{"id":"9ce508e8-50c9-4823-92fb-4bcc43a91e79","arxiv_id":"2507.21213","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The CHIMERA survey detects three CO(1-0) clouds in Leo T, a 10^5 M_sun dwarf galaxy, with virial masses around 5e3 M_sun each and CO-to-H2 conversion factors up to ~155, and finds one cloud that may be escaping the galaxy.","lead":"Astronomers report the first detection of carbon monoxide (CO) in Leo T, the lowest-mass galaxy known to still hold cold molecular gas. The finding tests how molecular gas and star formation behave in extremely metal-poor, low-mass dwarfs near the Milky Way.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Foreground Milky Way association of the CO clouds, especially the north cloud, is excluded only by an alphaCO argument (App. G) that extrapolates the MW GMC scaling relation to LCO~0.4 and ignores CO-dark/diffuse gas; this leaves the Leo T association insecure.","rationale":"The reader's weakest assumption is the foreground Milky Way association of the CO emission, and this is indeed the most load-bearing uncertainty. The detection significance of the individual clouds is reasonably established by the S/N analysis in Appendix B, but the conversion of those detections into 'CO in Leo T' requires a distance/association argument. Appendix G is the only quantitative step toward excluding foreground contamination, and it has a specific flaw: it rejects a foreground location because the inferred alphaCO is two orders of magnitude above the MW GMC relation, yet that relation is not applicable to a cloud with LCO~0.4 K km/s pc2 and does not exclude CO-dark or diffuse MW gas. Thus the north cloud's association with Leo T remains genuinely uncertain, and the derived masses, alphaCO up to ~155, and the gas-expulsion conclusion are conditional on that association. The west and south clouds are less exposed because their velocities are within 2 sigma of Leo T's HI systemic velocity, but no chance-coincidence estimate is given, so the ensemble association is not fully quantified. This does not warrant rejection: the data, reduction, and cloud-detection methodology are standard, the velocities of the west and south clouds are physically reasonable for Leo T, and the authors present the physical quantities as upper limits. The appropriate outcome is the same CONDITIONAL verdict: the paper should be published with the caveat that the foreground scenario, particularly for the north cloud, must be addressed with Galactic HI/dust data or a proper chance-coincidence calculation before the headline claims are taken as established.","tokens_in":15340,"tokens_out":12897,"duration_ms":173299,"concrete_test":"Extract the HI4PI/GALFA-HI 21-cm spectra at the exact positions of the three CO clouds and run a source-finder over v=-100 to +100 km/s to identify all Galactic HI velocity components; compare with the CO velocities (53 and 97 km/s). Also map the HI spatial distribution at v~53 km/s to see whether it tracks Leo T's own HI extent or a broader Galactic component. If a compact Galactic HI clump or dust counterpart exists at v~97 (or v~53) at the cloud positions, the foreground scenario is viable and the Leo T association fails; if no Galactic HI exists at those velocities and the 53 km/s HI is confined to Leo T, the association is supported. Additionally, compute the a priori chance of finding three faint CO clumps within the 2.2' ACA FoV from the high-latitude Galactic cloud population.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that Leo T hosts the first CO detection in the lowest-mass gas-rich dwarf known depends on the three CO clumps being at d=409 kpc and not in Milky Way foreground gas. The paper's only quantitative foreground test is Appendix G, applied to the north cloud (v_los~97 km/s), which lies close to the Galactic HI velocities of 80-90 km/s toward Leo T. The test computes that if the cloud were at the far side of the MW HI disk (~40 kpc), its virial-based alphaCO would be ~1120, and compares this with the Bolatto et al. (2013) MW relation alphaCO~(LCO/1e5)^-0.185, which gives ~10 at LCO~0.4. This comparison is not decisive: the relation is calibrated on resolved Milky Way GMCs with LCO>>0.4, and the existence of CO-dark/diffuse gas in the MW demonstrates that low-LCO clouds can have very high effective alphaCO, so a value ~1000 is not physically implausible for a foreground cloud. The argument also assumes the foreground cloud is virialized. For the west/south clouds, no quantitative foreground or chance-coincidence test is presented at all; their +13 km/s offset from Leo T's HI is suggestive but not a distance proof. The abstract's 'three clouds' and the gas-expulsion interpretation depend on the north cloud, and the general detection claim rests on an unquantified association, so this is the load-bearing weakness.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports new ACA 12CO(J=1-0) observations of the dwarf galaxy Leo T and identifies three compact CO emission regions (north, west, south) within the field of view. Using 2D Gaussian fitting and circular apertures, the authors derive CO luminosities of ~32-43 K km/s pc^2, upper-limit virial masses of ~5e3 M_sun per cloud (total 1.4e4 M_sun), CO-to-H2 conversion factors of 107-156 M_sun (K km/s pc^2)^-1, and mean molecular surface densities of ~9 M_sun pc^-2. The clouds show spatial and velocity offsets from the stellar populations and HI; two clouds have velocity offsets of +13 km/s relative to the HI, while the north cloud has +57.7 km/s. The authors argue that the first two are likely bound, while the north cloud is likely being expelled from Leo T, and interpret the high alphaCO values as consistent with the galaxy's very low metallicity. The central claim is that this is the first CO detection in the lowest stellar-mass gas-rich dwarf galaxy known.","tokens_in":15646,"tokens_out":7206,"duration_ms":78702,"significance":"If the association of the CO emission with Leo T is confirmed, this would be a landmark result: it would establish cold molecular gas in a galaxy with M*~1e5 M_sun and [M/H]~-1.7, with conversion factors among the highest measured, providing a strong test of molecular cloud scaling relations in the metal-poor regime. The paper is commendably transparent about the resolution limits: it reports that the sources are barely resolved, provides a spaxel-based S/N analysis in Appendix B, and labels the derived masses and alphaCO as upper limits. These strengths make the detection claim internally coherent. However, the scientific impact of the paper depends critically on the physical association of the detected CO with Leo T, and the current treatment of the Milky Way foreground hypothesis is not sufficiently robust. The reported properties and the gas-expulsion interpretation therefore remain conditional.","major_comments":[{"comment":"The exclusion of a Milky Way foreground origin for the north cloud is not decisive. In Appendix G the authors compute that if the cloud were at the far side of the MW HI disk (D~40 kpc), its virial-based alphaCO would be ~1120, and they reject this because the Bolatto et al. (2013) relation alphaCO=(LCO/1e5)^-0.185 gives ~10 for LCO~0.4 K km/s pc^2. This comparison is not valid because the relation is calibrated on resolved, luminous Milky Way GMCs with LCO several orders of magnitude higher, and CO-dark/diffuse gas in the Milky Way demonstrates that low-luminosity CO clouds can have effective alphaCO values far above that relation, so a value near 1120 is not physically implausible for a foreground cloud. The argument also assumes virial equilibrium for the foreground cloud. Since the abstract's 'three clouds', the total molecular mass, and the gas-expulsion interpretation all assume the Leo T distance, the association remains the load-bearing uncertainty. For the west and south clouds, no quantitative foreground or chance-coincidence test is provided; their +13 km/s offset from Leo T's HI is suggestive but not a distance proof.","section":"Appendix G / Sect. 3"},{"comment":"The CO sources are not resolved: the 2D Gaussian FWHM values of ~10-14 arcsec are comparable to or smaller than the beam major axis of 13.18 arcsec, so the adopted radii, virial masses, alphaCO, and Sigma_mol are upper limits that depend on the assumed source shape. More importantly, the velocity dispersions reported in Table C.1 are 0.59-0.63 km/s with 1-sigma uncertainties of 0.62-1.04 km/s, i.e., the line-width measurements are not formally significant and are comparable to the 0.62 km/s channel width. The virial masses Mmol~5e3 M_sun are therefore not robust, and the total mass of 1.4e4 M_sun should be presented with a much stronger caveat or re-derived from a more careful spectral analysis.","section":"Sect. 3 and Table C.1"},{"comment":"The conclusion that the north cloud is being expelled from Leo T rests on the assumption that the cloud is at the Leo T distance, on a projected separation of ~60 pc, and on using only the line-of-sight velocity component. While using only the los velocity is conservative for concluding unbound orbits, the model halo-mass thresholds (Mh>2e9 M_sun for extended cores; Mh>3.5e8 M_sun for higher central densities) span the range of recent dynamical mass estimates, and the interpretation is not independent of the foreground question raised in Major Comment 1. The phrase 'providing evidence for molecular gas depletion' in the abstract is therefore stronger than the current analysis warrants; the paper should either present additional evidence against foreground contamination (e.g., a search of existing MW CO surveys toward this direction, or HI self-absorption) or soften the claim.","section":"Sect. 3 (north cloud)"}],"minor_comments":[{"comment":"Placeholder references to 'Sect.H' (e.g., in the Introduction and in Sect. 3) and the garbled passages 'WLMS ect.H', 'HST S ect.H', and 'WSRT S ect.H' should be corrected.","section":"Throughout"},{"comment":"The reference list entry 'et al., A. 2016, A&A, 588, A23' is incomplete; the first author's name is missing.","section":"References"},{"comment":"The note refers to a 2D Gaussian fit of 'MOM1 maps', but the sizes are derived from the intensity (MOM0) maps; please correct the terminology.","section":"Table C.1, note (3)"},{"comment":"The abstract quotes M*~1e5 M_sun while the text gives M*~1.4e5 M_sun; please make the values consistent.","section":"Abstract and Sect. 1"},{"comment":"The reference list entry 'Toomre, a. 1964' should be 'Toomre, A. 1964'.","section":"References"},{"comment":"The Gaussian model for the spectral fit lacks an explicit normalization factor; as written it is not a normalized probability density, although this does not affect the derived line parameters.","section":"Appendix A"}],"recommendation":"major_revision","confidential_remarks":"The manuscript appears to have been submitted with unfinished cross-references and an incomplete reference entry, which suggests incomplete proofreading; this is secondary to the scientific content but should be fixed. The main scientific risk is the foreground association; the editor may want to solicit a referee with expertise in Milky Way CO surveys. The paper is otherwise well structured and the data products are described with useful transparency."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague — the headline is: this is a genuinely new detection of CO in the faintest dwarf yet, and the west/south clouds look plausible; but the north cloud, which carries the expulsion claim, is not convincingly separated from Milky Way foreground gas.\n\nWhat's new and good: the ACA data are fresh, the spaxel S/N selection is transparent, and the authors are careful to flag most quantities as upper limits. The comparison to WLM and DDO 70 is useful, and running an orbit code rather than just asserting binding/unbinding is a thoughtful step. They also put the CO-to-H2 conversion factor in a proper metallicity context, including the DDO 154 lower limit, which is the right comparison set.\n\nSoft spots: the load-bearing issue is the association of the north cloud at v~97 km/s. Galactic HI in that direction sits at 80–90 km/s, so the paper has to exclude a foreground cloud. Their only quantitative test (Appendix G) takes a Bolatto et al. (2013) relation calibrated on bright Milky Way GMCs and extrapolates it down to LCO~0.4 K km/s pc^2. That relation was never meant for that regime, and the existence of CO-dark gas means effective alphaCO can be orders of magnitude above the relation without violating anything we know. The argument also assumes the foreground cloud is virialized. So the foreground hypothesis is not actually excluded. The west and south clouds are less suspicious—their velocities are far from the Galactic HI peak—but the paper offers no chance-coincidence estimate for them either. And with the clouds unresolved and sigma errors comparable to the values, the virial masses and alphaCO are only order-of-magnitude estimates, which the body text does say.\n\nThe abstract, by contrast, states 'three clouds' and 'likely being expelled' with more confidence than the body supports. If the referee pushes on the north cloud, the strong conclusions will need to be trimmed or the foreground option retired with better data (higher-resolution CO, HI self-absorption, or an independent distance handle).\n\nWho should read this: anyone working on CO in low-mass dwarfs or on faint-line astronomy, because the detection methodology is a good template and the foreground trap is a reminder. It deserves a real referee: the observation is new, the analysis is mostly careful, and the flaw is addressable rather than fatal. I'd send it out, but the referee should not let the abstract's strong version of the expulsion claim through without the association being fixed.","headline":"New CO detection in Leo T is plausible for the west/south clouds, but the north cloud's foreground exclusion relies on an uncalibrated alphaCO extrapolation, so the expulsion claim is not secure.","tokens_in":16243,"tokens_out":2933,"would_cite":false,"duration_ms":37605,"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":"First CO detection in the dwarf galaxy Leo T, the lowest-stellar-mass galaxy known to host cold molecular gas, reveals three compact molecular clouds, one of which appears to be escaping.","keywords":["Galaxies: Local Group","Galaxies: individual: Leo T","galaxies: dwarf","galaxies: star formation","submillimeter: galaxies","molecular clouds","CO(1-0) emission","CO-to-H2 conversion factor"],"falsifier":"Measure the distance of the northern cloud directly, for example with a CO($J=2-1$)/CO($J=1-0$) line ratio or H I self-absorption against background continuum sources. If the cloud lies in the Milky Way disk at about 40 kpc rather than at Leo T's 409 kpc, the paper's own calculation shows its luminosity would drop to $L_{\\rm CO}\\approx0.41\\ \\mathrm{K\\,km\\,s^{-1}\\,pc^2}$ and the virial-based $\\alpha_{\\rm CO}$ would rise to about $1.1\\times10^3\\,M_\\odot\\,(\\mathrm{K\\,km\\,s^{-1}\\,pc^2})^{-1}$, two orders of magnitude above Milky Way values; that would shift the interpretation from expelled Leo T gas to foreground contamination.","tokens_in":15170,"feed_emoji":"☁️","tokens_out":20728,"duration_ms":179238,"temperature":0.7,"pith_summary":"Using new observations with the Atacama Compact Array, the authors report the first detection of carbon monoxide ($^{12}$CO($J=1-0$)) in Leo T, an extremely faint and metal-poor dwarf galaxy near the Milky Way with a stellar mass of only about $1.4\\times10^5\\,M_\\odot$. They identify three compact molecular clouds whose total virial mass is about $1.4\\times10^4\\,M_\\odot$, roughly 3% of Leo T's gas budget, and derive CO-to-H$_2$ conversion factors $\\alpha_{\\rm CO}$ as high as $\\sim155\\,M_\\odot\\,(\\mathrm{K\\,km\\,s^{-1}\\,pc^2})^{-1}$, far above Milky Way values and consistent with the galaxy's very low metallicity ($[M/H]\\sim-1.7$). Two of the clouds appear gravitationally bound to the dwarf, while the northern cloud has a line-of-sight velocity offset of about $+57\\ \\mathrm{km\\,s^{-1}}$ relative to the H I gas, which the authors' orbital models say means it is likely being expelled. If the detections hold up, Leo T becomes the lowest-stellar-mass galaxy known to host cold molecular gas, pushing CO surveys to a new extreme and supporting a picture in which such dwarfs lose their star-forming gas in episodic bursts.","feed_headline":"First CO detection in Leo T, the smallest galaxy with cold gas","feed_subtitle":"Three compact molecular clouds total ~14,000 solar masses of gas; one cloud appears to be escaping.","key_machinery":"The load-bearing object is the $^{12}$CO($J=1-0$) emission line, mapped with the Atacama Compact Array in a single pointing over about 2.2 arcmin, at roughly 26 pc resolution with 0.62 km s$^{-1}$ channels and a median rms of about 20 mJy beam$^{-1}$. From the line data the authors measure each cloud's CO luminosity $L_{\\rm CO}$ using the standard Solomon & Vanden Bout (2005) formula, and its virial mass via $M_{\\rm vir}=1044\\,R\\sigma^2$ (MacLaren et al. 1988), assuming a $1/r$ density profile; the ratio of these two quantities defines the conversion factor $\\alpha_{\\rm CO}=M_{\\rm mol}/L_{\\rm CO}$, the diagnostic for extreme metal-poor conditions. The orbital analysis of the northern cloud uses the delorean code of Bla\\~na et al. (2020) with gravitational potentials for the stellar, gaseous, and dark matter components plus ram pressure from the dwarf's ISM and the Milky Way's circumgalactic medium, with parameters from Bla\\~na et al. (2024), to test whether the cloud stays inside the Jacobi radius. Comparison with the WSRT H I cube of Adams & Oosterloo (2018) supplies the velocity offsets and the atomic-gas context.","core_discovery":"The paper claims the first detection of $^{12}$CO($J=1-0$) emission in Leo T, with the Atacama Compact Array resolving three compact molecular clouds (radii of $5.7$\\,--\\,$7.1$ arcsec, equivalent to less than 13 pc at Leo T's distance of 409 kpc) located in the central region but offset by roughly 60\\,--\\,100 pc from the centers of the stellar populations. Adopting the virial mass estimator with a $1/r$ density profile, each cloud carries an upper-limit mass of about $5\\times10^3\\,M_\\odot$, for a total of $1.4\\pm0.4\\times10^4\\,M_\\odot$, corresponding to about 3% of the galaxy's total gas mass. The derived CO-to-H$_2$ conversion factors $\\alpha_{\\rm CO}=M_{\\rm mol}/L_{\\rm CO}$ \\,--\\, the ratio of molecular mass to CO luminosity \\,--\\, fall in the range $107$\\,--\\,$156\\,M_\\odot\\,(\\mathrm{K\\,km\\,s^{-1}\\,pc^2})^{-1}$, among the highest measured and presented as consistent with the extremely low metallicity of Leo T. The south and west clouds have velocity offsets near $+13\\ \\mathrm{km\\,s^{-1}}$ relative to the H I and are probably bound; the north cloud, at about $+58\\ \\mathrm{km\\,s^{-1}}$, is unbound in orbital models with halo masses below roughly $10^9\\,M_\\odot$, from which the authors conclude the cloud is likely being expelled and that Leo T is experiencing molecular gas depletion as it evolves toward quenching.","pith_inferences":["A cleaner arbitration of the foreground-cloud question for the north cloud would be a distance-sensitive measurement such as H I self-absorption against Milky Way background continuum or a CO($J=2-1$) line ratio; the paper's exclusion relies on the virial estimator holding at both candidate distances.","If the clouds are genuinely bound to Leo T, the combination of a large H I reservoir with inefficient CO formation suggests molecule formation is triggered by localized compression (for example AGB winds or dynamical perturbations) rather than by the global gas surface density, which sits below the usual critical threshold of about $10\\,M_\\odot\\,\\mathrm{pc^{-2}}$.","Because Leo T shows no detected massive stars, H II regions, or dust, these clouds probe the $\\alpha_{\\rm CO}$\\,--\\,metallicity relation at nearly zero ambient radiation field; comparing them with dwarfs of similar metallicity but active star formation would isolate the role of photodissociation in setting $\\alpha_{\\rm CO}$.","Higher-resolution CO($J=2-1$) plus dust-continuum observations that resolve the clouds' cores could turn the upper-limit masses into direct measurements and test whether the clouds are gravitationally self-bound or transient, pressure-confined structures."],"forward_implications":["Leo T becomes the lowest-stellar-mass galaxy known to host cold molecular gas, extending the frontier of CO detection in dwarf galaxies down to $M_\\star\\approx10^5\\,M_\\odot$.","The extreme conversion factors ($\\alpha_{\\rm CO}=107$\\,--\\,$156$ in the paper's units) imply that CO strongly under-tracers H$_2$ in very metal-poor systems, so the true molecular gas content of Leo T could be far larger than the CO-derived mass.","If the north cloud is genuinely unbound, it is direct evidence that low-mass dwarfs expel molecular gas, supporting the episodic star formation and eventual quenching scenario for Leo T.","The spatial offsets between the CO clouds and the stellar and H I centers indicate that molecular gas forms away from the galaxy's dynamical center, possibly in H I compressed by stellar winds.","The detection shows CO($J=1-0$) is a viable tracer in the lowest-mass dwarfs, motivating systematic surveys of the molecular ISM at the faint end of the galaxy population."],"supporting_citations":[{"why":"Supplies the WSRT H I data cube, gas mass, line-of-sight velocity, and velocity dispersion of Leo T against which the CO clouds' offsets and total gas fraction are measured.","marker":"Adams & Oosterloo (2018)"},{"why":"The WLM molecular cloud study whose cloud luminosities, virial masses, and $\\alpha_{\\rm CO}$ values form the main comparison sample for the Leo T detections.","marker":"Rubio et al. (2015)"},{"why":"Provides the $\\alpha_{\\rm CO}$\\,--\\,metallicity framework and the Milky Way $\\alpha_{\\rm CO}$ reference, including the empirical relation used to argue the north cloud is not foreground gas.","marker":"Bolatto et al. (2013)"},{"why":"Supplies the virial mass formula $M_{\\rm vir}=1044\\,R\\sigma^2$ with a $1/r$ density profile used to derive all cloud masses.","marker":"MacLaren et al. (1988)"},{"why":"Provides the CO luminosity equation used to convert integrated line fluxes into $L_{\\rm CO}$.","marker":"Solomon & Vanden Bout (2005)"},{"why":"The delorean orbital-modeling code used to test whether the north CO cloud is gravitationally bound to Leo T.","marker":"Blaña et al. (2020) (B20)"},{"why":"Supplies the dark-matter halo mass models, ram-pressure parameters, and AGB-wind perturbation scenario used in the orbital and ISM analysis.","marker":"Blaña et al. (2024) (B24)"},{"why":"Provides Leo T's stellar mass, stellar population ages and centers, and the star formation history used for comparison.","marker":"Weisz et al. (2012)"},{"why":"Supplies the distance to Leo T that converts angular sizes to parsecs and enters the CO luminosity calculation.","marker":"Clementini et al. (2012)"},{"why":"The HI4PI survey of Galactic H I used to evaluate the foreground Milky Way cloud alternative for the north CO cloud.","marker":"Westmeier (2018)"}],"fun_headline_variants":["First CO detection in Leo T, the smallest galaxy with cold gas","Leo T sets record: smallest galaxy with CO molecules","Tiny Leo T hosts CO, one cloud may be escaping","Smallest galaxy with CO: Leo T's molecular clouds","CO found in Leo T, a dwarf with a possible gas outflow"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim stands or falls on the assumption that the detected CO emission, especially the northern cloud at about 97 km s$^{-1}$, truly belongs to Leo T and not to a foreground Milky Way molecular cloud; Milky Way H I toward Leo T moves at 80\\,--\\,90 km s$^{-1}$, and the paper's argument against that scenario is indirect, resting on the virial mass estimator being valid at both candidate distances.","fun_headline_variants_meta":{"raw":{"variants":["First CO detection in Leo T, the smallest galaxy with cold gas","Leo T sets record: smallest galaxy with CO molecules","Tiny Leo T hosts CO, one cloud may be escaping","Smallest galaxy with CO: Leo T's molecular clouds","CO found in Leo T, a dwarf with a possible gas outflow"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000312,"raw_usage":{"total_tokens":1991,"prompt_tokens":1376,"completion_tokens":615,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":992,"completion_tokens_details":{"reasoning_tokens":531}},"tokens_in":992,"tokens_out":615,"duration_ms":6888,"temperature":1.0,"reasoning_tokens":531,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T12:59:44.492813+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the distance of the northern cloud directly, for example with a CO($J=2-1$)/CO($J=1-0$) line ratio or H I self-absorption against background continuum sources. If the cloud lies in the Milky Way disk at about 40 kpc rather than at Leo T's 409 kpc, the paper's own calculation shows its luminosity would drop to $L_{\\rm CO}\\approx0.41\\ \\mathrm{K\\,km\\,s^{-1}\\,pc^2}$ and the virial-based $\\alpha_{\\rm CO}$ would rise to about $1.1\\times10^3\\,M_\\odot\\,(\\mathrm{K\\,km\\,s^{-1}\\,pc^2})^{-1}$, two orders of magnitude above Milky Way values; that would shift the interpretation from expelled Leo T gas to foreground contamination.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the WSRT H I data cube, gas mass, line-of-sight velocity, and velocity dispersion of Leo T against which the CO clouds' offsets and total gas fraction are measured."},{"cited_title":"G., Hunter, D","cited_arxiv_id":null,"evidence_quote":"The WLM molecular cloud study whose cloud luminosities, virial masses, and $\\alpha_{\\rm CO}$ values form the main comparison sample for the Leo T detections."},{"cited_title":"D., Wolfire, M., & Leroy, A","cited_arxiv_id":null,"evidence_quote":"Provides the $\\alpha_{\\rm CO}$\\,--\\,metallicity framework and the Milky Way $\\alpha_{\\rm CO}$ reference, including the empirical relation used to argue the north cloud is not foreground gas."},{"cited_title":"M., & Wolfendale , A","cited_arxiv_id":null,"evidence_quote":"Supplies the virial mass formula $M_{\\rm vir}=1044\\,R\\sigma^2$ with a $1/r$ density profile used to derive all cloud masses."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the CO luminosity equation used to convert integrated line fluxes into $L_{\\rm CO}$."},{"cited_title":"R., Zucker, D","cited_arxiv_id":null,"evidence_quote":"Provides Leo T's stellar mass, stellar population ages and centers, and the star formation history used for comparison."},{"cited_title":"C., et al","cited_arxiv_id":null,"evidence_quote":"Supplies the distance to Leo T that converts angular sizes to parsecs and enters the CO luminosity calculation."},{"cited_title":"2018, , 474, 289","cited_arxiv_id":null,"evidence_quote":"The HI4PI survey of Galactic H I used to evaluate the foreground Milky Way cloud alternative for the north CO cloud."}],"review_version":1}