Pith. sign in

REVIEW 4 major objections 8 minor 12 references

A Three-dimensional Reconstruction of Cosmic Ray Events in IceCube

T0 review · 4 major / 8 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read IceCube's new three-dimensional reconstruction, which fits surface and in-ice signals with one unified likelihood, recovers cosmic-ray showers whose cores land outside IceTop and points them as well as or better than either array alone.

desk verdict Useful methods paper from IceCube on a joint 3-D reconstruction, but the headline IT-Uncontained gain is asserted on an uncontrolled QC comparison and needs a matched-sample check. read the letter →

arxiv 1908.07582 v1 pith:OSJBYXWE submitted 2019-08-20 astro-ph.HE

Xinhua Bai , Emily Dvorak , Dennis Soldin , Javier Gonzalez (for the IceCube Collaboration) This is my paper · ORCID
classification astro-ph.HE
keywords cosmicrayairshowersIceCubeTopin-icearrayeventreconstructionlikelihoodmaximizationhybriddetectoranalysismuonbundles
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper introduces a three-dimensional reconstruction for cosmic-ray air showers detected by IceCube's surface array IceTop and the deep in-ice array, jointly fitting signals from both detectors by minimizing one unified likelihood function. The authors claim that this combined fit reconstructs the shower core and arrival direction of events whose cores land inside, at the edge of, or outside IceTop, recovering a substantial fraction of IT-Uncontained events that standard single-array reconstructions lose. A sympathetic reader would care because cosmic-ray rates fall steeply with energy, so every recovered event matters for spectrum, anisotropy, and composition studies at high energy. On Monte Carlo events with the 2012 detector configuration, the 3-D reconstruction achieves pointing resolution better than or comparable to IceTop-alone or in-ice-alone reconstructions, extends usable zenith angles beyond those of contained events, and adds new shower parameters that may sharpen energy and mass estimators.

What carries the argument

The load-bearing object is the unified likelihood function $L = L^{\rm IceTop}_Q \otimes L^{\rm IceTop}_t \otimes L^{\rm InIce}_{\rm SPE}$, whose three terms describe the signal charge and arrival time in IceTop tanks and the photon arrival times in the in-ice DOMs. The shower signal is modeled by the lateral distribution function $S(r) = S_{\rm ref}(r/r_{\rm ref})^{-\beta - 0.30264\log_{10}(r/r_{\rm ref})}$ and a curvature term $\delta t(r) = c_t r^2 + 19.41(1-e^{-r^2/(2(118.1)^2)})$, with $c_t$ left free per event; the per-station time fluctuation is redefined as $\sigma_{t_i}=C_1\sqrt{\sum_{j=1}^2(t_{ij}-\frac{t_{i1}+t_{i2}}2)^2}/(\sum_j Q_{ij})+C_2$ with MC-derived constants. A likelihood combiner feeds these terms to Minuit or SIMPLEX, seeded by parameter and boundary services, and the reconstructed track is then passed to a Millipede-based energy estimator for muon energy losses. The mechanism works because the in-ice term provides a long lever arm that anchors the direction, while the IceTop terms constrain the core and lateral profile, so events outside the surface array are no longer stranded.

What would settle it

Apply the 3-D reconstruction to real IceCube data for IT-Uncontained events and compare reconstructed core positions against an independent in-ice-only muon track: if reconstructed cores still cluster toward the IceTop center, or if the expected gain in recovered events over in-ice-alone reconstruction does not appear, the central claim is refuted. A sharper test would check the improved pointing resolution using the Moon shadow or a known source as an independent arrival-direction reference.

Watch

Extended reading notes

Core claim

The paper's central claim is that a single likelihood function combining IceTop charge and time information with in-ice photon-arrival information reconstructs the full three-dimensional air-shower footprint more reliably than either detector alone. On Monte Carlo showers from CORSIKA with Sibyll2.1, the 3-D reconstruction determines the shower core, arrival direction, lateral distribution parameter $S_{125}$, slope $\beta$, and curvature parameter $c_t$ by maximizing this combined likelihood. For cores contained in IceTop it gives the best 68% pointing resolution of the three methods across the studied energy range, while its core resolution is about 2-3 m worse than IceTop-alone, which widens the $S_{125}$-energy relation. For cores on or outside IceTop, where IceTop-alone systematically pulls reconstructed cores inward and in-ice-alone is biased outward, the combined fit finds the true core and raises the successful reconstruction rate after quality cuts from 13.7% (in-ice alone) to 25.3% in the highest energy bin, with similar large gains at lower energies. This event recovery is the paper's central discovery: the two arrays together see the three-dimensional footprint that neither alone reconstructs reliably.

Load-bearing premise

The load-bearing premise is that the CORSIKA/Sibyll2.1 Monte Carlo and the 2012 IceCube detector simulation reproduce the real spatial and temporal signals of showers with cores outside IceTop, so that the gains measured on Monte Carlo events transfer to real data.

Editorial extensions

If this is right

  • For IT-Contained proton Monte Carlo events, the 3-D reconstruction gives the best 68% pointing resolution of the three methods over the whole studied energy range, at the cost of a 2-3 m worse core resolution than IceTop-alone.
  • For IT-Uncontained events, the successful reconstruction rate after quality cuts rises from 13.7% (in-ice alone) to 25.3% in the $10^9$-$10^{9.5}$ GeV bin, directly reducing statistical uncertainty at high energy.
  • Recovered IT-Uncontained events extend to zenith angles near 60 degrees, versus about 30 degrees for contained events, letting the detector sample higher-energy showers near their $X_{\rm max}$.
  • The reconstruction yields per-event curvature $c_t$, an electromagnetic/muonic LDF option, and muon-bundle energy-loss information, opening new handles on primary mass and composition.
  • The widened $S_{125}$-energy relation caused by the slightly worse core resolution marks the main price of the method and a target for future optimization.

Reading between the lines

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

  • If the Monte Carlo-demonstrated pointing improvement holds in data, the 3-D reconstruction should sharpen IceCube's cosmic-ray anisotropy maps, especially for events whose cores are outside IceTop, since those events currently enter anisotropy analyses with much larger directional errors.
  • A testable extension would be a two-stage fit that first obtains the direction from the unified likelihood and then re-fits the core using only IceTop charge information; this could recover the IceTop-alone core resolution while keeping the improved pointing.
  • The paper describes the likelihood-combiner architecture as open to additional observables, so the same approach could be applied to future surface/deep hybrid arrays where the surface density and in-ice lever arm differ.
  • Because the mean $S(r_{\rm ref})$ versus energy for IT-Uncontained events is shown to separate by primary mass and zenith, a mass estimator built from these relations may extend composition measurements to energies and angles previously inaccessible.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 8 minor

Summary. This paper presents a new three-dimensional (3-D) maximum-likelihood reconstruction for cosmic-ray events in IceCube that jointly uses IceTop surface signals and in-ice muon signals. The algorithm combines a charge and time likelihood for IceTop tanks with an in-ice photon-arrival likelihood, and is tested on CORSIKA/Sibyll2.1 Monte Carlo with proton and iron primaries for the 2012 detector configuration. Two event classes are considered: IT-Contained (core inside IceTop, muon contained in the in-ice detector) and IT-Uncontained (core outside/at edge, muon contained). The paper reports that the 3-D reconstruction achieves better 68% pointing resolution than IceTop-alone or InIce-alone reconstructions for both event classes (except below ~10^6.6 GeV for IT-Uncontained), recovers more IT-Uncontained events, and provides additional EAS parameters, at the cost of a slightly worse core position resolution (2–3 m) and a wider S125–energy spread.

Significance. If the reported performance holds, the 3-D reconstruction would increase the number of usable high-energy cosmic-ray events in IceCube by recovering events with cores outside IceTop, and would improve angular resolution through the IceTop–in-ice lever arm. The paper clearly describes the software architecture and makes concrete, falsifiable MC-based predictions. However, the central performance comparisons for IT-Uncontained events are not controlled for event selection, and the success-rate claims lack a definition and uncertainties. With the missing identical-QC analysis added, the paper would be a useful methods contribution to the field.

major comments (4)
  1. [Section 3, Figure 3 (right)] The IT-Uncontained pointing resolution comparison is uncontrolled: the solid lines are results after each reconstruction's own QC, and the identical-QC dotted-line comparison shown for IT-Contained events (left panel) is absent for the IT-Uncontained case. Because the 3-D QC requires at least five triggered IceTop stations while the InIce-alone reconstruction has no such requirement, the apparent 3-D advantage above about 10^6.6 GeV may be a selection effect rather than a property of the unified likelihood. Please provide the identical-QC comparison for IT-Uncontained events, or explicitly limit the claim to the separate-QC scenario.
  2. [Section 3, 'successful reconstruction rates' paragraph] The quoted success rates for IT-Uncontained events (e.g., 62.6% vs 63.5% in the first bin) are not defined, are quoted without statistical uncertainties, and are evaluated after different QC for each reconstruction. This prevents a quantitative assessment of the claimed event-recovery improvement. Please define what constitutes a 'successful' reconstruction, provide uncertainties (the lower-energy bins appear to have limited MC statistics), and present a comparison on a common event set.
  3. [Section 3, Figure 2] The statement that 'only the 3-D reconstruction ... is able to correctly find the core' is based on events passing each method's own QC; for example, IceTop-alone retains only 2.4% of IT-Uncontained events in the lowest energy bin. The figure thus conflates reconstruction accuracy with selection effects. A comparison on an identical event sample is required to support the claim that the 3-D method removes the systematic core bias.
  4. [Section 2, 'Likelihood Combiner'] The combined likelihood function is not explicitly defined: the paper does not state whether the IceTop charge, IceTop time, and in-ice likelihoods are summed in log-space with unit weights, or how the relative normalizations are determined. Since the combination is the core of the new method, a precise definition is needed for reproducibility. Also, the derivation of C1 and C2 in Eq. (2.3) from 'a MC study' is described only briefly; please specify the procedure and whether these constants are fixed before evaluating the performance metrics.
minor comments (8)
  1. [Equation (2.2)] The notation '2· (118.1))2' is ambiguous; please clarify the intended expression, likely 2(118.1)^2.
  2. [Figure 3 caption] Please state explicitly that the right panel does not include the identical-QC comparison, or add it.
  3. [Section 2, after Eq. (2.3)] The sentence 'The values of C1 and C2 derived from a MC study' should state the MC sample and fitting procedure.
  4. [Section 3] The phrase 'Events are weighted to a E^-2.7 spectrum' should read 'an E^-2.7 spectrum'.
  5. [Section 3, Figure 2 text] The phrase 'extraordinary improvement' is subjective; replace it with a quantitative statement of the systematic-core-bias reduction.
  6. [Abstract and Section 4] The abstract and Section 4 state the advantages in present tense without noting that all performance numbers are from Monte Carlo; please add a qualifier such as 'in simulation'.
  7. [Reference [11]] Reference [11] (arXiv:1906.04317) should be cited with its title and venue if available.
  8. [Figure 5] The colorbar label 'relative intensity after the spectrum is weighted to E^-2.7' is unclear; please define the normalization.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the 3-D reconstruction is an independent likelihood-based method whose performance is evaluated against Monte Carlo truth.

full rationale

The paper's central claim is that a new 3-D reconstruction, combining IceTop and in-ice likelihoods, improves cosmic-ray event reconstruction relative to IceTop-alone or InIce-alone reconstruction. This claim is tested against Monte Carlo truth, not against the algorithm's own fitted outputs: reconstructed directions, cores, and S125 values are compared directly with true MC geometry and energy (Section 3, Figures 2-6). The only explicitly fitted quantities, C1 and C2 in Eq. (2.3), are calibration constants internal to the time-fluctuation likelihood and are not renamed as predictions; no result in the paper reduces to them by construction. The paper's self-citations to IceCube and AMANDA detector/reconstruction papers provide standard signal models and likelihood descriptions, but none is invoked as a uniqueness theorem or as the sole justification for a claimed prediction. The IT-Uncontained comparison uses different quality cuts for 3-D versus InIce-alone, and the paper does not show an identical-QC dotted-line comparison for that subset, so the comparison may be affected by event selection; however, this is a statistical fairness concern, not a circular derivation. The paper does show identical-QC comparisons for IT-Contained events (Figure 3 left, Figure 4 dotted lines), demonstrating that the method's advantage there is not solely an artifact of its own selection. Therefore the derivation chain is self-contained with respect to circularity, and the appropriate finding is no significant circularity.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central claim rests on the fidelity of the Monte Carlo simulation chain and the validity of previously published detector response models. The only new fitted constants are C1 and C2 in the time fluctuation model. No new physical entities are introduced.

free parameters (2)
  • C1 = 4.0 VEM
    Coefficient in the per-event time fluctuation model (Eq. 2.3), derived from a Monte Carlo study.
  • C2 = 1.22 ns
    Additive constant in the same time-fluctuation model, derived from Monte Carlo.
assumptions (3)
  • domain assumption CORSIKA with Sibyll2.1 and the IceCube detector simulation of the 2012 configuration accurately model air showers and detector response.
    The performance comparisons and claimed resolutions in Section 3 are evaluated entirely against these MC simulations, so any mismatch between MC and real detector response would propagate into the reported improvement.
  • domain assumption The IceTop LDF, curvature, and time-fluctuation models of Eqs. (2.1)-(2.3) are valid signal models for EAS particles in the surface tanks.
    The combined likelihood's IceTop terms inherit these models from [4] with a modified free curvature parameter and a per-event time fluctuation fitted to MC.
  • domain assumption The in-ice likelihood functions (SPE/MPE) from [5,8] and the Millipede muon energy loss estimator from [7,8] accurately describe photon arrival times and muon energy losses in deep ice.
    The InIce term of the combined likelihood and the energy estimator rely on these referenced models without re-derivation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Three-dimensional Reconstruction of Cosmic Ray Events in IceCube." pith.science (2026). https://pith.science/paper/OSJBYXWE

@misc{pith2026190807582,
  author       = {Pith},
  title        = {Pith review of: A Three-dimensional Reconstruction of Cosmic Ray Events in IceCube},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OSJBYXWE}},
  note         = {Machine review of arXiv:1908.07582}
}
read the original abstract

The IceCube Neutrino Observatory at the geographic South Pole consists of two components, a km2 surface array IceTop and a km3 in-ice array between 1.5 and 2.5 km below the surface. Cosmic ray events with primary energy above a few tens of TeV may trigger both the IceTop and in-ice array and leave a three-dimensional footprint of the electromagnetic and muonic components in the extensive air shower. A new reconstruction based on the minimization of a unified likelihood function involving quantities measured by both IceTop and in-ice detectors was developed. This report describes the new reconstruction algorithm and summarizes its performance tested with Monte Carlo events under two different containment conditions. The advantages of the new reconstruction are discussed in comparison with reconstructions that use IceTop or in-ice data separately. Some possible improvements are also summarized.

Figures

Figures reproduced from arXiv: 1908.07582 by the authors.

Figure 1
Figure 1. The key software components of the 3-D reconstruction and the intrinsic data/information flow through them. The final reconstructed EAS quantities (from the "Fitter") and muon energy losses (from the "InIce Energy Estimator") are used for physics analysis. See details of some of the indicated variables and functions in the text. 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Reconstructed versus Monte Carlo true EAS core distance from the center of the IceTop array after QC for all processed proton events. The diago￾nal black dash dotted lines indicate a perfect match between the reconstructed and the true core values. Points above/below the dash dotted line correspond to EAS cores dragged artificially away/toward IceTop ar￾ray center. To compare how the MC events can be effectively rec… view at source ↗
Figure 3
Figure 3. Pointing resolution by three reconstructions for IT-Contained (left) and IT-Uncontained (right) proton events. Solid lines are the results from each reconstruction after their own QC separately. The dotted lines in the plot on the left represent a comparison in which all three reconstructions use the same set of events after identical QC. contains 68% of the reconstructed tracks. As shown by the red lines in the plo… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Comparison of EAS core res￾olution from IceTop-alone, InIce-alone and the new 3-D reconstructions for IT￾Contained proton events. Solid lines are the results from each reconstruction af￾ter their own QC separately. The dotted lines in the plot are the results from thes…
Figure 5
Figure 5. Figure 5: Reconstructed S125 versus CR true primary energy for IT-Contained proton events after exactly the same QC for both methods. Plots on the left and right are for the IceTop-alone reconstruction and the 3-D reconstruction respectively. Numbers on the colorbars represent t…
Figure 6
Figure 6. Figure 6: Value of EAS ground particle LDF at two core distances (rre f =125 m, 175 m) versus CR pri￾mary energy for IT-Uncontained proton events. Solid lines (with error bars for standard deviation) are for events that have zenith angle 0°<θ<10° and dashed lines (with shaded ba…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 9 canonical work pages

  1. [1]

    IceCube: South Pole Neutrino Observatory, https://icecube.wisc.edu

    work page

  2. [2]

    Dawn Williams, for the IceCube Collaboration, PoS(ICRC2019)016

    work page

  3. [3]

    Dennis Soldin, for the IceCube Collaboration, PoS(ICRC2019)014

    work page

  4. [4]

    Abbasi, et al

    R. Abbasi, et al. (IceCube Collaboration), Nucl. Instrum. Meth. A700 (2013) 188-220, arXiv:1207.6326

    arXiv 2013

  5. [5]

    Muon Track Reconstruction and Data Selection Techniques in AMANDA

    J. Ahrens, et al. (AMANDA Collaboration), Nucl. Instrum. Meth. A 524 (2004) 169-194, arXiv:astro-ph/0407044

    work page Pith review arXiv 2004

  6. [6]

    James, M

    F. James, M. Roos, Computer Physics Communications 10 (1975) 343-367; Fred James and Matthias Winkler, MINUIT User’s Guide, INSPIRE-1258345

    work page 1975

  7. [7]

    The IceCube Collaboration: contributions to the 30th International Cosmic Ray Conference (ICRC 2007)

    S. Grullon, D. J. Boersma, G. Hill, K. Hoshina, K. Mase, in Proceedings of the 30 th International Cosmic Ray Conference, Merida, Yucatan, Mexico, 2007, pp. 1457-1460, arXiv:0711.0353

    work page Pith review arXiv 2007

  8. [8]

    Aartsen, M.G. et al. (IceCube Collaboration), JINST 9 (2014) P03009, arXiv:1311.4767

    arXiv 2014

Show all 12 references
  1. [9]

    CORSIKA: An Air Shower Simulation Program, http://www-ik.fzk.de/corsika/

  2. [10]

    E. Ahn, R. Engel, T. Gaisser, P. Lipari, and T. Stanev, Phys. Rev. D 80, 94003 (2009)

    2009

  3. [11]

    M. G. Aartsen, et al. (IceCube Collaboration), arXiv:1906.04317

    1906 arXiv

  4. [12]

    dissertation, Ghent University, 2013; Sam De Ridder, Ph.D

    See details of the formulism and previous work on muon bundle energy loss in deep ice in Tom Feusels, Ph.D. dissertation, Ghent University, 2013; Sam De Ridder, Ph.D. dissertation, Ghent University, 2019. 7

    2013

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.