Pith. sign in

REVIEW 3 major objections 5 minor 16 references

Simultaneous observations of multiple ELVES and SPRITES at the Pierre Auger Observatory

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A cosmic-ray observatory reports the first simultaneous camera and fluorescence-detector observations of sprites and ELVES.

desk verdict Honest commissioning report; the first TLECAM-FD sprite detections are real, but timing accuracy makes the ELVES-SPRITE coincidences suggestive rather than established. read the letter →

arxiv 2507.11641 v1 pith:X2CC7Q3L submitted 2025-07-15 astro-ph.IM physics.ao-phphysics.plasm-ph

Roberto Mussa (for the Pierre Auger Collaboration) This is my paper · ORCID
classification astro-ph.IMphysics.ao-phphysics.plasm-ph
keywords transientluminouseventsELVESSPRITEShalosfluorescencedetectorDBSCANsimultaneousobservationsatmosphericelectricity
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 reports the first events in which two ground-based cameras and the fluorescence detector of a large cosmic-ray observatory recorded the same transient luminous events — fast expanding rings at the base of the ionosphere (ELVES) and slower discharges from 40 to 70 km altitude (sprites) — at the same time. The anchor event, at GPS time 1416802034 on 28 November 2024, was seen by two fluorescence telescopes and both cameras: a double ELVES followed by a sprite with a ring-like halo about 130 km across. The paper also reports five sprites on 7 January 2024 that occurred within about 0.1 s of ELVES, and seven sprites on 9 March 2024 with no ELVES in the same second, suggesting the two phenomena are not always causally tied. A Python pipeline based on DBSCAN is described that finds sprites by clustering bright pixels in frame differences and cuts the stored video by two orders of magnitude, so sprite studies no longer need to wait for a fluorescence trigger. If the detections hold up, this gives a new way to study how fast ionospheric flashes and slower sprite discharges are connected.

What carries the argument

The central object is the combined FD-plus-TLECAM observation chain. On one side, the fluorescence detector's dedicated ELVES trigger and 0.9 ms trace-length readout sample the expanding-ring light curves at microsecond scale, and the second-level trigger (T2) records clusters of five adjacent pixels hit in a row, providing time stamps and durations for the slower sprite pulses. On the other side, TLEcam-1 (a Sony α7-III with a 50 mm f/0.95 lens) and TLEcam-2 (a ZWO ASI294MC with a 20 mm f/1.4 lens) record video at 12.5 and 16 frames per second, respectively, in the same field of view. The automatic sprite-finding mechanism is DBSCAN, a density-based clustering algorithm that groups nearby bright pixels in subtracted frames into a single sprite candidate and keeps only about sixteen frames around each candidate, cutting 10 GB of video per 10 minutes to a manageable size. The time link between the two data streams is currently the weak point: frame times come from a PC clock and an assumed constant frame rate, with synchronization no better than 100 ms, so the paper uses T2 timestamps as a millisecond-scale reference for the camera events.

What would settle it

Record a GPS-locked LED pulse train in the TLECAM field of view while the FD triggers on ELVES; if the difference between camera frame timestamps and FD trigger timestamps jitters by more than about 100 ms or drifts with time since the start of a 5-minute file, the claimed ELVES-sprite coincidences would not be established.

Watch

Extended reading notes

Core claim

The paper's central claim is that the observatory can now observe sprites and ELVES simultaneously with two complementary instruments: the fluorescence detector, which samples light traces with high time resolution, and the TLECAMs, which add spatial resolution and longer integration. The load-bearing observation is the event of 28 November 2024, in which two FD telescopes (HEAT and Los Leones) and both cameras recorded a double ELVES followed by a sprite with a ring-like halo; from the FD light curves and second-level trigger data, the paper concludes that the halo and sprite began more than 1 ms after the ELVES. The paper further claims that the five near-coincident sprite-ELVES pairs seen on 7 January 2024, together with the sprites observed with no ELVES in the same second on other nights, indicate that the causal connection between ELVES and sprites may depend on thunderstorm type. It also claims that a DBSCAN-based automatic detector can identify sprites from camera frame differences alone, reducing the data volume by at least a factor of 100, and can operate without an FD trigger.

Load-bearing premise

The strongest temporal claims rest on the assumption that camera frame times and fluorescence-trigger times can be aligned well enough to call events simultaneous, even though the stated synchronization is no better than 100 ms, sprites last only about 5 to 100 ms, and TLEcam-1 frame times are derived from an assumed constant 12.5 frames per second rate.

Editorial extensions

If this is right

  • The 28 November 2024 event implies that a single lightning stroke can produce a double ELVES and, more than a millisecond later, a sprite surrounded by a ring-like halo about 130 km in diameter.
  • FD second-level trigger (T2) clusters give a millisecond-scale clock for sprite durations and for their offset from camera frame times, which is the best available substitute for the current 100 ms synchronization.
  • The DBSCAN pipeline can build sprite catalogs from camera data alone, without FD triggers, while shrinking stored data by two orders of magnitude.
  • Comparing near-coincident sprite-ELVES pairs with sprites that have no ELVES in the same second provides a direct test of whether the two TLE types share a common lightning driver.
  • With the cameras and FD running through a full storm season, the sample of simultaneous events should grow from a handful to a statistically usable set.

Reading between the lines

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

  • With millisecond-grade synchronization, the same setup could directly measure the delay between an ELVES and its accompanying sprite, testing whether both are driven by the same electromagnetic pulse or by a later component of the lightning discharge.
  • The frame-difference-plus-DBSCAN recipe is generic enough that it could be extended to halos and to blue or gigantic jets by adjusting thresholds and frame windows, giving a wider census of transient luminous events.
  • The ~130 km ring seen around the sprite on 28 November, if confirmed with better timing, would be a useful geometric constraint on mesospheric electromagnetic-pulse heating models, because the FD reconstruction fixes the source distance at about 950 km.
  • A full-season sample could quantify how often sprites occur with no ELVES in the same second, placing an upper bound on how tightly coupled the two phenomena are.
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

3 major / 5 minor

Summary. This paper reports on the first year of commissioning of two optical cameras (TLEcam-1 and TLEcam-2) at the Pierre Auger Observatory, intended to complement the Fluorescence Detector's existing ELVES and halo observations with spatially resolved, longer-integration imaging of Transient Luminous Events. The central claim is the first simultaneous detection of SPRITEs and ELVES by both the TLECAMs and the FD, in particular the 28 November 2024 event at GPS time 1416802034, where a double ELVES recorded by two FD telescopes (HEAT and Los Leones) is followed by a SPRITE with a ring-like halo captured by the cameras. The paper also reports earlier December 2023 and January 2024 SPRITE observations from TLEcam-1, and describes a Python/DBSCAN-based algorithm for automatic offline SPRITE detection that reduces the recorded data volume by two orders of magnitude. The authors transparently state that the present FD–camera synchronization is not better than 100 ms, that only one 2024 event had both an ELVES and a SPRITE in coincidence, and that the halo interpretation is based on the expected sensitivity limit of the cameras to ELVES at large distances.

Significance. If the simultaneity claim holds, the paper demonstrates a new multi-instrument capability for TLE studies at the Pierre Auger Observatory, combining the FD's sub-microsecond timing of ELVES with the cameras' spatial resolution and longer integration, and it provides a practical data-reduction pipeline for sprite surveys. The paper is commendably transparent about its limitations: the small event sample, the 100 ms synchronization uncertainty, and the preliminary nature of the halo identification are all explicitly acknowledged. The DBSCAN-based automatic selection method, while not yet real-time, is a useful contribution that appears to reduce stored data by more than two orders of magnitude. However, the central claim of simultaneity depends on camera timestamps whose accuracy is not demonstrated against an absolute reference, and the halo interpretation is not quantitatively supported. These are correctable issues, but they are load-bearing for the paper's main assertion, so revision is needed before the claim can be taken as established.

major comments (3)
  1. [Section 6 (and Section 3)] The claimed simultaneity of the ELVES and SPRITE observations rests on the camera timestamps, but the paper states that FD–camera synchronization is 'not better than 100 ms' while sprite durations are 5–100 ms. For TLEcam-1, every frame time is derived by assuming a constant 12.5 fps from the start of a 5-minute file, with the PC clock taken as Unix time; no GPS-disciplined time transfer or calibration against FD T2 timestamps is shown for this camera. Over one 5-minute file, a 0.1% frame-rate error or clock drift would accumulate to about 300 ms, larger than the stated coincidence window. The residual distribution in Fig. 7 is shown only for TLEcam-2 and lacks an RMS or drift estimate. Please provide a timing-calibration check, or alternatively demonstrate that the 28 November 2024 sequence and the January 2024 coincidences remain intact when camera times are shifted by ±100 ms (or by ±1 frame), and state the resulting uncertainty on each claimed coincidence.
  2. [Section 5, Fig. 6] The identification of the ring of light around the SPRITE as a halo rests on the statement that the cameras are not sensitive to ELVES beyond about 350 km, while the 28 November 2024 source was at about 950 km. No sensitivity estimate, image calibration, or comparison with the expected surface brightness of a halo at that distance is provided. Since the ring diameter is quoted as about 130 km, and since ELVES can appear as rings, the possibility that this feature is an ELVES or an ELVES-related scattering effect should be quantitatively excluded or the interpretation should be explicitly labeled as a tentative hypothesis rather than a conclusion.
  3. [Section 4] The DBSCAN-based automatic SPRITE detection algorithm is presented as a key deliverable, but the paper does not give the parameter values (e.g., eps, min_samples, the frame-difference threshold, or the lower-edge image cutoff) or any measure of detection efficiency and false-positive rate. For the method to be reproducible and its stated data-reduction factor meaningful, please specify the parameters and report the algorithm's performance on a set of known SPRITE events and on background-only data.
minor comments (5)
  1. [Abstract] 'higher space resolution' should be 'higher spatial resolution'.
  2. [Section 3] 'the first four SPRITEs events' should be 'the first four SPRITE events'.
  3. [Throughout] The camera names are written inconsistently as 'TLEcam-1', 'TLEcam-2', 'TLECAM-1', and 'TLECAM-2'; please unify the notation.
  4. [Section 5] The description 'HEAT positioned pointing downwards [10, 11], i.e. between 2 and 30 degrees' is unclear; please specify the elevation range and what 'downwards' means in this context.
  5. [Section 6, Fig. 7] The left panel of Fig. 7 would benefit from an explicit statement of the number of events, the RMS of the distribution, and whether any outliers or drifts were observed; this would directly address the synchronization concern.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is an observational commissioning report whose central claims rest on newly acquired camera and FD data, not on fitted inputs or self-citation chains.

full rationale

This paper presents first simultaneous observations of ELVES and SPRITES with new TLECAM cameras and the Auger FD. There is no derivation of a predicted quantity from fitted parameters. The only self-references are to previous Auger work on the ELVES trigger and reconstruction geometry, used as background or as a standard method for locating the ELVES center; these are not the paper's target result, and the claimed coincidences are new observations. The paper explicitly discloses the timing limitations (synchronization not better than 100 ms, TLEcam-1 frame timing assumed at 12.5 fps), which is an uncertainty/correctness concern rather than circularity. No equation reduces to another by construction, no fitted quantity is renamed as a prediction, and no argument rests on an unverified self-citation. Therefore the circularity score is 0.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The central claim does not depend on a fitted physical model; the free parameters are all in the detection pipeline and were not calibrated or reported. The main domain assumptions are the geometric geometry and instrument sensitivity limits used to classify the events.

free parameters (4)
  • Frame-difference threshold
    Section 4: pixels above 'a certain threshold' are converted to 2D vectors; the threshold value and its dependence on sky brightness are not given.
  • DBSCAN cluster parameters
    Section 4: DBSCAN is used to cluster thresholded pixels, but eps and min_samples are not stated, so detection efficiency is not reproducible.
  • Lower-edge image cutoff
    Section 4: the dominant background from Rayleigh-scattered lightning is reduced by 'properly adjusting the lower edge of the analysed image', a hand-tuned spatial cut.
  • TLEcam-1 frame rate assumption = 12.5 fps
    Section 6: frame times are derived by assuming a constant 12.5 frames per second recording speed, which affects any timing comparison with the FD.
assumptions (3)
  • domain assumption Earth's curvature blocks direct lightning light for sources more than 250 km away, so FD triggers see only scattered or longer light paths.
    Section 1: this geometric premise justifies why ELVES are observed as rings rather than direct flashes.
  • domain assumption TLEcam sensitivity to ELVES is limited to less than 350 km, while sensitivity to halos and sprites extends farther.
    Section 5: the ring seen by TLEcam-2 for a source at about 950 km is interpreted as a halo because the cameras are assumed insensitive to ELVES beyond 350 km; the sensitivity limit is not measured in the paper.
  • domain assumption The FD-TLEcam time offset can be treated as roughly constant within a GPS second, and T2 clusters mark sprite pulses.
    Section 6: timing distributions use the first T2 in the same GPS second as reference, while absolute synchronization is stated to be no better than 100 ms.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Simultaneous observations of multiple ELVES and SPRITES at the Pierre Auger Observatory." pith.science (2026). https://pith.science/paper/X2CC7Q3L

@misc{pith2026250711641,
  author       = {Pith},
  title        = {Pith review of: Simultaneous observations of multiple ELVES and SPRITES at the Pierre Auger Observatory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/X2CC7Q3L}},
  note         = {Machine review of arXiv:2507.11641}
}
read the original abstract

Since 2014, the Pierre Auger Observatory has exploited a dedicated trigger and its very high time resolution to study ELVES and harvest record samples of multiple ELVES using the Fluorescence Detector (FD). In 2017, after extending the readout of trace lengths to 0.9 ms, we started observing other types of light transients from the base of the ionosphere, such as HALOS, which deserved further investigation. In December 2023 and April 2024, we installed two additional cameras (TLECAMs), which allow us to perform simultaneous detection of these transients with higher space resolution and longer integration times. Here, we present our first simultaneous observations of SPRITES and ELVES by both TLECAMs and FD. Furthermore, we describe the Python algorithm based on DBSCAN to automatically detect SPRITES in the videos recorded by our TLECAMs and acquire data efficiently without needing the FD trigger.

Figures

Figures reproduced from arXiv: 2507.11641 by the authors.

Figure 1
Figure 1. The new TLECAMs: Sony 𝛼7-III (left), ZWO ASI294MC (center), mounted in parking position with the field of view on the back (right). The events triggered by the Auger Observatory are produced by lightning sources far enough from the FD, i.e. more than 250 km away, so that the earth’s curvature prevents the direct light from reaching our sensors. In 2013, we developed a dedicated trigger and readout scheme to study th… view at source ↗
Figure 2
Figure 2. Sony camera alignment with brightest stars. Green crosses indicate the elevation of a potential ELVES center (h=90 km) as a function of the distance of the lightning source. Yellow crosses indicate the elevation of the top of a 15 km high cloud as a function of the distance of the lightning source. The colored hexagons show the field of view of each FD pixel. Various other types of “transient luminous events” (TLEs)… view at source ↗
Figure 3
Figure 3. The SPRITE image recorded by TLEcam-1 at GPStime 1388643594 overlapped with the final frame of the ELVES detected by the Coihueco FD. The Sony camera (named TLEcam-1) was installed in December 2023 and could take data during the Austral summer 2023-2024. The ZWO camera (named TLEcam-2) was installed in April 2024, after the storm season, and the first observation of a TLE (consisting of an ELVE, followed by a SPRITE… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Left: SPRITE image, overlapped to the pixels grid. Right: the traces corresponding to the indicated pixels, showing the initial part of the SPRITE pulse. 4. Python code for automatic detection of TLE From the first months of commissioning, the observation of a very lar…
Figure 5
Figure 5. Figure 5: Six SPRITEs recorded in less than half an hour on March 9, 2024. and (b) finally to numerical arrays (NUMPY.array), (c) to create 2D-vectors above a certain threshold (NUMPY.argwhere), and (d) to search for clusters of points in the resulting arrays (SKlearn.DBSCAN). T…
Figure 6
Figure 6. Figure 6: Left: Sprite detected at GPS time = 1416802034, overlapped on the last image recorded by the Los Leones FD, corrected for the parallax relative to Coihueco; Right: the result of the subtraction of two adjacent frames shows not just the SPRITE but also a ring of light w…
Figure 7
Figure 7. Figure 7: Left: the difference between the time stamp of the SPRITE labeled by TLEcam-2 and the first T2 recorded by the FD. Right: the time duration of a SPRITE as measured from the T2s in the FD. the TLECAMs described in this paper have been manually controlled to follow the e…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 15 canonical work pages

  1. [1]

    U. S. Inan, T. F. Bell, J. V. Rodriguez, Geophys. Res. Lett.18, (1991) 705

    work page 1991

  2. [2]

    W. L. Boecket al., Geophys. Res. Lett.19, (1992) 99

    work page 1992

  3. [3]

    Abrahamet al

    J. Abrahamet al. [Pierre Auger Coll.], Nucl. Instrum. Meth. A620 (2010) 227

    work page 2010

  4. [4]

    Mussaet al

    R. Mussaet al. [Pierre Auger Coll.],PoS(ICRC2019) 1131

    work page

  5. [5]

    Vásquez Ramírezet al

    A. Vásquez Ramírezet al. [Pierre Auger Collaboration],PoS(ICRC2021) 327

    work page

  6. [6]

    A. S. Tonachiniet al. [Pierre Auger Coll.] (2011) doi:10.7529/ICRC2011/V11/0878

    work page doi:10.7529/icrc2011/v11/0878 2011

  7. [7]

    Mussaet al

    R. Mussaet al. [Pierre Auger Coll.], Eur. Phys. J. Plus127(2012) 94

    work page 2012

  8. [8]

    Aabet al

    A. Aabet al. [Pierre Auger Coll.], arXiv:1307.5059 [astro-ph.HE]

    arXiv

Show all 16 references
  1. [9]

    Aabet al

    A. Aabet al. [Pierre Auger Coll.], Earth Space Sci.7, no.4, (2020) e2019EA000582

    2020

  2. [10]

    Abdul Halimet al

    A. Abdul Halimet al. [Pierre Auger], PoSICRC2023(2023), 372

    2023

  3. [11]

    T. H.-J. Matheset al. [Pierre Auger Coll.], doi:10.7529/ICRC2011/V03/0761

    doi

  4. [12]

    C . T. R. Wilson, Proc. Phys. Soc. London 37 (1924) 32D

    1924

  5. [13]

    R. C. Franz, R. J. Nemzek, and J. R. Winckler, Science249(1990) 48

    1990

  6. [14]

    E. M. Wescottet al., Geophys. Res. Lett.,22 (1995) 1209

    1995

  7. [15]

    Nicoraet al., Proc

    G. Nicoraet al., Proc. 35th Int. Conf. on Lightning Protection (ICLP),1 (2021) 01-06

    2021

  8. [16]

    Tor Vergata

    Y. Zhuet al.Remote Sens.14 (2021) 2209. 8 Investigating ELVES, SPRITES and halos with the Pierre Auger FDs Roberto Mussa The Pierre Auger Collaboration A. Abdul Halim13, P. Abreu72, M. Aglietta54,52, I. Allekotte1, K. Almeida Cheminant70, A.Almela 7,12,R.Aloisio 45,46,J.Alvare...

    2021

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.