{"id":"f9dd8cd3-3bb6-428c-be1d-4b41dffd7d65","arxiv_id":"2507.13451","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Long-term TRAPPIST, UVES, and CRIRES+ observations show C/2017 K2's volatile-driven activity transitioning from CO/CO2 to water near 3 au, with typical-to-enriched composition and HCN as the main CN parent.","lead":"An eight-year photometric and spectroscopic campaign tracked comet C/2017 K2 from 15 au before perihelion to 8.5 au after perihelion. The data reveal activity shifting from CO and CO2 driven outgassing to water driven sublimation near 3 au.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The G/R drop is attributed to a CO2-to-H2O switch, but the paper does not model collisional quenching of [OI] in the denser inner coma, and the ratio is inconsistent with the same-epoch CRIRES+ CO/H2O measurements.","rationale":"The reader's weakest assumption correctly identifies the G/R calibration as the load-bearing step, but my concern sharpens it into a specific, testable failure mode: density-dependent quenching and an internal inconsistency with the directly measured volatile production rates. The paper's own CRIRES+ data at 2.1 au give CO/H2O about 13 percent, which would raise G/R above the pure-water value if CO contributed to the oxygen lines; the observed G/R of 0.08 is below the nominal water value of 0.1, suggesting that either the calibration values are not applicable at this density, or a green-line suppression mechanism (quenching) is active. The same mismatch appears at 2.73 au in the opposite direction. These tensions mean that the G/R trend cannot be cleanly attributed to composition without a collisional-radiative model, and the conclusion that K2 was CO2-dominated at 3.23 au is not established. The broader claim that water sublimation takes over inside 3 au remains supported by the OH production increase and the light curve, so the verdict stays CONDITIONAL. My concern does not change the reader's verdict, but it adds a specific required revision: an acknowledgment of quenching and a quantitative reconciliation of G/R with the measured H2O, CO, and CO2 constraints.","tokens_in":31028,"tokens_out":13496,"duration_ms":149461,"concrete_test":"Re-extract the UVES long-slit data for the three 2022 epochs and measure the 5577/6300 ratio as a function of nucleocentric distance along the slit. If G/R decreases toward the nucleus while line fluxes rise, collisional quenching is biasing the aperture-integrated ratio; if G/R is flat, quenching is negligible. Separately, compute the predicted G/R from the CRIRES+ H2O and CO production rates using a photochemical model with modern branching ratios, and compare with the observed G/R at 2.73 and 2.1 au. If the predicted G/R matches without CO2, then the CO2 domination claim is unsupported; if an extra oxygen source is still required, the compositional interpretation survives this test.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3.2.1 interprets the G/R drop (0.25 at 3.23 au, 0.14-0.15 at 2.73 au, 0.08 at 2.1 au) as a clear switch from CO2 to H2O dominance, using the Festou & Feldman (1981) calibration. The paper does not test two assumptions of this diagnostic. First, collisional quenching: gas production rises from about 1e28 to about 1e29 mol/s between the May and September 2022 epochs, so the inner-coma density approaches the critical density for O(1S) quenching (about 1e10 cm-3) within the 0.45 arcsec by 10 arcsec slit. O(1S) quenching suppresses the green line and lowers G/R without any compositional change. Second, the same-epoch CRIRES+ data provide a quantitative contradiction: at 2.1 au (September 2022), CO/H2O is about 13 percent (Table 11), which would predict G/R around 0.11 if CO contributes with G/R about 0.2, yet the observed G/R is 0.08; at 2.73 au, the observed G/R of 0.14-0.15 is higher than the roughly 0.11 predicted from H2O and CO alone. These discrepancies are not resolved in the paper. The CO2 domination attribution at 3.23 au is particularly fragile because CO is a plausible alternative that was not detected at that epoch, only due to sensitivity, and CO2 is not directly measured. The central claim therefore overstates the certainty of the compositional switch.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents an eight-year photometric and spectroscopic study of the dynamically new Oort cloud comet C/2017 K2 (PanSTARRS) from 15.18 au pre-perihelion to 8.46 au post-perihelion, using TRAPPIST broadband and narrowband imaging, VLT/UVES optical echelle spectroscopy, and VLT/CRIRES+ near-infrared spectroscopy. The authors derive light curves, colors, Haser-model production rates of OH, NH, CN, C3, and C2, the dust proxy A(0)fρ, and production rates and mixing ratios of infrared parent species. The central result is a heliocentric-distance trend in the [OI] green-to-red (G/R) ratio, from 0.25 at 3.23 au to 0.08 at 2.1 au, which they interpret as a transition from CO/CO2-driven activity to water-driven sublimation inside ~3 au, with a specifically CO2-dominated phase at the largest distances. Secondary results include constant dust colors, a 'typical' C2/CN taxonomy, a high dust-to-gas ratio, iron and nickel abundances, and parent-daughter associations (HCN to CN, C2H2 to C2).","tokens_in":31372,"tokens_out":8956,"duration_ms":95857,"significance":"If the central interpretation holds, the paper offers a valuable long-baseline view of a dynamically new comet crossing the water sublimation line, with simultaneous optical and infrared coverage on several epochs. The dataset is substantial: 271 nights of TRAPPIST imaging, three epochs of high-resolution UVES spectra, and three epochs of CRIRES+ spectra with contemporaneous optical data. The G/R measurements are new for this object and are consistent with independent MUSE and other reported values for K2, which strengthens the observational basis. The paper also makes useful quantitative comparisons with the A'Hearn et al. (1995) taxonomy and the Lippi et al. (2021) infrared statistics. However, the central compositional-switch claim is not fully supported by the same-epoch infrared measurements, and the specificity of the 'CO2 domination' wording exceeds what the data demonstrate. The archival value of the light curve and production-rate tables is clear, but the interpretive conclusion needs to be either better quantified or appropriately qualified.","major_comments":[{"comment":"The interpretation of the G/R drop as a clean compositional switch is not quantitatively reconciled with the simultaneous CRIRES+ CO/H2O mixing ratios. Using the Festou & Feldman (1981) calibration cited by the authors, a linear decomposition of the observed G/R at rh=2.12 au from H2O (G/R≈0.1) and CO at 13.9% (if CO has G/R≈0.2 as implied by the text) predicts G/R≈0.11, whereas the observed value is 0.08±0.004. At rh=2.73 au, the same exercise with CO at 10.7% predicts G/R≈0.11, whereas the observed values are 0.14–0.15. The paper does not perform this check. The residual could indicate CO2 or an error in the adopted parent-specific G/R values, but without an explicit quantitative comparison the claimed 'clear' transition from CO2 to H2O is not established. Please add this decomposition with propagated uncertainties and discuss the residual explicitly.","section":"§3.2.1, Tables 8 and 11"},{"comment":"The attribution of the high G/R=0.25 at 3.23 au specifically to CO2 is fragile. CO is not detected at that epoch (only an upper limit, Table 11), CO2 is not directly measured, and the detection of CO2+ ions does not quantify the neutral CO2 production rate. The text itself acknowledges in §3.2.1 that 'we cannot then argue it was especially rich in CO or CO2 compared to others', yet the Summary/Conclusion (Section 4) states that 'The switch from a CO2 domination to an H2O dominated coma is also clear'. Please either derive quantitative constraints on the CO and CO2 contributions from the G/R value and the non-detections, or rephrase the conclusion to say 'CO/CO2-dominated' activity, consistent with the abstract and with the caveats stated earlier in the paper.","section":"§3.2.1 and §4"},{"comment":"The potential effect of collisional quenching of O(1S) on the green line is not assessed. The gas production rates in Table A.1 rise from roughly 10^28 to 10^29 mol/s between the May and September 2022 epochs, and within the innermost part of the 0.45''×10'' slit the coma density can approach the critical density for quenching (n_crit ~10^10 cm^-3). If quenching suppresses the green line preferentially at the later epochs, the G/R drop could be partially attributable to density effects rather than composition. A simple order-of-magnitude estimate, e.g., n(r)=Q/(4π v r^2) with v~0.5 km/s evaluated across the slit footprint, should be presented to show that quenching is negligible or to quantify a correction. This is a standard check for the forbidden-oxygen ratio diagnostic and is needed to support the compositional reading.","section":"§3.2.1"}],"minor_comments":[{"comment":"The sentence 'The nondetection of CO and the detection of CO2+ might point to CO2 to be the main contributor at larger heliocentric distances than 2.5 au' is appropriately hedged, but the phrase 'might point' is then abandoned in Section 4. Please keep the language consistent throughout.","section":"§3.2.1"},{"comment":"The statement that C2 is 'more likely to be dissociated from C2H2 than from C2H6' is based on upper limits for C2H2 (Table 11). The upper limits are consistent with C2H2 being the dominant parent, but they do not positively identify it. Please clarify that this is an inference from upper limits and the relative abundances, not a detection.","section":"§3.3 and Table 10"},{"comment":"Several entries in Table A.1 have quoted uncertainties larger than the central value, e.g., 2022-04-27 OH = 1.66±2.67e+28 mol/s and 2023-09-20 OH = 3.51±53.70e+26 mol/s. These values appear to be typographical errors or are not scientifically meaningful. Please correct or remove them.","section":"Table A.1"},{"comment":"The phrase 'Johnson-Cousin' should be 'Johnson-Cousins'.","section":"§2.1"},{"comment":"The four panels of Figure 3 use different x-axis ranges and do not all show the rh scale. Adding the heliocentric-distance range to each panel caption or axis would improve readability.","section":"Figure 3"},{"comment":"In Equation (2), the expression 'm⊙,R−HR' should have a clear separation (e.g., m⊙,R − HR) and HR should be explicitly defined as the absolute magnitude in the R band.","section":"Equation (2)"}],"recommendation":"major_revision","confidential_remarks":"The paper builds on a rich recent literature on C/2017 K2 (Jewitt et al. 2019; Kwon et al. 2023, 2024; Ejeta et al. 2025; Woodward et al. 2025), and its main new elements are the long-baseline TRAPPIST light curve and the simultaneous UVES/CRIRES+ epochs. The quantitative inconsistency between the G/R trend and the CRIRES+ CO/H2O ratios is, in my view, the key technical issue. I recommend major revision rather than rejection because the dataset is valuable and the interpretation can be repaired by an explicit decomposition and a more cautious wording. I also note that the conclusion section overstates the certainty of the CO2 attribution relative to the body of the paper; this is likely an oversight in wording rather than a fundamental flaw."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a solid, data-heavy observational paper on C/2017 K2, and it deserves a serious referee. What's genuinely new: an eight-year homogeneous TRAPPIST light curve covering 15 au to 8 au, the first simultaneous optical (UVES) and near-infrared (CRIRES+) parent-daughter snapshot for this comet, and a first NiI/FeI ratio for it. The photometric reduction is standard Haser-model work, carefully compared to independent datasets (Schleicher, Combi et al.), and the derived C2/CN 'typical' taxonomy and constant colors are credible. I don't see manufacturing or over-claiming in the basic measurements.\n\nThe soft spot is the G/R interpretation. The paper reads the drop from 0.25 at 3.23 au to 0.08 at 2.1 au as a clean CO2-to-H2O switch. But the G/R diagnostic is being treated as more decisive than it is. There is no discussion of collisional quenching of O(1S) in the inner coma, which matters at these production rates and would also push G/R down. And the same-epoch CRIRES+ data actually complicate the story: CO/H2O is ~11% at 2.73 au and ~14% at 2.1 au, so a simple H2O+CO mixture would predict G/R near 0.11, neither matching 0.14-0.15 at 2.73 nor 0.08 at 2.1. The higher-than-predicted value at 2.73 could require CO2 or another oxygen source, but CO2 is not measured. The lower-than-predicted value at 2.1 could be quenching. The paper does not address either. Also, Section 3.2.2's claim of near-surface CO depletion before 2.8 au sits oddly with the direct CO detection at 2.73 au in their own Table 11.\n\nThese are revision-level problems, not fatal. The broad conclusion—activity transition near 3 au from supervolatiles to water—is supported by the light curve, Afrho, and the appearance of OH. The specific attribution to CO2 over CO is not established. Minor issues: some OH entries in Table A.1 look mis-formatted, and the C2 parent claim rests on upper limits for C2H2.\n\nWho should read it: anyone working on long-period comet activity, Oort cloud chemistry, or Comet Interceptor target selection. It's a useful reference even if the parent assignment is provisional. I would accept it for peer review and push for a revision that directly addresses quenching and the CO/CO2 degeneracy.","headline":"A careful, data-rich study of C/2017 K2 that deserves review, but the G/R-based CO2-to-water switch is overinterpreted and needs a revision.","tokens_in":31973,"tokens_out":4365,"would_cite":true,"duration_ms":44922,"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":"Forbidden oxygen line ratios in C/2017 K2 show its coma switched from supervolatile CO/CO$_2$ to water as the comet crossed roughly 3 au.","keywords":["comet C/2017 K2 (PanSTARRS)","dynamically new comet","Oort cloud comet","forbidden oxygen lines","green-to-red ratio","water sublimation boundary","CO/CO2-driven activity","comet composition"],"falsifier":"Measure K2's G/R ratio beyond about 3.5 au with high signal-to-noise and simultaneously measure direct production rates of CO, CO$_2$, and O$_2$ (for example from infrared spectra); if the ratio does not stay high outside 3 au, or if the parent fractions inferred from G/R disagree with the directly measured CO$_2$ and CO production rates, the activity-switch claim would be refuted.","tokens_in":30823,"feed_emoji":"☄️","tokens_out":8769,"duration_ms":82966,"temperature":0.7,"pith_summary":"An eight-year campaign on the Oort-cloud comet C/2017 K2 (PanSTARRS), from 15.18 au before perihelion to 8.46 au after it, shows that the comet's activity changed power source as it crossed the water-sublimation boundary near 3 au. The paper argues that carbon monoxide and carbon dioxide drove the coma at larger distances, while water vapor took over inside 3 au. The key evidence is the green-to-red forbidden oxygen line ratio, which fell from 0.25 at 3.23 au to 0.08 at 2.12 au in the way the Festou & Feldman (1981) calibration predicts when water photodissociation becomes dominant. The result matters because K2 is a dynamically new comet, likely on its first passage into the inner solar system, so the measurements show how a pristine nucleus sheds its most volatile surface ice and then reawakens as water sublimation begins.","feed_headline":"Oxygen lines show comet K2's power source shifts at 3 au","feed_subtitle":"Eight years of data track a pristine Oort-cloud comet losing its super-volatile surface ice before water takes over.","key_machinery":"The central diagnostic is the green-to-red ratio of the forbidden oxygen lines, $$G/R = I(5577\\,\\mathrm{\\AA})/(I(6300\\,\\mathrm{\\AA})+I(6364\\,\\mathrm{\\AA})),$$ which tracks which parent molecules feed oxygen atoms in the coma. Festou & Feldman (1981) calibrated this ratio against the relative contributions of H$_2$O, CO$_2$, and CO photodissociation, with water near $G/R \\approx 0.1$ and higher values for CO and CO$_2$. Around this diagnostic the paper assembles an eight-year TRAPPIST light curve, Haser-model production rates of OH, NH, CN, C$_3$ and C$_2$, the dust proxy $A(0)f\\rho$, and simultaneous CRIRES+ infrared spectra of parent volatiles. The drop in $G/R$ across 3 au is the load-bearing piece of evidence for the activity-switch claim.","core_discovery":"The paper establishes that C/2017 K2's coma underwent a composition-driven activity transition. Using high-resolution UVES spectra at three epochs in 2022, the authors measured the forbidden oxygen line ratio $G/R = I(5577\\,\\mathrm{\\AA})/(I(6300\\,\\mathrm{\\AA})+I(6364\\,\\mathrm{\\AA}))$ and found a monotonic drop with decreasing heliocentric distance: 0.25 at 3.23 au, 0.14–0.15 at 2.73 au, and 0.08 at 2.12 au. Following Festou & Feldman (1981), they interpret this as a switch from a coma whose oxygen atoms come mostly from CO and CO$_2$ photodissociation to one dominated by H$_2$O photodissociation once water sublimation begins inside roughly 3 au. Supporting pieces are the non-detection of OH before the second epoch, CO emission too faint to measure in May 2022, the detection of CO$_2^+$ without CO$^+$, and a temporary stall in brightness and dust production between 3.6 and 2.7 au attributed to depletion of near-surface CO and CO$_2$. The paper also finds that K2's dust colors stayed constant across the entire campaign, that its gas composition is “typical” with a high dust-to-gas ratio, and that HCN is the likely parent of CN while C$_2$ is more likely from C$_2$H$_2$ than from C$_2$H$_6$.","pith_inferences":["If the G/R drop at 3 au is a generic property of comets, archival spectra of other long-period comets observed between 2.5 and 3.5 au should show the same steep decline; this can be checked without new observations.","A direct test that goes beyond the paper would be measuring O$_2$ in K2's coma in the 2–4 au range; if O$_2$ is abundant, the G/R-based H$_2$O/CO$_2$/CO partition would need revision, since O$_2$ photodissociation also produces oxygen atoms.","The inferred near-surface depletion of CO and CO$_2$ suggests that “dynamically new” comets may be classified by their outer-layer composition rather than their bulk composition, and future missions to pristine comets could look for a layered volatile structure rather than uniform ice.","One extension of the authors' method is to follow the post-perihelion branch: if the same G/R pattern appears as the comet recedes beyond 3 au, it would confirm that the transition follows the local sublimation temperature rather than an irreversible depletion of the nucleus."],"forward_implications":["If the G/R interpretation holds, K2's distant activity, detected out to 23.75 au, was driven by supervolatile sublimation, and the comet's modest brightness near perihelion reflects a real loss of CO/CO$_2$ surface ice before water sublimation could take over.","The plateau and subsequent steep fading in the light curve imply that dynamically new comets can undergo a volatile-exhaustion phase on their first inner-solar-system passage, so brightness alone may underestimate their nuclear volatile content.","The constant dust colors over eight years and across roughly 6 au of heliocentric distance indicate that the dust grain properties in the coma did not change, supporting comparisons of K2's dust with other long-period comets regardless of observing epoch.","The parent–daughter abundance matching (HCN→CN, C$_2$H$_2$→C$_2$) means simultaneous optical and infrared spectroscopy can be used to trace specific ice abundances in K2 and similar comets.","The similarity of K2's NiI/FeI ratio to other comets, despite its dynamically new status, supports the idea that metal abundance ratios are set at formation rather than by subsequent processing."],"supporting_citations":[{"why":"Supplies the calibration that maps the G/R forbidden-oxygen ratio to the relative contributions of H$_2$O, CO$_2$, and CO photodissociation, the basis for the central activity-switch claim.","marker":"Festou & Feldman (1981)"},{"why":"Sub-millimeter detection of CO in K2's coma at 6.72 au anchors the claim that supervolatiles drive the distant activity.","marker":"Yang et al. (2021)"},{"why":"Prediscovery images showing K2 active at 23.75 au establish the extreme-distance activity that the paper explains with CO/CO$_2$ sublimation.","marker":"Meech et al. (2017)"},{"why":"Reports K2's activity at 23.8 au and attributes such distant activity to supervolatile ices, the interpretation the paper adopts.","marker":"Jewitt et al. (2017)"},{"why":"Defines the “typical” versus “depleted” taxonomic classes and the Haser-model scale lengths used to classify K2's C$_2$/CN composition.","marker":"A'Hearn et al. (1995)"},{"why":"Provides the MUSE-based G/R value at 2.53 au that the paper reproduces, supporting the G/R trend and the water-onset interpretation.","marker":"Kwon et al. (2023)"},{"why":"Reports G/R = 0.28 at 2.8 au for K2, a comparison point for the paper's finding of high G/R outside the water sublimation region.","marker":"Cambianica et al. (2023)"},{"why":"Defines the roughly 3 au water sublimation boundary used to place the transition in heliocentric distance.","marker":"Crovisier & Encrenaz (2000)"},{"why":"Gives SOHO Lyman-alpha water production rates and the 28,000 au original semi-major axis, supporting both the production-rate comparison and K2's dynamically new classification.","marker":"Combi et al. (2025)"},{"why":"JWST measurements of K2's water production at 2.35 au provide an independent infrared comparison for the production rates and mixing ratios.","marker":"Woodward et al. (2025)"}],"fun_headline_variants":["Comet K2 switches from CO2 to water inside 3 au","Oxygen line ratios reveal K2's activity transition","Eight-year data show K2's volatile shift at 3 au","Pristine comet K2's activity driven by water inside 3 au"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The G/R ratio is read through the Festou & Feldman (1981) calibration, which assumes the coma's oxygen atoms come only from photodissociation of H$_2$O, CO$_2$, and CO and that no other oxygen source (such as O$_2$ or dust sputtering) and no collisional quenching perturbs the line ratios between 2 and 3 au.","fun_headline_variants_meta":{"raw":{"variants":["Comet K2 switches from CO2 to water inside 3 au","Oxygen line ratios reveal K2's activity transition","Eight-year data show K2's volatile shift at 3 au","Pristine comet K2's activity driven by water inside 3 au"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000272,"raw_usage":{"total_tokens":1811,"prompt_tokens":1304,"completion_tokens":507,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":920,"completion_tokens_details":{"reasoning_tokens":432}},"tokens_in":920,"tokens_out":507,"duration_ms":5631,"temperature":1.0,"reasoning_tokens":432,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T16:24:32.521160+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure K2's G/R ratio beyond about 3.5 au with high signal-to-noise and simultaneously measure direct production rates of CO, CO$_2$, and O$_2$ (for example from infrared spectra); if the ratio does not stay high outside 3 au, or if the parent fractions inferred from G/R disagree with the directly measured CO$_2$ and CO production rates, the activity-switch claim would be refuted.","supporting_citations":[{"cited_title":"& Feldman , P","cited_arxiv_id":null,"evidence_quote":"Supplies the calibration that maps the G/R forbidden-oxygen ratio to the relative contributions of H$_2$O, CO$_2$, and CO photodissociation, the basis for the central activity-switch claim."},{"cited_title":"2021, , 914, L17","cited_arxiv_id":null,"evidence_quote":"Sub-millimeter detection of CO in K2's coma at 6.72 au anchors the claim that supervolatiles drive the distant activity."},{"cited_title":"J., Kleyna , J","cited_arxiv_id":null,"evidence_quote":"Prediscovery images showing K2 active at 23.75 au establish the extreme-distance activity that the paper explains with CO/CO$_2$ sublimation."},{"cited_title":"2017, , 847, L19","cited_arxiv_id":null,"evidence_quote":"Reports K2's activity at 23.8 au and attributes such distant activity to supervolatile ices, the interpretation the paper adopts."},{"cited_title":"F., Millis , R","cited_arxiv_id":null,"evidence_quote":"Defines the “typical” versus “depleted” taxonomic classes and the Haser-model scale lengths used to classify K2's C$_2$/CN composition."},{"cited_title":"G., Opitom , C., & Lippi , M","cited_arxiv_id":null,"evidence_quote":"Provides the MUSE-based G/R value at 2.53 au that the paper reproduces, supporting the G/R trend and the water-onset interpretation."},{"cited_title":"2023, , 674, L14","cited_arxiv_id":null,"evidence_quote":"Reports G/R = 0.28 at 2.8 au for K2, a comparison point for the paper's finding of high G/R outside the water sublimation region."},{"cited_title":"& Encrenaz , T., eds","cited_arxiv_id":null,"evidence_quote":"Defines the roughly 3 au water sublimation boundary used to place the transition in heliocentric distance."},{"cited_title":"Water Production Rates from SOHO/SWAN Observations of Comets C/2017 K2 (PanSTARRS) and C/2022 E3 (ZTF)","cited_arxiv_id":"2505.11699","evidence_quote":"Gives SOHO Lyman-alpha water production rates and the 28,000 au original semi-major axis, supporting both the production-rate comparison and K2's dynamically new classification."},{"cited_title":"A JWST Study of the Remarkable Oort Cloud Comet C/2017 K2 (PanSTARRS)","cited_arxiv_id":"2504.19849","evidence_quote":"JWST measurements of K2's water production at 2.35 au provide an independent infrared comparison for the production rates and mixing ratios."}],"review_version":1}