{"id":"0d30c532-9cd9-47d1-bdf8-83dd0f9dedca","arxiv_id":"2505.05390","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Combined AKARI and Spitzer spectra of four protostars are decomposed into ice components, yielding new ice column densities and abundances and a tentative CO-ice decline from Class 0 to I.","lead":"This paper combines archived AKARI and Spitzer infrared spectra of four young stars to measure the frozen ices on dust grains around them. It is a useful reference for astronomers who want to turn infrared spectra into ice abundances, including upcoming James Webb observations.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The CO ice abundance trend rests on fixed, untested laboratory mixture ratios and a silicate-subtracted continuum that excludes systematic errors; the paper's own text admits the four-source sample is insufficient to establish such a trend.","rationale":"The fitting procedure is standard for the field, and the comparison with earlier AKARI and Spitzer results in Table 4 is a useful cross-check. The H2O libration fit appears to cover the observed band (Fig. 5), and the CO2 bending-mode analysis avoids the saturated AKARI stretching mode. However, the central evolutionary claim is not protected against the two largest known systematics in ice-profile fitting: the choice of laboratory mixture templates and the continuum/silicate baseline under the libration band. The paper itself acknowledges both the limited sample (Section 5) and a significant unattributed residual (Section 4.2), so the gaps are internal, not invented. Because the quoted errors in Table 3 are statistical only, the current error bars understate the true uncertainty of X(CO). A sensitivity sweep over the fixed mixture ratios is a direct, low-cost test: if the Class 0-to-I ordering survives the sweep, the qualitative conclusion stands; if not, the Abstract and Summary should be softened to match the tentative language already used in Section 5. This does not require rejecting the paper; it requires conditional acceptance with the sensitivity analysis as a revision condition, which is what the reader's CONDITIONAL verdict already reflects.","tokens_in":37327,"tokens_out":8572,"duration_ms":95861,"concrete_test":"Re-run the fitting for all four sources using at least two alternative published mixture sets — e.g., the Table 2 H2O-rich mixture versus a polar-ice mixture with CH3OH/H2O = 0.3 and NH3/H2O = 0.1 as in Boogert et al. (2008/2015) — and allow the scaling of each CO-bearing mixture component in the 4.67 µm fit to be a free parameter rather than fixed by the Table 2 ratios. Recompute X(CO) using the libration-based N(H2O). If the Class 0-to-I ordering is not preserved, or if any source's X(CO) shifts by more than its quoted 1σ uncertainty, the headline trend should be downgraded to a tentative suggestion.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central evolutionary result — CO ice abundance decreasing from Class 0 to Class I (§5, Table 4) — is asserted more strongly than the evidence supports, and the authors' own text acknowledges this. Section 5 states: \"It is insufficient to reveal clear trends linked to the differences in ice abundances across only the four samples and the evolutionary stages of protostars,\" yet the Abstract and §6 repeat the decrease as a finding. The quantitative support depends on two unpropagated systematics. First, N(CO) is decomposed at 4.67 µm using fixed laboratory mixtures (H2O-rich, CO:CO2 = 100:70, CO:CH3OH = 1:1; Table 2), with the mixture ratios never varied; different published protostellar ice mixtures would redistribute N(CO) among components and alter the total, since Eq. (1) uses mixture-specific band strengths. Second, the denominator N(H2O) comes from the libration-mode fit (§4.1), which is derived after silicate-template subtraction whose 18–30 µm continuum anchors and scaling are degenerate with the broad libration band (§3.1); Table 3 quotes only statistical errors. The paper also reports a significant unattributed 6–8 µm residual (§4.2), so the fitted model is incomplete. With four sources, and Ced 110 IRS4 (0/I) overlapping the Class I values within errors, the ordering could easily change under a plausible re-fit.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents combined 2.5–30 μm AKARI/IRC and Spitzer/IRS spectra of four low-mass protostars (Per-emb 25, Ced 110 IRS4, B1 a, and RNO 91) and analyzes the ice absorption features by fitting experimental laboratory profiles. The authors determine continua, subtract silicate absorption using a GCS 3 template and a synthetic amorphous olivine/pyroxene profile, and then globally fit ice components including H2O (stretch, bend, and libration), CO2, CO, NH3, CH3OH, HCOOH, CH3CHO, CH4, and NH4+. They derive column densities and report ice abundances relative to H2O (calculated from the libration mode), compare with previous AKARI and Spitzer results, identify a tentative decrease in CO ice abundance from Class 0 to Class I, and suggest that pure CO2 in RNO 91 traces a past accretion burst. The paper also states the method can be applied to JWST ice spectra.","tokens_in":37585,"tokens_out":4388,"duration_ms":41923,"significance":"If the derived column densities are accurate, this work provides new ice inventories for four protostars, including the first simultaneous use of the H2O stretch and libration modes in this sample and an analysis of the CO2 bending mode rather than the unresolved AKARI stretching mode. The comparison with dense-cloud JWST observations is useful, and the identification of pure CO2 in RNO91 is a concrete, potentially interesting result. The authors are also honest about several known limitations, such as the RNO91 stretch-mode mismatch and the unattributed 6–8 μm residual. However, the central evolutionary claim (CO ice decreasing from Class 0 to I) is not supported by the four-source sample and the quoted statistical-only uncertainties, and the fixed laboratory mixture ratios used in the decomposition are never tested. The paper's strengths are the careful data combination and a transparent fitting approach, but the overreach in the abstract and summary requires revision.","major_comments":[{"comment":"Section 5 states that \"It is insufficient to reveal clear trends linked to the differences in ice abundances across only the four samples and the evolutionary stages of protostars,\" yet the Abstract and Section 6 repeat the CO ice abundance decrease as a finding. With only four sources, and with Ced 110 IRS4 (Class 0/I) at X(CO)=0.18±0.04 overlapping B1-a (0.10±0.01) and RNO 91 (0.14±0.06) within 2σ, the ordering is not statistically robust. Please either remove the trend claim from the Abstract/Summary or support it with a quantitative significance test (e.g., bootstrap or weighted regression) that includes systematic uncertainties.","section":"§5, Table 4, Abstract, §6"},{"comment":"The CO ice decomposition at 4.67 μm uses fixed laboratory mixture ratios (H2O-rich 100:20:20:14:10, CO:CO2=100:70, CO:CH3OH=1:1; Table 2) that are never varied, although Eq. (1) uses mixture-specific band strengths. Different published protostellar ice mixtures would redistribute N(CO) among components and change the total. The authors should test the sensitivity by refitting with alternative mixtures (e.g., those in Boogert et al. 2015 or Rocha et al. 2025) and report the range of resulting CO abundances, since this is load-bearing for the claimed evolutionary trend.","section":"§4.2, Table 2, Eq. (1)"},{"comment":"The uncertainties in Table 3 are explicitly based only on statistical errors in the spectra. The H2O libration-mode column densities, which are the denominators for all abundances in Table 4, depend on the silicate-template scaling and the 18–30 μm continuum anchoring described in Section 3.1, and these systematic errors are not propagated. The authors should quantify the continuum/silicate systematic uncertainty (e.g., by varying the silicate template, the fitting ranges, and the AKARI-to-Spitzer scaling) and add it in quadrature to the reported values, or at least provide a quantitative estimate of its magnitude.","section":"§3.1, Table 3"},{"comment":"The paper reports that \"a significant fraction of the 6−8 μm absorption remains unattributed,\" so the fitted model does not fully describe the spectra. This incomplete modeling can bias the scaling of other components, particularly CH3OH and NH4+, and its potential effect on the derived column densities should be discussed or bounded. Even if the 6–8 μm residual does not directly affect the CO measurement at 4.67 μm, it indicates that the global fit is not closed, which should be acknowledged as a systematic limitation in the abundance analysis.","section":"§4.2"}],"minor_comments":[{"comment":"The manuscript contains many corrupted character sequences (e.g., lines of \"/uni00000015/uni00000011/...\" after the references and in Figures 1–9 captions). These must be cleaned before publication, as they currently appear as broken glyph placeholders.","section":"Throughout text and figure captions"},{"comment":"\"six-order polynomial\" should be \"sixth-order polynomial,\" and \"third-to-fourth-order\" should be \"third- to fourth-order.\"","section":"§3.1"},{"comment":"\"We summarized their positions\" should be \"We summarize their positions\" for consistency with the present-tense style used elsewhere.","section":"§2.1"},{"comment":"The phrase \"The fitted laboratory profiles well cover the broad 13 μm absorption\" is awkward; consider \"The fitted laboratory profiles cover the broad 13 μm absorption well.\"","section":"§3.1, Figure 2 caption"},{"comment":"The table lists comparison values with footnote markers (e.g., \"0.19b−0.42c\") but the footnote letters are not explained in the caption; please specify which references each marker corresponds to.","section":"Table 4 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper's ice reduction and fitting follow standard methodology and the authors are transparent about several limitations, but the abstract and summary overstate the CO evolutionary trend beyond what the four-source sample and statistical-only errors can support. The manuscript would likely be publishable after the authors (1) tone down or remove the trend claim, (2) add a sensitivity test for the fixed mixture ratios, and (3) propagate continuum/silicate systematics. The corrupted glyph sequences suggest a production/processing error that must be fixed. No concerns about author conduct or citation behavior are apparent."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Honest take: this is a solid, field-standard ice analysis of four low-mass protostars, and the genuinely new part is the combined AKARI/Spitzer 2.5–30 μm spectra and the resulting ice inventories in Tables 3 and 4. The silicate-subtraction recipe with synthetic olivine/pyroxene profiles plus GCS 3 is a practical contribution. The simultaneous H2O stretch and libration fitting is a nice technical step, and the tentative pure CO2 double-peak in RNO91 is interesting. The paper earns a serious referee.\n\nThe soft spots are real but not fatal. The error bars in Tables 3 and 4 are statistical only; the dominant systematics — continuum placement, silicate subtraction, and the fixed laboratory mixture ratios — are not propagated. The CO abundance trend from Class 0 to I is presented in the abstract and summary as a finding, but Table 4 shows overlapping values (Per-emb 25 at 0.23, Ced 110 IRS4 at 0.18, B1a at 0.10, RNO91 at 0.14), and Section 5 explicitly says the four targets are insufficient to reveal clear trends. That internal contradiction should be fixed by softening the abstract and summary, not by adding more spin. The claim of deriving H2O stretch and libration simultaneously is also slightly overstated, since the 3 μm core is excluded for half the targets.\n\nI disagree with any suggestion that the paper is circular or that the method is flawed beyond repair. The decomposition is a standard fit of laboratory profiles to observed spectra; the fixed mixtures are a real limitation but not a hidden agenda. The unattributed 6–8 μm residual is honestly reported and does not by itself break the qualitative conclusions.\n\nWho is this for? Astrochemists working on ice inventories, especially anyone combining Spitzer and AKARI archival data or preparing JWST ice fitting. They will find the recipe and the comparison values useful. The evolutionary trend should be treated as tentative until it is re-derived with a larger sample and tested mixture variations.\n\nRecommendation: send it to peer review. A good referee will ask for systematic-error propagation, a sensitivity test on the mixture ratios, and a toned-down evolutionary claim. Those are addressable, and the underlying data deserve to be in the literature.","headline":"A competent, incremental ice-fitting study of four protostars with genuinely new combined spectra, undermined mainly by an overclaimed evolutionary trend that the authors' own text disowns.","tokens_in":38206,"tokens_out":1916,"would_cite":false,"duration_ms":23249,"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":"The paper derives consistent water-ice column densities from the 3.05 and 13.6 µm bands and finds that CO ice abundance drops as protostars evolve from Class 0 to Class I.","keywords":["astrochemistry","interstellar ices","low-mass protostars","infrared spectroscopy","AKARI IRC","Spitzer IRS","ice column density","CO ice sublimation"],"falsifier":"A JWST spectrum of one of these sources that resolves both water bands without saturation, combined with a laboratory band-strength measurement of the water-rich mixture at 10–20 K, would settle whether the simultaneous H$_2$O column densities and the derived abundances are correct.","tokens_in":37078,"feed_emoji":"❄️","tokens_out":9588,"duration_ms":90913,"temperature":0.7,"pith_summary":"This paper combines 2.5–5 µm AKARI spectra with 5–30 µm Spitzer spectra of four low-mass protostars at different evolutionary stages and fits the full infrared absorption with laboratory ice profiles to derive, for the first time, H$_2$O ice column densities from the 3.05 µm stretching mode and the 13.6 µm libration mode simultaneously. The authors use the same global fit to separate blended absorptions of CO$_2$, CO, NH$_3$, CH$_3$OH, and candidate organic ices, producing complete ice inventories for Per-emb 25, Ced 110 IRS4, B1 a, and RNO 91. Their central astrophysical result is that CO ice abundance declines from Class 0 to Class I, which they attribute to sublimation as the protostar heats its envelope. The work matters because ice mantles are the raw material for planet formation and because the same decomposition method is directly applicable to JWST spectra, where higher resolution can resolve what remains unattributed here.","feed_headline":"CO ice abundance drops from Class 0 to Class I protostars","feed_subtitle":"Combined AKARI and Spitzer spectra also give water-ice columns from two bands at once.","key_machinery":"The load-bearing device is the global spectral decomposition: laboratory ice absorbance profiles, taken from published measurements at fixed temperatures and mixture ratios, are convolved to the telescope resolution and summed to reproduce the observed optical-depth spectrum; each component's column density then follows from the band-strength relation $N = (1/A)\\int \\tau\\,d\\nu$, where $A$ is the laboratory band strength. The key named ingredients are the water-rich mixture H$_2$O:CO:NH$_3$:CO$_2$:CH$_3$OH = 100:20:20:14:10 at 10 K, the CO:CO$_2$ = 100:70 and CH$_3$OH:CO$_2$ = 1:2 mixtures used for the CO$_2$ bending feature, and a silicate-subtraction step that couples the GCS 3 silicate template with synthetic amorphous olivine and pyroxene profiles. This machinery lets the authors attribute each absorption band to an intrinsic ice carrier and converts the fitted profiles into the column densities of Tables 3 and 4. The H$_2$O libration mode (10–20 µm) serves as the anchor because it is less blended than the stretch and bend modes, and the agreement between stretch- and libration-derived columns is what validates the whole decomposition.","core_discovery":"On its own terms, the paper establishes a reliable decomposition of the 2.5–30 µm ice absorption toward four low-mass protostars. After removing the continuum and the 9.7 and 18 µm silicate bands, using both a template and synthetic amorphous olivine and pyroxene profiles, the authors fit all remaining absorption with laboratory ice absorbance data convolved to each instrument's resolution. The fit includes H$_2$O-rich mixtures (H$_2$O:CO:NH$_3$:CO$_2$:CH$_3$OH = 100:20:20:14:10 at 10 K), CO- and CH$_3$OH-mixed CO$_2$ ices, pure CO$_2$, pure and mixed CO with warm gas CO, and pure NH$_3$, CH$_3$OH, HCOOH, CH$_3$CHO, CH$_3$CH$_2$OH, and NH$_4^+$ profiles. The simultaneous derivation of the H$_2$O column from the 3.05 µm stretch and 13.6 µm libration modes yields values that agree within uncertainties, justifying the use of the libration mode as the abundance reference. The same fit gives CO$_2$ to H$_2$O ratios near 0.3 in all sources and a CO to H$_2$O ratio that decreases from 0.23 in the Class 0 source to 0.10–0.14 in the Class I sources. Toward RNO 91, a resolved double-peaked 15.2 µm CO$_2$ feature is interpreted as pure CO$_2$ ice left behind after thermal evaporation of CO, making it a possible infrared tracer of recent accretion bursts.","pith_inferences":["If the same stretch-versus-libration water check were run on a large JWST sample, systematic differences between the two bands would quantify how much NH$_3$ and CH$_3$OH contamination actually hides inside the 3 µm water band.","Because the decomposition fixes the mixture ratios without testing alternatives, the abundances in Tables 3 and 4 should be read as conditional on those laboratory recipes; new band-strength measurements for realistic multi-component ices could re-scale all of them.","The pure CO$_2$ signature in RNO 91 predicts that other protostars with known recent accretion bursts should also show the double-peaked 15.2 µm feature, while quiescent low-mass protostars should not.","The unattributed 6–8 µm residual suggests the ice inventory is not closed; high-resolution MIRI spectra of the same targets could identify additional carriers such as CH$_4$ or NH$_4^+$ in that region."],"forward_implications":["The combined 2.5–30 µm analysis yields new ice inventories for Per-emb 25, Ced 110 IRS4, B1 a, and RNO 91, with column densities and abundances listed in Tables 3 and 4.","The agreement between the H$_2$O stretch- and libration-mode columns supports using the 13.6 µm libration band as the abundance reference, avoiding contamination by 3 µm absorptions from NH$_3$ and CH$_3$OH.","The decrease in CO ice abundance from the Class 0 source to the Class I sources supports the scenario in which protostellar heating sublimates CO ice into the gas phase as the system evolves.","The resolved double-peaked 15.2 µm CO$_2$ feature toward RNO 91 indicates pure CO$_2$ ice left after thermal processing, a potential infrared tracer of episodic accretion and recent accretion bursts.","The silicate-subtraction and global-fitting method applies directly to JWST ice spectra, where higher spectral resolution can address the absorption that remains unattributed in the 6–8 µm range."],"supporting_citations":[{"why":"Supplies the experimental ice absorbance spectra and band strengths used as the fitting templates for the major ice species.","marker":"Gerakines et al. 1995"},{"why":"Provides the laboratory profiles for the water-rich and mixed-ice compositions that carry the global fit.","marker":"Ehrenfreund et al. 1996"},{"why":"Gives the pure-ice absorbance profiles and band strengths for the H$_2$O libration-mode anchor and other pure components.","marker":"Fraser & van Dishoeck 2004"},{"why":"Supplies the GCS 3 silicate template used as the baseline for subtracting the 9.7 and 18 µm dust absorption.","marker":"Kemper et al. 2004"},{"why":"Provides the laboratory amorphous olivine and pyroxene optical constants used to build the synthetic silicate subtraction.","marker":"Dorschner et al. 1995"},{"why":"Establishes the CO$_2$ bending-mode analysis at 15 µm that the paper relies on for robust CO$_2$ column densities.","marker":"Pontoppidan et al. 2008"},{"why":"Sets the comparison ice abundances for embedded protostars that the paper's results are validated against.","marker":"Boogert et al. 2008"},{"why":"Provides the interpretation of pure CO$_2$ ice as a thermal-processing tracer tied to episodic accretion.","marker":"Kim et al. 2012"},{"why":"Supplies the JWST Chamaeleon I ice abundances used to compare prestellar and protostellar CO ice evolution.","marker":"McClure et al. 2023"},{"why":"Establishes the AKARI IRC data reduction and previous AKARI ice abundances that this paper extends with Spitzer IRS spectra.","marker":"Kim et al. 2022"}],"fun_headline_variants":["CO ice abundance drops from Class 0 to I","Water-ice columns from two bands simultaneously","New ice decomposition method ready for JWST","Protostellar ice inventory: CO depletes, organics hinted","AKARI + Spitzer spectra trace ice evolution"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole analysis rests on the assumption that the real ice mantles are exactly described by the laboratory mixture profiles and band strengths chosen here, with fixed ratios such as H$_2$O:CO:NH$_3$:CO$_2$:CH$_3$OH = 100:20:20:14:10 at 10 K that the paper never varies; if the true ice composition, temperature, or morphology differs, every derived column density and abundance in Tables 3 and 4 shifts, and the paper's own 6–8 µm residual shows the decomposition is not closed.","fun_headline_variants_meta":{"raw":{"variants":["CO ice abundance drops from Class 0 to I","Water-ice columns from two bands simultaneously","New ice decomposition method ready for JWST","Protostellar ice inventory: CO depletes, organics hinted","AKARI + Spitzer spectra trace ice evolution"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000254,"raw_usage":{"total_tokens":1668,"prompt_tokens":1145,"completion_tokens":523,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":761,"completion_tokens_details":{"reasoning_tokens":448}},"tokens_in":761,"tokens_out":523,"duration_ms":5687,"temperature":1.0,"reasoning_tokens":448,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T23:05:29.885268+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A JWST spectrum of one of these sources that resolves both water bands without saturation, combined with a laboratory band-strength measurement of the water-rich mixture at 10–20 K, would settle whether the simultaneous H$_2$O column densities and the derived abundances are correct.","supporting_citations":[],"review_version":1}