{"id":"85b3dff8-6cba-4a9c-b7af-46729112b049","arxiv_id":"2507.12756","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Comet C/2023 A3 ejected dust at about 170 kg/s before perihelion, with ejection speeds near (65±5) β^1/2 m/s and activity onset around 9.1 au.","lead":"This study measures the dust output of comet C/2023 A3 (Tsuchinshan-ATLAS) before it reached the Sun, using archival Zwicky Transient Facility images. It finds a steady dust loss rate near 170 kilograms per second and estimates the activity began at 9.1 au from the Sun, likely driven by a phase change in water ice.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Degeneracy between a_min and s in the FWHM fit, plus sensitivity of Eq. 8 to the chosen aperture rate, leaves the 170 kg/s dust loss rate under-constrained.","rationale":"The reader identifies the two-parameter fit to a single FWHM observable as the weakest point, and I agree; I further note that Eq. 8 inherits aperture choice because the slopes in Table 3 vary strongly with aperture, and the paper selects the largest-aperture slope without demonstrating that it equals the total production rate. The paper itself labels the activity-onset extrapolation an estimate, so that is not the central vulnerability. Independent support exists: the two velocity estimates agree, the phase function is checked against a ZTF-aligned subsample, and the size index agrees reasonably with Moreno et al. (2025). The central number may be correct within a factor of two, but the stated uncertainty is likely too small. A CONDITIONAL recommendation is appropriate, asking for a sensitivity analysis of Eq. 8 over the degenerate (a_min, s) region and an aperture-growth check.","tokens_in":16295,"tokens_out":7901,"duration_ms":73005,"concrete_test":"Recompute the dust-loss rate using dC_e/dt from the two largest apertures (80,000 and 160,000 km) and from a weighted mean across all apertures, propagating the reported 1-sigma slope uncertainties; then vary (a_min, s) across the grid shown in Figure 7 (a_min = 10-100 um, s = 2.8-3.8) and recompute a_bar from Eq. 7 for each pair. If the resulting dM/dt ranges by more than a factor of three or the central value shifts below 100 kg/s, the headline rate is not robust; the paper should then either quote a broader systematic range or add a forward-model check that the simulated coma simultaneously matches the observed aperture growth rates.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The headline dust-loss rate is set by Eq. 8, dM/dt = (4 a_bar rho / 3) dC_e/dt, with a_bar = 0.4 mm computed from Eq. 7 using a_min = 20 um and s = 3.4. These two parameters are fitted to a single scalar, the FWHM of the radial brightness profile at one epoch (Sec. 4.2, Fig. 7), with no uncertainty quantification. The contour plot likely contains a curved valley: a_min and s are degenerate, so a_min = 10-50 um with an adjusted s is not excluded. Because a_bar shifts by a factor of order two across that range, the 170 kg/s central value inherits a systematic error not captured by the quoted +/-80 kg/s. In addition, Eq. 8 uses dC_e/dt = 55 +/- 27 km^2/day from only the largest 160,000-km aperture (Table 3); smaller apertures give systematically smaller slopes (e.g., 20,000 km: 19.6 +/- 5.1 km^2/day). A fixed-aperture cross-section rate is not obviously the total production rate: it mixes freshly produced dust with older grains leaving the aperture, and the simple factor 4/3 applies to the total ejected cross-section, not to the portion inside an arbitrary projected radius. The paper does not justify that the largest-aperture slope equals the total production rate, so the headline rate may be biased. The activity-onset extrapolation is explicitly flagged by the authors as an estimate (Sec. 4.1), so it is not the most load-bearing claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes archival ZTF images of long-period comet C/2023 A3 (Tsuchinshan-ATLAS) obtained between 2024 February and May, before perihelion. From the coma surface brightness profiles the authors infer a steady-state inner coma with slope q ≈ -1 and a sunward turnaround distance that yields an ejection velocity v_ej ~ (88 ± 3) β^{1/2} m/s; from the tail width perpendicular to the orbital plane they derive a consistent velocity v_perp ~ (65 ± 5) β^{1/2} m/s. Photometry in fixed projected apertures gives the dust scattering cross-section and its rate of change, a nucleus radius upper limit of 5.9 ± 0.2 km at albedo 0.04, and an extrapolated activity onset around 25 July 2022 at 9.1 au. A dust dynamics simulation is used to fit the minimum grain radius a_min = 20 μm and size distribution index s = 3.4 from the FWHM of the radial brightness profile at one epoch. These values feed an average grain radius a_bar ≈ 0.4 mm and, via Eq. (8), the headline dust loss rate (1.7 ± 0.8) × 10^2 kg/s. The paper also argues that the nucleus is stable against tidal disruption, sublimation erosion, and rotational instability.","tokens_in":16700,"tokens_out":5766,"duration_ms":69085,"significance":"If the central dust-loss-rate result is robust, the paper provides an important pre-perihelion characterization of a great comet and useful constraints on its activity before the dramatic brightening. The authors should be credited for several independent or carefully checked ingredients: the v_perp measurement from the geometric tail width is independent of the photometric modeling; the phase-function slope is checked by refitting with ZTF-aligned dates; the activity-onset extrapolation is explicitly flagged as an estimate; and the simulated tail morphology is compared with an observed linear feature. The paper also engages with independent work by Moreno et al. (2025) and identifies a plausible crystallization mechanism for early activity. However, the headline dust loss rate depends on a two-parameter fit to a single scalar observable and on a fixed-aperture cross-section rate whose relation to the total production rate is not demonstrated. These issues currently leave the central quantitative claim under-constrained.","major_comments":[{"comment":"The two free parameters a_min and s are adjusted to match a single scalar, the FWHM of the radial brightness profile at one epoch (30 May 2024), and no uncertainty is quoted for the best-fit pair. The text says 'we prefer not to overinterpret the result', but the fitted values are nevertheless used without error in the average grain size a_bar = 0.4 mm and then in Eq. (8). Because Eq. (7) is sensitive to a_min for s near 3.4, and Figure 7 shows only a normalized absolute difference without confidence levels, a degenerate (a_min, s) valley — for example a_min of order 10–50 μm with a slightly adjusted s — is not excluded. That would change a_bar by roughly a factor of two and directly scale the dust loss rate. The authors should provide uncertainty contours (e.g., ΔFWHM levels corresponding to the measurement error of the observed FWHM) and propagate the resulting a_bar uncertainty into Eq. (8).","section":"Section 4.2, Figure 7 and Eq. (7)"},{"comment":"The dust production rate is evaluated using only the 160,000-km aperture slope k = 55 ± 27 km²/day, while the slopes for the smaller apertures are systematically lower (8.3 ± 4.0, 19.6 ± 5.1, 28.2 ± 22.3, and 45.4 ± 30.9 km²/day) and the linear fits have low R² values (0.35–0.83, as stated in the Figure 6 caption). A fixed projected aperture does not directly measure the total ejected cross-section: fresh grains enter and older grains leave the aperture, so dC_e/dt inside a fixed projected radius is not obviously equal to the total production rate. The factor 4/3 in Eq. (8) applies to the total ejected population under a spherical-grain assumption, not automatically to the portion inside an arbitrary aperture. The authors need to justify, or correct with a model, the conversion from aperture-residence cross-section to total dust production; otherwise the central 170 kg/s value has an unquantified aperture-dependent bias.","section":"Section 4.2, Eq. (8) and Table 3"},{"comment":"The quoted uncertainty of ±80 kg/s on the dust loss rate appears to propagate only the statistical uncertainty of dC_e/dt. It does not include the systematic uncertainty in a_bar from the a_min–s fit, the assumed bulk density ρ = 500 kg/m³, or the choice of aperture slope. For example, if a_bar were 0.2 mm instead of 0.4 mm the rate would halve, and ρ in the plausible range 300–800 kg/m³ would scale the rate by 0.6–1.6. A total systematic error budget should be presented before the abstract quotes (1.7 ± 0.8) × 10² kg/s, which currently implies a precision that is not supported by the analysis.","section":"Section 4.2, Eq. (8) and abstract"}],"minor_comments":[{"comment":"The text states that on 23 February 2024 the tail fades at about 30 arcsec from the nucleus, while on 30 May 2024 the tail extends beyond the 60-arcsec field of view; the projected distance quoted for 60 arcsec should be checked, since at Δ = 1.79 au the linear scale is about 78,000 km rather than 80,000 km.","section":"Section 3.1 and Figure 1 caption"},{"comment":"There is a typo: 'The the average size...' should read 'The average size...'.","section":"Section 4.2, before Eq. (7)"},{"comment":"Please clarify the units used for r_n in Eq. (6). The text says r_n is 'used in kilometers', but for β to be dimensionless the numerical constants in the equation need to be consistent; a reader trying to reproduce β_min = 0.0001 will otherwise obtain a different value if r_n is converted to meters.","section":"Section 4.2, Eq. (6)"},{"comment":"The horizontal axis label appears to be missing the micro sign: the grid is described in the text as spanning 1 μm to 1.5 mm, so the axis should read 'Minimum radius (μm)' rather than 'Minimum radius ( m)'.","section":"Figure 7"},{"comment":"The abstract and conclusions quote the activity onset as '25 July 2022', but Table 3 lists per-aperture onset times from -452 to -692 days with a weighted mean of -524 ± 104 days. The text should state explicitly which aperture or combination is used for the quoted value and why the weighted mean is preferred over the individual fits.","section":"Table 3 and Section 4.1"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is likely to be of interest to the comet-dust community, but the headline dust-loss rate is not yet robust because the a_min–s fit and the aperture-residence conversion are not quantified. The issues are fixable within the scope of the paper: adding uncertainty contours for Figure 7, propagating systematic errors through Eq. (8), and either modeling the aperture effect or clearly labeling the rate as aperture-dependent would strengthen the central claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a careful, mostly solid analysis of pre-perihelion ZTF imaging of C/2023 A3, and the headline dust loss rate of ~170 kg/s is plausible but softer than the error bar suggests. The v_perp measurement and the nucleus radius upper limit are independently grounded and worth taking seriously. The load-bearing number, however, rests on a two-parameter fit to a single FWHM and on a fixed-aperture cross-section slope that hasn't been reconciled with the total production rate.\n\nWhat's new: they give the first quantitative dust loss parameters for A3 from ZTF: v_perp ~65±5 β^1/2 m/s, a_min=20 µm, s=3.4, dust loss ~170 kg/s, activity onset at 9.1 au. These numbers don't appear in the Moreno et al. paper, which they cite and compare with. The surface brightness profiles show a steady coma (q≈-1), and they cross-check the phase function with ZTF-aligned COBS dates. The stability analysis is a bonus, not the main point, but it's done sensibly.\n\nThe biggest issue is the dust loss rate. The average grain size 0.4 mm comes from a fit of a_min and s to the FWHM of the radial profile at one epoch. That fit has no uncertainties, and the contour plot in Fig 7 likely contains a degenerate valley—a_min between 10 and 50 µm with adjusted s would shift a_bar by a factor of two. Second, Eq. 8 uses dC_e/dt = 55±27 km²/day from only the 160,000-km aperture. The slope at 20,000 km is 19.6±5.1, which would cut the derived rate to about 60 kg/s. A fixed-aperture cross-section rate mixes fresh dust with grains that are leaving the aperture, so it's not obviously the total production rate. The paper doesn't justify why the largest-aperture slope is the right one. These issues don't undermine the v_perp or nucleus radius results, but they do mean the 170 kg/s central value is less certain than the ±80 kg/s suggests.\n\nAlso, the activity onset at 9.1 au is a linear extrapolation from a short arc with low R², but the authors flag it as an estimate, so I don't hold that against them.\n\nWho it's for: anyone working on cometary dust activity or long-period comet evolution. It's a useful data point, and the comparison with Moreno et al. is valuable. It deserves peer review—the methods are standard but applied carefully, and the questions about the dust loss rate are exactly what a referee should ask for. I'd engage with it, and I'd want to see a revised version that addresses the aperture dependence and the fit degeneracy.","headline":"Plausible and useful pre-perihelion dust characterization of A3, but the headline mass-loss rate is softer than the quoted uncertainty suggests.","tokens_in":17261,"tokens_out":4142,"would_cite":true,"duration_ms":43875,"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 measures pre-perihelion dust loss from comet C/2023 A3 at roughly 170 kg/s, with a steady coma, grains as small as 20 microns, and a nucleus that survives the encounter.","keywords":["comets: individual: C/2023 A3 (Tsuchinshan-ATLAS)","dust loss rate","cometary coma","scattering cross-section","long-period comet","dust dynamics simulation","nucleus stability","amorphous-crystalline ice transition"],"falsifier":"Observe the coma at several additional pre-perihelion epochs and fit the dust dynamics model to the full two-dimensional surface brightness rather than to one radial FWHM; if no single set of ($a_{\\rm min}$, $s$) reproduces all epochs simultaneously, or if the implied dust-to-gas mass ratio is inconsistent with the measured water production rate, the 170 kg/s estimate would be ruled out.","tokens_in":16045,"feed_emoji":"☄️","tokens_out":5230,"duration_ms":54836,"temperature":0.7,"pith_summary":"The paper tries to establish how much dust the Great Comet C/2023 A3 was shedding before its late-2024 perihelion and whether it was in danger of disintegrating. From archival images taken between February and May 2024, it infers a steady-state coma, an ejection velocity that scales as $v_\\perp\\sim(65\\pm5)\\beta^{1/2}$ m s$^{-1}$, dust grains from 20 microns to about 10 mm with a size distribution index $s=3.4$, and a dust loss rate of $(1.7\\pm0.8)\\times10^2$ kg s$^{-1}$. It also extrapolates the scattering cross-section back in time to estimate that activity began around 25 July 2022 at 9.1 au, and argues that the amorphous-to-crystalline ice transition, not water sublimation, triggered it. The paper further argues that the roughly 6 km nucleus is stable against tidal, sublimation, and rotational breakup, consistent with the comet having survived perihelion.","feed_headline":"Comet A3 shed about 170 kg of dust per second before perihelion","feed_subtitle":"Archival images reveal a steady coma, 20-micron grains, and a nucleus that survived the encounter.","key_machinery":"The central quantities are the effective scattering cross-section $C_e$ and the dimensionless radiation-pressure parameter $\\beta$, which is inversely proportional to grain radius. The observational machinery consists of three pieces: the logarithmic slope $q$ of the surface brightness profile, whose value near $-1$ marks a steady-state coma; the sunward turnaround distance $l_{\\rm coma}$ and the out-of-plane tail width $\\theta_\\perp$, which translate projected distances into ejection velocities of the form $v\\propto\\beta^{1/2}$; and a Monte Carlo dust dynamics simulation that propagates grains under solar gravity plus radiation pressure and compares the simulated radial FWHM with the observed FWHM on 30 May 2024, fixing $a_{\\rm min}=20\\,\\mu$m and $s=3.4$. The dust loss rate follows from $dM_d/dt = (4\\bar{a}\\rho/3)\\,dC_e/dt$, with the average grain radius $\\bar{a}\\approx0.4$ mm obtained from the size distribution.","core_discovery":"Using archival ZTF images, the authors measure the coma's surface brightness profile and find a logarithmic slope $q\\approx -1$ in the inner 7 arcseconds, the signature of a steady-state coma in which dust loss is roughly constant over time. A dust dynamics simulation that integrates solar gravity and radiation pressure, matched to the observed radial FWHM on 30 May 2024, fixes the minimum grain radius at 20 microns and the power-law size distribution index at $s=3.4$, with grains extending up to about 10 mm. Combining the resulting average grain radius of about 0.4 mm with the measured rate of change of the scattering cross-section, $dC_e/dt=55\\pm27$ km$^2$ day$^{-1}$, yields a dust production rate of $(1.7\\pm0.8)\\times10^2$ kg s$^{-1}$. The same data place an upper limit of $5.9\\pm0.2$ km on the nucleus radius (assuming a geometric albedo of 0.04) and, by backward extrapolation, an activity onset around 25 July 2022 at 9.1 au, which the authors attribute to the amorphous-to-crystalline ice phase transition. A simplified stability analysis rules out tidal disruption ($\\epsilon\\sim10^{-5}$ at perihelion), sublimation erosion (about 700 years to erode the nucleus), and sublimation-induced rotational spin-up (timescale greater than about 50 years), leading the authors to conclude that the nucleus was unlikely to fragment near perihelion.","pith_inferences":["The extrapolated onset date of 25 July 2022 rests on a linear fit to only five photometric epochs and assumes the scattering cross-section grew linearly before the observations; if activity began later or in bursts, the 9.1 au onset and the amorphous-ice interpretation would be weakened.","Because the two free dust parameters, minimum grain size and size distribution index, are fitted to a single scalar observable, the FWHM at one epoch, the reported 170 kg/s should be regarded as order-of-magnitude until the fit is checked against additional epochs or against the full two-dimensional coma morphology.","The steady-coma interpretation of $q\\approx -1$ could be tested directly by measuring the surface brightness profile in other filters or at more epochs; a slope that deviates systematically from $-1$ would indicate nonsteady dust emission and would change the mass-loss extrapolation.","If the amorphous-to-crystalline ice trigger is correct, other long-period comets first becoming active in the 5-17 au range should show a similar onset-distance pattern, a testable prediction for future distant comet observations."],"forward_implications":["The nucleus of C/2023 A3 is inferred to have a radius of at least several kilometers, making tidal breakup at its 0.39 au perihelion very unlikely, and its survival is consistent with the stability analysis.","Pre-perihelion dust production was relatively modest and steady, implying that the comet's great brightness was achieved with a small total mass loss of about $10^{10}$ kg, roughly $10^{-4}$ of the nucleus mass.","If activity indeed began at 9.1 au, the amorphous-to-crystalline ice phase transition becomes a plausible trigger for distant dust activity in long-period comets, complementing water-ice sublimation at 4-5 au.","A size distribution index of $s=3.4$ with a 20 micron minimum grain size implies that small, highly scattering grains dominate the observed coma, consistent with reported high polarization measurements.","The derived velocity-size relation, $v_\\perp\\sim(65\\pm5)\\beta^{1/2}$ m s$^{-1}$, provides a quantitative constraint for models of dust ejection from long-period comets."],"supporting_citations":[{"why":"Supplies the formula relating the sunward coma turnaround distance $l_{\\rm coma}$ to ejection velocity and radiation pressure, used to derive $v_{ej}\\sim(88\\pm3)\\beta^{1/2}$ m s$^{-1}$.","marker":"Jewitt & Meech 1987"},{"why":"Supplies the relation between out-of-plane tail width and perpendicular ejection velocity, used to obtain $v_\\perp\\sim(65\\pm5)\\beta^{1/2}$ m s$^{-1}$.","marker":"Jewitt et al. 2015"},{"why":"Provides the dust dynamics simulation framework that numerically integrates solar gravity and radiation pressure to generate synthetic coma images.","marker":"Liu et al. 2016"},{"why":"Refines the dust model and supplies the radiation-pressure constant $C_{pr}$ used to convert $\\beta$ values into grain radii.","marker":"Liu & Liu 2024"},{"why":"Provides an independent dust environment model of C/2023 A3 used for comparison; the paper's velocities and production rates align with it, while its size distribution index is slightly higher.","marker":"Moreno et al. 2025"},{"why":"Reports high polarization degrees that the paper cites as independent evidence for a pre-perihelion population of small, highly scattering dust grains.","marker":"Lim et al. 2025"},{"why":"Supplies the typical cometary dust geometric albedo range (0.03-0.06) that justifies the adopted albedo of 0.04 in converting scattering cross-section to nucleus radius.","marker":"Hanner 2003"},{"why":"Supplies the adopted solar $r$-band magnitude used in the scattering cross-section photometry equation.","marker":"Willmer 2018"},{"why":"Provides the water production rate on 31 May 2024 that the nucleus stability analysis uses to estimate the sublimation-driven mass loss and rotational spin-up timescale.","marker":"Ahuja et al. 2024"},{"why":"Raises the competing claim that C/2023 A3 may have undergone a disintegration event, which the paper's stability analysis specifically argues against.","marker":"Sekanina 2024"}],"fun_headline_variants":["Comet A3's steady coma yields 170 kg/s dust loss, nucleus safe","A3's pre-perihelion dust rate: 170 kg/s, nucleus stable","Comet A3: dust loss 170 kg/s, nucleus intact after perihelion","Steady coma, 170 kg/s dust loss, nucleus survived: comet A3","A3's dust rate 170 kg/s, nucleus stable, activity onset at 9 au"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything hinges on the dust simulation's fit of two free parameters, the minimum grain size and the size distribution index, to a single number, the width of the radial brightness profile on one night; if that fit is not unique, the inferred average grain size and the 170 kg/s dust loss rate change proportionally.","fun_headline_variants_meta":{"raw":{"variants":["Comet A3's steady coma yields 170 kg/s dust loss, nucleus safe","A3's pre-perihelion dust rate: 170 kg/s, nucleus stable","Comet A3: dust loss 170 kg/s, nucleus intact after perihelion","Steady coma, 170 kg/s dust loss, nucleus survived: comet A3","A3's dust rate 170 kg/s, nucleus stable, activity onset at 9 au"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000951,"raw_usage":{"total_tokens":4188,"prompt_tokens":1207,"completion_tokens":2981,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":823,"completion_tokens_details":{"reasoning_tokens":2866}},"tokens_in":823,"tokens_out":2981,"duration_ms":21817,"temperature":1.0,"reasoning_tokens":2866,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T16:39:40.030806+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe the coma at several additional pre-perihelion epochs and fit the dust dynamics model to the full two-dimensional surface brightness rather than to one radial FWHM; if no single set of ($a_{\\rm min}$, $s$) reproduces all epochs simultaneously, or if the implied dust-to-gas mass ratio is inconsistent with the measured water production rate, the 170 kg/s estimate would be ruled out.","supporting_citations":[{"cited_title":"& Meech, K","cited_arxiv_id":null,"evidence_quote":"Supplies the formula relating the sunward coma turnaround distance $l_{\\rm coma}$ to ejection velocity and radiation pressure, used to derive $v_{ej}\\sim(88\\pm3)\\beta^{1/2}$ m s$^{-1}$."},{"cited_title":"2015, The Astro- physical Journal, 798, 109","cited_arxiv_id":null,"evidence_quote":"Supplies the relation between out-of-plane tail width and perpendicular ejection velocity, used to obtain $v_\\perp\\sim(65\\pm5)\\beta^{1/2}$ m s$^{-1}$."},{"cited_title":"2016, Journal of Geophysical Re- search: Planets, 121, 1141","cited_arxiv_id":null,"evidence_quote":"Provides the dust dynamics simulation framework that numerically integrates solar gravity and radiation pressure to generate synthetic coma images."},{"cited_title":"& Liu, X","cited_arxiv_id":null,"evidence_quote":"Refines the dust model and supplies the radiation-pressure constant $C_{pr}$ used to convert $\\beta$ values into grain radii."},{"cited_title":"2025, The Astrophysical Journal Let- ters, 983, L19","cited_arxiv_id":null,"evidence_quote":"Reports high polarization degrees that the paper cites as independent evidence for a pre-perihelion population of small, highly scattering dust grains."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the typical cometary dust geometric albedo range (0.03-0.06) that justifies the adopted albedo of 0.04 in converting scattering cross-section to nucleus radius."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the adopted solar $r$-band magnitude used in the scattering cross-section photometry equation."},{"cited_title":"2024, The Astronomer’s Telegram, 16637, 1","cited_arxiv_id":null,"evidence_quote":"Provides the water production rate on 31 May 2024 that the nucleus stability analysis uses to estimate the sublimation-driven mass loss and rotational spin-up timescale."}],"review_version":1}