Pith. sign in

REVIEW 2 major objections 5 minor 7 references

Performance and User Response of Android's Smartphone-Based Alerts in the 2025 Marmara Ereglisi Earthquake

T0 review · 2 major / 5 minor · reviewed 2026-07-13 · grok-4.5

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

Pith's one-line read Android phones detected the 2025 Marmara quake in 5.31 seconds and gave 16 million users a median 56-second warning for weak shaking.

desk verdict Solid large-N observational audit of AEA on a real M6.2: 5.31 s detection, 90 % TP / 99 % precision for Be Aware, median 56 s warning, and clear lead-time → action/trust links; soft spots are the composite MMI-III contour and proprietary data, not fatal. read the letter →

arxiv 2607.08975 v1 pith:4XHYVSYK submitted 2026-07-09 physics.geo-ph

S. Mostafa Mousavi , Patrick Robertson , Richard M. Allen , Alexei Barski , Robert Bosch , Nivetha Thiruverahan , Youngmin Cho , Tajinder Gadh
show 5 more authors
Steve Malkos Boone Spooner Greg Wimpey Edward Shi Marc Stogaitis
This is my paper · ORCID
classification physics.geo-ph
keywords earthquakeearlywarningAndroidAlertsmartphoneseismologycrowd-sourcedsensingtimeuserresponseMarmaraSeaMMIintensity
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

This paper evaluates how Google’s Android Earthquake Alert system performed during the Mw 6.2 Marmara Ereğlisi earthquake. The phone network detected the event 5.31 seconds after origin, faster than the sparse local seismic stations near the offshore epicenter, and issued alerts to more than 16 million users. For weak (MMI III) shaking the system delivered a median 56-second warning (up to 150 seconds), with 90 percent true-positive coverage of the population and 99 percent precision. User surveys show that people who received the alert before shaking were far more likely to take protective action, rate the alert useful, and say they would trust future alerts. The authors argue that dense, crowd-sourced smartphone sensing can both speed detection and, when the alert is timely, drive constructive public response at continental scale.

What carries the argument

The AEA detection-and-alert pipeline: on-phone accelerometers that flag P- or S-wave arrivals, a backend that fuses the resulting sparse triggers into real-time magnitude and intensity polygons, and two-tier delivery (“Be Aware” for MMI III–IV, “Take Action” for MMI ≥ V) whose spatial coverage is scored against a composite ground-truth contour using population-weighted true-positive / false-negative metrics.

What would settle it

An independent, instrument-based intensity map that places a substantially different fraction of the regional population outside the AEA “Be Aware” polygons would drop the true-positive rate well below 90 percent and falsify the claimed precision.

Watch

Extended reading notes

Core claim

During the Mw 6.2 Marmara Ereğlisi earthquake the Android Earthquake Alert system detected the event 5.31 s after origin, alerted more than 16 million users, achieved a 90 percent true-positive rate and 99 percent precision for MMI-III “Be Aware” alerts with a median 56 s warning, and produced substantially higher rates of protective action, perceived usefulness and future trust when the alert arrived before shaking.

Load-bearing premise

The composite MMI-III contour built from ShakeMaps, felt reports and Google Search “yes” ratios correctly defines the population that should have been alerted, so that the reported true-positive and false-negative rates are unbiased.

Editorial extensions

If this is right

  • In dense coastal or urban corridors, phone networks can issue usable early warnings even for offshore events that traditional stations detect later.
  • Alert-before-shaking is the dominant driver of protective action, usefulness ratings and future trust; late alerts actively erode credibility.
  • Hybrid phone-plus-station architectures could reduce detection latency and improve magnitude estimates for complex aftershock sequences.
  • “Take Action” breakthrough alerts remain essential for nighttime or Do-Not-Disturb scenarios where standard notifications fail to wake users.
  • Public-education campaigns that reinforce Drop-Cover-Hold-On remain necessary so that alerts trigger prescribed actions rather than panic.

Reading between the lines

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

  • Because magnitude saturation limited the final “Take Action” polygon, the paper’s high true-positive figure is driven almost entirely by the lower-tier “Be Aware” alerts; a larger event would expose a more severe performance gap.
  • The elevated active-response rate relative to earlier studies in other countries may reflect both the recent 2023 Kahramanmaraş disaster and cultural norms rather than a universal property of the AEA interface.
  • Self-selected survey respondents who actively searched for earthquake information may over-represent people who already felt shaking, inflating the apparent correlation between alert receipt and protective behavior.
  • If phone density continues to grow faster than traditional station density, the relative detection advantage of crowd-sourced networks will widen for most inhabited coastal and inland settings.
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.

Referee Report

2 major / 5 minor

Summary. This manuscript evaluates Google’s Android Earthquake Alert (AEA) system for the 23 April 2025 Mw 6.2 Marmara Ereğlisi earthquake. It reports detection 5.31 s after origin time, issuance of alerts to >16 million users, 90 % true-positive rate and 99 % precision for MMI-III “Be Aware” alerts (median 56 s warning time, up to ~150 s), faster phone-network detection than nearby AFAD stations despite the offshore epicenter, and large-N user-feedback analyses linking pre-shaking alert arrival to higher rates of protective action (DCHO or warning others), perceived usefulness, and future trust. Performance is quantified on population-weighted ~20 km grids against a multi-source MMI-III ground-truth contour; behavioral associations are tested with chi-square statistics and Cramer’s V effect sizes (Table 1).

Significance. If the reported metrics hold, the paper supplies a rare, large-scale observational audit of a production smartphone-based EEW system under real seismic loading, combining technical performance (detection latency, magnitude evolution, warning-time distributions) with behavioral response data at N~67 000. The multi-source ground-truth construction, population-weighted TP/LA/FN/FP definitions, explicit effect sizes, and open station-processing code are clear strengths. The work is of direct interest to seismologists, EEW operators, and risk-reduction communities because it quantifies both the technical reach of crowd-sourced sensing and the conditions under which alerts translate into protective action and trust.

major comments (2)
  1. [Methods, Alert performance evaluation; Fig. S3] Methods (“Alert performance evaluation”) and Fig. S3: the composite MMI-III ground-truth contour that underpins the 90 % TP / 99 % precision claim mixes USGS/INGV ShakeMaps, DYFI/EMSC felt reports, and Google Search “yes” ratios. The Search survey samples only users who actively sought earthquake information and therefore already perceived shaking; the paper itself notes this selection. Because the outer edges of the TP region (e.g., Izmir) rest partly on these ratios, a sensitivity test that recomputes the spatial metrics after excluding the Search layer is needed to confirm that the headline rates are not inflated.
  2. [User Feedback Analysis; Table 1; Discussion] User Feedback Analysis and Table 1: the chi-square associations between alert timing, shaking intensity, active response, usefulness, and future trust rest on self-selected in-alert respondents (N=67 056) plus a tiny independent DYFI subsample (N=55). While the paper correctly flags non-random sampling, the claim that AEA “elicited substantially higher rates of protective action” relative to prior studies (Goltz, Nakayachi, Vinnell) requires a more quantitative discussion of how self-selection and the recent 2023 Kahramanmaraş sequence may bias the 46 % active-response figure upward.
minor comments (5)
  1. [Discussion] Discussion paragraph comparing prior studies: “greater percentage of infections among responders” is a clear typographical error for “inactions.”
  2. [Abstract; title] Abstract and title: “Googles” and “Marmara Ereglisi” lack the correct apostrophe/diacritics used elsewhere; standardize to “Google’s” and “Marmara Ereğlisi.”
  3. [Results, Detection and Strong Motion Arrivals; Fig. 1] Figure 1 caption and text: the claim that phones detected the event “sooner than conventional seismic stations” is supported for the nearest coastal phones, but the two AFAD stations at 25–29 km are the only stations inside 30 km; a short quantitative comparison of first-trigger times (phones vs. stations) would make the statement more precise.
  4. [Supplementary Table S1] Table S1 lists successive magnitude estimates and alert radii; adding the corresponding alert-issue times relative to origin would help readers reconstruct the timeline shown in Fig. 3 without cross-referencing the text.
  5. [Data availability] Ethics / Data availability: the statement that AEA phone-trigger and feedback data “cannot be made publicly available” is understandable, yet the paper would benefit from a brief note on whether any aggregated, privacy-preserving summary tables (e.g., binned warning-time histograms) can be released to support independent verification of the population-weighted metrics.

Circularity Check

1 steps flagged · score 1.0 of 10

No significant circularity: empirical performance audit against multi-source external ground truth; only minor non-load-bearing self-citation to the authors' global AEA overview.

  1. self citation load bearing [Introduction / Discussion (refs to Allen et al. 2025)]
    "From April 1, 2021, to April 30, 2025, AEA successfully identified a total of 11,231 earthquakes... (see Allen et al., 2025 for AEA performance for these events). ... A comprehensive, global evaluation of the system's technical limitations across more than 11,000 earthquakes is detailed in Allen et al.14."

    Overlapping-author citation supplies background on global AEA performance and system architecture. It is not load-bearing for the present paper's event-specific metrics (detection time, TP/precision, warning times, user-behavior statistics), which rest on external USGS/AFAD/EMSC ground truth and the authors' own anonymized telemetry for this single quake. Minor and non-circular under the stated criteria.

full rationale

This is an observational case study of system performance and user behavior for one earthquake, not a first-principles derivation or predictive model. Detection latency (5.31 s), alert counts (>16 M), spatial-grid TP/FP/FN rates (90 % TP / 99 % precision for Be Aware), and warning-time distributions (median 56 s) are computed by comparing AEA alert polygons and delivery timestamps against independent external references: USGS/INGV ShakeMaps, AFAD station waveforms, USGS/EMSC felt reports, and a Google Search survey whose sampling bias is explicitly noted. Population-weighted grid metrics (Methods) do not reduce to free parameters of the AEA magnitude estimator. User-response associations (timeliness vs. active response / usefulness / trust) are descriptive chi-square tests on voluntary survey data (Table 1) with reported effect sizes; they are not fitted predictions. The sole self-citation (Allen et al. 2025) supplies system description and global context and is not required to close any logical loop for the event-specific numbers. No equation, uniqueness claim, or ansatz is smuggled in. Score 1 reflects only the presence of that non-load-bearing self-citation; the central claims remain independently falsifiable against the cited external catalogs.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The work is an observational case study; it inherits standard seismological conventions (MMI intensity scale, S-wave group velocities from Cochran et al. 2022, USGS ShakeMap relations) and Google’s internal AEA detection thresholds (Mw > 4.5 for alerting). No free parameters are fitted to produce the central performance numbers; the composite MMI-III contour is an ad-hoc but documented synthesis of public sources. No new physical entities are postulated.

assumptions (4)
  • domain assumption MMI III and MMI V intensity thresholds correspond to approximate PGA levels of 0.3 % g and 6.2 % g, respectively, and can be used to mark “weak” and “moderate-to-strong” shaking on both phone and station waveforms.
    Used throughout Results and Methods to define alert targets and ground-truth regions.
  • domain assumption Group velocities defined in Cochran et al. (2022) correctly describe the spatial extent of the “Shaken Zone” at any given time after origin.
    Invoked to color the green “timely alert” polygons in Figure 3.
  • ad hoc to paper Population counts on ~20 km s2cells are the appropriate unit for computing true-positive / false-negative rates rather than device counts.
    Explicit methodological choice stated in “Alert performance evaluation”; changes the numerical TP rate relative to a pure device-based metric.
  • ad hoc to paper Self-selected in-alert survey respondents and the small DYFI EEW subsample are sufficiently representative for chi-square tests of association between alert timing, shaking intensity, protective action, usefulness, and future trust.
    Acknowledged as non-random in the User Feedback section; underpins all behavioral claims.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Performance and User Response of Android's Smartphone-Based Alerts in the 2025 Marmara Ereglisi Earthquake." pith.science (2026). https://pith.science/paper/4XHYVSYK

@misc{pith2026260708975,
  author       = {Pith},
  title        = {Pith review of: Performance and User Response of Android's Smartphone-Based Alerts in the 2025 Marmara Ereglisi Earthquake},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4XHYVSYK}},
  note         = {Machine review of arXiv:2607.08975}
}
read the original abstract

This study presents a comprehensive evaluation of Googles Android Earthquake Alert (AEA) system during the Mw 6.2 Marmara Ereglisi, Turkiye earthquake. AEA detected the event 5.31 seconds after its initiation, alerting over 16 million users. Warning times for weak shaking (MMI III) reached up to 150 seconds, with a median of 56 seconds. While near-source warning windows were shorter, the system achieved 90% true positives and 99% precision overall. The high density of the phone network enabled faster detection than traditional stations, even for this offshore epicenter. Feedback data shows AEA recipients were highly likely to take protective actions, such as drop, cover, and hold on, or warn others. Timely alerts substantially increased user engagement, perceived usefulness, and future trust. These results highlight how crowd-sourced technology and behavioral insights can effectively enhance seismic resilience on a massive scale.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

7 extracted references · 5 canonical work pages

  1. [1]

    A., Husker, A

    Vaiciulyte, S., Novelo-Casanova, D. A., Husker, A. L., and Garduño-González. (2022). Population response to earthquakes and earthquake early warnings in Mexico. Int. J. Disaster Risk Reduct. 72, 102854. doi: 10.1016/j.ijdrr.2022.102854

    work page doi:10.1016/j.ijdrr.2022.102854 2022

  2. [2]

    Did You Feel It?

    Saunders, J. K. and D. J. Wald (2025). Quantitative Evaluations of Earthquake Early Warning Performance Using “Did You Feel It?” and Post-Alert Surveys, The Seismic Record . 5(2), 239–249, doi: 10.1785/0320250018

    work page doi:10.1785/0320250018 2025

  3. [3]

    S., Potter, S

    Nakayachi, K., Becker, J. S., Potter, S. H., and Dixon, M. (2019). Residents’ reactions to earthquake early warnings in Japan. Risk Analysis 39, 1723–1740. doi: 10.1111/risa.13306

    work page doi:10.1111/risa.13306 2019

  4. [4]

    Vinnell, L. J., M. L. Tan, R. Prasanna, and J.S. Becker (2023). Knowledge, perceptions, and behavioral responses to earthquake early warning in Aotearoa New Zealand, Front. Commun. 8, 1229247, doi: 10.3389/fcomm.2023.1229247

    work page doi:10.3389/fcomm.2023.1229247 2023

  5. [5]

    Yokoi, and J

    Nakayachi, K., R. Yokoi, and J. D. Goltz (2024). Human behavioral response to earthquake early warnings (EEW): Are alerts received on mobile phones inhibiting protective actions? Int. J. Disaster Risk Reduct. 105, 104401

    2024

  6. [6]

    McBride, S. K., H. Smith, M. Morgoch, D. Sumy, M. Jenkins, L. Peek, A. Bostrom, D. Baldwin, E. Reddy, R. de Groot, et al. (2022). Evidence-based guidelines for protective actions and earthquake early warning systems, Geophysics 87, no. 1, WA77–WA102

    2022

  7. [7]

    Be Aware

    McBride, S., J. Becker, and D. M. Johnston (2019). Exploring the barriers for people taking protective action during 2012 and 2015 New Zealand ShakeOut drills, Int. J. Disaster Risk Reduct. 37, 101150, doi: 10.1016/j.ijdrr.2019.101150. Acknowledgements We would like to thank Brian Williams, Mohammad Khidar, Ed Fernandez, for insightful remarks and suggest...

    work page doi:10.1016/j.ijdrr.2019.101150 2019

Pith tools

Reviewed July 13, 2026 · model on record in the stance chip above.