{"id":"3841a2a6-6f10-4985-ae13-f12f183086eb","arxiv_id":"2504.15757","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":14,"one_line_summary":"First high-resolution emission spectroscopy of TOI-2109b finds Fe I and CO dayside emission, a thermal inversion, and nightside temperature upper limits that the authors interpret as inefficient heat transport.","lead":"Astronomers detected iron and CO emission from the dayside of the ultra-hot Jupiter TOI-2109b, which shows a thermal inversion over roughly 3200 K to 4600 K. The study adds a benchmark data point in an unexplored temperature gap between known hot Jupiters and the extreme planet KELT-9b.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'inefficient heat transport' conclusion rests on an unquantified 'slightly closer' comparison to 1D HELIOS models and a non-constraining line-broadening measurement; the data likely cannot distinguish the two heat-redistribution scenarios.","rationale":"The reader's weakest_assumption identifies the same load-bearing premise: that the retrieved T-p profile's position between the two HELIOS models can be read as a measure of heat-transport efficiency. My stress-test agrees and sharpens the concern. The quantitative values in Table 3 show that the retrieval uncertainties are large compared with the expected difference between the two HELIOS models, and the paper provides no statistical comparison (no delta-chi2, no evidence ratio, no posterior overlap calculation). The rotation-broadening argument is also non-constraining because the recovered v_eq is consistent with, and not an upper limit on, tidally locked rotation. The paper's own hedged wording ('slightly closer', 'may have', 'further observations are needed') indicates the authors recognize the provisional nature of the conclusion, but the title and abstract state it as a main finding. The core detections of Fe I (S/N=4.3, tentative) and CO (S/N=6.3) and the thermal inversion are likely robust, so the paper remains valuable. The appropriate editorial outcome is unchanged from the reader's CONDITIONAL verdict: the heat-transport interpretation should be reworded or supported by a quantitative model-comparison before the claim is promoted. I found no new load-bearing concern beyond the one identified in the reader's verdict, and I set verdict_should_be to UNCHANGED because the reader's CONDITIONAL recommendation already captures the needed revision.","tokens_in":24334,"tokens_out":7501,"duration_ms":73231,"concrete_test":"Re-run the dayside retrieval with the T-p profile fixed to each of the two HELIOS profiles (no and full heat redistribution) using the publicly released data, and compute the Bayesian evidence ratio or delta-BIC for the two scenarios. Separately, overlay the 68% credible interval of the retrieved T-p profile at each pressure in Fig. 6 and check whether both HELIOS profiles fall inside this envelope across the inversion region. If the evidence ratio is below ~3, or both HELIOS profiles lie within the 1-sigma credible interval, then the data cannot distinguish the two heat-redistribution scenarios and the 'inefficient heat transport' conclusion should be removed or heavily qualified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline conclusion of inefficient day-to-night heat transport rests primarily on the statement in Sect. 5.2.1 (Fig. 6) that the retrieved dayside T-p profile lies 'slightly closer' to the no-redistribution HELIOS model than to the full-redistribution one. This comparison is not quantified, and the retrieval uncertainties are too large to support it: Table 3 gives T1 = 4601 +1557/-609 K, T2 = 3173 +300/-727 K, and log p1 = -4.92 +1.99/-1.89 dex. The credible envelope of the retrieved profile is therefore several hundred to over a thousand kelvin wide, while the separation between the two HELIOS end-member profiles is expected to be only a few hundred kelvin in the relevant pressure range. Without a model-comparison statistic, 'slightly closer' is an anecdote, not evidence. The supporting line-broadening argument is similarly non-constraining: the retrieved v_eq = 12.07 +5.75/-5.78 km/s (1-sigma range 6.3-17.8 km/s) contains the tidally locked value of 10 km/s and extends to 17.8 km/s, so it cannot rule out strong dynamical processes. Thus the heat-transport efficiency claim is not established at the level suggested by the title and abstract, even though the individual detections of Fe and CO emission are likely robust.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents the first in-depth atmospheric characterization of the ultra-hot Jupiter TOI-2109b using high-resolution emission spectroscopy from CARMENES (dayside) and CRIRES+ (day- and nightside). Cross-correlation analysis yields a CO detection at S/N=6.3 and a tentative Fe I signal at S/N=4.3, with no H2O detection on the dayside and no line detections on the nightside. A Bayesian retrieval that combines the high-resolution spectra with TESS and Palomar/WIRC secondary-eclipse photometry constrains the dayside T-p profile to show a thermal inversion between roughly 3200 K and 4600 K, gives [M/H] consistent with the stellar value, places only a lower limit on C/O, and retrieves a rotation broadening parameter consistent with tidal locking. The nightside retrieval provides upper limits on nightside temperature and temperature gradient. The authors compare the retrieved dayside profile with HELIOS models with no and full heat redistribution and, together with the broadening result and day-night temperature contrast, conclude that heat transport is inefficient.","tokens_in":24718,"tokens_out":4557,"duration_ms":44463,"significance":"If the conclusions hold, this paper is significant because TOI-2109b occupies the poorly explored equilibrium-temperature gap between cooler ultra-hot Jupiters and KELT-9b, and the detection of Fe I and CO emission with a thermal inversion provides an important data point for thermochemical and dynamical models. The study also demonstrates a useful framework for combining high-resolution emission spectroscopy with broadband photometry in a Bayesian retrieval, and it makes reduced spectra publicly available. The nightside upper limits and the careful treatment of SYSREM iteration choice are commendable. However, the headline claim of inefficient heat transport rests on a qualitative model comparison and a non-constraining broadening measurement, and the Fe I detection is described as tentative in the text but as a firm detection in the abstract and conclusions; these issues need to be addressed before the central claims can be considered established.","major_comments":[{"comment":"The inference of inefficient heat transport rests on the statement that the retrieved dayside T-p profile is \"slightly closer\" to the no-redistribution HELIOS model than to the full-redistribution one, but no quantitative comparison statistic is provided. Given the retrieved T1 = 4601 +1557/-609 K, T2 = 3173 +300/-727 K, and log p1 = -4.92 +1.99/-1.89 dex (Table 3), the credible envelope of the retrieved profile is several hundred to over a thousand kelvin wide, while the separation between the two HELIOS end-member profiles is expected to be only a few hundred kelvin in the relevant pressure range. The data therefore do not discriminate between the two redistribution scenarios. I recommend adding a formal model-comparison statistic (e.g., a likelihood ratio, information criterion, or direct distance metric between the retrieved profile and each HELIOS profile) or explicitly downgrading the heat-transport conclusion to a tentative suggestion, and adjusting the abstract, title, and Conclusions accordingly.","section":"Sect. 5.2.1 (Fig. 6) and Conclusions item 3"},{"comment":"The claim that the retrieved line broadening indicates \"the absence of strong dynamical processes\" is not supported by the measurement. The retrieved v_eq = 12.07 +5.75/-5.78 km/s gives a 1-sigma interval of roughly 6.3-17.8 km/s, which contains the tidally locked value of about 10 km/s and extends to 17.8 km/s; this is consistent with tidal locking but does not exclude substantial additional broadening from winds or turbulence. The caution expressed later in the same section is appropriate, but the abstract (\"indicating the absence of strong dynamical processes\") and Conclusions item 3 overstate what the broadening measurement can show. The dynamical contribution should either be modeled explicitly or this supporting argument should be removed or qualified.","section":"Sect. 5.2.1, v_eq argument"},{"comment":"The Fe I detection is described in the text as \"tentative evidence\" with a maximum S/N of 4.3, yet the abstract and Conclusions state that Fe I emission lines were identified or detected. At S/N=4.3, with searches over Kp, vsys, and the number of SYSREM iterations, a false-alarm probability that accounts for the number of independent trials should be reported before Fe I is promoted to a detection in the headline results. If the trials-adjusted significance is not robust, the manuscript should consistently describe the Fe I signal as tentative, which would also affect the title's claim of \"Detection of Fe and CO emission lines\".","section":"Sect. 4.4.1, abstract, and Conclusions item 1"},{"comment":"The comparison with HELIOS models is used as the primary evidence for limited heat transport, but the manuscript does not specify how the retrieved median T-p profile is compared with the model profiles (e.g., over which pressure range, with what weighting, and whether the full posterior is used). The statement in the text that the retrieved profile is \"slightly closer\" to one model is anecdotal rather than quantitative, especially given the large asymmetric uncertainties on the low-pressure anchor. I request that the authors either supply a quantitative comparison, such as a distribution of distances between posterior samples and each HELIOS profile, or clearly state that the data cannot distinguish the two scenarios.","section":"Sect. 5.2.1, HELIOS comparison methodology"}],"minor_comments":[{"comment":"Consider labeling the Fe I signal as \"tentative\" in the abstract and in the Conclusions list, for consistency with the body of the paper.","section":"Abstract and Sect. 4.4.1"},{"comment":"The Wong et al. (2021) dayside temperature point is placed outside the T-p space with no associated pressure; this makes the comparison with the retrieved profile difficult to interpret. Consider placing the point at a representative photospheric pressure or removing it from the T-p panel.","section":"Fig. 6"},{"comment":"The Gaussian likelihood in Eq. (3) includes the noise-scaling parameter beta, but the manuscript does not state whether beta is marginalized or profiled and does not report the number of effectively independent spectral channels used in the calculation; a brief clarification would help the reader assess the reported uncertainties.","section":"Sect. 5.1, Eq. (3)"},{"comment":"The nightside columns list [M/H] and C/O as \"0\" and \"0.55\", but the text states that these were fixed to solar values; the table should clearly distinguish fixed parameters from free parameters rather than using prior-like notation.","section":"Table 3"},{"comment":"No information is given about MCMC burn-in, thinning, or convergence checks (e.g., autocorrelation times) for the 32 walkers with 30,000 steps; a short statement on convergence would strengthen the reported parameter intervals.","section":"Sect. 5.1"},{"comment":"The notation \"3sys\", \"3eq\", and \"3p\" appears to be an encoding artifact for the systemic velocity, equatorial velocity, and planetary velocity; please ensure the typeset symbols (v_sys, v_eq, v_p) are correct in the final version.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"I do not see grounds for rejection: the CO detection, the thermal inversion, and the retrieval framework are credible, and the nightside upper limits are a useful contribution. The main risk is the heat-transport conclusion, which is presented in the title and abstract as evidence but rests on a qualitative comparison that the data do not yet support. I would require either a quantitative model comparison or a clearly softened claim. The Fe I detection should also be labeled tentative unless a trials-adjusted significance is provided."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is the first atmospheric study of TOI-2109b, a planet sitting in the unexplored Teq gap, and the CO detection and dayside T-p constraints are probably solid. The headline conclusion about inefficient heat transport is not supported by the evidence as presented. The paper deserves serious refereeing, but the authors need to rephrase or strengthen that claim.\n\nWhat's genuinely new: high-resolution emission spectroscopy of both hemispheres. CO at S/N 6.3 is a clean detection, and the thermal inversion inferred from emission lines is consistent with the dayside being hot. The joint retrieval with published TESS and Palomar/WIRC eclipse depths gives a useful dayside T-p profile and nightside temperature upper limits around 2400 K. The data are public on Zenodo. The SYSREM iteration handling and telluric removal follow the established playbook, and the injection-recovery tests are a good check against over-cleaning.\n\nWhere it's soft: the Fe i detection is S/N=4.3 and the text in Sect. 4.4.1 calls it 'tentative', but the abstract and conclusions list it as a detection. That's a mismatch that should be fixed—either run the analysis to justify a detection or label it tentative throughout. More importantly, the heat transport conclusion rests on a 'slightly closer' comparison of the retrieved T-p profile to a no-redistribution HELIOS model versus a full-redistribution one, with no model comparison statistic. Given the retrieval uncertainties (T1 = 4601 +1557/-609 K, T2 = 3173 +300/-727 K), the two end-member models are likely within the credible envelope. And the rotation broadening, v_eq = 12.07 +5.75/-5.78 km/s, brackets the tidally locked value of 10 km/s, so it cannot support the claim of 'absence of strong atmospheric dynamics' on its own. The authors do include the caveat 'further observations are needed' in Sect. 5.2.1, but that caveat does not make its way into the abstract or title with appropriate force.\n\nMinor: the nightside retrieval fixes [M/H], C/O, and v_eq to solar/tidally locked values, which is reasonable for upper limits but does constrain the inference. No retrieval code is shipped, though the reduced spectra are public.\n\nWho this is for: anyone working on UHJ atmospheres or high-resolution cross-correlation retrievals. The target is well chosen, and the paper is a solid data paper for the dayside CO and temperature structure. The heat transport framing needs a dose of humility before it should appear in an abstract.\n\nRecommendation: send it to peer review. It is a new target with public data and standard methods; referees can sort out the wording. If this were my call I'd request a revision: report Fe i as tentative throughout, and demote the heat transport claim to a suggestion, or add a quantitative model comparison.","headline":"First atmospheric characterization of TOI-2109b; the CO detection and dayside T-p constraints look solid, but the 'inefficient heat transport' headline is not supported by the quantitative evidence.","tokens_in":25520,"tokens_out":3825,"would_cite":true,"duration_ms":33266,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"TOI-2109b, an ultra-hot Jupiter in an unexplored temperature gap, emits Fe I and CO lines from an inverted dayside while its nightside stays featureless, which the authors read as evidence that it transports little heat from day to night.","keywords":["ultra-hot Jupiter","exoplanet atmospheres","high-resolution spectroscopy","thermal inversion","heat redistribution","emission spectroscopy","atmospheric retrieval","TOI-2109b"],"falsifier":"When a phase-resolved thermal light curve of TOI-2109b is measured, the claim will be settled: a nightside brightness temperature above roughly 2400 K, or the appearance of nightside line features in emission or absorption, would contradict the paper's picture of a nearly isothermal nightside and inefficient day-night heat transport.","tokens_in":24092,"feed_emoji":"🪐","tokens_out":15115,"duration_ms":124939,"temperature":0.7,"pith_summary":"TOI-2109b is an ultra-hot Jupiter that sits in the temperature gap between 3500 K and 4500 K, a regime where no other gas giant has been characterized in detail. Using high-resolution emission spectra of both hemispheres, the paper detects Fe I and CO lines in emission on the dayside, proving that the upper atmosphere is hotter than the lower atmosphere, a thermal inversion extending from about 3200 K to 4600 K. No water is seen, and no line emission at all is detected from the nightside, where the retrieval sets an upper temperature limit near 2400 K and a maximum temperature gradient of roughly 700 K/log bar. Comparing the retrieved dayside temperature-pressure profile with model atmospheres that assume either no or full heat redistribution, the paper argues that TOI-2109b transports heat inefficiently from day to night and represents a transitional atmospheric state between cooler ultra-hot Jupiters and the extreme object KELT-9b.","feed_headline":"Iron and CO emission lines expose TOI-2109b's heat imbalance","feed_subtitle":"Dayside Fe and CO lines show a hot upper layer; a featureless nightside under 2400 K means almost no heat crosses.","key_machinery":"The central machinery is the two-point temperature-pressure profile, a curve fixed by a low-pressure temperature point and a high-pressure temperature point connected linearly in log-pressure, used to model each hemisphere. Emission lines found by cross-correlating the reduced spectra against Fe I, CO, and H2O templates are the direct evidence for the dayside inversion, since only an atmosphere that gets hotter with altitude produces lines in emission. The same profile, forward-modeled with a radiative transfer code and fit jointly to high-resolution spectra and eclipse photometry, converts line strengths into temperatures and abundances. Finally, placing the retrieved dayside profile between model atmospheres computed with no and with full day-to-night heat redistribution is the step that turns the retrieved temperatures into a conclusion about heat transport efficiency.","core_discovery":"The paper's central claim is that the atmosphere of TOI-2109b is lopsided in a specific, measurable way. Detected Fe I and CO emission lines place the dayside in a thermal inversion: temperatures rise from around 3200 K at depth to about 4600 K in the upper atmosphere, and a Bayesian retrieval that combines the high-resolution spectra with TESS and Palomar/WIRC secondary-eclipse photometry finds the high-pressure layers near 3600 K. The absence of H2O emission follows naturally if the dayside is hot enough to dissociate water. On the nightside, the absence of any line signal yields only upper limits, temperatures below roughly 2400 K at pressures above $10^{-4}$ bar and a maximum temperature gradient of about 700 K/log bar, consistent with a nearly isothermal, featureless nightside. Because the retrieved dayside profile sits between, and slightly closer to, the model with no day-to-night heat redistribution, and because the measured line broadening matches tidal locking rather than strong winds, the authors conclude that heat transport across the planet is inefficient, and that TOI-2109b fills a transitional thermochemical regime between cooler ultra-hot Jupiters and KELT-9b.","pith_inferences":["Because the submitted claim of inefficient heat transport rests on the retrieved profile being only slightly closer to the no-redistribution model, a direct phase-curve measurement of the planet's day-night brightness contrast would settle the question more cleanly than the current model comparison.","The 1-sigma range on the rotation velocity, about 6.3 to 17.8 km/s, contains the tidally locked value of about 10 km/s, so the paper's absence-of-strong-dynamics conclusion is weaker than its own wording suggests; a higher signal-to-noise detection of Fe I or CO could tighten this constraint and either confirm or overturn the low-dynamics interpretation.","If the transitional-regime picture is right, other ultra-hot Jupiters in the 3500-4500 K gap should show CO emission without H2O, so a small survey of such planets would directly test the thermochemical transition proposed here.","The featureless nightside could equally arise from a temperature structure that combines emission and absorption features, which the two-point profile cannot represent; using a more flexible nightside temperature parameterization would distinguish these cases."],"forward_implications":["TOI-2109b becomes the first characterized benchmark in the 3500-4500 K equilibrium-temperature gap, giving observers a target for testing how ultra-hot Jupiter thermochemistry changes across this regime.","The retrieved dayside profile implies that a strong upper-atmosphere absorber drives the thermal inversion and that water is thermally dissociated, predicting H2O will remain undetectable in similar planets at or above these temperatures.","The nightside upper limits imply that high-resolution line searches on TOI-2109b's nightside will stay empty unless the nightside temperature-pressure profile is more complex than a single-gradient structure, so nondetections should be interpreted as thermal structure rather than missing chemistry.","If line broadening is indeed only tidal rotation, then any future measurement of extra broadening in Fe I or CO lines would signal atmospheric dynamics, making the $v_{\\rm eq}$ constraint a direct diagnostic of circulation.","The results place TOI-2109b between cooler ultra-hot Jupiters, which show CO and H2O emission, and KELT-9b, which shows no molecular features, implying the 3500-4500 K range is where molecular signatures drop out of ultra-hot Jupiter emission spectra."],"supporting_citations":[{"why":"Discovery paper that supplies the system parameters, the TESS and Palomar/WIRC secondary-eclipse depths, and the nightside photometric flux used as retrieval constraints.","marker":"Wong et al. 2021"},{"why":"Provides the radiative-transfer code used to generate cross-correlation templates and to forward-model the emission spectra in the retrieval.","marker":"Mollière et al. 2019"},{"why":"Fe I opacity line list from which the iron emission template is computed.","marker":"Kurucz 2018"},{"why":"CO line list used to build the carbon monoxide emission template.","marker":"Li et al. 2015"},{"why":"FastChem equilibrium chemistry, used to compute volume mixing ratios for the model atmospheres.","marker":"Stock et al. 2022"},{"why":"Establishes the Gaussian likelihood framework for high-resolution atmospheric retrievals that the paper's Bayesian analysis follows.","marker":"Brogi & Line 2019"},{"why":"Provides the median-normalization and filtering treatment that accounts for SYSREM distortions when comparing models to residual spectra.","marker":"Gibson et al. 2022"},{"why":"Presents the HELIOS radiative-convective model whose no- and full-redistribution profiles bracket the retrieved dayside temperature-pressure curve.","marker":"Malik et al. 2017"},{"why":"SYSREM algorithm used to remove stellar and telluric lines from the spectral time series before cross-correlation.","marker":"Tamuz et al. 2005"},{"why":"Theoretical prediction that strong absorption of stellar irradiation creates thermal inversions in hot Jupiter atmospheres, which the emission-line detection confirms.","marker":"Hubeny et al. 2003"}],"fun_headline_variants":["TOI-2109b's lopsided atmosphere: hot dayside, cold nightside","Ultra-hot Jupiter shows inefficient heat transport across hemispheres","Fe and CO lines reveal dayside inversion on TOI-2109b","TOI-2109b: dayside scorches, nightside stays dark and cold","Transitional hot Jupiter: no heat sharing between day and night"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim that TOI-2109b transports heat inefficiently rests on assuming that the one-dimensional model atmospheres with no and full heat redistribution bracket the real atmosphere, so the retrieved dayside temperature-pressure profile lying between them, and slightly closer to the no-redistribution case, can be converted into a measure of heat transport efficiency; if the models do not bracket reality, or if the two-point profile is too simple to represent the true structure, the inefficiency conclusion does not follow.","fun_headline_variants_meta":{"raw":{"variants":["TOI-2109b's lopsided atmosphere: hot dayside, cold nightside","Ultra-hot Jupiter shows inefficient heat transport across hemispheres","Fe and CO lines reveal dayside inversion on TOI-2109b","TOI-2109b: dayside scorches, nightside stays dark and cold","Transitional hot Jupiter: no heat sharing between day and night"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000345,"raw_usage":{"total_tokens":2017,"prompt_tokens":1195,"completion_tokens":822,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":811,"completion_tokens_details":{"reasoning_tokens":722}},"tokens_in":811,"tokens_out":822,"duration_ms":5871,"temperature":1.0,"reasoning_tokens":722,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:18:31.511285+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"When a phase-resolved thermal light curve of TOI-2109b is measured, the claim will be settled: a nightside brightness temperature above roughly 2400 K, or the appearance of nightside line features in emission or absorption, would contradict the paper's picture of a nearly isothermal nightside and inefficient day-night heat transport.","supporting_citations":[{"cited_title":"2021, AJ, 162, 256","cited_arxiv_id":null,"evidence_quote":"Discovery paper that supplies the system parameters, the TESS and Palomar/WIRC secondary-eclipse depths, and the nightside photometric flux used as retrieval constraints."}],"review_version":1}