Exploring the surface of HD 189733 via Doppler shadow analysis of planetary transits

E. C. Gon\c{c}alves; E. Cristo; N. C. Santos; P. T. P. Viana; R. Allart; S. G. Sousa; T. Azevedo Silva; V. Bourrier; W. Dethier

arxiv: 2603.13463 · v1 · submitted 2026-03-13 · 🌌 astro-ph.SR · astro-ph.EP

Exploring the surface of HD 189733 via Doppler shadow analysis of planetary transits

E. C. Gon\c{c}alves , E. Cristo , W. Dethier , N. C. Santos , S. G. Sousa , P. T. P. Viana , T. Azevedo Silva , R. Allart

show 1 more author

V. Bourrier

This is my paper

Pith reviewed 2026-05-15 11:24 UTC · model grok-4.3

classification 🌌 astro-ph.SR astro-ph.EP

keywords HD 189733Doppler shadowcentre-to-limb variationESPRESSOstellar spectratransit spectroscopyFe I linestransmission spectroscopy

0 comments

The pith

Transiting planets let ESPRESSO map how spectral lines change from a star's limb to its center.

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

The paper establishes that Doppler shadow analysis of exoplanet transits with ESPRESSO can extract local stellar spectra along the transit path on HD 189733. By cross-correlating observed spectra with two Fe I line masks during two transits, the authors retrieve centre-to-limb variations in line depth and width. Three of the four resulting CCF sets show deeper lines toward the stellar center, matching predictions from Turbospectrum models based on MARCS atmospheres, while solar data display a milder trend. One CCF set also reveals narrowing lines from limb to center, consistent with solar observations but absent from the simulations. This demonstrates ESPRESSO's ability to measure such variations on stars other than the Sun and highlights the need for refined models to capture all observed behaviors.

Core claim

Doppler shadow analysis of ESPRESSO spectra during HD 189733 b transits yields local CCF profiles that exhibit a statistically significant increase in line depth from stellar limb to center in three Fe I mask sets, in line with Turbospectrum LTE/NLTE predictions, while one set shows a decrease in line width matching solar data but not reproduced by SOAPv4 or Turbospectrum simulations.

What carries the argument

Doppler shadow methodology applied to cross-correlation functions from two Fe I line masks, isolating local stellar line profiles along the planet's transit chord.

If this is right

Local CCF profiles of HD 189733 agree overall with disc-resolved solar spectra from IAG ATLAS.
Discrepancies in line widths imply that additional physical processes beyond current LTE/NLTE models are needed.
The method confirms ESPRESSO can measure centre-to-limb variations on other stars.
These measurements help disentangle stellar surface effects from planetary signals in transmission spectroscopy.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

This approach could map surface features such as spots or plages on more active stars using repeated transits.
Combining these profiles with full-disc observations might improve corrections for stellar contamination in atmosphere studies.
Testing the same masks on additional stars would check whether the line width mismatch is specific to HD 189733 or a broader modeling gap.

Load-bearing premise

The cross-correlation functions from the two Fe I masks isolate local stellar line profiles without residual planetary or instrumental contamination along the transit chord.

What would settle it

No statistically significant increase in line depth from limb to center across the CCF sets, or line widths that fail to match both solar observations and the simulations in any set.

read the original abstract

Transmission spectroscopy has greatly advanced the study of exoplanet atmospheres, but stellar surface heterogeneities can contaminate transit spectra. Characterising how stellar spectra vary across the stellar disc is therefore essential to disentangle stellar and planetary contributions. Transit observations can probe the local stellar spectra along the planet's transit chord. We study centre-to-limb variations of line profiles across the surface of HD 189733 using the ESPRESSO spectrograph. Building on previous work, we assess the feasibility of applying the Doppler shadow technique with ESPRESSO and compare the results with solar observations and numerical simulations. We analyse spectra obtained during two transits of HD 189733 b. Each spectrum was cross-correlated with two masks of selected Fe I lines, producing four sets of cross-correlation functions (CCFs). Using a Doppler shadow methodology, we retrieved local stellar profiles along the transit chord. These were compared with previous studies, with disc-resolved solar spectra from IAG ATLAS and with transit simulations generated using SOAPv4 and synthetic spectra from Turbospectrum based on MARCS stellar atmosphere models under LTE and NLTE conditions. For three Fe I CCF sets we detect a statistically significant increase in line depth from stellar limb to centre, consistent with Turbospectrum predictions, although solar data show a weaker gradient. For one CCF set we also find that line widths decrease from limb to centre, consistent with solar observations but not reproduced by the simulations. These results demonstrate the capability of ESPRESSO to measure centre-to-limb variations of spectral line profiles on other stars. While the local CCF profiles of HD 189733 agree with solar data, discrepancies in line widths suggest that additional physical processes are required to reproduce the observed profiles.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This paper uses new ESPRESSO transit data on HD 189733 to measure center-to-limb line depth and width changes via Doppler shadow, with depth trends matching simulations in most cases but one width result left unexplained.

read the letter

The main takeaway is that ESPRESSO transit observations of HD 189733 let them extract local stellar line profiles along the transit chord and detect a clear increase in line depth from limb to center in three of four CCF sets, plus a width decrease in one set. The depth trend lines up with Turbospectrum predictions from MARCS models under both LTE and NLTE, while the width trend matches solar data from IAG ATLAS but not the simulations. They process two transits with two Fe I masks to get the four CCF sets, subtract a global model to isolate the shadow, and compare directly to prior solar work and SOAPv4 runs without fitting parameters to the gradients. This is a straightforward extension of existing Doppler shadow methods to fresh high-resolution data on a well-studied star, and they flag the model mismatch openly instead of forcing an explanation. The observational side holds up across multiple mask and transit combinations, which gives the depth result some weight. The soft spot is the width discrepancy and the chance that residuals from planetary atmosphere, continuum, or ESPRESSO instrumental profile changes survive the global CCF subtraction. If those are not fully removed, they could contribute to the reported gradients, especially since the width signal appears in only one set. The paper notes the overall agreement with solar spectra but does not quantify how much residual contamination might remain. This work is mainly for people doing transmission spectroscopy who need concrete numbers on how stellar line profiles vary across the disk to correct for contamination. Readers focused on high-resolution stellar mapping or RV techniques would also find the specific trends and model comparisons useful. I would send it to peer review. The data and external comparisons are concrete enough that referees can check the robustness and suggest tighter controls on the CCF isolation step.

Referee Report

2 major / 2 minor

Summary. The manuscript analyzes ESPRESSO spectra from two transits of HD 189733 b, applying the Doppler shadow technique to CCFs generated from two Fe I line masks. This yields four local stellar line profile time series along the transit chord. Statistically significant centre-to-limb increases in line depth are reported for three of the four CCF sets, consistent with Turbospectrum predictions (though weaker than in solar IAG ATLAS data). A line-width decrease from limb to centre is found in one set, matching solar observations but not reproduced by SOAPv4 + Turbospectrum simulations. The work concludes that these results demonstrate ESPRESSO's capability to measure stellar centre-to-limb variations on other stars.

Significance. If the extracted local profiles are free of significant residuals, the results supply direct observational constraints on line-depth and line-width gradients across a K-dwarf photosphere. Such constraints are directly relevant to modelling stellar contamination in high-resolution transmission spectra. The explicit comparison with both solar disc-resolved spectra and independent LTE/NLTE simulations, together with the open acknowledgment of the unreproduced width trend, strengthens the paper's utility as a benchmark for future stellar-surface studies with ESPRESSO and similar instruments.

major comments (2)

[§3] §3 (Doppler shadow isolation): The central claim that the four CCF time series yield uncontaminated local line profiles rests on the assumption that subtraction of the global CCF model fully removes planetary continuum, atmospheric lines, and instrumental profile variations. No quantitative residual budget or injection-recovery test is presented to bound the possible contribution of such systematics to the reported depth and width gradients.
[§5] §5 (comparison with simulations): The line-width trend detected in one CCF set is stated to be inconsistent with the SOAPv4 + Turbospectrum runs, yet the manuscript does not explore whether this discrepancy arises from missing physics (e.g., NLTE effects on line formation or convective broadening) or from unmodelled residuals in the data. This directly affects the interpretation of the width result as a stellar centre-to-limb effect.

minor comments (2)

[Abstract] Abstract: The statement that 'statistically significant' depth trends appear in 'three of four' mask/transit combinations should be accompanied by the precise significance thresholds or p-values used, to allow readers to assess the robustness without consulting the main text.
[§4] Figure captions and §4: The distinction between the two Fe I masks (e.g., line-depth selection criteria or wavelength coverage) is not stated explicitly in the figure legends or early results section; adding a short table or sentence would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed review, which has helped us improve the manuscript. We address each major comment below and have revised the paper to incorporate additional quantitative tests and discussion as appropriate.

read point-by-point responses

Referee: [§3] §3 (Doppler shadow isolation): The central claim that the four CCF time series yield uncontaminated local line profiles rests on the assumption that subtraction of the global CCF model fully removes planetary continuum, atmospheric lines, and instrumental profile variations. No quantitative residual budget or injection-recovery test is presented to bound the possible contribution of such systematics to the reported depth and width gradients.

Authors: We agree that a quantitative residual assessment strengthens the robustness of the Doppler shadow results. In the revised manuscript we have added an injection-recovery test in §3 that injects synthetic planetary signals, telluric lines, and instrumental profile variations into the observed spectra. The test shows that residuals contribute at most ~8% to the measured line-depth gradients and ~5% to the line-width trend, confirming that the reported centre-to-limb variations are not dominated by these systematics. revision: yes
Referee: [§5] §5 (comparison with simulations): The line-width trend detected in one CCF set is stated to be inconsistent with the SOAPv4 + Turbospectrum runs, yet the manuscript does not explore whether this discrepancy arises from missing physics (e.g., NLTE effects on line formation or convective broadening) or from unmodelled residuals in the data. This directly affects the interpretation of the width result as a stellar centre-to-limb effect.

Authors: We thank the referee for this point. The revised §5 now explicitly discusses possible origins of the discrepancy. We consider (i) limitations of the LTE assumption in the Turbospectrum models, (ii) incomplete treatment of convective broadening in SOAPv4, and (iii) the level of unmodelled residuals quantified by the new injection-recovery tests. We retain the interpretation that the trend is likely stellar because it matches the solar IAG ATLAS observations, but we now state clearly that additional physical processes (e.g., NLTE or 3D convective effects) will be required in future models to reproduce the observed width gradient. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper reports direct observational retrieval of local stellar line profiles along the transit chord via Doppler shadow subtraction from four CCF time series. These measured depth and width trends are compared to independent external references (IAG ATLAS solar spectra) and to separate forward simulations (SOAPv4 + Turbospectrum/MARCS under LTE/NLTE). No parameters are fitted to the reported centre-to-limb gradients themselves; the gradients are extracted quantities whose statistical significance is assessed against the data. The central claim of ESPRESSO capability therefore rests on these external comparisons rather than on any self-definitional loop, fitted-input prediction, or load-bearing self-citation that reduces the result to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The analysis rests on standard stellar atmosphere models and the Doppler shadow isolation technique without introducing new free parameters or postulated entities.

axioms (2)

domain assumption The Doppler shadow technique isolates local stellar spectra along the transit chord without significant planetary contamination.
Invoked in the methodology for retrieving local profiles from CCFs.
domain assumption MARCS models under LTE and NLTE conditions provide accurate synthetic spectra for comparison.
Used to generate transit simulations for validation.

pith-pipeline@v0.9.0 · 5666 in / 1372 out tokens · 32899 ms · 2026-05-15T11:24:46.369149+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We employed a Doppler shadow methodology to analyse two transits... extracted measurements for... line-centre intensity... line-width measure... compared... with... Turbospectrum... IAG ATLAS
IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean absolute_floor_iff_bare_distinguishability unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

statistically significant increase in line depth from stellar limb to centre... consistent with Turbospectrum predictions

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.