{"id":"9e58d62d-955d-47ea-9238-086fba00df3c","arxiv_id":"2502.00117","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A spectroscopic survey of 44 ultra-diffuse galaxy candidates finds 7 post-starburst systems, including 3 isolated ones, indicating diverse quenching pathways.","lead":"Using Keck spectroscopy, astronomers found that about 20% of the ultra-diffuse galaxies they observed recently stopped forming stars, showing a young stellar population but little ongoing star formation. Several of these 'caught in the act' systems are isolated, suggesting quenching can happen without a massive neighbor galaxy.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 20% post-starburst fraction hinges on an unvalidated two-template spectral classification; if template mismatch biases A/K or the K+A subtraction removes Hbeta emission, several of the seven systems could be star-forming dwarfs, and the SDSS check does not cover UDG masses.","rationale":"The reader's weakest assumption correctly identifies the post-starburst classification as the load-bearing element. The paper is otherwise well structured: the UDG confirmation via KCWI redshifts is solid, the sample selection is transparent, and the environmental analysis is careful about projected distances and backsplash arguments. The central existence claim—that some UDGs show post-starburst spectral features—is plausible and consistent with standard techniques, which supports the conditional accept. However, the specific two-template fitting procedure introduces a systematic risk that is not checked at the relevant stellar masses. The SDSS sanity check at 1e9–1e11 Msun is the natural validation attempt, but it does not cover the UDG mass regime where spectral signal-to-noise, stellar population properties, and dust attenuation may differ. The proposed pPXF/Prospector re-fit is a concrete, feasible check that would settle whether the 7/35 classification is an artifact of the template choice. Since the reader already set the verdict to CONDITIONAL, my stress test does not change that verdict; it reinforces the need for the requested verification before the isolated-quenching conclusion is accepted as definitive.","tokens_in":17661,"tokens_out":4736,"duration_ms":58333,"concrete_test":"Re-fit all 42 KCWI spectra using pPXF with the full MILES stellar library (or Prospector with a non-parametric SFH), fitting stellar continuum and emission lines simultaneously, then recompute the analogous A/K ratio and the Hbeta emission EW after subtracting the best-fit stellar model. If fewer than 7 of the 35 UDGs still satisfy the Section 3.2 criteria, the claimed post-starburst population is not robust to template choice and the central conclusion would need to be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that 7/35 confirmed UDGs are post-starburst, including 3 isolated systems—rests entirely on the Section 2.4 spectral fitting and selection criteria in Section 3.2. The method fits a linear combination of a single A7 star and a single K0/K1 star to continuum-subtracted spectra, then measures Hbeta EW after subtracting that same best-fit template. This is the load-bearing risk: the two templates are single points in metallicity and age, the fit is unregularized, and no statistical uncertainties are propagated to A/K or Hbeta EW. If the true stellar populations are lower-metallicity or composite, the fitted A/K can be biased upward; if the template subtraction overcorrects the Balmer absorption profile, real Hbeta emission can be removed. Either effect would shift star-forming dwarfs into the post-starburst region defined by A/K > 0.3 and Hbeta EW < (A/K) + 0.5. The SDSS validation sample (Section 2.5) has stellar masses 1e9–1e11 Msun, well above the UDG masses of ~1e7–1e9 Msun, so the ~1% post-starburst fraction there does not validate the selection at the masses and signal-to-noise of interest. Because the environment analysis and the 'isolated quenching' conclusion are built entirely on the identity of these seven objects, an unverified classification bias is the most load-bearing concern.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"Sandoval Ascencio et al. present KCWI spectroscopy of 44 UDG candidates from SMUDGes, confirming 35 as UDGs. Fitting a linear combination of A7 and K0/K1 stellar templates to continuum-subtracted spectra and measuring Hbeta and [OII] equivalent widths, they identify 7 post-starburst UDGs, or ~20% of the confirmed UDG sample. They further find that 3 of these 7 are isolated (or at large projected distance from massive hosts) and argue that these systems demonstrate quenching mechanisms independent of environment. The paper frames this as evidence for diverse quenching pathways in UDGs.","tokens_in":17954,"tokens_out":8409,"duration_ms":78967,"significance":"If the classification is correct, this is a valuable discovery: post-starburst UDGs are rare and provide a direct view of recent quenching, and the isolated examples would challenge purely environmental quenching models. The paper's strengths are its use of a standard K+A classification approach, presentation of the spectra (Figure 5 indeed shows deep Balmer absorption), and a documented comparison to SDSS. However, the quantitative claims—the 20% fraction and the isolation of 3 systems—rest on a two-template decomposition whose uncertainties are not quantified at the masses and signal-to-noise of the UDG sample, and the SDSS validation is at much higher stellar masses. The conclusions are plausible but need additional robustness tests before the population fraction and isolation counts can be fully accepted.","major_comments":[{"comment":"The A/K ratio and Hbeta EW measurements are presented without statistical uncertainties. The only robustness test described is variation of the extraction window by ±20%, which tests aperture choice but not template mismatch, continuum placement, noise, or the two-population simplification. Given the selection boundary A/K > 0.3 and Hbeta EW < A/K + 0.5, the quoted ~20% scatter in A/K and ~25% in Hbeta EW could move systems across the selection cut. Please provide bootstrap or Monte Carlo uncertainties that include alternative stellar templates (the text says alternative templates give 'qualitatively similar' results, but no quantitative comparison is shown) and show error bars in Figure 4. This is load-bearing because the count of 7/35 depends on the exact placement of these measurements relative to the selection.","section":"§2.4; Fig. 4"},{"comment":"The SDSS comparison sample used to validate the post-starburst selection spans stellar masses 10^9–10^11 Msun, whereas the UDGs studied here have masses ~10^7.5–10^8.5 Msun. The paper itself cites literature showing that the post-starburst fraction increases toward lower masses, so the ~1% fraction measured in the SDSS sample does not demonstrate that the selection is clean at the masses or signal-to-noise of the UDG sample. A low-mass comparison sample, or synthetic spectra with realistic S/N and metallicity, is needed to show that star-forming low-mass galaxies do not scatter into the post-starburst region defined by A/K > 0.3 and Hbeta EW < A/K + 0.5.","section":"§2.5"},{"comment":"The central fraction of 20% (7/35) is quoted without a statistical uncertainty and without a discussion of selection effects. With only seven objects, the Poisson 1-sigma uncertainty is roughly ±7 percentage points, so the 20% is not strongly distinguished from, e.g., 10% or 30%. In addition, the sample is not volume-limited: it is selected by g < 20, R_e > 5 arcsec, and mu_0,g < 25 mag arcsec^-2, and it excludes the Coma and Virgo cluster regions. Please provide a binomial or Poisson confidence interval for the fraction and clarify how the selection biases (e.g., the brightness/size bias visible in Fig. 1) affect the population-level claim.","section":"§4.1 and Abstract"},{"comment":"The environmental classification of SMDG0224231-031059 is contradictory. The text first states that the galaxy is 'likely unassociated' with the NGC 936 group (projected distance 0.7 Mpc, group virial radius <400 kpc), but later describes it as 'possibly a backsplash galaxy' that was recently quenched by the environment of NGC 936. Because the number of isolated post-starburst UDGs (3 of 7) is a central result, this contradiction must be resolved. The isolation analysis also relies on projected distances to massive galaxies in the NSA; please state explicitly whether the three 'isolated' systems lie within the NSA footprint and discuss the impact of NSA incompleteness (four KCWI targets were already excluded near the NSA edge).","section":"§4.2"},{"comment":"The reduction from 13 post-starburst candidates to 7 post-starburst UDGs is ambiguous. The text states that 8 of the 13 candidates are UDGs, then says that the [OII] < 6 Å cut removes one candidate, yielding 7 post-starburst UDGs, but it is not explicitly stated that the removed source is one of the 8 UDG candidates. The gold cross in Figure 4 should be identified in Table 1 or the text, and the denominator (35 confirmed UDGs) should be clearly distinguished from the 42 secure-redshift sources. Clarifying this step is necessary for the reader to verify the 20% figure.","section":"§3.2"}],"minor_comments":[{"comment":"The fraction is given as '20%' in the abstract and '~20% (7/35)' in the text; please standardize the notation.","section":"Abstract and §4.1"},{"comment":"35/42 is 83%, not ~85%; adjust the stated confirmation fraction or round explicitly.","section":"§3.1"},{"comment":"The median errors shown in red are for the SDSS sample only; if KCWI error bars are omitted for clarity, state this in the caption or add them.","section":"Figure 4 caption"},{"comment":"The adopted velocity width of 75 km/s appears smaller than the KCWI instrumental resolution at R~1800 (c/R ~ 170 km/s at 4500 Å); clarify whether this is the intrinsic dispersion or a typo.","section":"§2.4"},{"comment":"Table 1 does not include the measured A/K, Hbeta EW, and [OII] EW values; adding these columns would make the classification transparent and reproducible.","section":"Table 1"},{"comment":"The statement that the target population was 'largely unbiased' relative to SMUDGes is made in the context of galaxy color; Figure 1 shows a clear brightness/size bias, so please restrict the unbiased claim to color or add a caveat.","section":"§4.1"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the journal's scope. The main technical risk is the unquantified spectral classification; I believe the authors can address it with additional robustness tests and error estimates, so I recommend major revision rather than rejection. No concerns about novelty or citation pattern beyond the usual need to compare with low-mass post-starburst fractions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is the first spectroscopically confirmed population of post-starburst UDGs, and the three isolated ones are the headline. The classifications look credible to me—the Figure 5 spectra show deep Balmer absorption and no strong emission—and the method is standard, so the central existence claim should hold up.\n\nWhat's new: earlier work had individual quenched/passive UDGs in the field (Papastergis 2017; Román 2019), but this paper turns that into a population: 7/35 confirmed UDGs, with 4 in groups/clusters and 3 at >1.5 Mpc from any massive host. The environmental split is the genuinely interesting result. The SDSS sanity check giving ~1% is a useful guard against an over-inclusive selection.\n\nNow the soft spots, roughly in order. First, no uncertainties on A/K and Hβ EW. The extraction-window scatter (~20–25%) is not a measurement error, and template mismatch could shift objects across the cuts. Without error bars, I can't tell if any of the seven are borderline. Second, the SDSS validation sample is at 1e9–1e11 Msun, well above UDG masses; the criteria are from the literature and look fine, but the ~1% baseline is not a direct calibration at the masses and S/N of interest. Third, the 20% fraction is for a brightness/size-selected subsample, not a volume-limited census; they acknowledge the bias toward bright, large sources, but readers should not take 20% as a universal UDG fraction. Fourth, the 'absence of environmental influence' for the isolated cases rests on projected distances and back-splash arguments; two at 2–3 Rvir is suggestive but not definitive. With seven objects, Poisson uncertainty is large, and the isolated subset could plausibly be 1 or 5.\n\nThe stress-test worry about the two-template fit removing Hβ emission is worth taking seriously but I don't think it sinks the paper. The spectra look like K+A systems, and the authors say alternative templates give similar results. I'd want the authors to release the full measurements with uncertainties and show the selection in an error-barred A/K–EW plane.\n\nBottom line: this deserves a serious referee. It's a small-sample discovery paper; the right referee will push on the error bars and the environment catalog. I'd engage with it and cite it if I worked on dwarf quenching.","headline":"First spectroscopic population of post-starburst UDGs; the isolated members are intriguing, but missing error bars and a small sample keep the 20% fraction at the teaser level.","tokens_in":18575,"tokens_out":4194,"would_cite":true,"duration_ms":40806,"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 reports that roughly one in five spectroscopically confirmed ultra-diffuse galaxies is a post-starburst system, and that three of these seven systems are isolated, implying UDGs can quench without a dense environment.","keywords":["ultra-diffuse galaxies","post-starburst galaxies","galaxy quenching","dwarf galaxies","integral field spectroscopy","SMUDGes","KCWI","galaxy evolution"],"falsifier":"Take the seven candidate galaxies and look for Hα emission or ultraviolet light; if strong Hα or bright UV is detected in the three isolated systems, star formation has not actually stopped and the post-starburst classification fails. A second check is to refit the same spectra with full stellar-population models that include dust and varied metallicity; if the best-fit A-star fractions drop below the adopted threshold, the identified population disappears.","tokens_in":17437,"feed_emoji":"🔭","tokens_out":7195,"duration_ms":65225,"temperature":0.7,"pith_summary":"Using Keck Cosmic Web Imager spectra of 44 candidate ultra-diffuse galaxies from the SMUDGes survey, the authors show that about 20% of the 35 confirmed UDGs display post-starburst signatures: strong Balmer absorption from recently formed A-type stars with little or no Hβ emission. Three of these seven systems sit more than 1.5 Mpc in projection from any massive galaxy, beyond the reach of likely backsplash or tidal quenching. The paper argues this is evidence that ultra-diffuse galaxies can quench without environmental assistance, most plausibly through star-formation feedback after a recent burst. If right, local UDGs do not follow a single quenching channel, and isolated quenched UDGs are not forbidden by current formation models.","feed_headline":"20% of ultra-diffuse galaxies are caught quenching","feed_subtitle":"Spectra catch seven ultra-diffuse galaxies right after quenching; three sit isolated, beyond any environmental explanation.","key_machinery":"The central object is the post-starburst selection in the (A/K, Hβ EW) plane. A/K is the ratio of the best-fit A-type star to K-type star contribution in a linear combination of one K0/K1 and one A7 MILES template, fit to the continuum-subtracted KCWI spectrum over 3700–5500 Å. Galaxies with A/K > 0.3, negligible Hβ emission, and weak [OII] form the post-starburst spur: they have recent star formation (A stars) but no ongoing star formation. This two-population decomposition, inherited from classic E+A searches, is what lets the paper catch the short-lived transition phase; the environment analysis then measures projected distances to massive galaxies to test whether the quenching can be blamed on a host.","core_discovery":"On the paper's own terms, the discovery is a population of seven post-starburst ultra-diffuse galaxies in the local universe, identified from Keck/KCWI spectra of 44 SMUDGes candidates. After fitting a linear combination of A7 and K0/K1 stellar templates to continuum-subtracted spectra, the authors select galaxies with A/K > 0.3, Hβ EW < A/K + 0.5 Å, and [OII] EW < 6 Å; seven of the 35 confirmed UDGs meet these criteria. These galaxies show deep Balmer absorption from young A stars but little or no ongoing star formation, the spectroscopic signature of a galaxy caught shortly after quenching. Four are plausibly satellites of massive groups or clusters, but three lie more than 1.5 Mpc in projection from any massive galaxy, and two of those are more than 2–3 virial radii from any potential host. The paper reads these three as evidence that quenching can occur in isolation, likely through star-formation feedback after a recent burst.","pith_inferences":["An editor-level extension: the single isolated post-starburst with residual [OII]/Hβ emission (SMDG0015208+200027) is a natural target for deep Hα or HI follow-up to test whether quenching is complete or only a lull.","If the same selection is applied to a larger spectroscopic sample of SMUDGes UDGs, a 20% post-starburst fraction would imply UDGs cycle through bursts and quiescence more violently than normal dwarfs, a claim the present sample is too small to confirm.","A testable prediction follows: isolated post-starburst UDGs should show disturbed or extended stellar morphologies from recent feedback, and their gas reservoirs may be partially retained, so HI mapping could distinguish feedback-driven quenching from gas exhaustion.","Applying the same A/K and Hβ method to intermediate-redshift UDGs would show whether this isolated quenching channel was more common in the past, when gas fractions and burst rates were higher."],"forward_implications":["If the ~20% fraction holds for the broader SMUDGes UDG population, local UDGs are not simply a mix of star-forming field dwarfs and quenched cluster dwarfs; a substantial fraction are caught mid-transition.","The three isolated post-starburst UDGs imply that at least some UDG quenching does not require ram-pressure stripping, tidal stripping, or backsplash from a massive host.","Because the A-star phase lasts only of order a gigayear, finding 20% of confirmed UDGs in this phase implies a high rate of recent quenching, or of recycling between star-forming and quenched states, in UDGs.","The result supports feedback-driven size growth and quenching in low-mass diffuse galaxies, since a recent burst can both puff up the stellar distribution and temporarily shut off star formation."],"supporting_citations":[{"why":"Defines the SMUDGes candidate UDG catalog from which the 44 KCWI targets were drawn.","marker":"Zaritsky et al. (2023)"},{"why":"Supplies the MILES K0/K1 and A7 stellar templates used in the A/K spectral decomposition.","marker":"Sánchez-Blázquez et al. (2006)"},{"why":"Establishes the Hβ emission measurement after K+A template subtraction and the post-starburst selection approach.","marker":"Quintero et al. (2004)"},{"why":"Provides the classic local E+A identification criteria and the low post-starburst fraction baseline.","marker":"Zabludoff et al. (1996)"},{"why":"Documents the ~1% post-starburst fraction in SDSS used as the comparison baseline.","marker":"Goto et al. (2003)"},{"why":"Shows that isolated UDGs are typically star-forming, making the isolated quenched systems anomalous.","marker":"Leisman et al. (2017)"},{"why":"Provides low-density UDG population data used to frame the environmental contrast.","marker":"Rong et al. (2020)"},{"why":"Supplies the NASA-Sloan Atlas from which massive-neighbor distances and environments are measured.","marker":"Blanton et al. (2011)"}],"fun_headline_variants":["Seven UDGs caught in the act of quenching","Post-starburst UDGs sit isolated, quench alone","Keck finds UDGs that just stopped forming stars","Ultra-diffuse galaxies quench without help"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything rests on the assumption that fitting a galaxy's spectrum with one young A-type star and one old K-type star, after removing the continuum, truly measures how many A stars are present without deleting real hydrogen emission from the data.","fun_headline_variants_meta":{"raw":{"variants":["Seven UDGs caught in the act of quenching","Post-starburst UDGs sit isolated, quench alone","Keck finds UDGs that just stopped forming stars","Ultra-diffuse galaxies quench without help"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000177,"raw_usage":{"total_tokens":1353,"prompt_tokens":1068,"completion_tokens":285,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":684,"completion_tokens_details":{"reasoning_tokens":218}},"tokens_in":684,"tokens_out":285,"duration_ms":3597,"temperature":1.0,"reasoning_tokens":218,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T20:07:21.226820+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the seven candidate galaxies and look for Hα emission or ultraviolet light; if strong Hα or bright UV is detected in the three isolated systems, star formation has not actually stopped and the post-starburst classification fails. A second check is to refit the same spectra with full stellar-population models that include dust and varied metallicity; if the best-fit A-star fractions drop below the adopted threshold, the identified population disappears.","supporting_citations":[],"review_version":1}