map survives earthquakes that kill tens of thousands elsewhere.” An unsourced generalisation ("survives", "tens of thousands") replaced by the comparison the ICEF report actually makes, with its own hedge ("relative to its size", "in large part").

Claim 17 · checked 24 September 2026 · AI-assisted audit

Claim: Six weeks later a magnitude 8.8 struck Chile, which enforces high seismic safety standards, and caused substantially less damage and loss of life for its size.

Kind
Historical claim. A statement about the past that historical evidence can test.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.E, value of forecasts

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: the MW 8.8 Chile earthquake of 27 February 2010 caused substantially less damage and loss of life relative to its size, in large part because Chile enforces high seismic safety standards

What this does not establish

The report attributes the difference "in large part" to building standards, not wholly; the two earthquakes also differed in location and depth, which the page does not compare.

Corrected

24 September 2026. Previously: “” Added with claim 17, replacing an unsourced generalisation.

Claim 18 · checked 24 September 2026 · AI-assisted audit

Claim: The maps have critics as well. Stein, Geller and Liu point to Wenchuan in 2008 and Tohoku in 2011, both in areas the maps showed as relatively safe, and argue that the maps’ uncertainties should be stated plainly and the maps tested against simple alternatives.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Why earthquake hazard maps often fail and what to do about itAbstract

    Cited · Link answered, 19 October 2026 · The cited passage was read, 24 September 2026 · Earth and its record · Review of the field

    The source reads: The 2011 Tohoku earthquake is another striking example – after the 2008 Wenchuan and 2010 Haiti earthquakes – of highly destructive earthquakes that occurred in areas predicted by earthquake hazard maps to be relatively safe

  • Supports

    Why earthquake hazard maps often fail and what to do about itAbstract

    Cited · Link answered, 19 October 2026 · The cited passage was read, 24 September 2026 · Earth and its record · Review of the field

    The source reads: the uncertainties in hazard map predictions should be assessed and clearly communicated to potential users ... hazard maps should undergo rigorous and objective testing to compare their predictions to those of null hypotheses, including ones based on uniform regional seismicity or hazard

Other readings

Stein, Geller and Liu also list Haiti among the failures, where the ICEF report says large earthquakes were anticipated (claim 17). A published comment by Frankel disputes the paper; it was found only as a record and its argument is not reported here.

Corrected

24 September 2026. Previously: “” The passage reported only the case for hazard maps. A review by Stein, Geller and Liu argues they often fail; the page now says so and who says it.

Claim 19 · checked 24 September 2026 · AI-assisted audit

Prediction of individual events has not been demonstrated

Claim: Prediction of individual events has not been demonstrated

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: This lack of understanding is reflected in the inability to reliably predict large earthquakes in seismically active regions on short time scales

  • Supports

    Can you predict earthquakes?FAQ answer, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: Neither the USGS nor any other scientists have ever predicted a major earthquake

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Claim 20 · checked 24 September 2026 · AI-assisted audit

Claim: A prediction, in the sense that matters, specifies place, time and magnitude narrowly enough that it could fail.

Kind
About a word. What a word in the original language means, or how it is translated.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Review of the field

    The source reads: a deterministic statement that a future earthquake will or will not occur in a particular geographic region, time window, and magnitude range

  • Supports

    Can you predict earthquakes?FAQ answer, second and third paragraphs

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: An earthquake prediction must define 3 elements : 1) the date and time, 2) the location, and 3) the magnitude ... Their predictions are so general that there will always be an earthquake that fits

Claim 21 · checked 24 September 2026 · AI-assisted audit

Claim: No method has produced those reliably.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Claim 22 · checked 24 September 2026 · AI-assisted audit

Claim: In 1997 a paper in Science put the case that earthquakes cannot be predicted in principle: any small earthquake has some chance of cascading into a large one, and whether it does depends on unmeasurably fine details of conditions in the Earth’s interior.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Earthquakes Cannot Be Predictedp. 1616, second column

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Argument

    The source reads: any small earthquake has some probability of cascading into a large event ... Whether any particular small earthquake grows into a large earthquake depends on a myriad of fine details of physical conditions throughout a large volume

  • Supports

    Earthquakes Cannot Be Predictedp. 1616, second column

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Argument

    The source reads: The prediction of individual large earthquakes would require the unlikely capability of knowing all of these details with great accuracy.

  • Supports

    Earthquakes Cannot Be PredictedJournal summary of the Perspective

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Argument

    The source reads: they argue that any small earthquake has some chance of cascading into a large event ... whether or not this happens depends on unmeasurably fine details of conditions in Earth’s interior. Earthquakes are therefore inherently unpredictable

What this does not establish

An argument by four authors, not a finding. The Perspective says individual earthquakes are "probably" inherently unpredictable, and that in practice it does not matter whether prediction is impossible or only very difficult.

Corrected

24 September 2026. Previously: “on the grounds that the state of stress deep in the crust is not measurable and that a rupture’s final size depends on conditions it encounters as it propagates. A small slip either arrests or grows into a large one depending on details nobody can see in advance.” Restated in the terms the only readable account of the paper uses. The propagation mechanism was not in anything read of it; the ICEF’s version of that point is now a separate, attributed hypothesis (claims 24 and 25).

Claim 23 · checked 24 September 2026 · AI-assisted audit

Claim: Strainmeters near several large earthquakes, L’Aquila among them, recorded no change before the rupture

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, strain-rate changes

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: Strainmeters have thus far shown no significant precursory changes in strain rate during the nucleation of large earthquakes ... Before L’Aquila earthquake, two laser strainmeters located about 20 km from the epicenter did not record precursory signals ... Similarly stringent constraints on nucleation precursors have come from strainmeters in California

What this does not establish

The instruments constrain precursory slip to below what they can detect; they do not show there was none.

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Corrected

24 September 2026. Previously: “” Added, from the ICEF report, to carry the propagation point with a source.

Claim 24 · checked 24 September 2026 · AI-assisted audit

Claim: which fits the hypothesis that large and small earthquakes start alike and that size is settled as a rupture spreads and stops.

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, strain-rate changes

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: consistent with the hypothesis that this scale is comparable for all earthquakes, big and small; i.e., most seismogenic ruptures start in more or less the same way ... earthquake magnitude depends more on the physical processes that propagate and arrest, rather than nucleate, the rupture

  • Qualifies

    Do large earthquakes start with a precursory phase of slow slip?Abstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Observational study: associations, not causes

    The source reads: Even though the results of our tests do not irrefutably demonstrate the existence of a precursory phase of slow slip, they do support its existence

What this does not establish

A hypothesis the strainmeter data are consistent with, not one they prove. Bletery and Nocquet report a precursory slip signal before large earthquakes that, if real, would sit uneasily with it (claims 52 and 53).

Corrected

24 September 2026. Previously: “” Replaces "A small slip either arrests or grows into a large one depending on details nobody can see in advance", which stated as fact what the ICEF calls a hypothesis.

Claim 25 · checked 24 September 2026 · AI-assisted audit

Claim: That argument has not been refuted and it has not been proved.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s summary of where the 1997 argument stands, which follows from the ICEF calling predictability poorly understood and from the continuing debate Bletery and Nocquet describe.

  • Context

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Review of the field

    The source reads: Earthquake predictability, the degree to which the future occurrence of earthquakes can be determined from the observable behavior of earthquake systems, is poorly understood

  • Context

    Do large earthquakes start with a precursory phase of slow slip?Introduction

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Observational study: associations, not causes

    The source reads: The search for precursory signals to large earthquakes has been a long-standing pursuit and the existence of such signals has generated a multi-decadal debate in the earthquake science community

Claim 26 · checked 24 September 2026 · AI-assisted audit

Claim: The US Geological Survey says it does not expect to know how to predict earthquakes in the foreseeable future

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Can you predict earthquakes?FAQ answer, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: Neither the USGS nor any other scientists have ever predicted a major earthquake. We do not know how, and we do not expect to know how any time in the foreseeable future

Corrected

24 September 2026. Previously: “It has been the working assumption of the field for a quarter of a century.” No source read says the 1997 argument is the field’s working assumption; the ICEF calls predictability poorly understood and recommends research on it, and the debate continues (Bletery and Nocquet 2024). Replaced with what two named bodies say. "A quarter of a century" was also out of date.

Claim 27 · checked 24 September 2026 · AI-assisted audit

Claim: and an international commission in 2011 called earthquake predictability poorly understood.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

Claim 28 · checked 24 September 2026 · AI-assisted audit

Seconds of warning are real

Claim: Seconds of warning are real

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

How the method works; not a finding later work revises.

Claim 29 · checked 24 September 2026 · AI-assisted audit

Claim: Early warning systems work on a different principle. They detect the first, faster, less damaging waves at instruments near the epicentre and send an electronic signal ahead of the slower, more damaging ones.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Earthquake Early Warning: How can an earthquake alert be announced before tremors are felt?Leaflet, panel on seismic waves

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: P-waves are the first to travel outward. They are followed by S-waves, which cause stronger tremors. Most earthquake-induced damage results from these S-waves ... After a tremor (P-wave) is detected the earthquake early warning will be transmitted immediately. This is essentially a race against the seismic wave

  • Supports

    What is an Earthquake Early Warning?What is an Earthquake Early Warning?

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: based on prompt analysis of the focus and magnitude of the earthquake using wave form data observed by seismographs near the epicenter

Later work looked for

How the method works; not a finding later work revises.

Claim 30 · checked 24 September 2026 · AI-assisted audit

Claim: That buys seconds to tens of seconds, which is enough to slow trains, bring lifts under control and let people take cover.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

How the method works; not a finding later work revises.

Corrected

24 September 2026. Previously: “which is enough to stop trains, open lift doors and halt surgery.” Aligned with what the Japan Meteorological Agency lists. Halting surgery is in no source read, and the JMA says trains are slowed and lifts controlled.

Claim 31 · checked 24 September 2026 · AI-assisted audit

Claim: Close to the epicentre the shaking can arrive with the warning or before it.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    What is an Earthquake Early Warning?Limitations of the Earthquake Early Warning, Timing

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: In areas that are close to the focus of the earthquake, the warning may not be transmitted before strong tremors hit

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection III.D, Japan

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Review of the field

    The source reads: informs the public of a strong tremor before it arrives at more distant sites

Later work looked for

How the method works; not a finding later work revises.

Corrected

24 September 2026. Previously: “” Added: "seconds to tens of seconds" is not what everyone gets, and the JMA and the ICEF both say so.

Claim 32 · checked 24 September 2026 · AI-assisted audit

Claim: It is not prediction: the earthquake has already begun.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

How the method works; not a finding later work revises.

Claim 33 · checked 24 September 2026 · AI-assisted audit

Claim: Three different things called prediction

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Scientists can tell you an earthquake is coming next week.

Kind
Framing. Signposting. It asserts nothing a source could check.

The misconception, stated as one; the page answers it in claims 36 and 37.

Claim: No method has ever done this reliably.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Can you predict earthquakes?FAQ answer, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: Neither the USGS nor any other scientists have ever predicted a major earthquake

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: The search for diagnostic precursors has not yet produced a successful short-term prediction scheme

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Claim 36 · checked 24 September 2026 · AI-assisted audit

Claim: Claims that a particular quake was predicted usually rest on a statement so broad that some earthquake was bound to fit it.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Can you predict earthquakes?FAQ answer, third paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Reference work

    The source reads: Their predictions are so general that there will always be an earthquake that fits

What this does not establish

The USGS gives this as one of several reasons such claims fail; others are that they omit time, place or magnitude, or claim success for a quake that differs widely from the prediction.

Later work looked for

A description of how claims fail; not a finding later work revises.

Corrected

24 September 2026. Previously: “almost always describe a forecast so broad it could not fail.” "Almost always" was stronger than the USGS, which lists breadth as one of several failings; and such a claim is not a forecast in the ICEF’s sense, which is a probability.

Claim 37 · checked 24 September 2026 · AI-assisted audit

Claim: Earthquake warnings on a phone are predictions.

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: They are detections. The rupture has started, and the alert is racing the slower waves to you.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

How the method works; not a finding later work revises.

Claim 39 · checked 24 September 2026 · AI-assisted audit

Claim: Nothing useful can be said about future earthquakes.

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Rates and locations are forecast well enough to set building codes, and building to them is the most effective protection a community has.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Crossref, 2012 on, "building codes seismic hazard mitigation": nothing that changes it.

Corrected

24 September 2026. Previously: “which is what actually determines how many people die.” The ICEF calls building codes and retrofitting the most effective measures; "what actually determines how many people die" claimed more than that.

Claim 41 · checked 24 September 2026 · AI-assisted audit

What might yet be

Claim: What might yet be

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: A rupture may not know how big it is going to be when it starts.

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, strain-rate changes

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: consistent with the hypothesis that this scale is comparable for all earthquakes, big and small; i.e., most seismogenic ruptures start in more or less the same way

What this does not establish

The hypothesis of claim 25, put figuratively. It is not established.

Other readings

Bletery and Nocquet argue that large earthquakes may begin with a precursory phase of slow slip, which the page reports in claims 52 and 53.

Corrected

24 September 2026. Previously: “A rupture does not know how big it is going to be when it starts.” The ICEF states this as a hypothesis the data are consistent with, and it is contested.

Claim 43 · checked 24 September 2026 · AI-assisted audit

Whether short-term probabilities can be raised usefully

Claim: Whether short-term probabilities can be raised usefully

Kind
Open question. Something the page says is not known.
Review
Presented as open. The page says this is unresolved, and the review found that it is.

Claim 44 · checked 24 September 2026 · AI-assisted audit

Claim: After a moderate earthquake the probability of a larger one nearby rises for days, by a factor that can be estimated from aftershock statistics.

Kind
What a model says. What a theory or model predicts or describes, which is not the same as what has been observed.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

What clustering models forecast. The ICEF calls the gains they report nominal, highly uncertain and largely unvalidated, though some have been confirmed in prospective tests.

Claim 45 · checked 24 September 2026 · AI-assisted audit

Claim: Operational forecasting turns that into an official, updated number.

Kind
About a word. What a word in the original language means, or how it is translated.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

In 2011 the ICEF found that no country surveyed had fully implemented it, regularly updated and on a national scale.

Later work looked for

Crossref, 2012 on, "operational earthquake forecasting decision making": the 2014 Varenna workshop report (Marzocchi, Jordan and Woo 2015) restates the ICEF findings and nothing found overturns them.

Claim 46 · checked 24 September 2026 · AI-assisted audit

Claim: The problem is what to do with a probability that goes from one in ten thousand to one in a hundred.

Kind
What a model says. What a theory or model predicts or describes, which is not the same as what has been observed.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.A.5

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Theory · Review of the field

    The source reads: a forecast based on long-term seismicity rates might yield a 0.01% chance that a target earthquake will occur in a small subdomain of space and time ... might increase the event probability to, say, 1%, in the same subdomain

What this does not establish

The ICEF’s illustrative example, not a measured case. Its measured case, the day before L’Aquila, stayed well below 1 per cent in a day.

Claim 47 · checked 24 September 2026 · AI-assisted audit

Claim: It is a large relative change and a small absolute one.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Crossref, 2012 on, "operational earthquake forecasting decision making": the 2014 Varenna workshop report (Marzocchi, Jordan and Woo 2015) restates the ICEF findings and nothing found overturns them.

Claim 48 · checked 24 September 2026 · AI-assisted audit

Claim: At a 2014 workshop on exactly this, some argued that such forecasts are useless in practice, because acting on them means a false alarm nearly every time, and others that even small probabilities can nudge people toward sensible precautions.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Varenna workshop report. Operational earthquake forecasting and decision makingAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Review of the field

    The source reads: A workshop on Operational earthquake forecasting and decision making was convened in Varenna, Italy, on June 8-11, 2014

  • Supports

    Varenna workshop report. Operational earthquake forecasting and decision makingSection 3, Decision making in a low-probability environment

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Review of the field

    The source reads: Some attendees argued that current short-term forecasting methods are effectively useless for practical purposes, owing to the high false alarm rates implied by the low-probability rates ... Others countered that even small probabilities can be helpful in nudging ... towards beneficial self-protection actions

Corrected

24 September 2026. Previously: “and there is no agreement on what action it justifies.” The disagreement was asserted without a source; the Varenna workshop report records both sides, which the page now names.

Claim 49 · checked 24 September 2026 · AI-assisted audit

Whether any precursor is real

Claim: Whether any precursor is real

Kind
Open question. Something the page says is not known.
Review
Presented as open. The page says this is unresolved, and the review found that it is.

  • Supports

    Do large earthquakes start with a precursory phase of slow slip?Introduction

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Observational study: associations, not causes

    The source reads: The search for precursory signals to large earthquakes has been a long-standing pursuit and the existence of such signals has generated a multi-decadal debate in the earthquake science community

Claim 50 · checked 24 September 2026 · AI-assisted audit

Claim: Slow slip events, changes in seismic wave speed and patterns of small quakes are all under investigation, and some have statistical support in some regions.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

The statistical support is for seismicity-pattern methods; for changes in wave speed the ICEF found the data limited and no causal link shown.

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Claim 51 · checked 24 September 2026 · AI-assisted audit

Claim: In 2023 Bletery and Nocquet stacked the GPS records from before 90 large earthquakes and found about two hours of accelerating slip before the ruptures.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The precursory phase of large earthquakesAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Observational study: associations, not causes

    The source reads: We summed the displacements measured by 3026 high-rate GPS time series ... during 48 hours before 90 (moment magnitude ≥7) earthquakes. Our approach reveals a ≈2-hour-long exponential acceleration of slip before the ruptures

  • Supports

    The precursory phase of large earthquakesRevised manuscript, results, stack of the GPS time series (Fig. 2A)

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Observational study: associations, not causes

    The source reads: the distribution of S as a function of time shows no obvious coherent pattern from 48 to 2 hours before the earthquakes (Fig. 2A). However, in the 2 hours preceding the events, the stack reveals a positive trend

What this does not establish

The signal is an average over all 90 earthquakes and, by the authors’ account in their 2024 follow-up, cannot be seen before any single one. Read in the authors’ revised manuscript on HAL, not the version of record.

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Corrected

24 September 2026. Previously: “” The currency search found the most prominent precursor result since the page’s sources, and the dispute over it; the page now reports both.

Claim 52 · checked 24 September 2026 · AI-assisted audit

Claim: Others found that the signal vanishes once noise shared across a network is removed; the authors reply that the correction can remove a real signal too, and call the question unsettled.

Kind
Disputed. A contested point, where the page reports a side or the disagreement.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Do large earthquakes start with a precursory phase of slow slip?Abstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Observational study: associations, not causes

    The source reads: it has been proposed that correcting GPS time series from network common-mode noise makes the signal vanish. We confirm this result, but we show that this common-mode filtering procedure may inadvertently remove an existing tectonic signal

  • Supports

    Do large earthquakes start with a precursory phase of slow slip?Abstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Observational study: associations, not causes

    The source reads: Even though the results of our tests do not irrefutably demonstrate the existence of a precursory phase of slow slip, they do support its existence

What this does not establish

The critics’ side is reported as the authors describe it. Their own analyses, posted informally rather than in a journal, were not read.

Corrected

24 September 2026. Previously: “” Added with claim 52, so the result is not reported without the dispute over it.

Claim 53 · checked 24 September 2026 · AI-assisted audit

Claim: Pattern-recognition methods tested prospectively have beaten chance, but only by a factor of two to four, with alarms lasting months or years over areas as large as Switzerland, which leaves them little use as predictions.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, seismicity patterns, and Figure 2.4

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: When an adequate sample of target earthquakes is available (N > 10), these prediction methods show skill that is statistically significant with respect to time-independent forecasts ... the prediction methods achieve only small probability gains, in the range 2-4 relative to these time-independent reference forecasts

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection II.B.2, seismicity patterns

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: alarms are issued for periods of months or years in geographically extended areas; e.g. the CN alarms for Italy cover areas comparable to or greater than the area of Switzerland ... limit the practical utility of these methods as deterministic prediction tools

What this does not establish

"Chance" here is a time-independent reference forecast, and the ICEF notes that the testing method and that reference are themselves disputed.

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Corrected

24 September 2026. Previously: “None has produced a prospective prediction that succeeded where a chance guess would have failed.” The cited ICEF report says the opposite for pattern-recognition methods: in prospective tests they showed statistically significant skill over a time-independent reference, with small gains and alarms too broad to act on.

Claim 54 · checked 24 September 2026 · AI-assisted audit

Claim: For many proposed precursors, a 2011 review found, the rates of false alarms and missed earthquakes are unknown, so whether the signal means anything cannot be judged.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Operational Earthquake Forecasting: State of Knowledge and Guidelines for UtilizationSection IV.C

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Earth and its record · Review of the field

    The source reads: In many cases of purported precursory behavior ... the rate of false negatives (earthquake but no precursory signal) cannot be ascertained ... the rate of false positives (signal but no earthquake) is unknown. Without constraints on these error rates, the diagnostic properties of the signal cannot be evaluated

  • Context

    Earthquakes Cannot Be PredictedJournal summary of the Perspective

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Argument

    The source reads: prediction claims are not stated as objectively testable scientific hypotheses

Later work looked for

Crossref, 2020 on, "precursory slip large earthquakes GPS" and "earthquake prediction precursors review": Bletery and Nocquet 2023 and their 2024 follow-up, now on the page; no precursor with a successful prospective test was found.

Corrected

24 September 2026. Previously: “The field treats retrospective identification of a precursor as a hypothesis rather than a result, which is the lesson of a long history of the opposite.” What "the field" treats as what was unsourced as worded; replaced with the specific finding of the ICEF review, attributed.

Claim 55 · checked 24 September 2026 · AI-assisted audit

What was read