Constraining the geometry and kinematics of the quasar broad emission line region using gravitational microlensing. II. Comparing models with observations in the lensed quasar HE0435-1223

D. Hutsem\'ekers; D. Sluse; L. Braibant; R. Goosmann

REVIEW 1 major objections 5 minor 15 references

Constraining the geometry and kinematics of the quasar broad emission line region using gravitational microlensing. II. Comparing models with observations in the lensed quasar HE0435-1223

T0 review · 1 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 Gravitational microlensing of the lensed quasar HE0435-1223 favors a flattened, disk-like broad emission line region over a biconical polar wind.

desk verdict Solid application of the Paper I microlensing framework to HE0435-1223; the flattened-geometry conclusion holds up, but the continuum-size argument overreaches the simulated grid. read the letter →

arxiv 1908.04178 v1 pith:YBWZDM6Q submitted 2019-08-12 astro-ph.GA

D. Hutsemékers , L. Braibant , D. Sluse , R. Goosmann This is my paper · ORCID

classification astro-ph.GA

keywords gravitationalmicrolensingbroademissionlineregionquasarHαprofiledistortionsKepleriandiskHE0435-1223magnificationmaps

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

Gravitational microlensing by stars in the foreground lens galaxy magnifies different parts of the broad emission line region (BLR) of the quadruply lensed quasar HE0435-1223 by different amounts, distorting the observed H$\alpha$ line profile. The paper measures four indices of this distortion—$\mu_\mathrm{cont}$, $\mu_\mathrm{BLR}$, $\mathrm{RBI}$, and $\mathrm{WCI}$—and compares them with simulated microlensing events computed for three representative BLR geometries: a Keplerian disk, an equatorial wind, and a biconical polar wind. It finds that flattened geometries (disk and equatorial wind) reproduce the observed red/blue asymmetry of the H$\alpha$ profile much more readily than the polar wind. When an independent estimate of the continuum source size is added, the Keplerian disk becomes the slightly preferred model. The paper is careful that a single-epoch signal cannot robustly discriminate the models, so the result is a likelihood ranking rather than a definitive detection.

What carries the argument

The argument is carried by four microlensing observables defined in the companion Paper I: the continuum magnification $\mu_\mathrm{cont}$, the total line magnification $\mu_\mathrm{BLR}$, the red/blue index $\mathrm{RBI}$ that measures asymmetric red versus blue deformation of the line profile, and the wings/core index $\mathrm{WCI}$ that compares line-wing to line-core magnification. Simulated line profiles are generated by convolving monochromatic images of each BLR model (Keplerian disk, polar wind, equatorial wind, with varying inclination, inner radius, emissivity index, and continuum disk size) with microlensing magnification maps produced with a ray-shooting code tuned to the macro-model parameters of image D. The $(\mathrm{WCI}, \mathrm{RBI})$ diagnostic diagrams and the marginalised likelihood ratio $P(G,i\,|\,d)$ then rank how easily each geometry reproduces the four observed values.

What would settle it

Multi-epoch monitoring of image B that reveals microlensing variability comparable to image D would falsify the reference-spectrum assumption and invalidate the model ranking, as would an independent reverberation-mapping measurement of the H$\alpha$ BLR size in HE0435-1223 that is incompatible with the favored Keplerian disk configuration.

Watch

Extended reading notes

Core claim

The observed microlensing-induced amplification and distortion of the H$\alpha$ line in image D of HE0435-1223, quantified by the continuum magnification $\mu_\mathrm{cont} = 1.68 \pm 0.10$, the line magnification $\mu_\mathrm{BLR} = 1.30 \pm 0.17$, the red/blue asymmetry index $\mathrm{RBI} = 0.15 \pm 0.02$, and the wings/core index $\mathrm{WCI} = 1.09 \pm 0.17$, can be reproduced by convolving BLR emission models with a caustic network magnification map. Comparing the relative likelihoods of the models, the paper concludes that flattened geometries (Keplerian disk and equatorial wind) more easily reproduce the observed line profile deformations than a biconical polar wind, with no strong preferred inclination. Adding the independent constraint that the continuum source radius $r_s \geq 0.6\,r_E$, derived from published accretion-disk size estimates at the H$\alpha$ wavelength, slightly reinforces the Keplerian disk as the model that most easily matches the four observed indices. The authors stress that the single-epoch microlensing signal does not allow unambiguous discrimination, so the result is a comparative likelihood ranking among the models rather than a unique determination.

Load-bearing premise

The entire measurement chain assumes that image B of HE0435-1223 is free of microlensing and that the D/B macro-magnification ratio is $M = 0.47 \pm 0.03$; if image B is itself microlensed or $M$ is in error, all four observed indices would be biased and the model comparison would no longer be valid.

Editorial extensions

If this is right

If flattened geometries are correct, the H$\alpha$-emitting gas in HE0435-1223 is more likely orbiting in a disk (Keplerian or equatorial wind) than flowing in a biconical polar outflow.
Multi-epoch spectroscopy of the microlensing signal should further discriminate the BLR models, because different geometries sample the caustic pattern differently as the source moves across the magnification map.
Simultaneously modelling the microlensing distortions of several emission lines, such as C IV and H$\alpha$ in the Einstein Cross, would produce stronger constraints on the BLR geometry and kinematics.
With the additional continuum-size constraint, the Keplerian disk is the single most favored model, implying rotation-dominated kinematics is marginally preferred over a radially accelerated equatorial wind.

Reading between the lines

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

If the same four-index comparison is applied to other lensed quasars, it could build a statistical sample of BLR geometries and reveal whether the disk-versus-wind dichotomy depends on quasar luminosity or redshift.
A sharper measurement of the H$\alpha$ continuum-emitting region's size could turn the slight preference for a Keplerian disk into a robust discrimination, because the $r_s \geq 0.6\,r_E$ cut is what produces the preference.
The paper marginalises over the orientation of the caustic network relative to the BLR symmetry axis, so a joint fit that keeps orientation as a free parameter and uses multi-epoch data could also recover the disk's position angle.
The smooth, axisymmetric emissivity laws assumed for the BLR may miss clumpy or spiral structure, which could mimic or mask microlensing distortions in the observed indices.

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

Damien,

Worth a look if you work on quasar microlensing. The paper takes the simulation framework from Braibant et al. (Paper I) and applies it to a new target, HE0435-1223. The new results are the measured microlensing indices for Hα (µcont, µBLR, RBI, WCI), the comparison of simulated (WCI,RBI) distributions, and the likelihood-based probabilities for three BLR geometries. The measurement section is careful: error propagation is explicit, the choice of image B as reference is justified, and the smoothing checks are reassuring. The simulation setup is transparent, with the caustic network parameters from the literature and five map rotations to reduce alignment bias.

The main finding—flattened geometries (Keplerian disk and equatorial wind) reproduce the observed line-profile distortion more easily than a biconical polar wind—is well supported by the figures and the probability table (KD+EW ≈ 80% vs PW ≈ 20% for the all-rs case). That part holds up.

The soft spot is the “slightly favored” Keplerian disk after adding the continuum-size constraint. The stress-test note is correct: the quoted independent estimates of the continuum half-light radius correspond to rs from 0.7 to 4 rE, but the simulation grid stops at 0.7 rE. So the constraint is only applied as a lower bound at rs ≥ 0.6 rE; the upper end of that range is never simulated. The prior over rs is therefore not the one implied by the independent measurements, and the KD-favored result is not established. It is a marginal effect in a truncated parameter space. This is a moderate issue, not a deal-breaker, because the flattened-vs-polar conclusion does not rely on it.

Two minor notes. First, image B being unmicrolensed and the macro-magnification M = 0.47 ± 0.03 are load-bearing assumptions; the paper cites prior work for them, which is reasonable, but a direct test against image A or C would strengthen the analysis. Second, no code or magnification maps are released, so independent verification of the likelihood computation is harder than it should be.

Who should read it: anyone using microlensing to constrain BLR structure. It deserves a serious referee. I’d send it out with a request to either extend the grid to larger rs or explain why the upper range is irrelevant, and to soften the KD claim accordingly. The central geometry conclusion is worth publishing.

Best, [Your name]

Referee Report

1 major / 5 minor

Summary. The paper analyzes microlensing-induced distortions of the Halpha line in image D of the quadruply lensed quasar HE0435-1223. It measures four observables defined in Paper I: the continuum magnification mu_cont, the line magnification mu_BLR, the red-blue asymmetry index RBI, and the wings-to-core index WCI, using image B as a non-microlensed reference and a macro-magnification ratio M = 0.47 +/- 0.03. These measurements are compared with simulations of three representative BLR geometries (Keplerian disk KD, polar wind PW, and equatorial wind EW) convolved with a microlensing magnification map appropriate to image D. The paper concludes that flattened geometries (KD and EW) reproduce the observed (WCI, RBI) and magnification constraints more easily than the biconical polar wind, and that adding a literature-based constraint on the continuum source size slightly favors the Keplerian disk.

Significance. If the comparison is valid, this is a useful step toward using single-epoch microlensing line-profile distortions to constrain BLR geometry and kinematics. The paper is transparent: the simulation grid, the index definitions, and the likelihood normalization are clearly described, and the diagnostic (WCI, RBI) diagrams make the comparison easy to interpret. The qualitative conclusion that flattened geometries out-produce the polar wind appears reasonably supported by the simulations. However, the strongest quantitative claim in the abstract and Section 3.3, that an independent continuum-size constraint favors the Keplerian disk, is currently not established because the simulations do not cover the full range of source sizes allowed by that constraint.

major comments (1)

[Section 3.3 and Table 1] The 'additional independent constraint' on the continuum source size is implemented only as a lower-bound cut (rs >= 0.6 rE), but the independent size estimates quoted in Section 3.3 translate to rs ~ 0.7-4 rE. The simulation grid in Section 3 only goes up to rs = 0.7 rE, and hence rin <= 0.75 rE, so no simulated model explores the upper range of the independent constraint. The conclusion that the Keplerian disk model is 'slightly favored' is therefore not established over the full domain allowed by the independent measurement; it may be an artifact of the truncated grid. I ask the authors to extend the simulations to larger rs values (up to about 4 rE) and recompute Table 1, or alternatively to show quantitatively that the upper range is incompatible with the observed mu_cont and can be excluded.

minor comments (5)

[Section 3.2] The word 'Nervertheless' after Eq. (8) should be 'Nevertheless'.
[Appendix A] The statement that low-inclination PW models 'can generate a small number of unrealistic simulations' should specify whether those simulations are excluded from the probability counts in Table 1; if they are retained, a brief justification is needed because their inclusion could bias the comparison against the polar wind.
[Section 3.3] The conversion R1/2(Halpha) = R1/2(UV) * (lambda_Halpha/lambda_UV)^p with an assumed p = 4/3 when p was not measured introduces an additional systematic uncertainty in the independent size constraint that is not propagated into the final comparison.
[Section 2 and Eq. (6)] The errors on mu_BLR, RBI, and WCI are propagated from the same flux-density uncertainties, so the four observables are likely correlated; reporting the covariance matrix or at least acknowledging this correlation in the Gaussian likelihood of Eq. (6) would strengthen the statistical interpretation.
[Eq. (7)] The summation over microlensing parameters is written as a sum over 'n' without an explicit index; using a notation such as sum_eta or defining eta_n would remove the ambiguity.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the observed microlensing indices are compared to a pre-computed simulation grid, with no parameter fitted to force the conclusion.

full rationale

The derivation chain is self-contained against external observations. The four observables (µcont, µBLR, RBI, WCI) are measured from HE0435-1223 spectra (Sect. 2) and compared with a pre-existing grid of microlensed BLR simulations from Paper I (Sect. 3). No model parameter is fitted to the observed indices; the posterior probabilities are likelihood comparisons with shared priors (Eqs. 5–8), and the quoted errors are propagated from spectral uncertainties. The only self-citations are to Paper I for the simulation framework and to Braibant et al. (2014) for the macro-magnification ratio M and the identification of image D as microlensed. These supply methods or prior measurements, not the target conclusion, and the Paper I simulations are parameter-free grids not tuned to HE0435-1223. The 'additional independent constraint' on continuum source size comes from external literature estimates and is applied as a restriction rs ≥ 0.6 rE; the paper's use of that constraint is incomplete because the simulation grid extends only to rs = 0.7 rE while the quoted estimates reach 4 rE, but that is a correctness risk, not a definitional equivalence or a fitted-input prediction. The paper explicitly concedes that the single-epoch constraints are not robust and that models cannot be unambiguously discriminated. No prediction in the paper reduces by construction to its inputs, so the circularity score is 0.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claim is a model-comparison result, so the free parameters are the grid choices and thresholds; the key axioms are the representativeness of the three BLR models, the integrity of the reference image, and the accuracy of the magnification map and continuum size constraint.

free parameters (6)

BLR inner radius rin
Grid of nine values from 0.1 to 0.75 rE; marginalized over with uniform prior, not fitted.
BLR outer radius rout = 10 rin
Fixed multiplicative factor chosen by hand; affects the line profile width and magnification.
Emissivity index q = 1.5 or 3
Two chosen values; results are insensitive to q, as noted in Sect. 3.2.
Continuum source radius rs
Grid of nine values from 0.1 to 0.7 rE; marginalized, with an additional threshold rs >= 0.6 rE in one scenario.
Inclination i
Four values (22, 34, 44, 62 degrees); marginalized.
Microlensing source position and map rotation
Positions over the magnification map and 5 rotation angles; marginalized over via the sum over n in Eq. 7.

assumptions (5)

domain assumption The three BLR models (Keplerian disk, polar wind, equatorial wind) with emissivity laws q = 1.5 and 3 are representative of real BLRs.
Sect. 3 states these are 'representative BLR models' from Paper I; the conclusion is relative to this set.
domain assumption Image B is unaffected by microlensing and the macro-magnification ratio M = 0.47 ± 0.03 is correct.
Sect. 2: image B is used as the non-microlensed reference; a microlensed B or inaccurate M would bias all four indices.
domain assumption The macro-model parameters for image D (kappa_s = 0.124, kappa_c = 0.466, gamma = 0.640) and the computed caustic map are accurate.
Sect. 3.1: the magnification map is built from these values; if the macro-model is wrong, the simulated indices would be wrong.
domain assumption The continuum source is a uniform disk with the same inclination as the BLR and radius rs.
Sect. 3: this is the source model used for mu_cont and for the rs constraint; real sources may have limb darkening and different geometry.
ad hoc to paper The independent continuum size estimates (R1/2 ~ 6-30 light days) map to rs >= 0.6 rE.
Sect. 3.3: the conversion uses R1/2 = 1.4 rs, an assumed p = 4/3, and M = 0.3 Msun, and the threshold 0.6 rE is a round cut on the derived range.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Constraining the geometry and kinematics of the quasar broad emission line region using gravitational microlensing. II. Comparing models with observations in the lensed quasar HE0435-1223." pith.science (2026). https://pith.science/paper/YBWZDM6Q

@misc{pith2026190804178,
  author       = {Pith},
  title        = {Pith review of: Constraining the geometry and kinematics of the quasar broad emission line region using gravitational microlensing. II. Comparing models with observations in the lensed quasar HE0435-1223},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YBWZDM6Q}},
  note         = {Machine review of arXiv:1908.04178}
}

abstract

The quadruply lensed quasar HE0435-1223 shows a clear microlensing effect that affects differently the blue and red wings of the H$\alpha$ line profile in its image D. To interpret these observations, and constrain the broad emission line region (BLR) properties, the effect of gravitational microlensing on quasar broad emission line profiles and their underlying continuum has been simulated considering representative BLR models and microlensing magnification maps. The amplification and distortion of the H$\alpha$ line profile, characterized by a set of four indices, can be reproduced by the simulations. Although the constraints on the BLR models set by the observed single-epoch microlensing signal are not very robust, we found that flattened geometries (Keplerian disk and equatorial wind) can more easily reproduce the observed line profile deformations than a biconical polar wind. With an additional independent constraint on the size of the continuum source, the Keplerian disk model of the H$\alpha$ BLR is slightly favored.

Figures

Figures reproduced from arXiv: 1908.04178 by the authors.

**Figure 1.** Flux density ratio µ(v) = F l D /(M × F l B ) computed for the Hα emission line observed in the lensed quasar HE0435-1223. This ratio is illustrated as a function of the Doppler velocity over the useful [−8600, 8600] km s−1 velocity range (red lines). The continuumsubtracted spectra of images B and D (M × F l B and F l D , respectively) are superimposed, on an arbitrary flux scale (thin black and blue lines). The… view at source ↗

**Figure 2.** Two-dimensional histograms of simulated (WCI, RBI). These indices were measured from simulated line profiles that arise from the BLR models KD, PW, and EW seen at inclinations 22◦ , 44◦ , and 62◦ . The BLR models that have an emissivity ǫ0 (rin/r) q that sharply decreases with radius, i.e., q = 3, are illustrated in the left panel, while those characterized by a slowly decreasing emissivity, i.e., q = 1.5, are illus… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

15 extracted references · 12 canonical work pages

[1]

A., Kochanek, C

Blackburne, J. A., Kochanek, C. S., Chen, B., Dai, X., & Chart as, G. 2014, ApJ, 789, 125

work page 2014
[2]

A., Pooley, D., Rappaport, S., & Schechter, P

Blackburne, J. A., Pooley, D., Rappaport, S., & Schechter, P . L. 2011, ApJ, 729, 34

2011
[3]

Braibant, L. 2018, Probing the Broad Emission Line Region of Quasars using Gravitational Microlensing, PhD Thesis, University of Liège (http://hdl.handle.net/2268/221427) Article number, page 4 of 7 D. Hutsemékers et al.: Microlensing of the quasar broad line region. II. HE0435-1223

work page 2018
[4]

2016 , A&A, 592, A23

Braibant, L., Hutsemékers, D., Sluse, D., & Anguita, T. 2016 , A&A, 592, A23

work page 2016
[5]

Braibant, L., Hutsemékers, D., Sluse, D., Anguita, T., & Gar cía-V ergara, C. J. 2014, A&A, 565, L11

work page 2014
[6]

2017, A&A, 607, A32

Braibant, L., Hutsemékers, D., Sluse, D., & Goosmann, R. 2017, A&A, 607, A32

2017
[7]

Fian, C., Mediavilla, E., Jiménez-Vicente, J., Muñoz, J. A. , & Hanslmeier, A. 2018, ApJ, 869, 132

work page 2018
[8]

20 13, ApJ, 764, 160 Jiménez-Vicente, J., Mediavilla, E., Kochanek, C

Guerras, E., Mediavilla, E., Jimenez-Vicente, J., et al. 20 13, ApJ, 764, 160 Jiménez-Vicente, J., Mediavilla, E., Kochanek, C. S., & Muñoz, J. A. 2015, ApJ, 799, 149 Jiménez-Vicente, J., Mediavilla, E., Kochanek, C. S., et al . 2014, ApJ, 783, 47

work page 2015

Show all 15 references

[9]

M., Muñoz, J

Mosquera, A. M., Muñoz, J. A., Mediavilla, E., & Kochanek, C. S. 2011, ApJ, 728, 145

2011
[10]

2017, ApJ, 835, 1 32 O’Dowd, M., Bate, N

Motta, V ., Mediavilla, E., Rojas, K., et al. 2017, ApJ, 835, 1 32 O’Dowd, M., Bate, N. F., Webster, R. L., Wayth, R., & Labrie, K. 2011, MNRAS, 415, 1985

2017
[11]

T., Keeton, C

Richards, G. T., Keeton, C. R., Pindor, B., et al. 2004, ApJ, 6 10, 679

2004
[12]

2012, A&A, 544, A62

Sluse, D., Hutsemékers, D., Courbin, F., Meylan, G., & Wambs ganss, J. 2012, A&A, 544, A62

2012
[13]

2011, A&A, 528, A1 00

Sluse, D., Schmidt, R., Courbin, F., et al. 2011, A&A, 528, A1 00

2011
[14]

1999, Journal of Computational and Applied M athematics, 109, 353

Wambsganss, J. 1999, Journal of Computational and Applied M athematics, 109, 353

1999
[15]

B., O’Dowd, M., & Webster, R

Wayth, R. B., O’Dowd, M., & Webster, R. L. 2005, MNRAS, 359, 56 1 Article number, page 5 of 7 A&A proofs: manuscript no. aa31087 Appendix A: Examples of simulated magniﬁcation proﬁles µ(u ) Given the di ﬃculty of accurately measuring the whole magniﬁ- cation proﬁle µ(v) and th...

2005

Pith tools

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

[1] [1]

A., Kochanek, C

Blackburne, J. A., Kochanek, C. S., Chen, B., Dai, X., & Chart as, G. 2014, ApJ, 789, 125

work page 2014

[2] [2]

A., Pooley, D., Rappaport, S., & Schechter, P

Blackburne, J. A., Pooley, D., Rappaport, S., & Schechter, P . L. 2011, ApJ, 729, 34

2011

[3] [3]

Braibant, L. 2018, Probing the Broad Emission Line Region of Quasars using Gravitational Microlensing, PhD Thesis, University of Liège (http://hdl.handle.net/2268/221427) Article number, page 4 of 7 D. Hutsemékers et al.: Microlensing of the quasar broad line region. II. HE0435-1223

work page 2018

[4] [4]

2016 , A&A, 592, A23

Braibant, L., Hutsemékers, D., Sluse, D., & Anguita, T. 2016 , A&A, 592, A23

work page 2016

[5] [5]

Braibant, L., Hutsemékers, D., Sluse, D., Anguita, T., & Gar cía-V ergara, C. J. 2014, A&A, 565, L11

work page 2014

[6] [6]

2017, A&A, 607, A32

Braibant, L., Hutsemékers, D., Sluse, D., & Goosmann, R. 2017, A&A, 607, A32

2017

[7] [7]

Fian, C., Mediavilla, E., Jiménez-Vicente, J., Muñoz, J. A. , & Hanslmeier, A. 2018, ApJ, 869, 132

work page 2018

[8] [8]

20 13, ApJ, 764, 160 Jiménez-Vicente, J., Mediavilla, E., Kochanek, C

Guerras, E., Mediavilla, E., Jimenez-Vicente, J., et al. 20 13, ApJ, 764, 160 Jiménez-Vicente, J., Mediavilla, E., Kochanek, C. S., & Muñoz, J. A. 2015, ApJ, 799, 149 Jiménez-Vicente, J., Mediavilla, E., Kochanek, C. S., et al . 2014, ApJ, 783, 47

work page 2015