{"id":"02aeb427-9775-488e-9986-a11bb93ebd1e","arxiv_id":"2607.03668","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":2,"one_line_summary":"Presence of ~mm-sized H2O ice grains reduces abundance differences between linear and Hapke RT mixture models to ~2 percent, while RT remains preferred overall for Europa.","lead":"Lab tests of water-ice mixtures with mixed grain sizes show linear and Hapke radiative-transfer spectral models both recover abundances within about 10 percent, and agree with each other to within 2 percent when millimeter grains are present. The result guides which unmixing method to trust for Europa’s surface maps from upcoming spectrometers.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"The fixed 2/3 shape-factor correction applied to the irregular ~1060 µm grains is load-bearing for both the reported RT accuracy and the claimed LM–RT agreement when coarse grains are present.","rationale":"The reader correctly isolated the single scalar that most directly controls the tabulated abundance residuals on which the strongest claim rests. All other simplifications (isotropic phase function, binary H2O-only mixtures, intermediate phase angles) are either already validated by Mustard & Pieters or explicitly caveated; only the shape factor is both untested for these particular particles and capable of moving the RT numbers relative to both laboratory truth and the LM baseline. Because the reader already assigned CONDITIONAL status precisely for this reason, no verdict adjustment is required. The proposed re-runs constitute a minimal, fully reproducible check that would either confirm robustness or quantify the sensitivity.","tokens_in":18280,"tokens_out":584,"duration_ms":25944,"concrete_test":"Re-run the identical MCMC RT inversions of the six laboratory spectra (three mixing ratios × two temperatures) after replacing the shape factor 2/3 by the bounding values 1.0 (no correction) and 0.5; recompute the mean |true – RT| residuals and the |RT – LM| differences. If either residual set exceeds the paper’s stated ±10 % / ±2 % thresholds for any mixture, the grain-size-dependent agreement claim is sensitive to the ad-hoc factor.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2.2 converts mass fractions of the irregular ~1060 µm grains into geometric cross-sections for the Hapke single-scattering albedo mixture (Eq. 7) by multiplying by an ad-hoc shape factor of 2/3 (citing the 0.2–0.9 range of Shkuratov & Grynko 2005 and the same choice in Berdis et al. 2025). The MCMC then retrieves those adjusted fractional abundances, which are reported as %wt and compared with laboratory mass ratios (Tables 1–2) and with the unadjusted LM results. Because LM operates directly on reflectance and never applies the factor, both the |true – RT| residuals (claimed ≤ ±10 %, tighter when fines dominate) and the average |RT – LM| difference (claimed ≤ ±2 %) are direct functions of this single scalar. If the true effective diameter of the laboratory particles lies substantially outside ~2/3 of the measured 1060 µm, the RT posteriors shift systematically while LM remains fixed, so the central numerical claim that “the presence of coarse H2O ice grains minimizes abundance differences between LM and RT modeling” no longer holds under the same laboratory truth values.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"This paper tests linear mixture (LM) and Hapke-based radiative-transfer (RT) intimate-mixture modeling on laboratory NIR reflectance spectra of binary H2O-ice mixtures that combine ~70 µm spherical grains with ~1 mm irregular grains at three mass ratios and two Europa-relevant temperatures (100 K, 120 K). Reflectance is converted to single-scattering albedo under a simplified isotropic Hapke model (Eqs. 3–6); the irregular grains are assigned a fixed shape factor of 2/3 when converting mass fractions to geometric cross-sections for the linear SSA mixture (Eq. 7). MCMC retrievals show that both methods recover laboratory abundances to within ±10 % (tightening to ±5 % when fines dominate) and that the average |RT–LM| difference stays within ±2 % whenever coarse grains are present. The author therefore concludes that RT remains the preferred approach for Europa regardless of grain size, while LM is still reliable for terrains that contain ~mm-sized ice.","tokens_in":18547,"tokens_out":1109,"duration_ms":8616,"significance":"If the numerical claims hold, the work supplies a practical, laboratory-anchored guideline for choosing between LM and RT when analyzing upcoming MAJIS and MISE spectra of Europa. The experimental design is strong: true mass fractions are known a priori, both models are applied to the same spectra, and posterior uncertainties are reported. The demonstration that the presence of even modest fractions of coarse ice largely erases the LM–RT discrepancy previously found for ~100 µm grains (Emran 2026) is a useful, falsifiable result for the community. The explicit caution that RMSE does not track abundance accuracy is also a valuable methodological reminder.","major_comments":[{"comment":"Section 2.2 and Eq. 7: the relative fractional cross-section of the irregular ~1060 µm grains is scaled by a fixed shape factor of 2/3 before the MCMC retrieval. Because LM never applies this factor, both the reported |true–RT| residuals (±10 %/±5 %) and the average |RT–LM| difference (±2 %) are direct functions of this single scalar. A short sensitivity test (e.g., 0.5 and 0.8) is needed to show that the central claim—“presence of coarse H2O ice grains minimizes abundance differences between LM and RT”—survives plausible variations of the effective diameter; without it the numerical agreement remains under-constrained.","section":null},{"comment":"Abstract and §4: the strong claim that RT is preferred “regardless of grain size or compositional mixture” rests on pure-H2O binary mixtures plus the earlier H2O–SAO results of Emran (2026). The discussion itself notes that Europa hosts multi-component mixtures (CO2, H2O2, salts, NH3-bearing species). The preference statement should be qualified to the grain-size and binary-composition regimes actually tested, or the multi-component caveat should be elevated from a future-work remark to a limitation of the present conclusion.","section":null}],"minor_comments":[{"comment":"Tables 1–2 report abundances as “%wt” even for the RT column; after the 2/3 shape-factor correction the retrieved quantities are geometric cross-sections, not mass fractions. Clarify the conversion (or re-label the RT columns) so that the comparison with laboratory mass ratios is unambiguous.","section":null},{"comment":"Figs. 2–3 list RMSE values that are systematically lower for LM than for RT, yet the text correctly notes that lower RMSE does not imply better abundance accuracy. Adding a short sentence in the figure captions that reiterates this decoupling would prevent casual readers from over-interpreting the fit metrics.","section":null},{"comment":"The isotropic-phase-function and B(g)=0 assumptions (Eq. 3) are justified by the laboratory geometry and known grain sizes, but a one-sentence reminder that remote-sensing geometries may require the full Hapke parameter set would strengthen the bridge to spacecraft applications.","section":null},{"comment":"Minor typographical inconsistencies appear (e.g., “Emrana” in the author line, “used used” in Data Availability, occasional missing spaces around µm). A careful proof-read will remove them.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The manuscript is a solid, incremental laboratory validation that usefully complements Emran (2026). The heavy self-citation is natural for a Part II paper but should not obscure the independent contribution of the mixed-grain experiments. I see no ethical or scope issues; minor revision addressing the shape-factor sensitivity and the scope of the preference claim should be sufficient for acceptance."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The new piece is the first quantitative LM-versus-Hapke-RT abundance recovery on laboratory H2O-ice mixtures that deliberately combine ~70 µm spherical and ~1 mm irregular grains at 100–120 K. That grain-size dependence is missing from the prior literature, including the author’s own same-size binary work. Everything else—end-member conversion, isotropic simplification, MCMC pipeline—is carried over cleanly.\n\nWhat it does well is straightforward: known mass fractions, public spectra (Stephan et al. 2021), identical application of both models, and tabulated residuals that stay inside the claimed ±10 % (tightening to ±5 % when fines dominate). Temperature dependence is minor, as expected. The practical takeaway for MISE/MAJIS teams is usable: RT remains the safer default across grain sizes; LM is still reliable once mm-scale ice is present even in modest amounts. Citation pattern is heavy on the author’s 2026 paper for the small-grain baseline, but that is legitimate scaffolding, not circularity—the mixed-grain numbers are externally anchored.\n\nThe soft spot the stress-test flags is real: the fixed 2/3 shape factor that converts irregular-grain mass fraction into geometric cross-section for Eq. 7. Because LM never sees that factor, both the |true–RT| residuals and the claimed |RT–LM| ≤ ±2 % difference are functions of it. The paper cites the 0.2–0.9 range and Berdis et al. 2025, but never varies the scalar. If the true effective diameter sits far outside ~2/3, the RT posteriors shift while LM stays put. That is a genuine limitation, not a fatal one; the laboratory particles are well-characterized and the factor is conventional, so the reported numbers remain informative under the stated assumptions. Isotropic scattering and binary-only mixtures are secondary and already flagged by the author.\n\nThis is for icy-moon spectroscopists who need a decision rule before the next data drop. It deserves a serious referee; the core comparison is clean enough that the shape-factor sensitivity can be tested or bounded in revision. I would cite the tables and the grain-size conclusion.","headline":"Solid lab validation of LM vs Hapke RT on mixed-grain H2O ice; the 2/3 shape factor is a real but bounded soft spot, not a collapse of the result.","tokens_in":19169,"tokens_out":548,"would_cite":true,"duration_ms":5452,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Coarse water-ice grains make linear and Hapke radiative-transfer mixture models agree within a few percent for Europa, yet radiative transfer remains preferred overall.","keywords":["Europa","spectral mixture modeling","Hapke theory","linear unmixing","radiative transfer","water ice grain size","near-infrared spectroscopy"],"falsifier":"Repeat the same laboratory binary mixtures with independently measured effective diameters (for example by laser diffraction or micro-CT) and re-run both models; a systematic offset larger than a few percent would falsify the claimed equivalence of the two methods for coarse-grain mixtures.","tokens_in":19149,"feed_emoji":"🌍","tokens_out":641,"duration_ms":5811,"temperature":0.7,"pith_summary":"The paper tests how grain size changes the accuracy of two standard ways of turning near-infrared spectra into material abundances on Europa: simple linear (areal) mixing and Hapke radiative-transfer intimate mixing. Using laboratory spectra of water ice mixed from ~70 µm spherical grains and ~1 mm irregular grains at Europa-relevant temperatures, both methods recover the true laboratory fractions to within roughly ±10 percent, tightening to ±5 percent when fine grains dominate. When any coarse grains are present the two methods differ from each other by only ~2 percent on average; when only fine grains are present the radiative-transfer model is clearly more accurate. The author therefore concludes that Hapke-based radiative transfer should be the default tool for Europa composition work, while linear mixing remains serviceable wherever millimetre-sized ice grains occur.","feed_headline":"Coarse ice grains make Europa mixture models nearly agree","feed_subtitle":"Hapke radiative transfer still preferred, but linear mixing works where millimetre ice is present","key_machinery":"The Hapke single-scattering albedo of an intimate mixture, formed as a mass- and size-weighted linear combination of the end-member albedos (with a fixed 2/3 shape factor applied to the irregular millimetre grains) and inverted under isotropic scattering to recover fractional abundances.","core_discovery":"Across laboratory water-ice mixtures that contain both ~70 µm and ~1 mm grains, linear-mixture and Hapke radiative-transfer abundance estimates stay within ±10 percent of the true laboratory values and within ~2 percent of each other; when only fine grains are present the radiative-transfer model recovers abundances more accurately, so Hapke radiative transfer is preferred for Europa regardless of grain size, yet linear mixing remains reliable wherever millimetre ice is present.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["Coarse grains shrink Europa ice model gaps to ~2%","Millimetre ice makes linear and Hapke estimates nearly match","Fine grains alone make Hapke recover ice abundances better","Hapke preferred for Europa ice at any grain size","Linear mixing stays reliable where mm-scale ice grains occur"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The irregular millimetre grains are assigned a fixed shape factor of two-thirds when converting mass fraction into geometric cross-section; if that factor is wrong the radiative-transfer abundances shift systematically.","fun_headline_variants_meta":{"raw":{"variants":["Coarse grains shrink Europa ice model gaps to ~2%","Millimetre ice makes linear and Hapke estimates nearly match","Fine grains alone make Hapke recover ice abundances better","Hapke preferred for Europa ice at any grain size","Linear mixing stays reliable where mm-scale ice grains occur"]},"model":"grok-4.5","effort":"low","cost_usd":0.00433,"raw_usage":{"total_tokens":1403,"prompt_tokens":926,"num_sources_used":0,"completion_tokens":65,"cost_in_usd_ticks":43300000,"prompt_tokens_details":{"text_tokens":926,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":412,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":926,"tokens_out":65,"duration_ms":3821,"temperature":1.0,"reasoning_tokens":412,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-12T00:48:18.196243+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Repeat the same laboratory binary mixtures with independently measured effective diameters (for example by laser diffraction or micro-CT) and re-run both models; a systematic offset larger than a few percent would falsify the claimed equivalence of the two methods for coarse-grain mixtures.","supporting_citations":[],"review_version":1}