REVIEW 2 major objections 4 minor 31 references
Polarization Studies of Black Widows PSRs B1957+20, J2055+3829 and J1544+4937
T0 review · 2 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read Using FAST observations, this paper shows that the linear polarization position angle shifts at the eclipse boundary of three black widow pulsars, implying lower-limit line-of-sight magnetic fields of a few to tens of milligauss, with…
desk verdict First direct B-field estimate for B1957+20's eclipse medium and reversal evidence in two spiders, but the core inference rests on untested Faraday assumptions and a fixable internal inconsistency. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The mechanism carrying the argument is Faraday rotation of the pulsar signal as it traverses the ionized eclipse medium. The paper measures, for each 30-second sub-integration, the shift $\Delta\mathrm{PA}$ in the linear polarization position angle relative to the out-of-eclipse average profile, and converts it to a rotation-measure change using $\Delta\mathrm{PA} = \Delta\mathrm{RM}\,\lambda^2$, where $\lambda$ is the observing wavelength; the line-of-sight magnetic field then follows from $\Delta B_{\parallel} = 1.23\,\mu\mathrm{G} \times \Delta\mathrm{RM}/\Delta\mathrm{DM}$. A second element is the depolarization model $L/L_0 = \exp(-2\lambda^4 \sigma_{\rm RM}^2)$, used to attribute the loss of linear polarization at eclipse boundaries to rapid rotation-measure fluctuations within a sub-integration rather than to changes in the pulsar's intrinsic emission. The eclipse edges are located by Fermi-Dirac fits to the flux density, following the approach of Polzin et al. (2019).
What would settle it
Observe the three pulsars at two or more widely spaced frequencies, such as 600 MHz and 2 GHz, and test whether the eclipse-boundary PA shift scales as $\lambda^2$ with the same $\Delta\mathrm{RM}$; a deviation from the $\lambda^2$ law would mean part of the shift is intrinsic and the inferred fields are spurious. The single-sub-integration detection for J2055+3829 could also be checked with a longer, higher-signal-to-noise observation.
Extended reading notes
Core claim
On the paper's own terms, the central discovery is that the position angle of linearly polarized emission shifts at the eclipse boundary of all three black widow pulsars, and that this shift is best explained as Faraday rotation produced in the eclipse medium. From $\Delta\mathrm{PA} = \Delta\mathrm{RM}\,\lambda^2$, the authors obtain rotation-measure changes of about $-9$ to $+30$ rad m$^{-2}$ for PSR B1957+20 and about $+25$ to $-10$ rad m$^{-2}$ for PSR J1544+4937; combined with a conservative choice of $\Delta\mathrm{DM}$ (taken to be the DM measurement uncertainty), these imply lower-limit line-of-sight fields of roughly $-8$ to $+23$ mG for B1957+20 and $+7.3$ to $-4.8$ mG for J1544+4937. For PSR J2055+3829 a single sub-integration gives $\Delta\mathrm{RM} = -7 \pm 1$ rad m$^{-2}$ and a lower-limit field of $-2.2 \pm 0.3$ mG. The sign changes in $\Delta\mathrm{RM}$ for B1957+20 and J1544+4937 are presented as evidence that the magnetic field reverses inside the eclipse medium. The paper explicitly notes that the field-strength values are lower limits because the DM change is not independently resolved, and that the J2055+3829 detection rests on one sub-integration.
Load-bearing premise
The load-bearing premise is that the position-angle shifts come entirely from Faraday rotation in the eclipse medium, with the pulsar's own position-angle swing unchanged at those orbital phases; if that premise fails, the rotation measures and field strengths are not real.
Editorial extensions
If this is right
- If the field estimates hold, the eclipse media of black widow pulsars are threaded by ordered magnetic fields of order a milligauss or more, not just by dense, unstructured plasma.
- The sign reversals seen in PSRs B1957+20 and J1544+4937 imply that the magnetic field structure in the eclipse medium is coherent over at least a fraction of the orbit and changes direction within it.
- Depolarization setting in before measurable DM changes points to magnetic-field fluctuations, rather than density fluctuations, as the dominant cause of the loss of linear polarization at eclipse boundaries.
- PSR J1544+4937, whose radio emission survives throughout the eclipse at 1250 MHz, offers a way to map the magnetic field across the full eclipse, making it a useful laboratory for eclipse-medium structure.
Reading between the lines
- A direct test of the Faraday-rotation interpretation would be to observe these pulsars at two widely separated frequencies and check that the same $\Delta\mathrm{RM}$ reproduces the PA shifts at both; if $\Delta\mathrm{PA}$ does not scale as $\lambda^2$, part of the shift is intrinsic to the pulsar.
- The inferred field reversals, if confirmed, suggest that the companion wind and the pulsar wind interact to create current sheets or magnetic loops; time-resolved mapping of $\Delta\mathrm{RM}$ across the orbit could constrain the geometry and scale of these structures.
- Because the depolarization scatter $\sigma_{\rm RM}$ values (13–20 rad m$^{-2}$) overlap those inferred for repeating fast radio bursts, the black widow eclipse medium may serve as a local laboratory for the magnetoionic environments that depolarize FRBs; bright lensed pulses near eclipse, like those seen in PSR B1744-24A, could test this connection.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper presents FAST 1250 MHz polarimetric observations of three black widow pulsars, PSRs B1957+20, J2055+3829, and J1544+4937, during and around their radio eclipses. For each pulsar, the authors measure flux density, dispersion measure, rotation measure, and linear/circular polarization as functions of orbital phase. They report position-angle (PA) shifts at the eclipse boundaries in all three systems, convert these shifts to rotation-measure changes via ΔPA = ΔRM λ², and then estimate lower limits on the line-of-sight magnetic field strength in the eclipse medium using ΔB_∥ = 1.23 μG ΔRM / ΔDM. They claim lower-limit field strengths of a few mG to tens of mG, and interpret sign changes of ΔRM in B1957+20 and J1544+4937 as evidence for magnetic field reversals. The paper also discusses depolarization at eclipse boundaries and compares the inferred σ_RM values with those seen in repeating fast radio bursts.
Significance. If the central claim is correct, this would provide the first direct measurement of the line-of-sight magnetic field strength in the eclipse medium of PSR B1957+20, and would add two more systems with evidence for ordered, sign-changing milligauss-level magnetic fields in black widow eclipses. Such measurements are important for understanding eclipse mechanisms and the magnetoionic environment of spider pulsars. The paper has several strengths: the PA shifts for B1957+20 and J1544+4937 are large compared with the quoted uncertainties and are directly visible in the figures; the observations are polarization-calibrated with standard tools; and the authors explicitly acknowledge the limitation of the J2055+3829 detection resting on a single sub-integration. However, the central inference depends on an untested assumption that the observed PA shifts are entirely due to Faraday rotation, and the lack of a frequency-dependence or profile-shape check is a significant gap.
major comments (2)
- [Section 3.2.1, Eq. (1)] The central inference that the observed ΔPA is purely Faraday rotation is not tested. A Faraday rotation measure would produce a PA shift that scales as λ² across the 400 MHz band, and a pure rotation would leave the shape of the PA swing unchanged (i.e., the difference between a sub-integration and the reference would be a constant offset modulo π at every pulse phase). The analysis uses only the full-band PA and a single effective wavelength of 24 cm, and does not check either prediction. If the intrinsic PA swing changes at the eclipse boundary (for example, because different emission regions or modes become visible), the inferred ΔRM and ΔB_∥ would be spurious. I request a frequency-resolved analysis (e.g., sub-band ΔPA values) and a per-pulse-phase residual analysis to support the Faraday interpretation before the magnetic field and reversal claims can be accepted.
- [Section 3.2.2] The detection of a PA shift for PSR J2055+3829 rests on a single sub-integration at orbital phase 0.402, with ΔPA = -23±4 deg and a 6% fractional linear polarization outside eclipse. The authors acknowledge this, but a single measurement is not sufficient to claim a PA shift at the eclipse boundary for this pulsar, especially given the low linear polarization fraction and the possibility of a noise fluctuation. The abstract and conclusions state that PA shifts are observed in all three pulsars; the J2055+3829 result should either be supported by additional independent sub-integrations or explicitly downgraded to a tentative detection.
minor comments (4)
- [Section 4.2] The σ_RM values quoted in the discussion (13, 16, and 20 rad/m² for PSRs B1957+20, J2055+3829, and J1544+4937) are inconsistent with the values derived in Section 3.2 (15, 5, and 13 rad/m², respectively). Please reconcile these numbers and clarify the order in which the pulsars are listed.
- [Section 3.2.3] The statement that linear polarization depolarizes at orbital phases 0.120-0.158 and 0.195-0.348 while it remains unchanged at 0.158-0.220 is confusing because 0.158-0.220 overlaps with the second depolarized interval. Please clarify whether 0.158-0.220 is a region where the linear polarization is measured to remain unchanged, and if so, how this is consistent with the claimed depolarization in the overlapping range.
- [Figure 3 caption] The caption refers to 'vertical filled red areas in the fourth panel', but the panels are not numbered in the figure. Please add explicit panel labels (e.g., a, b, c) or describe the panels by their content to avoid ambiguity.
- [Section 2] The sentence 'For PSRs J2055+3829, and J1544+4937, our observations cover the entire binary orbital phases' contains a grammatical error and should read 'For PSRs J2055+3829 and J1544+4937, our observations cover the entire binary orbit'.
Circularity Check
Headline magnetic-field inference is an independent Faraday-rotation application; the secondary depolarization sigma_RM values are constructed from an assumed 5% threshold and then called measured.
-
fitted input called prediction
[Section 3.2.1 (Equation 3) and Section 4.2]
"We arbitrarily assume that the fractional linear polarization decreases to less than 5%. Setting L0=21.5%, L=5%, and λ=24 cm, we estimate σ_RM∼15 rad m−2. ... In comparison, we measured σ_RM values of 13,16,and20 rad m−2 for PSRs B1957+20, J2055+3829, and J1544+4937, respectively, comparable to those observed in FRB repeaters."
The paper's only described route to σ_RM is Equation (3) with an arbitrarily imposed depolarization threshold L=5% and a measured L0; solving that equation for σ_RM makes σ_RM a function of the assumed threshold rather than an independent measurement. Section 4.2 then reports these constructed values as 'measured' and uses them to claim comparability with FRB repeaters. The comparison is thus an output of the assumption, not a new empirical result. This does not bear on the main B_parallel inference, but the secondary depolarization claim reduces by construction to the chosen threshold.
full rationale
The central claim is not circular: the observed PA shifts are compared with an out-of-eclipse reference PA swing, converted to ΔRM through the standard Faraday relation ΔPA=ΔRM λ² (Eq. 1), and then to a line-of-sight magnetic-field lower limit through ΔB∥=1.23 μG ΔRM/ΔDM (Eq. 2). The target quantity B∥ is not fitted or defined in terms of itself; it is inferred from independently measured PA and DM quantities using standard physics. The reliance on the assumption that intrinsic PA swings do not change at the eclipse boundary is an untested physical assumption, not a circularity. The one concrete circular element is in the depolarization discussion: Section 3.2.1 explicitly says the σ_RM estimate follows from an arbitrary L=5% assumption, yet Section 4.2 describes σ_RM values as 'measured' and compares them with FRB repeaters. That is an assumed input presented as a measured output, but it is a secondary comparison and does not undermine the independent main magnetic-field derivation. Hence a moderate score of 4.
Assumptions & free parameters
free parameters (2)
- Fermi-Dirac eclipse fit parameters a, p1, p2 (per pulsar) =
not reported numerically
- Depolarization threshold L/L0 =
5%
assumptions (4)
- domain assumption The PA shift at eclipse boundaries is entirely due to Faraday rotation in the eclipse medium: DeltaPA = DeltaRM * lambda^2.
- domain assumption The per-sub-integration DM uncertainty can stand in for the actual DeltaDM when no variation is detected, giving lower-limit B_parallel values.
- domain assumption The DM measured by multiband timing accurately traces the electron column through the eclipse medium.
- standard math Standard published relations: Faraday rotation (Eq 1), field from RM/DM (Eq 2), Gaussian-RM depolarization (Eq 3, You et al. 2018), and B_perp upper limit (Eq 4, Polzin et al. 2019).
Cite this review
Pith. "Pith review of Polarization Studies of Black Widows PSRs B1957+20, J2055+3829 and J1544+4937." pith.science (2026). https://pith.science/paper/KECQOSPO
@misc{pith2026250612272,
author = {Pith},
title = {Pith review of: Polarization Studies of Black Widows PSRs B1957+20, J2055+3829 and J1544+4937},
year = {2026},
howpublished = {\url{https://pith.science/paper/KECQOSPO}},
note = {Machine review of arXiv:2506.12272}
}
read the original abstract
We present an analysis of the polarization of three black widow pulsars, PSRs B1957+20, J2055+3829 and J1544+4937 at 1250\,MHz using the Five-hundred-meter Aperture Spherical radio Telescope (FAST). Radio eclipses for PSRs B1957+20 and J2055+3829 are detected, while the radio emission for J1544+4937 is detected throughout the eclipse. We study the polarization and dispersion measure properties of the pulsars near and during the eclipse. The position angle of the linear polarization is observed to shift at the eclipse boundary in all of these three pulsars implying a lower limit line-of-sight magnetic field strength of the eclipse medium of approximately a few mG to tens of mG. We also find evidence that the magnetic field in the eclipse medium of PSRs B1957+20 and J1544+4937 reverses.
Figures
Reference graph
Works this paper leans on
-
[1]
write newline
" write newline "" before.all 'output.state := FUNCTION fin.entry write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.state := if if FUNCTION not #0 #1 if FUNCTION and 'skip pop #0 if FUNCTION or pop #1...
-
[2]
Bai J. T., et al., 2022, @doi [ ] 10.1093/mnras/stac918 , https://ui.adsabs.harvard.edu/abs/2022MNRAS.513.1794B 513, 1794
-
[3]
Bhattacharyya B., et al., 2013, @doi [ ] 10.1088/2041-8205/773/1/L12 , https://ui.adsabs.harvard.edu/abs/2013ApJ...773L..12B 773, L12
-
[4]
Bilous A. V., Ransom S. M., Demorest P., 2019, @doi [ ] 10.3847/1538-4357/ab16dd , https://ui.adsabs.harvard.edu/abs/2019ApJ...877..125B 877, 125
-
[5]
Crowter K., et al., 2020, @doi [ ] 10.1093/mnras/staa933 , https://ui.adsabs.harvard.edu/abs/2020MNRAS.495.3052C 495, 3052
-
[6]
Feng Y., et al., 2022, @doi [Science] 10.1126/science.abl7759 , https://ui.adsabs.harvard.edu/abs/2022Sci...375.1266F 375, 1266
-
[7]
Fruchter A. S., Stinebring D. R., Taylor J. H., 1988, @doi [ ] 10.1038/333237a0 , https://ui.adsabs.harvard.edu/abs/1988Natur.333..237F 333, 237
doi:10.1038/333237a0 1988
-
[8]
Fruchter A. S., et al., 1990, @doi [ ] 10.1086/168502 , https://ui.adsabs.harvard.edu/abs/1990ApJ...351..642F 351, 642
Show all 31 references
-
[9]
Guillemot L., Octau F., Cognard I., Desvignes G., Freire P. C. C., Smith D. A., Theureau G., Burnett T. H., 2019, @doi [ ] 10.1051/0004-6361/201936015 , https://ui.adsabs.harvard.edu/abs/2019A&A...629A..92G 629, A92
2019 doi
-
[10]
B., Edwards R
Hobbs G. B., Edwards R. T., Manchester R. N., 2006, @doi [ ] 10.1111/j.1365-2966.2006.10302.x , https://ui.adsabs.harvard.edu/abs/2006MNRAS.369..655H 369, 655
2006
-
[11]
Hobbs G., et al., 2020, @doi [ ] 10.1017/pasa.2020.2 , https://ui.adsabs.harvard.edu/abs/2020PASA...37...12H 37, e012
2020 doi
-
[12]
W., van Straten W., Manchester R
Hotan A. W., van Straten W., Manchester R. N., 2004, @doi [ ] 10.1071/AS04022 , https://ui.adsabs.harvard.edu/abs/2004PASA...21..302H 21, 302
2004 doi
-
[13]
X., Main R., Pen U.-L., van Kerkwijk M
Li D., Lin F. X., Main R., Pen U.-L., van Kerkwijk M. H., Yang I. S., 2019, @doi [ ] 10.1093/mnras/stz374 , https://ui.adsabs.harvard.edu/abs/2019MNRAS.484.5723L 484, 5723
2019 doi
-
[14]
Li D., Bilous A., Ransom S., Main R., Yang Y.-P., 2023, @doi [ ] 10.1038/s41586-023-05983-z , https://ui.adsabs.harvard.edu/abs/2023Natur.618..484L 618, 484
2023 doi
-
[15]
Ng C., et al., 2020, @doi [ ] 10.1093/mnras/staa1658 , https://ui.adsabs.harvard.edu/abs/2020MNRAS.496.2836N 496, 2836
2020 doi
-
[16]
Ng C., et al., 2025, @doi [ ] 10.3847/1538-4357/adb0bc , https://ui.adsabs.harvard.edu/abs/2025ApJ...982..154N 982, 154
2025 doi
-
[17]
J., Breton R
Polzin E. J., Breton R. P., Stappers B. W., Bhattacharyya B., Janssen G. H., Os owski S., Roberts M. S. E., Sobey C., 2019, @doi [ ] 10.1093/mnras/stz2579 , https://ui.adsabs.harvard.edu/abs/2019MNRAS.490..889P 490, 889
2019 doi
-
[18]
J., Breton R
Polzin E. J., Breton R. P., Bhattacharyya B., Scholte D., Sobey C., Stappers B. W., 2020, @doi [ ] 10.1093/mnras/staa596 , https://ui.adsabs.harvard.edu/abs/2020MNRAS.494.2948P 494, 2948
2020 doi
-
[19]
Roberts M. S. E., 2013, in van Leeuwen J., ed., IAU Symposium Vol. 291, Neutron Stars and Pulsars: Challenges and Opportunities after 80 years. pp 127--132 ( @eprint arXiv 1210.6903 ), @doi 10.1017/S174392131202337X
2013 arXiv
-
[20]
F., Taylor J
Ryba M. F., Taylor J. H., 1991, @doi [ ] 10.1086/170613 , https://ui.adsabs.harvard.edu/abs/1991ApJ...380..557R 380, 557
1991 doi
-
[21]
Sotomayor-Beltran C., et al., 2013, @doi [ ] 10.1051/0004-6361/201220728 , https://ui.adsabs.harvard.edu/abs/2013A&A...552A..58S 552, A58
2013 doi
-
[22]
Spiewak R., et al., 2022, @doi [ ] 10.1017/pasa.2022.19 , https://ui.adsabs.harvard.edu/abs/2022PASA...39...27S 39, e027
2022 doi
-
[23]
W., Bailes M., Lyne A
Stappers B. W., Bailes M., Lyne A. G., Camilo F., Manchester R. N., Sandhu J. S., Toscano M., Bell J. F., 2001, @doi [ ] 10.1046/j.1365-8711.2001.04074.x , https://ui.adsabs.harvard.edu/abs/2001MNRAS.321..576S 321, 576
2001
-
[24]
Tavani M., Brookshaw L., 1991, @doi [ ] 10.1086/186187 , https://ui.adsabs.harvard.edu/abs/1991ApJ...381L..21T 381, L21
1991 doi
-
[25]
D., Evans C
Thompson C., Blandford R. D., Evans C. R., Phinney E. S., 1994, @doi [ ] 10.1086/173728 , https://ui.adsabs.harvard.edu/abs/1994ApJ...422..304T 422, 304
1994 doi
-
[26]
Q., et al., 2021, @doi [ ] 10.3847/2041-8213/ac365c , https://ui.adsabs.harvard.edu/abs/2021ApJ...922L..13W 922, L13
Wang S. Q., et al., 2021, @doi [ ] 10.3847/2041-8213/ac365c , https://ui.adsabs.harvard.edu/abs/2021ApJ...922L..13W 922, L13
2021 doi
-
[27]
Wang N., et al., 2023a, @doi [Science China Physics, Mechanics, and Astronomy] 10.1007/s11433-023-2131-1 , https://ui.adsabs.harvard.edu/abs/2023SCPMA..6689512W 66, 289512
-
[28]
Q., et al., 2023b, @doi [ ] 10.3847/1538-4357/acea81 , https://ui.adsabs.harvard.edu/abs/2023ApJ...955...36W 955, 36
Wang S. Q., et al., 2023b, @doi [ ] 10.3847/1538-4357/acea81 , https://ui.adsabs.harvard.edu/abs/2023ApJ...955...36W 955, 36
-
[29]
P., Manchester R
You X. P., Manchester R. N., Coles W. A., Hobbs G. B., Shannon R., 2018, @doi [ ] 10.3847/1538-4357/aadee0 , https://ui.adsabs.harvard.edu/abs/2018ApJ...867...22Y 867, 22
2018 doi
-
[30]
H., Breton R
van Kerkwijk M. H., Breton R. P., Kulkarni S. R., 2011, @doi [ ] 10.1088/0004-637X/728/2/95 , https://ui.adsabs.harvard.edu/abs/2011ApJ...728...95V 728, 95
2011 doi
-
[31]
van Straten W., Bailes M., 2011, @doi [ ] 10.1071/AS10021 , https://ui.adsabs.harvard.edu/abs/2011PASA...28....1V 28, 1
2011 doi
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.