{"id":"1ab1039c-dfe9-4411-a270-41ee3527a8d7","arxiv_id":"2607.27467","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":8,"one_line_summary":"Synthesizer now combines stellar population spectra with Cloudy photoionisation grids to predict nebular line and continuum emission for galaxies, with a systematic map of modelling choices.","lead":"This paper shows how the open-source synthesizer package attaches photoionised-gas (nebular) emission to stellar population models, letting users generate spectra, line fluxes, and photometry for both toy galaxies and full cosmological simulations. It also maps how choices like stellar model, IMF, gas density, dust, and ionisation parameter shift key emission-line diagnostics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Default per-particle HII-region mapping (fixed nH, Z_gas=Z_SSP, single zone) is unvalidated; the only observational check relies on a dust prescription calibrated with the same SPS model.","rationale":"The paper is a genuine software contribution: open-source, version-controlled, with reproducible grid construction and a thorough parameter exploration. The reader's conditional acceptance is appropriate. The strongest concern is not a bug but an unvalidated physical mapping: every star-forming region in a cosmological simulation is collapsed to one Cloudy run with fixed density, spherical geometry, and SSP-matched gas metallicity. This is exactly the weakest assumption identified by the reader. The only observational comparison (EAGLE Hα LF) does not isolate the photoionisation model, because the dust attenuation model was calibrated with the same SPS model. I see no internal inconsistency in the grid construction or the parameter exploration; the equations and figures are coherent. The concern is about external validity, and it is load-bearing because the EAGLE application is presented as a demonstration of 'correct' nebular predictions. A concrete, feasible test would be to rerun the EAGLE predictions with simulation-based gas properties or a two-zone prescription. If the results are insensitive, the concern is retired; if sensitive, the paper should present default results as one choice among several and reduce the strength of the physical-motivation claim. Thus the verdict remains CONDITIONAL, unchanged from the reader's assessment.","tokens_in":36444,"tokens_out":7631,"duration_ms":78425,"concrete_test":"Repeat the EAGLE z=2.237 Hα LF, EW distribution, and BPT diagnostics with two alternative per-particle mappings: (i) gas-phase metallicity and density taken from the simulation's star-forming gas (rather than Z_SSP and fixed 10^2.5 cm^-3); (ii) a two-zone model splitting the ionising luminosity between a dense birth-cloud component and a diffuse ISM component, keeping all other settings fixed. If the dust-attenuated Hα LF shifts by more than ~0.3 dex in any bin, or the median Hα EW shifts by more than ~0.2 dex, the default single-zone mapping is not robust and the 'physically motivated' claim needs qualification.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that SYNTHESIZER 'correctly predicts nebular line and continuum emission' rests on the assumption that each star particle in a simulation can be represented by a single radiation-bounded Cloudy HII region with fixed hydrogen density nH=10^2.5 cm^-3, gas-phase metallicity equal to the SSP metallicity (Section 3.3), spherical geometry with R0=0.01 pc (Section 3.1.2), and the cube-root ionisation-parameter scaling (Eq. 4). No test is provided for this mapping against the resolved gas distribution in EAGLE or TNG50. The only quantitative observational validation, the z=2.237 Hα luminosity function compared to Sobral et al. (2013), uses the FLARES dust attenuation prescription that was itself calibrated to the z=5 UV luminosity function with the same BPASS v2.2.1 SPS model (Section 5.6). This makes the agreement partially circular for the SPS/dust combination, though not for the photoionisation physics. The paper itself flags the Z_gas=Z_SSP assumption as 'may not be fully self-consistent' (Section 3.3) and the ionisation-bounded stopping criterion as potentially affecting attenuation (Section 3.4). Since the EAGLE Hα LF and the parameter-exploration conclusions depend on this single-zone mapping, the lack of any direct test of its validity is the most load-bearing gap.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents the implementation of photoionised gas emission in the SYNTHESIZER package, using Cloudy to process stellar population grids and produce nebular line and continuum predictions. It describes the default modelling assumptions (reference ionisation parameter, spherical geometry, constant density, gas-phase metallicity tied to the SSP, depletion and grain prescriptions, ionisation-bounded stopping) and systematically explores the sensitivity of UV continuum slopes, emission-line luminosities, and diagnostic ratios to SPS model, IMF, ionisation parameter, density, abundance pattern, depletion, and dust. Example applications include a parametric galaxy, a single TNG50 galaxy, and the full EAGLE simulation, where Hα luminosity functions and equivalent width distributions are compared to observational data. The authors claim that SYNTHESIZER provides a flexible, physically motivated framework for modelling stellar and nebular emission.","tokens_in":36984,"tokens_out":3847,"duration_ms":39770,"significance":"If the central claim is borne out, this is a valuable open-source contribution: the code and pre-computed grids are publicly available with version-stamped releases, and the systematic parameter exploration provides a useful baseline for interpreting synthetic observations. The range of applications—from toy models to cosmological simulations—demonstrates genuine versatility. However, the validation is limited and partially circular: the single observational check of the Hα luminosity function uses a dust attenuation prescription calibrated with the same BPASS v2.2.1 SPS model, and the per-particle single-zone HII-region mapping is not directly tested. The paper is transparent about several acknowledged approximations (Z_gas=Z_SSP, grain scheme self-consistency), which is commendable but does not remove the need for robustness tests.","major_comments":[{"comment":"The default mapping of every star particle to a single constant-density spherical Cloudy HII region (n_H=10^2.5 cm^-3, R0=0.01 pc, Z_gas=Z_SSP, and the cube-root ionisation-parameter scaling of Eq. 4) is a strong idealisation when applied to cosmological simulation particles. The paper does not test this mapping against the resolved gas distribution in EAGLE or TNG50, nor does it quantify how varying n_H, U_ref, or the stopping column would affect the predicted Hα luminosity function. The EAGLE Hα LF and the equivalent-width distributions are central demonstrations, so a robustness test with plausible alternative mappings or an explicit estimate of the systematic uncertainty is needed to support the claim that the tool 'correctly predicts nebular line and continuum emission.'","section":"§5.6, §3.1.2, §3.3"},{"comment":"The agreement of the default-model Hα LF at z=2.237 with Sobral et al. (2013) is partly circular for the validation of the photoionisation modelling: the FLARES dust attenuation prescription was calibrated using the same BPASS v2.2.1 SPS model to match the z=5 UVLF (Vijayan et al. 2021), as acknowledged in the text. The agreement can therefore be driven largely by the dust calibration and SPS choice rather than by the photoionisation modelling itself. The authors should either compare intrinsic (dust-free) Hα LFs, use an independently calibrated dust prescription, or vary the SPS model with a fixed dust prescription to isolate the photoionisation contribution. The current figure cannot distinguish between these degeneracies.","section":"§5.6"},{"comment":"The default assumption that the gas-phase metallicity equals the SSP metallicity is explicitly acknowledged as 'may not be fully self-consistent,' and the depletion-to-grain implementation is likewise stated to be 'not fully self-consistent.' These assumptions underlie every grid and every application in the paper. The parameter exploration in Section 4 varies the metallicity of the stellar population and the gas simultaneously, so it does not isolate the effect of decoupling Z_gas from Z_SSP. A quantitative test of this approximation—e.g., a small grid with Z_gas offset from Z_SSP, or a discussion of the expected bias from abundance decoupling—would materially strengthen the claim that the framework is physically motivated and would help users assess the default grids.","section":"§3.3, §3.3.4"}],"minor_comments":[{"comment":"The data availability statement says that all scripts to generate the plots 'will be made publicly available on Github on the acceptance of the paper.' For a software-centric paper, releasing plot scripts at submission would aid reproducibility and reviewer verification.","section":"Data Availability"},{"comment":"The discussion of the stopping criterion notes that ionisation-bounded models with dust produce attenuation that depends on the ionising photon rate, but the quantitative effect on the EAGLE Hα LF (e.g., comparing to a fixed column-density stop) is not shown. A brief figure or table would clarify the impact of this modelling choice.","section":"§3.4"},{"comment":"The visual agreement with Sobral et al. (2013) and Khostovan et al. (2024) would be easier to assess with residuals or a reduced chi-square statistic; for the luminosity function, plotting the observed error bars and model uncertainties would also help.","section":"Fig. 29"},{"comment":"Minor typographical and formatting issues: 'Hiiregions' appears without proper spacing in several places; the caption of Fig. 23 says 'The same as Figure 21 but showing the resulting nebular continuum spectra' but the axes are identical to Fig. 21—consider clarifying the difference.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"This is a software-driven technical paper rather than a discovery paper, so the validation bar is about demonstrating that the tool works as advertised. The open code and grids are strengths. The main concern is that the single observational validation of the Hα luminosity function is partially circular through the dust prescription, and the single-zone HII-region mapping is not robustly tested. Both are fixable with additional calculations or a more careful framing of claims. I would not reject, but the authors should address these before publication in a journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does what it says: it integrates Cloudy photoionisation grids into synthesizer, the open-source synthetic-observation package, and demonstrates the result across toy models, a single TNG50 galaxy, and the full EAGLE volume. The real contribution is public infrastructure, not new physics. The grids and pipeline (syncretize) are version-controlled and documented, the paper ships with extensive diagnostics, and the people behind synthesizer have a track record of actually maintaining this kind of tool. That deserves credit.\n\nThe parameter exploration is largely a re-run of established photoionisation behaviour, and the authors admit this. The value is in having it done consistently across multiple SPS models, IMFs, densities, depletion patterns, and geometries, all under one open interface. Anyone who has spent weeks creating Cloudy grids for a simulation will see the point immediately.\n\nThe soft spots are real but not fatal. First, plot scripts are promised only on acceptance, which makes independent reproduction of the figures impossible right now. That is a standard but increasingly annoying practice; easy referee fix. Second, the default mapping of every star particle to a single constant-density, radiation-bounded Cloudy sphere with gas-phase metallicity equal to the SSP metallicity is a strong idealisation. The paper flags the self-consistency issue but does not test the mapping against the resolved gas properties in the simulations. For a software paper this is tolerable, but the authors should state it as a known uncertainty in the abstract or summary, not just in Section 3.3. Third, the one observational check—the EAGLE H-alpha luminosity function—uses the FLARES dust attenuation prescription that was calibrated with the same BPASS model. So the agreement partially reflects the calibration, not an independent test of the photoionisation physics. The stress-test note is right about that, though it is not a damning flaw; it just means the validation is weaker than it appears.\n\nThe central claim, that synthesizer provides a flexible framework for modelling stellar and nebular emission, holds up. The limitations are explicit, the code is open, and the paper is honest about what is explored versus what is new.\n\nWho is this for? People in galaxy formation who need synthetic spectra and line maps from simulations, and anyone who wants to compare SPS models consistently. It is not a paper that changes a physical picture. It is a good tool paper. Yes, it deserves a serious referee. The referee should ask for plot scripts now, a clear statement of the single-zone limitation, and ideally a sanity check where grid predictions are compared to a simulation's actual gas density and metallicity distribution. With those, it is a solid contribution.","headline":"Solid, useful software paper; the single-zone Cloudy-per-particle mapping is a real limitation but exactly the kind of thing a referee can push on without sinking the paper.","tokens_in":694,"tokens_out":2464,"would_cite":true,"duration_ms":36022,"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":"SYNTHESIZER now integrates photoionised gas emission into synthetic galaxy spectra","keywords":["photoionisation modelling","nebular emission","synthetic observations","stellar population synthesis","HII regions","spectral diagnostics","cosmological simulations","emission lines"],"falsifier":"A comparison with resolved HII regions in the local Universe where both the ionising stellar population and the gas-phase abundances are measured: if the observed line ratios deviate systematically from the model grid by more than the grid's sensitivity range, the equal-metallicity or single-geometry assumptions would be falsified.","tokens_in":36396,"feed_emoji":"🔭","tokens_out":4808,"duration_ms":46237,"temperature":0.7,"pith_summary":"SYNTHESIZER, an open-source package for generating synthetic galaxy observations, now includes emission from photoionised gas. The paper integrates stellar population synthesis grids with the Cloudy photoionisation code, allowing parametric and simulation-based galaxies to be rendered with nebular line and continuum emission. A systematic exploration shows which modelling assumptions most affect line luminosities, diagnostic line ratios, and UV continuum slopes. The package is demonstrated on a toy galaxy, a TNG50 galaxy, and the full EAGLE simulation, where the default model reproduces observed Hα luminosity functions and equivalent width distributions at z≈2.2.","feed_headline":"Synthetic galaxies now glow with photoionised gas emission","feed_subtitle":"SYNTHESIZER predicts nebular lines and continua from cosmological simulations, matching Hα and EW surveys at z≈2.","key_machinery":"The central machinery is the pre-computed spectral grid: each simple stellar population (SSP) from a stellar population synthesis model, such as BPASS, is run through the Cloudy photoionisation code to store incident, transmitted, and nebular spectra plus line luminosities. The key identity is the reference ionisation parameter scaling, US = (QH/QH,ref)^(1/3) Uref, which sets the ionisation parameter for each grid point in spherical geometry based on the ionising photon production rate relative to a reference SSP at t=1 Myr and Z=0.01. This avoids imposing an artificial geometry evolution when age or metallicity changes.","core_discovery":"The central claim is that SYNTHESIZER provides a flexible, physically motivated framework for modelling stellar and nebular emissions, serving as a vital link between theory and observations. The paper argues that processing simple stellar population spectra through Cloudy with a carefully chosen default configuration—spherical geometry with inner radius 0.01 pc, a reference ionisation parameter that scales with the cube root of the ionising photon rate, gas-phase metallicity matching the stellar metallicity, Jenkins (2009) depletion with F*=0.5, and Orion-type grains—yields reliable predictions across a wide range of ages, metallicities, and galaxy types. The parameter exploration quantifie","pith_inferences":["A natural extension would be to replace the single-zone Cloudy model with a distribution of ionisation parameters or escape fractions per star particle, which could be tested against resolved HII region observations.","The strong sensitivity of the BPT diagram's high-metallicity locus to dust depletion suggests that metallicity calibrations derived without dust-depletion corrections could be systematically biased, a caution that applies beyond this package.","Coupling the grids to a machine-learning emulator, as the authors note is underway, could make full photoionisation models tractable in Bayesian parameter estimation and simulation-based inference.","One could test the equal-metallicity assumption by applying the package to galaxies with independently measured stellar and gas-phase metallicities from stacked spectra."],"forward_implications":["Synthetic spectra from cosmological simulations will now include physically motivated nebular lines and continua, enabling direct comparison with JWST, Euclid, and future ELT spectroscopy.","The framework can be used to calibrate dust attenuation prescriptions by matching predicted Hα luminosity functions to observed ones.","The sensitivity maps show which parameters (e.g., depletion, grain mixture) must be constrained before line-ratio diagnostics can be trusted for metallicity or ionisation parameter inference.","The reference ionisation parameter scaling makes predictions robust to SPS model choice for line ratios, while absolute line luminosities remain sensitive to the ionising photon budget.","The package's flexibility means the same grids can be used for SED fitting and forward modelling, unifying interpretation and prediction."],"fun_headline_variants":["SYNTHESIZER models photoionised gas in synthetic galaxies","Nebular emission added to synthetic galaxy toolkit","Galaxy simulations now glow with photoionised gas","Synthetic galaxies get photoionised gas emission","Tool links stellar and nebular emission in galaxy models"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that each star particle's unresolved gas can be represented by a single spherical Cloudy HII-region model whose gas-phase metallicity equals the stellar metallicity.","fun_headline_variants_meta":{"raw":{"variants":["SYNTHESIZER models photoionised gas in synthetic galaxies","Nebular emission added to synthetic galaxy toolkit","Galaxy simulations now glow with photoionised gas","Synthetic galaxies get photoionised gas emission","Tool links stellar and nebular emission in galaxy models"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000472,"raw_usage":{"total_tokens":2175,"prompt_tokens":729,"completion_tokens":1446,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":473,"completion_tokens_details":{"reasoning_tokens":1382}},"tokens_in":473,"tokens_out":1446,"duration_ms":10469,"temperature":1.0,"reasoning_tokens":1382,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T07:24:55.054298+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A comparison with resolved HII regions in the local Universe where both the ionising stellar population and the gas-phase abundances are measured: if the observed line ratios deviate systematically from the model grid by more than the grid's sensitivity range, the equal-metallicity or single-geometry assumptions would be falsified.","supporting_citations":[],"review_version":1}