Radio Flares from Collisions of Neutron Stars with Interstellar Asteroids

Abraham Loeb; Amir Siraj

REVIEW 3 major objections 5 minor 17 references

Radio Flares from Collisions of Neutron Stars with Interstellar Asteroids

T0 review · 3 major / 5 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 Collisions between neutron stars and interstellar asteroids could produce roughly ten observable one-jansky radio flares per day in the Milky Way.

desk verdict A transparent, testable rate prediction for a new class of Galactic radio transient, but the headline number is an order-of-magnitude guess built on a two-object extrapolation. read the letter →

arxiv 1908.11440 v2 pith:B4JESY3K submitted 2019-08-26 astro-ph.HE astro-ph.EP

Amir Siraj , Abraham Loeb This is my paper · ORCID

classification astro-ph.HEastro-ph.EP

keywords interstellarasteroidsneutronstarsmillisecondradioflaresfastburstsrotatingtransientstidaldisruptionOumuamua

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 argues that an interstellar asteroid falling onto a neutron star is tidally shredded, and the stripped electrons radiate a millisecond-long burst of coherent radio emission near 1 GHz. It estimates that collisions of this kind should occur often enough in the Milky Way to produce roughly ten flares per day at a flux threshold of about one jansky. That rate would make neutron star–interstellar asteroid impacts a new class of non-repeating, millisecond-duration radio transients, distinct from cosmological fast radio bursts. If the estimate holds, the flares would also provide a way to measure the abundances and kinematics of both neutron stars and interstellar asteroids.

What carries the argument

The rate estimate is carried by a chain of three ingredients: the coherent-curvature-radiation luminosity of a neutron star–asteroid impact (millisecond duration, ~1 GHz emission), the gravitational-focusing cross section for such impacts, and an interstellar-asteroid abundance factor $\zeta_{\mathrm{ISA}}$ defined as the ratio of interstellar asteroids to stars. That factor is calibrated from two known interstellar objects and assumed to be dominated by iron-rich bodies; the per-neutron-star event rate is $\dot N_{\mathrm{flare,NS}} = \zeta_{\mathrm{ISA}} n_\star \sigma_a v_{\mathrm{rel}}$, with $v_{\mathrm{rel}}$ drawn from a Monte Carlo sampling of stellar and neutron-star velocity distributions. The minimum asteroid radius that survives to be tidally disrupted, $r_{\min}\sim1$ m, sets the low-size cutoff and shapes the final fitting function.

What would settle it

A year-long all-sky radio survey at 1 GHz with sensitivity near 1 Jy that finds no non-repeating millisecond flares with dispersion measures consistent with Milky Way distances would contradict the claimed $\sim 10$ per day rate; alternatively, a direct measurement of the meter-scale interstellar asteroid density would rescale the prediction and test the same assumption.

Watch

Extended reading notes

Core claim

The central claim is that neutron star–interstellar asteroid collisions happen often enough in the Milky Way to produce a detectable all-sky rate of $\sim 10\,\mathrm{day}^{-1}$ at $\sim 1\,\mathrm{GHz}$ with a flux threshold of $\sim 1\,\mathrm{Jy}$. The calculation uses the existing model of an asteroid being tidally disrupted in the strong magnetic field of a neutron star, with the stripped electrons emitting coherent curvature radiation for about a millisecond. The event rate combines the impact cross section with the relative velocity of the two populations and an interstellar asteroid number density calibrated from the known interstellar objects 'Oumuamua and CNEOS 2014-01-08. The resulting rate is summarized by a piecewise fitting function of the minimum asteroid radius, and the paper notes that the same events would not produce detectable X-rays and would be far too rare to explain cosmological fast radio bursts.

Load-bearing premise

The predicted rate is directly proportional to the assumed number density of meter-sized iron interstellar asteroids in the Galaxy, a quantity extrapolated from just two known interstellar visitors; if that density is off by a factor of ten, the daily rate changes by the same factor.

Editorial extensions

If this is right

The rate of observable flares is roughly $10\,\mathrm{day}^{-1}$ at 1 Jy and 1 GHz, so the prediction is within reach of existing radio transient searches.
Each flare should be a one-off, millisecond-duration event, so non-repeating single radio pulses are the expected observational signature.
No X-ray counterpart is expected for meter-sized asteroids, distinguishing these flares from other neutron star transients.
The estimated rate is too low to explain cosmological fast radio bursts, so the events would constitute a separate, Galactic population.
Detections would simultaneously calibrate the interstellar asteroid number density and the neutron star population in the Milky Way.

Reading between the lines

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

A null result from archival single-pulse searches could be inverted to place an upper limit on the meter-scale interstellar asteroid abundance, because the predicted flare rate is linearly proportional to that abundance.
A direct measurement of the meter-scale interstellar asteroid density from future surveys would rescale the predicted daily rate by the same factor, making the headline number a testable population constraint rather than a fixed prediction.
The mechanism may offer a clean observational division: non-repeating bursts from isolated impacts versus repeating bursts from neutron stars passing through dense asteroid belts, which the paper notes as a prior explanation for repeaters.
Simultaneous X-ray and radio observations of any candidate flare could test the predicted absence of X-ray emission, because that absence is a specific consequence of the chosen impact model.

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.

Desk Editor's Note

read the letter

Two things to know. First, the paper is a clean, short proposal: apply Dai et al.'s radio flare mechanism to collisions between neutron stars and interstellar asteroids, compute a Galactic rate. The result is a new predicted transient class — non-repeating millisecond radio flares at ~1 GHz with no X-ray counterpart. That part is genuinely new and worth taking seriously as a target for CHIME-like surveys. Second, the headline "~10 day^-1" is not a robust prediction. It scales linearly with an interstellar asteroid abundance extrapolated from two objects down to meter scales, multiplied by an assumed 5% iron fraction. No error bars are propagated anywhere. A factor of ten in the meter-scale abundance changes the rate by the same factor, and a 1% iron fraction alone would drop it to ~0.5 day^-1.

What the paper does well: the derivation is transparent; each step is stated and referenced; the Monte Carlo for relative velocities is reasonable; the dependence on rmin is shown; and the authors are honest that the rate is too small to explain FRBs and that the RRAT abundance comparison is difficult. The emission mechanism and cross section are external, so the main calibration comes from their own earlier work — that is not a flaw in itself, since it is an independent calibration from Earth impact data, not from radio flares.

Soft spots: (1) No uncertainty analysis at all. A one-parameter sensitivity scan would have been easy and would honestly show the rate could be anywhere from a fraction of a day to tens per day. (2) The ISA kinematic model assumes thin-disk stellar velocities; interstellar objects are not necessarily born in the thin disk. This likely matters less than the abundance normalization but is still an stated assumption with no range. (3) The Discussion says the collisions "could reliably power" the flares. Given (1), "reliably" is too strong.

Is the central idea sound? Yes, as an order-of-magnitude estimate. The math checks out; the rate is built from a straightforward convolution of known-ish distributions. The paper would be stronger with an explicit error budget and a less confident abstract.

Who this is for: anyone working on radio transients, RRATs, FRB progenitors, or interstellar object populations. It deserves a serious peer review, not a desk reject. If I were an editor, I'd send it out with a request that the authors add a brief uncertainty discussion and temper the abstract.

Referee Report

3 major / 5 minor

Summary. This paper proposes that collisions between neutron stars (NSs) and interstellar asteroids (ISAs) such as 'Oumuamua produce millisecond-duration, ~1 GHz radio flares via the coherent curvature radiation mechanism of Dai et al. (2016). The authors derive a rate of such flares detectable at ~1 Jy by combining a Monte Carlo calculation of NS-ISA relative velocities, a model of the Milky Way's stellar and NS distribution, and an ISA abundance calibration from the authors' prior work based on two interstellar objects. They report an all-sky rate of ~10 day^-1 and suggest these events could constitute a subclass of non-repeating Rotating Radio Transients.

Significance. If the rate estimate holds, the paper identifies a new, potentially observable class of Galactic radio transients and a novel probe of both NS and ISA populations. The calculation is transparent, builds on published emission and cross-section results, and makes a falsifiable prediction (millisecond, non-repeating, ~1 Jy radio flares at ~10/day) that can be tested with current and upcoming radio surveys. The main weakness is that the dominant input—the meter-scale ISA number density—is extrapolated from two objects and carries an unquantified, likely order-of-magnitude uncertainty; the headline rate should therefore be read as an order-of-magnitude estimate rather than a precise prediction.

major comments (3)

[Section 3, Eq. (3), and Eq. (8)] The predicted all-sky rate is directly proportional to ζISA, the ISA-to-star abundance ratio, which is calibrated from only two objects ('Oumuamua and CNEOS 2014-01-08). The calibration in Eq. (3) extrapolates a power law with slope -3.4 from these objects down to r=1 m and multiplies by an assumed 5% iron fraction, with no error bars. Since Eq. (8) is linear in ζISA, the abstract's '~10 day^-1' inherits at least an order-of-magnitude uncertainty; for example, reducing the iron fraction from 5% to 1% alone lowers the rate to ~1.5 day^-1, and a factor-of-10 error in the meter-scale abundance changes the rate to ~1 day^-1. The authors should either propagate these uncertainties, present a sensitivity analysis over a plausible range of ζISA and iron fraction, or explicitly reframe the result as an upper limit under a stated set of assumptions.
[Section 4, Eq. (9), and Abstract] The abstract's '~10 day^-1' is not reproduced by Eq. (9) at the stated minimum radius rmin=1 m; the fitting function gives 7.4 day^-1 at f=1 Jy. Please clarify whether the headline is a rounded value, specify the values of rmin and f used, and show the underlying Monte Carlo points in Fig. 3. The abrupt break in the fitting function at rmin=3.4 m also needs physical justification.
[Section 3, Eq. (7)] The ISA abundance ζISA is calibrated locally from Earth impact rates, but the total rate is computed over the entire Galactic disk, with both nISA and nNS assumed proportional to n⋆. The resulting rate scales as the integral of n⋆^2 over the disk, which is dominated by the inner Galaxy where the stellar density is far from the solar-neighborhood calibration. The validity of assuming a constant ζISA throughout the disk—particularly in the inner Galaxy—should be discussed, as this assumption strongly affects the total rate.

minor comments (5)

[Eq. (1)] The magnetic dipole moment μNS is written with units of G cm^-2; the standard cgs unit for a magnetic dipole moment is G cm^3. The text and equation should be corrected.
[Eq. (2)] The Gaussian exponents are written as exp(v_NS^2/σ^2), which is positive and missing the factor 1/2. The correct form should be exp(-v_NS^2/(2σ^2)).
[References] The in-text citations 'Siraj & Loeb 2019a' and 'Siraj & Loeb 2019b' do not have matching labels in the reference list, which lists one 'submitted' paper and one arXiv paper without distinguishing letters. Also, the text cites 'Burbine 2002' but the reference list entry is 'Burbine 2001'.
[Fig. 3] Figure 3 shows only the fitting function (red line) and not the Monte Carlo data points; please include the actual simulated rates so the reader can judge the quality of the fit.
[Section 4] The phrase 'piece-wise fitting function' is used; the standard spelling is 'piecewise'.

Circularity Check

1 steps flagged · score 4.0 of 10

Predicted radio-flare rate inherits its ISA-abundance normalization from the authors' own two-object calibration; the headline number is a linear rescaling of that fitted input, though the emission and geometric content are external.

self citation load bearing [Section 3, Eqs. (3) and (8); abstract headline rate]
"The cumulative Earth impact rate for an ISA of radius r is estimated to be 2×10−4 (r/1 m)−3.4 (Siraj & Loeb 2019b). ... the number density of ISAs of radius ≥ r to be related to the number density of stars by a factor of, ζISA∼ 2.5×10^18 (1 pc−3/n⋆,⊙)(r/1 m)^−3.4. (3) ... We then find the rate at which each NS produces flares ... ˙Nflare,NS = ζISA n⋆ σa vrel. (8)"

The all-sky rate reported as ~10 day−1 in the abstract is obtained by multiplying ζISA, taken from the authors' own prior calibration (Siraj & Loeb 2019b), by external geometric and kinematic factors. Equation (8) is linear in ζISA, so the headline rate is a direct rescaling of that fitted normalization; it is not an independent test of the model. ζISA itself rests on an extrapolation from only two interstellar objects, 'Oumuamua and CNEOS 2014-01-08, down to meter scales, with an assumed 5% iron fraction, and no uncertainty is propagated. Consequently, the central predicted rate is load-bearing on the authors' own unpublished calibration rather than on a first-principles derivation, although the emission mechanism and cross section are external to this chain.

full rationale

The paper's derivation chain is mostly transparent and uses external inputs: the radio luminosity and millisecond duration come from Dai et al. (2016), the impact cross section from Safronov/Dai, the NS velocity distribution from Faucher-Giguere & Kaspi, and the Milky Way disk model is standard. Equation (8) is a forward convolution of these inputs with the ISA abundance ζISA. The only load-bearing self-citation is Eq. (3), where the ISA number-density normalization is adopted from Siraj & Loeb (2019b), an empirical fit to two interstellar objects. Because Eq. (8) is proportional to ζISA, the abstract's rate of ~10 day−1 is not independent of that fit; changing ζISA by an order of magnitude changes the headline rate by the same factor. However, the paper does not use the predicted radio rate to infer ζISA, and the emission and geometry content is external and falsifiable. Thus the circularity is partial: a central input is self-cited and the headline prediction is linearly contingent on it, but the central claim still has independent physical content. No other circular steps were found; the uncertainty in ζISA is a correctness concern rather than an additional circularity.

Assumptions & free parameters 7 free parameters · 6 assumptions · 0 invented entities

The central rate calculation combines external emission and cross-section models with a self-cited empirical ISA abundance. The largest uncertainty is the ISA number density normalization, which is derived from two objects and a fixed iron fraction.

free parameters (7)

Interstellar asteroid number density normalization zeta_ISA = 2.5e18 (1 pc^-3 / n_stars,solar) at r = 1 m
Empirical normalization from two interstellar objects in the authors' prior work (Siraj & Loeb 2019b). The predicted flare rate is directly proportional to zeta_ISA.
Iron fraction among asteroids = 5%
Assumed from Burbine (2002); used to convert total ISA abundance to iron-rich asteroids capable of the assumed tidal-disruption emission. Rate scales linearly with this fraction.
NS number density fraction zeta_NS = 1.7e-3
Adopted from Sartore et al. (2010); controls the number of available neutron stars in the Galaxy.
NS magnetic dipole moment mu_NS = 10^30 G cm^-2
Fiducial value used in Eq. (1); sets the luminosity scale and therefore the minimum visible asteroid radius in Eq. (5).
Asteroid tensile strength s = 10^10 dyn cm^-2
Assumed iron-asteroid strength in Eq. (1); enters the luminosity scaling as s^(2/3).
Asteroid mass density rho = 8 g cm^-3
Assumed iron density in Eq. (1); enters as rho^(-14/9).
Minimum asteroid radius r_min = about 1 m
Tidal disruption before melting threshold from Cordes & Shannon (2008) and Geng & Huang (2015); the final rate is sensitive to r_min, as shown in Fig. 3.

assumptions (6)

domain assumption Emission mechanism of Dai et al. (2016): tidally disrupted asteroid material produces coherent curvature radiation with luminosity Eq. (1).
The paper does not re-derive this model; the central claim depends on its validity.
domain assumption ISA population traces the stellar distribution of the Milky Way disk.
Used in Eq. (7) to distribute ISAs and NSs; interstellar asteroids are not necessarily bound to the disk, so this is a simplifying assumption.
domain assumption NS population traces the stellar distribution with zeta_NS = 1.7e-3.
Adopted from Sartore et al. (2010); supports the random NS positions used for the distance distribution.
domain assumption Five percent of asteroids are iron-rich.
Used to derive zeta_ISA in Eq. (3); from Burbine (2002).
domain assumption r_min is approximately 1 m for tidal disruption before melting.
Taken from Cordes & Shannon (2008) and Geng & Huang (2015); below this size the asteroid melts before disruption.
domain assumption Gravitational focusing cross section from Safronov (1972) via Dai et al. (2016).
Used to compute the NS-ISA impact cross section in Section 3.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Radio Flares from Collisions of Neutron Stars with Interstellar Asteroids." pith.science (2026). https://pith.science/paper/B4JESY3K

@misc{pith2026190811440,
  author       = {Pith},
  title        = {Pith review of: Radio Flares from Collisions of Neutron Stars with Interstellar Asteroids},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/B4JESY3K}},
  note         = {Machine review of arXiv:1908.11440}
}

abstract

We propose that collisions between neutron stars and interstellar asteroids, such as `Oumuamua, could power observable radio flares in the Milky Way galaxy. We find the rate of such events at $\sim 1 \mathrm{\; Jy}$ to be $\sim 10 \mathrm{\; day^{-1}}$.

Figures

Figures reproduced from arXiv: 1908.11440 by the authors.

**Figure 2.** Normalized probability function of the distance of NS from the Earth. yielding the minimum ISA radius that produces a visible flare, r ∼ 0.71 d kpc !3/4 [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** Rate of ∼ 1 Jy radio flares at ν ∼ 1 GHz from NS-ISA collisions, as a function of the minimum ISA radius, rmin. The red line shows the piece-wise fitting function in Equation (9). 4 RATE We find the all-sky rate of observable NS-ISA flares to be described as the following fitting function, N˙ ∼    (8.2 − 0.8 (rmin/1 m)) f /1 Jy−1.28 day−1 if rmin ≤ 3.4 m 370 (rmin/1 m) −3.4 f /1 Jy−1.28 day−1 if rmin > … view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

17 extracted references · 4 canonical work pages

[1]

Abbott B. P. et al., 2018, @doi [Physical Review Letters] 10.1103/PhysRevLett.121.161101 , 121, 16

doi:10.1103/physrevlett.121.161101 2018
[2]

Bland-Hawthorn J., Gerhard O., 2016, @doi [Annual Review of Astronomy and Astrophysics] 10.1146/annurev-astro-081915-023441 , 54, 529

doi:10.1146/annurev-astro-081915-023441 2016
[3]

Burbine T., 2001, Asteroids: Their composition and impact threat, Bulletin of the Czech Geological Survey, 77, 4

work page 2001
[4]

M., Shannon R

Cordes J. M., Shannon R. M., 2008, @doi [The Astrophysical Journal] 10.1086/589425 , 682, 1152

doi:10.1086/589425 2008
[5]

Dai Z. G. et al., 2016, @doi [The Astrophysical Journal] 10.3847/0004-637X/829/1/27 , 829, 27

doi:10.3847/0004-637x/829/1/27 2016
[6]

M., 2005, @doi [The Astrophysical Journal] 10.1086/501516 , 643, 332

Faucher-Giguere C-A., Kaspi V. M., 2005, @doi [The Astrophysical Journal] 10.1086/501516 , 643, 332

doi:10.1086/501516 2005
[7]

J., Huang Y

Geng J. J., Huang Y. F., 2015, @doi [The Astrophysical Journal] 10.1088/0004-637X/809/1/24 , 809, 24

doi:10.1088/0004-637x/809/1/24 2015
[8]

Greif S. K. et al., 2019, @doi [Monthly Notices of the Royal Astronomical Society] 10.1093/mnras/stz654 , 485, 5363

doi:10.1093/mnras/stz654 2019

Show all 17 references

[9]

I, 2019, @doi [Monthly Notices of the Royal Astronomical Society] 10.1093/mnras/stz1250 , 487, 491

Katz J. I, 2019, @doi [Monthly Notices of the Royal Astronomical Society] 10.1093/mnras/stz1250 , 487, 491

2019 doi
[10]

McLaughlin M. A. et al., 2006, @doi [Nature] 10.1038/nature04440 , 439, 817

2006 doi
[11]

J., et al., 2017, @doi [Nature] 10.1038/nature25020 , 552, 378

Meech K. J., et al., 2017, @doi [Nature] 10.1038/nature25020 , 552, 378

2017 doi
[12]

Micheli M., et al., 2018, @doi [Nature] 10.1038/s41586-018-0254-4 , 559, 223

2018 doi
[13]

et al., 2019, @doi [Physics Reports] 10.1016/j.physrep.2019.06.003

Platts E. et al., 2019, @doi [Physics Reports] 10.1016/j.physrep.2019.06.003

2019 doi
[14]

S., 1972, Evolution of the Protoplanetary Cloud and Formation of the Earth and Planets (Jerusalem: Keter Publishing House)

Safronov V. S., 1972, Evolution of the Protoplanetary Cloud and Formation of the Earth and Planets (Jerusalem: Keter Publishing House)

1972
[15]

et al., 2010, @doi [Astronomy & Astrophysics] 10.1051/0004-6361/200912222

Sartore N. et al., 2010, @doi [Astronomy & Astrophysics] 10.1051/0004-6361/200912222

2010 doi
[16]

& Loeb A., 2019, submitted to The Astrophysical Journal Letters

Siraj A. & Loeb A., 2019, submitted to The Astrophysical Journal Letters

2019
[17]

Siraj A., Loeb A., 2019, ( @eprint arXiv 1906.03270 )

2019 arXiv

Pith tools

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

[1] [1]

Abbott B. P. et al., 2018, @doi [Physical Review Letters] 10.1103/PhysRevLett.121.161101 , 121, 16

doi:10.1103/physrevlett.121.161101 2018

[2] [2]

Bland-Hawthorn J., Gerhard O., 2016, @doi [Annual Review of Astronomy and Astrophysics] 10.1146/annurev-astro-081915-023441 , 54, 529

doi:10.1146/annurev-astro-081915-023441 2016

[3] [3]

Burbine T., 2001, Asteroids: Their composition and impact threat, Bulletin of the Czech Geological Survey, 77, 4

work page 2001

[4] [4]

M., Shannon R

Cordes J. M., Shannon R. M., 2008, @doi [The Astrophysical Journal] 10.1086/589425 , 682, 1152

doi:10.1086/589425 2008

[5] [5]

Dai Z. G. et al., 2016, @doi [The Astrophysical Journal] 10.3847/0004-637X/829/1/27 , 829, 27

doi:10.3847/0004-637x/829/1/27 2016

[6] [6]

M., 2005, @doi [The Astrophysical Journal] 10.1086/501516 , 643, 332

Faucher-Giguere C-A., Kaspi V. M., 2005, @doi [The Astrophysical Journal] 10.1086/501516 , 643, 332

doi:10.1086/501516 2005

[7] [7]

J., Huang Y

Geng J. J., Huang Y. F., 2015, @doi [The Astrophysical Journal] 10.1088/0004-637X/809/1/24 , 809, 24

doi:10.1088/0004-637x/809/1/24 2015

[8] [8]

Greif S. K. et al., 2019, @doi [Monthly Notices of the Royal Astronomical Society] 10.1093/mnras/stz654 , 485, 5363

doi:10.1093/mnras/stz654 2019