{"id":"ec4842eb-1880-495d-a156-83dc520c5f37","arxiv_id":"2607.14251","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Twelve white dwarfs from DESI DR1 have accreted debris resembling inner-Solar-System rock, with two systems likely accreting water-rich planetesimals.","lead":"Astronomers measured the chemical composition of planetary debris being eaten by 12 white dwarf stars, finding rock similar to Earth, meteorites, and planetary cores, plus two candidates that may be water-rich planetesimals. This first detailed study of polluted white dwarfs from the DESI survey shows the survey can reliably measure the interiors of destroyed exoplanets.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Abstract overstates water-rich claim: 1352+0323 O lines poorly modelled but counted as water-rich planetesimal","rationale":"The reader's weakest assumption—diffusion timescales—is broad and acknowledged by the authors, but the most load-bearing concern for the paper's headline is the internal mismatch between the abstract's firm 'two water-rich planetesimals' claim and the body's explicit warning that one of those two (1352+0323) has poorly modelled O lines. This is a specific, author-acknowledged flaw that directly affects a key result highlighted in the abstract and conclusions. It is more concrete and more easily tested than a general systematic in diffusion theory. I still agree with the CONDITIONAL verdict: the abundance measurements and DESI validation are strong, and the water-rich claim for 0452−0214 appears robust, but the paper should be revised to temper the claim for 1352+0323 or present it as tentative. My proposed test would settle the issue with additional data; until then, the abstract is overclaimed.","tokens_in":37631,"tokens_out":4542,"duration_ms":51477,"concrete_test":"Obtain higher-S/N X-shooter or UVES spectra of 1352+0323 targeting the O I 7774 Å and 8446 Å lines, and rederive the O abundance using an independent atmosphere code (e.g., TLUSTY or PHOENIX). If the O abundance shifts by more than ~0.3 dex or becomes undetected, the water-rich classification for 1352+0323 is not supported and the abstract should be revised to state one secure water-rich candidate (0452−0214) and one tentative candidate.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract states that two of the six analyzed white dwarfs (0452−0214 and 1352+0323) accrete 'something akin to a water-rich planetesimal.' The body, however, explicitly cautions that 'the O lines in 1352+0323 are weak and not satisfactorily modelled (Fig. 12)' and that its O abundance 'requires a deeper spectrum for confirmation and should for now be treated with some care' (§4.6). The water-rich classification for 1352+0323 is therefore built on an O measurement the authors themselves flag as unreliable, yet the abstract presents it with the same confidence as the well-reproduced O lines in 0452−0214. If the O abundance for 1352+0323 is incorrect, the number of secure water-rich candidates drops from two to one, directly weakening the headline claim. This is a concrete, internally flagged limitation, not a speculative systematic. The diffusion-timescale dependence identified by the reader is also a valid concern, but it is global, explicitly acknowledged by the authors, and does not single out a specific object in the abstract.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a pilot abundance analysis of 12 highly metal-enriched white dwarfs selected from DESI DR1, with X-shooter follow-up for seven of them. The authors measure stellar parameters and photospheric metal abundances through an iterative photometric+spectroscopic modelling workflow, identify between three and ten metals per system, and convert the abundances into parent-body compositions using diffusion timescales and Bayesian accretion modelling. They report that most accreted bodies resemble primitive chondrites or bulk Earth, that 0850+3208 may be core-rich, that 1336−0337 resembles bulk silicate Earth, and that 0452−0214 and 1352+0323 show large oxygen mass fractions interpreted as water-rich planetesimals. A key quantitative result is the excellent agreement between DESI and X-shooter abundances, with a weighted mean difference of 0.007±0.007 dex.","tokens_in":37968,"tokens_out":3802,"duration_ms":47330,"significance":"If the abundance results are robust, this is a valuable demonstration that DESI medium-resolution spectroscopy can deliver reliable photospheric abundances for highly polluted white dwarfs, expanding the sample of systems with detailed compositional constraints. The cross-instrument comparison is a genuine strength, as is the explicit six-stage fitting workflow and the use of publicly available data with a Zenodo release. The compositional classification of the sample, including two water-rich candidates, is scientifically interesting but depends on external diffusion models and on the reliability of the oxygen detections. The paper also transparently flags its main limitations, including the diffusion-timescale uncertainty and the poor oxygen-line modelling for 1352+0323. These caveats are not fatal, but they need to be reflected in the abstract and conclusions.","major_comments":[{"comment":"The abstract and Conclusions count 0452−0214 and 1352+0323 as two water-rich planetesimals. However, §4.5 states that the O lines in 1352+0323 are 'weak and not satisfactorily modelled' and that its O abundance 'requires a deeper spectrum for confirmation and should for now be treated with some care'; §4.6.1 similarly cautions that the water-rich result for 1352+0323 'only holds for a reliably-measured O abundance, which may not be the case for this star.' The Bayesian water fraction for 1352+0323 (Fig. 14, p=0.999) is therefore built on an O measurement the authors themselves flag as unreliable. The abstract should either list only 0452−0214 as a secure water-rich candidate, or explicitly separate 1352+0323 as tentative. As written, the headline claim overstates the evidence.","section":"Abstract; §4.5; §4.6.1"},{"comment":"All parent-body inferences—accretion state, steady-state accretion rates, water fractions, oxygen budgets, and accreted masses in Tables 5–6 and Figs 9–14—depend on element-specific diffusion timescales from Koester (2009) Eq. 5. The authors acknowledge in §4.7 that diffusion models are not yet at consensus and that 'future refinements of the diffusion time scales could impact our conclusions.' This is a load-bearing assumption for the compositional classifications, not merely a peripheral caveat. I request a quantitative sensitivity test, e.g. recomputing the oxygen budgets and water-fraction posteriors with diffusion timescales varied coherently by ±0.2–0.5 dex, or with an alternative diffusion prescription. If that is not feasible, the abstract and conclusions should be tempered to state that the dry/wet and core-rich classifications are conditional on current diffusion models.","section":"§4.7; Table 5; Eq. (1)"},{"comment":"There is an unresolved internal tension for 1333+3254: the oxygen-budget analysis (§4.6.1) indicates an O excess corresponding to roughly 25 per cent water, while the Bayesian analysis (§4.6.1) disfavours a wet model at about the 1-σ level, and §4.6.2 shows that the all-H-as-H2O assumption yields an unphysical total mass smaller than the measured CVZ mass. The authors propose speculative resolutions (Fe2O3, CO2, or a decreasing accretion phase). This does not affect the headline water-rich pair, but it is a cautionary example that the oxygen-budget method and the Bayesian method can diverge for an object with weak/ambiguous O and H constraints. The paper should explicitly state that the dry/wet classification for 1333+3254 is uncertain rather than presenting the O budget as a water detection without immediate qualification.","section":"§4.6.1; §4.6.2"}],"minor_comments":[{"comment":"The abstract says the two stars accrete 'something akin to a water-rich planetesimal,' while the Conclusions say the large O mass fractions 'may tentatively indicate' water-bearing parent bodies. These two formulations should be harmonised, especially given the 1352+0323 caveat.","section":"Abstract vs Conclusions"},{"comment":"For stars with untraceable Si, Si is included in the models at CI-chondrite ratios relative to Mg but is not listed in Table 4. The text explains this, but the caption of Table 4 should state explicitly that the Si column does not include these fixed model inputs, to avoid readers interpreting the table as an upper limit or a detection.","section":"§3.1; Table 4"},{"comment":"The statement 'No new data were generated for this study' is confusing, since the paper presents new DESI coadds and new X-shooter observations. Rephrase to indicate that no new data products beyond the published spectra and the Zenodo figure data are released, or clarify that the raw data are public from the archives.","section":"Data Availability"},{"comment":"Minor typographical and formatting issues: the journal header reads 'MNRAS000, 1–34 (20262026)'; several figure captions use 'overplot' instead of 'overplotted'; and Table 5 would benefit from a clearer separator between the four star blocks to avoid misreading the columns.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The core measurement layer is strong and the cross-instrument agreement is convincing. The main issue is that the abstract overstates the water-rich claim for 1352+0323 against the authors' own text, and the diffusion-timescale dependence of all compositional classifications is acknowledged but not quantified. These are fixable with rewording and a sensitivity test. The paper is within scope for MNRAS and likely publishable after major revision. I would not recommend reject, since the abundance measurements and the DESI demonstration are valuable regardless of the water-rich interpretation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this is a useful, well-executed pilot study, and the abundance measurements are the strength. The authors selected 12 heavily polluted white dwarfs from DESI DR1, nine newly identified, got X-shooter follow-up for seven, and the cross-instrument agreement is genuinely good — weighted mean difference 0.007±0.007 dex between DESI and X-shooter abundances. That is a strong validation that DESI medium-resolution spectra can support this kind of science. The iterative fitting workflow (photometry + spectroscopy + custom grids) is transparent, the data are public, and the figures show the fits, including weak cases.\n\nThe new results: eight systems with six or more metals, which adds meaningfully to the small pool of well-characterized polluted white dwarfs. The parent-body classifications are plausible, and the two water-rich candidates (0452−0214 and 1352+0323) plus the core-like object (0850+3208) are worth following up. The oxygen-budget analysis is a nice, conservative sanity check, and the Bayesian accretion-state modelling is the right tool even if it leans on prior assumptions.\n\nNow the soft spots, in proportion. The biggest issue is the abstract's wording. The body explicitly says the O lines in 1352+0323 are \"weak and not satisfactorily modelled\" and that the O abundance \"requires a deeper spectrum for confirmation and should for now be treated with some care.\" But the abstract lumps 1352+0323 in with 0452−0214 as accreting \"something akin to a water-rich planetesimal,\" with no caveat. That is overstating the finding. The stress-test is right: if the O measurement for 1352+0323 is off, you drop from two water-rich candidates to one. The authors do soften it in the conclusions (\"may tentatively indicate\"), so this is an abstract/body mismatch, not a fundamentally dishonest claim.\n\nThe other soft spots are the usual: uncertainties are statistical-only, with ~0.2 dex systematics discussed but not propagated; and the whole parent-body interpretation sits on Koester diffusion timescales that the authors themselves note are still under development. That is not a fatal flaw — it is the standard uncertainty in the field, and the paper flags it (§4.7). But it means the most exciting conclusions (water fractions, core fractions) are more fragile than the measured abundances.\n\nBottom line: the measurement layer is solid, the interpretation is reasonable with caveats, and the abstract overreaches on one object. For anyone working on polluted white dwarfs or exoplanet compositions, this is a useful paper and it validates a new survey channel. I would send it to peer review. A good referee should push for the abstract to match the body — specifically, to flag 1352+0323 as tentative — and for a brief discussion of how the 0.2 dex systematics would shift the water fractions.","headline":"Solid DESI-era measurement paper: the abundance work is careful and cross-validated, but the abstract sells the weaker water-rich candidate harder than the data support.","tokens_in":38646,"tokens_out":2197,"would_cite":true,"duration_ms":24181,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"DESI spectra of 12 polluted white dwarfs measure the composition of their accreted planetesimals, including two that may be water-rich.","keywords":["white dwarfs","metal pollution","planetesimal composition","exoplanet interiors","accretion","DESI","water-rich planetesimals","oxygen budget"],"falsifier":"A high-resolution ultraviolet spectrum of 1352+0323, whose oxygen lines are weak and imperfectly fitted, would confirm or refute its oxygen abundance and settle whether the water-rich classification is real. More generally, if refined diffusion models change the relative sinking times of oxygen, magnesium, and iron by more than about a factor of two, the oxygen-budget excess in 0452−0214 could disappear. A search for a transit or a circumstellar disc around these two stars would also test whether accretion is actually ongoing.","tokens_in":37540,"feed_emoji":"🪐","tokens_out":6125,"duration_ms":59538,"temperature":0.7,"pith_summary":"The paper analyzes 12 white dwarfs whose atmospheres are polluted with metal debris from disrupted planetary bodies, using survey spectra plus follow-up ultraviolet spectroscopy for seven of them. It aims to show that medium-resolution surveys can yield reliable, precise elemental abundances for the accreted material. From those abundances, the authors reconstruct what the parent bodies were made of. They find that most match primitive, unprocessed rock similar to chondrite meteorites, one looks like a planetary core fragment, another resembles silicate mantle, and two show large oxygen fractions that tentatively indicate water-bearing parent bodies. If right, the work extends a method for reading exoplanet interiors from a few dozen well-studied stars to a survey scale of thousands.","feed_headline":"12 dead stars expose the make-up of their planets","feed_subtitle":"Two of the white dwarfs show hints of water-rich planetesimals; most look like primitive meteorites.","key_machinery":"The load-bearing mechanism is the conversion from measured photospheric abundances to parent-body compositions. Each element sinks out of the white dwarf's convection zone on a different diffusion timescale; assuming steady-state accretion, the photospheric abundance of element Z is proportional to its accretion rate times its sinking time. The paper combines these sinking-time corrections with an oxygen-budget accounting (oxygen apportioned among MgO, SiO2, CaO, FeO and similar oxides, with any leftover oxygen attributed to H2O) and a Bayesian model that marginalizes over accretion phase (increasing, steady, or decreasing) and water fraction. The central identity is the oxygen budget ratio","core_discovery":"On the paper's own terms, the central discovery is twofold. First, the survey spectra, when combined with iterative atmosphere modelling, give photospheric metal abundances that agree with independent higher-resolution follow-up to within about 1–2 per cent, with no systematic offset. Second, the 12 accreted bodies are compositionally diverse: mostly chondritic, with one core-like (iron-rich, oxygen-poor) body, one silicate-mantle-like body, and two (0452−0214 and 1352+0323) with oxygen budgets suggesting water fractions around 60–80 per cent, i.e. something akin to water-rich planetesimals.","pith_inferences":["If the water-rich interpretation for 0452−0214 and 1352+0323 holds, it would lend support to the idea that volatile-rich delivery can raise debris-disc mass flow and thereby explain the high accretion rates measured for helium-atmosphere white dwarfs; the paper raises this possibility but does not endorse it.","The tension in 1333+3254 — oxygen budget suggesting roughly 25 per cent water while the Bayesian analysis prefers dry and the hydrogen budget is too low — points toward alternative oxygen carriers such as CO2 or Fe2O3; this could be tested with ultraviolet spectroscopy searching for photospheric carbon.","A testable extension: the two water-rich candidates should show a circumstellar debris disc with a detectable infrared excess if they are currently accreting volatile-rich material, or atmospheric carbon and nitrogen if the water arrived as CO2 or NH3 ices; the paper finds no infrared excess but notes faint discs can escape detection.","The cross-instrument agreement of about 1–2 per cent suggests that a systematic abundance error of roughly 0.2 dex, as the paper adopts, is a practical floor for future large-sample surveys; that is enough to separate dry from water-rich bodies at the claimed level."],"forward_implications":["If the survey abundances are reliable at the claimed level, thousands of metal-polluted white dwarfs can be screened for detailed composition study without pre-selection bias.","The two water-rich candidates, if confirmed, imply that water-bearing planetesimals can survive to the white-dwarf stage and be accreted, with consequences for where water resides in evolved planetary systems.","Core-rich and mantle-rich accreted bodies (0850+3208 and 1336−0337) indicate that differentiated bodies are common enough to appear in a sample of twelve.","The diversity seen here supports the view that white-dwarf pollution traces the full range of planetesimal interiors, not just primitive chondrites.","Detecting three to ten metals per system, including the rock-forming species, sets a benchmark for what medium-resolution spectroscopy can extract from polluted white dwarfs."],"fun_headline_variants":["12 white dwarfs reveal planet interiors","Dead stars show meteorite-like and water-rich worlds","DESI finds 12 metal-rich white dwarfs with planet clues","White dwarf debris: primitive meteorites and water-rich bodies"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The parent-body compositions, water fractions, and accretion rates all rest on element-specific diffusion (sinking) timescales for each white dwarf; if those timescales are materially wrong, the inferred accretion phase, mass budget, and dry-versus-water-rich classifications all shift.","fun_headline_variants_meta":{"raw":{"variants":["12 white dwarfs reveal planet interiors","Dead stars show meteorite-like and water-rich worlds","DESI finds 12 metal-rich white dwarfs with planet clues","White dwarf debris: primitive meteorites and water-rich bodies"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000279,"raw_usage":{"total_tokens":1496,"prompt_tokens":746,"completion_tokens":750,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":490,"completion_tokens_details":{"reasoning_tokens":686}},"tokens_in":490,"tokens_out":750,"duration_ms":8457,"temperature":1.0,"reasoning_tokens":686,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T02:39:37.716388+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A high-resolution ultraviolet spectrum of 1352+0323, whose oxygen lines are weak and imperfectly fitted, would confirm or refute its oxygen abundance and settle whether the water-rich classification is real. More generally, if refined diffusion models change the relative sinking times of oxygen, magnesium, and iron by more than about a factor of two, the oxygen-budget excess in 0452−0214 could disappear. A search for a transit or a circumstellar disc around these two stars would also test whether accretion is actually ongoing.","supporting_citations":[],"review_version":1}