{"id":"88431893-25d4-4aff-aa32-83823c0a196e","arxiv_id":"2506.13627","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"New synthetic-photometry transformations connect JWST, Roman, Euclid, HST and ground-based NIR filters, with 0.01 mag typical scatter for most stellar types and 0.1 mag for AGB stars.","lead":"This paper computes near-infrared filter transformations between JWST, Roman, Euclid, HST and ground-based surveys, and uses them to predict how the tip of the red giant branch appears in the new space telescopes. It gives astronomers a practical cross-calibration toolkit for distance-scale measurements across upcoming missions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Sparse library coverage of upper RGB and extreme metallicities makes the Roman/Euclid TRGB predictions extrapolations; the JWST check covers only an intermediate metallicity range, so a sensitivity refit is needed.","rationale":"The reader's weakest-assumption analysis identified the same load-bearing concern: the spectral libraries are sparsely populated in exactly the upper-RGB and extreme-metallicity regimes where the TRGB predictions are made. I agree that this is the most important vulnerability. The paper itself flags the limitation in Section 4 and in Figure 4, which is to its credit, but flagging a limitation does not remove it. The JWST empirical comparison provides partial external validation, yet only over a restricted metallicity interval, and the Euclid/Roman predictions remain unchecked. I considered whether missing coefficient uncertainties or the 0.05 mag zeropoint offset relative to Euclid Collaboration et al. (2022) should be the primary objection; these are real completeness concerns, but they do not threaten the central method as directly as the extrapolation issue does. The proposed sensitivity test would settle whether the sparse coverage actually changes the predictions; until such a test is done, the Roman and Euclid TRGB numbers should be described as extrapolations rather than calibrated predictions. The reader's conditional verdict is therefore appropriate, and my read does not move it.","tokens_in":26679,"tokens_out":4628,"duration_ms":51082,"concrete_test":"Refit the transformations actually used for the TRGB predictions (for example YE-HE, F115W-F200W, F106-F158, and the F090W/F150W JWST paths) after excluding all upper-RGB/AGB stars from the fitting sample, and again after restricting to the metallicity range actually populated by the library. Then recompute the Table 5 and Table 6 predictions at [Fe/H] = -3.2, -1.5, and +0.06. If the predicted TRGB magnitudes shift by more than roughly 0.05 mag, the sparse-coverage caveat propagates into the headline Roman/Euclid predictions; if they shift by less than 0.02 mag, the extrapolation is robust to the stated library limitation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's most consequential product is the set of predicted TRGB color-magnitude relations for JWST, Roman, and Euclid. Those predictions are obtained by applying piecewise linear transformations to theoretical TRGB models, but the spectral libraries used to fit those transformations contain few upper RGB stars and no stars at the extreme low or high metallicities of the theoretical grid, exactly as the paper states in Section 4. This matters because Table 5 extends from [Fe/H] = -3.2 to +0.06, while the only external validation, the empirical comparison in Figure 3, covers only the intermediate range -1.9 <= [Fe/H] <= -0.7. The reported JWST residuals of -0.06 and -0.05 mag are reassuring in that range but do not constrain behavior at the extremes. For Roman and Euclid there is no independent empirical check at all, so the headline TRGB magnitudes in Tables 5 and 6 inherit whatever uncorrected systematic error the fits have in the sparsely sampled upper-RGB and metal-poor/metal-rich regimes. The residual scatters quoted in the abstract (0.01 mag for RGB stars) are in-sample scatter about the fitted lines and do not by themselves constrain extrapolation error. This is not an internal contradiction, but it is the load-bearing weakness of the central claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs empirical color transformations between near-infrared broadband filters on JWST/NIRCam, Euclid/NISP, Roman/WFI, HST, and ground-based DECam/OGLE-IV/2MASS systems, using synthetic photometry from CALSPEC, IRTF/EIRTF, and XSL spectral libraries. The authors fit continuous piecewise linear transformations for 105 filter combinations, report coefficients, segment boundaries, per-segment residual RMS, and stellar counts, and apply the transformations to theoretical TRGB models to predict TRGB absolute magnitudes and color-magnitude relations in JWST, Roman, and Euclid bands. The JWST predictions are compared with the empirical Newman et al. (2024a) calibration, with mean residuals of -0.06 and -0.05 mag in F090W and F150W over the overlapping metallicity range.","tokens_in":26987,"tokens_out":5885,"duration_ms":58468,"significance":"If the transformations hold up, this is a broadly useful calibration product for the distance-scale and stellar-population communities, and the TRGB predictions provide a concrete bridge between theoretical models and upcoming JWST/Roman/Euclid observations. The paper is careful in several respects: it tabulates Vega/AB/ST/Sirius-Vega conversions for all filters, reports residual scatter by stellar type, compares its Euclid transformation with an existing Euclid Collaboration transformation, and explicitly acknowledges library coverage gaps. The method is a well-defined extension of the authors' earlier published work, and the self-citation is not circular because the underlying method was published independently. The main limitations are the absence of uncertainties on the fitted coefficients and the extrapolated nature of the TRGB predictions at extreme metallicities and for Roman/Euclid, which currently have no empirical validation.","major_comments":[{"comment":"The fitted coefficients c0, c1 and segment breakpoints X0, X1 in Tables 2–4 are reported without uncertainties. Because the TRGB predictions in Table 5 and the color-magnitude relations in Table 6 are obtained by applying these coefficients to theoretical models, the precision of the central TRGB product is unquantified. Please provide coefficient uncertainties, for example via bootstrap resampling over the spectra or via covariance propagation from the piecewise least-squares fit, and propagate them to Tables 5 and 6, or state explicitly that the tabulated TRGB values are point estimates with no formal uncertainty.","section":"§2.2 and §3.1, Tables 2–4"},{"comment":"The empirical check against Newman et al. (2024a) covers only the intermediate metallicity range -1.9 ≤ [Fe/H] ≤ -0.7, while Table 5 extends from [Fe/H] = -3.2 to +0.06 and Table 6 spans large color ranges. Section 4 acknowledges that the spectral libraries contain few upper RGB stars and no extreme-metallicity stars, so the reported JWST residuals of -0.06 and -0.05 mag do not constrain behavior at the metal-poor and metal-rich ends, and Roman and Euclid have no independent empirical check at all. The skeptical concern therefore lands: the TRGB predictions in Tables 5 and 6 are extrapolations in precisely the regimes where the theoretical TRGB slope is least certain. Please add a sensitivity test, for example refitting the transformations using only stars in the TRGB-relevant stellar-parameter range and recomputing the predictions, or at minimum quantify the library coverage in the color-metallicity cells used by Tables 5 and 6 and explicitly mark extrapolated entries.","section":"§3.2, Figure 3, Tables 5–6"},{"comment":"The abstract's 'typical residual dispersion around these transformations of 0.01 mag for Cepheid and RR Lyrae variables and RGB stars' is the in-sample RMS scatter about the fitted piecewise lines (σfit in Tables 2–4). It is not an estimate of the prediction uncertainty for the TRGB magnitudes in Tables 5 and 6, which additionally includes extrapolation error and spectral-library coverage effects. The text should state this distinction explicitly so that readers do not carry the 0.01 mag in-sample value over as the precision of the TRGB predictions.","section":"Abstract and §3.1"}],"minor_comments":[{"comment":"The phrase 'regularization parameterl' appears to contain a typo; it should read 'regularization parameter λ'.","section":"§3.1"},{"comment":"The value of the regularization parameter λ is described as 'the minimum uncertainty on Y', but the reader cannot reproduce the analysis without knowing the per-star total uncertainties; please define the quantity explicitly.","section":"§3.1"},{"comment":"The manuscript states that individual output magnitudes and uncertainties are available upon request; for reproducibility, consider depositing the synthetic photometry and fitting code in a permanent archive.","section":"§2.2"},{"comment":"The residual distribution figure would be clearer with sample sizes and either error bars or percentile labels for each stellar type, so that the reader can judge the statistical weight of the 0.01 mag and 0.1 mag dispersions.","section":"Figure 4"},{"comment":"The quadratic terms with large coefficients (25.78 and 44.95) look surprising; a brief note explaining the curvature, or showing the underlying points, would help readers trust the extrapolation.","section":"Table 6"}],"recommendation":"major_revision","confidential_remarks":"This is a solid and useful calibration paper. The missing coefficient uncertainties and the unquantified extrapolation of the TRGB predictions are the two load-bearing issues; both are fixable and do not, in my view, warrant rejection. The self-citation of Durbin et al. (2023) is appropriate because the method was published there independently. The paper fits the journal's scope as a reference calibration product."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this is a solid, straightforward calibration paper. The 105 filter transformations between JWST, Euclid, Roman, HST, and ground-based izY+IJHKs are built from real spectra (CALSPEC, IRTF, XSL) with piecewise linear fits and honest residual scatter per segment. The residuals are genuinely small for Cepheids, RR Lyrae, and RGB stars (~0.01 mag), and the paper calls out that O- and C-rich AGB stars are worse (up to 0.1 mag). That is the right kind of paper to have on hand when Roman and Euclid data arrive.\n\nWhat's new: Roman/WFI and Euclid/NISP transformation coefficients, and the TRGB color-magnitude predictions for those bands plus JWST. The JWST predictions check out against Newman et al. 2024a in the metallicity range that has empirical coverage (-1.9 to -0.7) with mean residuals around -0.05 to -0.06 mag. That is reassuring. The Euclid comparison with Euclid Collaboration 2022 shows agreement in slope but a 0.05 mag zeropoint difference; they flag it.\n\nThe soft spots, in order of importance:\n\n1. No coefficient uncertainties in Tables 2-4. The paper reports residual RMS and star counts, but not errors on c0, c1, segment breaks. For a calibration paper whose whole purpose is to let people transform photometry, users need to propagate uncertainties. This is the clearest concrete gap.\n\n2. The TRGB predictions for Roman and Euclid are extrapolations. The spectral libraries have few upper RGB stars and none at the extreme low or high metallicities in the theoretical grid (the paper says this explicitly in Section 4). The JWST empirical comparison only constrains the intermediate range. So the Roman/Euclid TRGB magnitudes in Tables 5 and 6 carry unknown systematic error from the fits in those sparsely sampled regimes. This is a real limitation, but the paper is honest about it, and for a demonstration application that is probably acceptable as long as users are warned. A sensitivity refit (e.g., dropping the few upper-RGB stars and seeing how much the TRGB predictions move) would turn the caveat into a number, and that would strengthen the paper.\n\n3. The synthetic photometry catalog is 'available upon request'. For reproducibility in a paper with 105 transformations, the catalog should be published as a data product.\n\nMinor: the paper notes angle-of-incidence and chip-dependent throughput variations for Roman/Euclid/JWST but doesn't model them. That's a reasonable scope choice, not a flaw.\n\nCitation pattern is fine. The method comes from Durbin et al. 2023, and citing that is legitimate, not circular. They engage with the existing Euclid Collaboration transformations.\n\nWho it's for: people doing distance-scale work (TRGB, Cepheids) and stellar populations who need to cross-calibrate photometry across JWST, Roman, Euclid, HST, and ground-based surveys. They will use these coefficients.\n\nMy take: deserves serious refereeing, and I'd accept after the uncertainties and data-access issues are fixed. The extrapolation caveat should be stated more prominently in the abstract or conclusions, but it doesn't invalidate the main product.","headline":"Solid calibration paper; the transformations are the real deliverable and hold up, while the TRGB predictions are honestly labeled extrapolations that would benefit from a sensitivity refit and published coefficient uncertainties.","tokens_in":27499,"tokens_out":3097,"would_cite":true,"duration_ms":30083,"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":"New near-infrared color transformations put JWST, Euclid, Roman, HST, and ground-based photometry on one grid, with ~0.01 mag scatter for distance-scale stars and tip-of-the-red-giant-branch predictions that match JWST.","keywords":["calibration","infrared astronomy","photometric systems","stellar colors","stellar populations","near-infrared","tip of the red giant branch","synthetic photometry"],"falsifier":"Measure the tip of the red giant branch in the same nearby galaxies with both Roman/WFI and JWST/NIRCam once Roman data exist, or with Euclid/NISP in fields already observed by JWST, and check whether the predicted Roman F106-F158 and Euclid YE-HE tip relations reproduce the JWST-derived tip to within the claimed 0.01-0.05 mag; alternatively, recompute the transformations after adding upper-red-giant-branch spectra with $[\\mathrm{Fe/H}]$ below $-2$ and above $-0.3$ from an independent library and see whether the piecewise-linear extrapolation shifts by more than the quoted scatter.","tokens_in":26508,"feed_emoji":"🔭","tokens_out":28079,"duration_ms":214781,"temperature":0.7,"pith_summary":"The paper aims to give the community a single, consistent set of color transformations for converting near-infrared photometry among the major space and ground-based systems: JWST/NIRCam, Euclid/NISP, Roman/WFI, HST, and ground-based izY and JHKs. It derives these from synthetic photometry of over a thousand real stellar spectra spanning the full HR diagram, fitting continuous piecewise-linear relations for 105 unique filter combinations. For the stars most used in distance-scale astronomy—Cepheids, RR Lyrae variables, and red giants—the typical residual scatter is about 0.01 mag, while oxygen- and carbon-rich asymptotic-giant-branch stars scatter by up to 0.1 mag. The paper then applies the transformations to predict the tip of the red giant branch (TRGB) distance indicator in the JWST, Euclid, and Roman bands, and shows that the JWST predictions agree with an empirical calibration to within about 0.06 mag. This matters because Euclid and Roman will cover huge sky areas, and a common photometric grid lets their NIR measurements be compared directly with existing HST and ground-based data.","feed_headline":"105 color transforms put JWST, Euclid, and Roman on one grid","feed_subtitle":"Distance-scale stars switch filters with ~0.01 mag scatter; predicted tip of the red giant branch matches JWST.","key_machinery":"The central object is the continuous piecewise-linear color transformation, fit with the pwlf package, which expresses the magnitude difference between two filters $Y$ as a sum of linear segments in a color $X$, with breakpoints chosen by Bayesian optimization; the output is a table of slopes and intercepts per segment for each filter pair. The engine behind it is synthetic photometry: each library spectrum is integrated against each filter's transmission curve, with fluxes and covariances propagated, and Vega zero-points set by the CALSPEC alpha Lyrae spectrum. The same machinery that converts any star's colors between systems is then applied to theoretical TRGB models to predict the tip's absolute magnitude as a function of color in the new filters.","core_discovery":"Working with 1269 spectra of 1061 stars drawn from four spectral libraries, the authors fit continuous piecewise-linear transformations of the form $Y = c_0 + c_1(X - X_0) + \\cdots$ between the near-infrared filters of JWST/NIRCam, Euclid/NISP, Roman/WFI, HST, and ground-based izY+IJHKs, producing coefficients for 105 unique filter combinations. The typical residual scatter is 0.01 mag for Cepheid, RR Lyrae, and RGB stars, and rises to around 0.1 mag for oxygen- and carbon-rich thermally-pulsing AGB stars. Using these transformations, the authors convert two recent theoretical TRGB calibrations into the Euclid, JWST, and Roman bandpasses, yielding color-magnitude relations for the tip of the red giant branch in each system. When the JWST F090W and F150W predictions are compared with a recent empirical calibration, the mean residuals are -0.06 and -0.05 mag over the common color range.","pith_inferences":["One consequence the authors leave implicit is that the Roman and Euclid TRGB relations can be tested before those surveys deliver: compare the transformed JWST relations against TRGB measurements in galaxies with geometric distances, to separate a filter-transformation offset from a TRGB calibration offset.","The sparse upper-RGB coverage implies the ~0.01 mag scatter is a property of the fitted stars, not of the extrapolated TRGB regimes; a dedicated library of upper-RGB spectra at metallicities below -2 and above -0.3 in [Fe/H] would directly test whether the extrapolation holds.","Because the machinery is filter-agnostic, the same coefficient-generation workflow could be applied to any bandpass set with known transmission curves, allowing the tables to be extended as new filter systems are defined.","The piecewise breakpoints track known opacity features (the Paschen jump at 0.82 µm and the 1.65 µm H- continuum turnover), so the coefficient tables also serve as a map of where different stellar atmospheres drive the largest cross-instrument corrections."],"forward_implications":["Photometry from JWST, Roman, Euclid, HST, and ground-based surveys can be shifted onto one grid with ~0.01 mag accuracy for Cepheids, RR Lyrae stars, and RGB stars, enabling direct apples-to-apples comparisons of distance-scale measurements.","TRGB distances measured with Roman and Euclid survey data can be calibrated directly with the published color-absolute-magnitude relations, without waiting for new empirical anchors.","The near-equivalent filter pairs (F090W and z, F106 and YE, F115W and F110W, F150W/F158/F160W, F200W/F213/KS) can be treated as interchangeable at the ~0.1 mag level in planning cross-mission observations.","The larger scatter for O- and C-rich AGB stars (up to 0.1 mag) implies that cross-calibrating AGB photometry between these systems needs more careful, type-specific treatment rather than relying on the global color terms.","Future revisions of the fits, as JWST, Euclid, and Roman calibrations mature, will directly propagate into the predicted TRGB zero points, so the published relations are tied to the current calibration state of each instrument."],"supporting_citations":[{"why":"Companion paper that supplies the synthetic-photometry method, the input spectra, and the rejection criteria that this work extends to new filters.","marker":"M. J. Durbin et al. (2023)"},{"why":"Provides the IRTF spectral library, one of the four training samples that span the NIR HR diagram.","marker":"J. T. Rayner et al. 2009"},{"why":"Extends the IRTF library to cooler, more evolved stars, adding the regime where AGB scatter becomes visible.","marker":"A. Villaume et al. 2017"},{"why":"Provides the X-Shooter spectral library (XSL), another component of the training sample.","marker":"Y.-P. Chen et al. 2014"},{"why":"Supplies the 3D interstellar dust map used to deredden all library spectra before synthetic photometry.","marker":"R. Lallement et al. (2022)"},{"why":"Defines the Euclid/NISP filter transmission curves and documents passband variations across the field of view.","marker":"Euclid Collaboration et al. (2022)"},{"why":"Provides the pwlf code that fits the continuous piecewise-linear transformation coefficients.","marker":"C. F. Jekel & G. Venter 2019"},{"why":"Theoretical TRGB calibration in J-KS and F110W-F160W, transformed here into Euclid, JWST, and Roman bands.","marker":"A. Serenelli et al. (2017)"},{"why":"BaSTI model suite giving TRGB absolute IJHK magnitudes as a function of metallicity, converted to the new systems.","marker":"A. Pietrinferni et al. (2021)"},{"why":"Empirical JWST TRGB calibration against which the transformed F090W and F150W predictions are checked.","marker":"M. J. B. Newman et al. (2024a)"}],"fun_headline_variants":["105 NIR color transforms link five telescopes on one grid","JWST, Euclid, Roman colors unified by 105 transformations","One photometric grid for JWST, Euclid, Roman, HST, and more","0.01-mag scatter links distance-scale stars across 105 filters","Cepheid and RR Lyrae colors align to 0.01 mag across 105 filters"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the spectral libraries contain enough stars in the relevant regimes for the piecewise fits to be valid; they have almost no upper red-giant-branch stars and none at the extreme low or high metallicities where the tip-of-the-red-giant-branch color-magnitude relation is least certain, so the Euclid and Roman tip predictions rest on fits calibrated mostly on main-sequence, giant, and asymptotic-giant-branch stars.","fun_headline_variants_meta":{"raw":{"variants":["105 NIR color transforms link five telescopes on one grid","JWST, Euclid, Roman colors unified by 105 transformations","One photometric grid for JWST, Euclid, Roman, HST, and more","0.01-mag scatter links distance-scale stars across 105 filters","Cepheid and RR Lyrae colors align to 0.01 mag across 105 filters"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000885,"raw_usage":{"total_tokens":3801,"prompt_tokens":902,"completion_tokens":2899,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":518,"completion_tokens_details":{"reasoning_tokens":2798}},"tokens_in":518,"tokens_out":2899,"duration_ms":20882,"temperature":1.0,"reasoning_tokens":2798,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:58:01.561222+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the tip of the red giant branch in the same nearby galaxies with both Roman/WFI and JWST/NIRCam once Roman data exist, or with Euclid/NISP in fields already observed by JWST, and check whether the predicted Roman F106-F158 and Euclid YE-HE tip relations reproduce the JWST-derived tip to within the claimed 0.01-0.05 mag; alternatively, recompute the transformations after adding upper-red-giant-branch spectra with $[\\mathrm{Fe/H}]$ below $-2$ and above $-0.3$ from an independent library and see whether the piecewise-linear extrapolation shifts by more than the quoted scatter.","supporting_citations":[{"cited_title":"J., Beaton , R","cited_arxiv_id":null,"evidence_quote":"Companion paper that supplies the synthetic-photometry method, the input spectra, and the rejection criteria that this work extends to new filters."}],"review_version":2}