{"id":"588f15ba-190c-4a08-adf7-7fa0c319051a","arxiv_id":"2507.21231","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"In the Romulus25 simulation, classical low surface brightness galaxies form mainly through major mergers with aligned, co-rotating orbits that spin up and puff out the remnant galaxy.","lead":"This paper analyzes a large cosmological simulation and finds that low surface brightness galaxies, which are faint and spread out, are mostly produced when two galaxies merge in a special configuration, with the smaller galaxy orbiting in the same direction as the larger one's gas disk. The result matters because these dim galaxies are numerous and upcoming sky surveys should find many of them, so a formation mechanism is needed to interpret what they are.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Unsupported '>80% of stars formed after last major merger' statistic (Sec. 3.3.2) is inconsistent with the reported median merger lookback time and near-constant SFHs; the causal spin-up imprint claim is not established.","rationale":"The central causal claim requires that the last major merger's spin-up actually reshapes the stellar distribution. The only explicit bridge is the >80% statistic, which is unbacked and conflicts with the paper's own median LMM lookback time (9.52 Gyr for bulge+disk LSBs, Sec. 3.3.2) and the near-constant SFHs in Figure 6. Under constant SF, roughly a third of stars form after 9.5 Gyr, so the statistic appears to be an error or a misstatement. Without it, the post-merger spin-up acts on a minority stellar component; the present-day LSB structure could be a pre-merger property, and the orientation differences could be selection effects rather than causes. This is an internal consistency issue rather than a disagreement with outside consensus. The paper still has substantive evidence in Figure 16's evolutionary tracks and the KS-test differences in Figure 17, so rejection is not warranted. However, the concern reinforces the reader's conditional verdict: the authors should quantify merger fractions, make the 80% statistic reproducible, and provide access to the data needed for the test. The subgrid-resolution concern raised by the reader is real but secondary; the 80% claim is sharper and directly testable from existing simulation outputs.","tokens_in":41553,"tokens_out":6949,"duration_ms":84704,"concrete_test":"Using the same tangos merger trees and LMM definition (virial radii first overlap) used in Section 3.3.2, compute for each LSB and HSB galaxy the fraction f_post = M* formed after the last major merger divided by the z=0 stellar mass, and report medians per subsample. If medians are below 0.5, the '>80%' claim is false; additionally, repeat the Figure 16 central surface brightness tracks using only stars formed after LMM to test whether post-merger spin-up alone produces the low central surface brightness, or whether pre-existing old stars dominate.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3.3.2 argues that merger-driven spin-up permanently shapes LSB galaxies, relying on the sentence: 'Because more than 80% of the median galaxy’s stars are formed during or after this final major merger, the alteration in gas spin leaves a permanent imprint on the distribution of the galaxy’s stars.' This statistic is not shown in any figure or table, and it appears internally inconsistent with values given in the same section: bulge+disk LSB galaxies have a median last-major-merger lookback time of 9.52 Gyr (z≈1.5), and Figure 6 shows LSB star formation histories that are near-constant over a Hubble time. A constant SFR since early times forms only ~30% of z=0 stars after lookback 9.5 Gyr, not >80%. If the true fraction is much lower, the present-day low central surface brightness is dominated by stars that formed before the last major merger, so the spin-up of the post-merger gas cannot be the cause of the galaxy's LSB classification. The merger-orientation differences in Figure 17 could then be incidental rather than causal.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses the Romulus25 cosmological simulation to identify a sample of 326 classical low surface brightness (LSB) galaxies among 511 late-type central galaxies, split into irregular, pure disk, and bulge+disk categories. These are compared with mass-matched late-type high surface brightness (HSB) galaxies. The authors report that LSB galaxies are HI-rich, metal-poor, spatially extended, and have near-constant star formation histories, with older and redder bulges. They argue that classical LSB galaxies are predominantly produced by major mergers in which the secondary is co-rotating and aligned with the primary's gas disk and/or has high orbital angular momentum, causing merger-driven spin-up that makes star formation more extended and inefficient. The paper also claims that classical LSB galaxies constitute about 60% of all central galaxies in the quoted stellar mass range.","tokens_in":41794,"tokens_out":5082,"duration_ms":67706,"significance":"If the central claim holds, the paper provides a coherent and testable formation channel for classical LSB galaxies that goes beyond the traditional high-spin halo scenario and connects to earlier merger-driven explanations for ultra-diffuse galaxies. The analysis is based on a large, uniformly selected simulated sample, uses a selection procedure that mimics observational surveys, and includes careful mass-matching and statistical comparisons with several observed samples. The authors are also commendably explicit about known simulation limitations, such as the overall HI deficit and resolution effects on star formation. However, because the causal interpretation rests on a small number of specific quantitative claims, those claims need to be supported or corrected before the paper can be accepted.","major_comments":[{"comment":"The abstract and Section 3.1 state that classical LSB galaxies account for approximately 60% of all central galaxies with 8 <= log10(Mstar/Msun) <= 10, but the sample is explicitly restricted to late-type central galaxies (M_HI >= 10^6.5 Msun and B-V < 0.72). As written, the denominator in this fraction is ambiguous and likely inconsistent: the 511-galaxy parent sample contains only late-type centrals, so the 60% figure should refer to late-type central galaxies, not all central galaxies. Please revise the text, abstract, and Figure 2 caption to state the correct denominator, or recompute the fraction using all central galaxies if that is the intended claim.","section":"Abstract; Section 3.1; Section 4"},{"comment":"The sentence \"Because more than 80% of the median galaxy's stars are formed during or after this final major merger\" is a load-bearing step in the argument that merger-driven spin-up leaves a permanent imprint on the galaxy's stellar distribution, but no figure, table, or calculation supporting this statistic is provided. Given that the same section reports a median last-major-merger lookback time of 9.52 Gyr for bulge+disk LSB galaxies and that Figure 6 shows near-constant star formation histories, the 80% value is not self-evident and may depend on the assumed assembly epoch. Please add a quantitative derivation or a figure showing the distribution of the fraction of stars formed after the last major merger for each subsample, and use that to justify the causal statement.","section":"Section 3.3.2"},{"comment":"The merger orientation and orbital angular momentum analyses in Figure 17 are central to the claim that co-rotating, high-angular-momentum mergers produce LSB galaxies, but the paper does not state how galaxies without any major merger are treated. If such galaxies are excluded from the figure, the comparison between LSB and HSB galaxies could be biased; if they are included, it is unclear what value of phi_orb or orbital angular momentum they are assigned. Please report the number of galaxies contributing to each panel and explicitly describe the treatment of galaxies with no major merger.","section":"Section 3.3.2; Figure 17"},{"comment":"The authors note that the simulation resolution boosts star formation in low-density gas, which is precisely the regime that determines whether a galaxy is classified as LSB. This raises the question of whether the extended star formation and faint central surface brightness of LSB galaxies in Romulus25 are caused by merger-driven spin-up or are partly artifacts of the subgrid star formation prescription. Since the causal claim in Section 3.3.2 depends on this distinction, the paper should include a resolution or subgrid-parameter test, or at least an explicit quantitative argument that the merger-orientation signal is not driven by the resolution-dependent SF threshold.","section":"Section 3.2.2; Section 3.3.2"}],"minor_comments":[{"comment":"The resolution cut Mvir < 3e9 Msun is described as corresponding to ~10,000 dark matter particles, but with MDM,part = 3.39e5 Msun the number is closer to 8,850; please correct this estimate.","section":"Section 3.1"},{"comment":"The text says that a perfectly aligned co-rotating merger \"has phi_orb = 1\" and a counter-rotating merger \"has phi_orb = -1\", but Eq. (5) defines phi_orb as an angle; it is cos(phi_orb) that takes these values. Please fix the notation throughout the paragraph.","section":"Section 3.3.2"},{"comment":"The phrase \"(Alejandra in prep.)\" is informal for a journal submission; please replace it with a proper citation or remove it.","section":"Section 3.3.1"},{"comment":"The caption does not state whether the distributions are normalized or how the KS test sample sizes were determined; adding the number of galaxies and the normalization convention would improve reproducibility.","section":"Figure 17"},{"comment":"The statement that the LSB versus HSB metallicity differences are significant uses p < 0.02 for all subsamples, but the pure disk difference of 0.19 dex would be more informative with confidence intervals on the median difference.","section":"Section 3.2.3"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a topic well within the scope of astrophysics journals and the core merger-orientation result is interesting. The main issues are internal consistency of the 60% population fraction and missing support for the 80% post-merger star formation statistic; both should be resolvable with a focused revision. The resolution-dependence concern is more serious but can be addressed with additional analysis or a clearly scoped caveat. I do not see grounds for rejection, but the current text overstates some conclusions relative to what is demonstrated."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is the strongest simulation-based case I've seen that merger-driven spin-up builds classical LSB galaxies at masses up to 10^10–10^11 Msun. The paper deserves a serious referee. But there are two things to fix before acceptance: the unsupported '>80% of stars' claim in Section 3.3.2, and the ambiguous '60% of all central galaxies' framing.\n\nWhat's new: previous work (Di Cintio 2019, Wright 2021) had the co-rotating merger idea, but mostly in dwarfs or zoom-ins. This paper shows it in a uniform cosmological volume, with a mass-matched HSB control, and extends it to bulge+disk systems. The merger-orientation distributions are statistically distinct (KS p < 1e-6 for pure disks) and the higher orbital angular momentum at infall is a genuinely new ingredient. The HI, metallicity, and color trends are all consistent with observations, and the authors are honest about the simulation being HI-poor and the resolution boosting SFRs in low-density gas.\n\nSoft spots: the 'more than 80% of the median galaxy's stars are formed during or after this final major merger' sentence is a load-bearing claim for the 'permanent imprint' argument, but it appears nowhere in the figures or tables. The math in the stress-test note looks right: with a median lookback time of 9.5 Gyr and near-constant SFH, you'd expect roughly 60–70%, not >80%. That overstatement weakens the causal claim. It's not fatal—the post-merger evolutionary tracks in Figure 16 do show sustained spin-up and outward star formation—but the authors need to show that statistic. Also, the abstract says '60% of all central galaxies,' but the sample is selected to be late-type (HI mass and color cuts). If the denominator is really all central galaxies, that's misleading; if it's late-type centrals, say so. Finally, the simulation's known HI deficiency and SFR boosting at low density are acknowledged, but they should be discussed as possible biases in the LSB classification itself.\n\nBottom line: this is a serious, citable paper for anyone working on LSB galaxies or merger-driven disk formation. Send it to referees, but insist on the missing statistics and denominator clarity.","headline":"Strong simulation-based case for merger-driven LSB formation, but the '>80% of stars' statistic is unsupported and the sample denominator needs tightening.","tokens_in":42349,"tokens_out":2490,"would_cite":true,"duration_ms":28733,"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":"Classical low surface brightness galaxies in the Romulus25 simulation are predominantly formed by major mergers that spin up the remnant, spreading star formation outward and keeping the galaxy gas-rich, metal-poor, and faint.","keywords":["low surface brightness galaxies","galaxy mergers","galaxy spin","angular momentum","cosmological simulations","star formation","neutral hydrogen","Romulus25"],"falsifier":"Re-run Romulus25 with a star formation model that prevents star formation in low-density gas (for example, by raising the density threshold) and check whether co-rotating major mergers still produce LSB remnants; if the spin-up-to-LSB connection disappears or weakens sharply, the central claim is an artifact of the subgrid recipe.","tokens_in":41365,"feed_emoji":"🌌","tokens_out":8162,"duration_ms":83951,"temperature":0.7,"pith_summary":"This paper argues that classical low surface brightness (LSB) galaxies — the gas-rich, late-type galaxies with faint central surface brightness — are not mostly products of quiet, isolated evolution but of a specific kind of major merger. In the Romulus25 cosmological simulation, when the merging companion co-rotates with the primary's gas disk and/or arrives with high orbital angular momentum, the merger spins up the remnant; star formation becomes more spread out and inefficient, so the galaxy keeps a large HI reservoir and low metallicity, yielding a faint center. LSB galaxies made this way account for about 60% of central galaxies with stellar masses $10^8$–$10^{10} \\, M_\\odot$ in the simulation. The paper also shows that LSB bulges have the same masses as HSB bulges but are more diffuse and redder, and that LSB galaxies are part of a continuum of merger outcomes rather than a separate species.","feed_headline":"Co-rotating mergers spin up most low surface brightness galaxies","feed_subtitle":"Aligned major mergers spread out star formation, leaving galaxies gas-rich and faint.","key_machinery":"The argument is carried by merger orientation and spin. The orientation is quantified by $\\cos \\phi_{\\rm orb} = \\hat{J}_{\\rm orb} \\cdot \\hat{J}_{\\rm gas,primary}$ at infall: +1 means the secondary co-rotates in the primary's gas disk plane, -1 counter-rotates, and 0 is perpendicular. Orbital angular momentum of the secondary at infall is the second input. The output is the gas spin parameter $\\lambda'_{\\rm gas}$ (the Bullock spin parameter without explicit energy dependence), which rises after the last major merger for LSB galaxies and falls for HSB galaxies. Because more than 80% of the median galaxy's stars form during or after that merger, the spin change leaves a permanent imprint on the stellar distribution: the radius containing 80% of star formation ($r_{80}$) grows to 2–3 times its pre-merger value in LSB galaxies while HSB remnants compactify. The classification itself uses the exponential disk fit central surface brightness $\\mu_0 \\ge 22.5$ mag/arcsec$^2$.","core_discovery":"The central claim is that classical LSB galaxies in Romulus25 are typically significantly spun up by their last major merger, and that this spin-up is what makes them low surface brightness. In mergers that produce LSB galaxies, the secondary is typically co-rotating and aligned with the primary's gas disk (cos $\\phi_{\\rm orb} > 0.25$) and/or has higher orbital angular momentum at infall, so more orbital angular momentum is converted into internal spin. As a result the post-merger galaxy has high spin, star formation is extended and inefficient, central surface brightness stays faint, and the galaxy accumulates unenriched gas. The paper further claims that LSB galaxies' blue observed colors arise from low dust reddening (low metallicity), not from young stellar populations; their stellar populations are actually slightly older and redder than those of HSB galaxies. The same analysis finds that LSB bulges are as massive as HSB bulges but lower surface brightness, redder, and more diffuse, and that LSB galaxies inhabit similar environments to HSB galaxies, with the last major merger being the main differentiating event.","pith_inferences":["If the same merger-spin-up mechanism operates in the real universe, faint stellar streams and disturbed HI morphologies around classical LSB galaxies should be detectable with deep imaging and HI mapping long after the last major merger.","Because the Romulus25 LSB fraction (about 60%) slightly exceeds the roughly 50% often quoted from observations, a higher-resolution simulation with less star formation in low-density gas might yield a lower LSB fraction; this is testable by varying the subgrid star formation threshold.","The orientation dependence is strongest for pure disk LSB galaxies and weakest for irregular dwarfs, suggesting the merger-spin-up channel becomes the dominant LSB formation route above dwarf masses, while feedback may remain more important in dwarfs."],"forward_implications":["If classical LSB galaxies form this way, major mergers do not always destroy disks and quench star formation: aligned, gas-rich mergers can produce extended, star-forming, disk-dominated remnants.","The observed combination of HI-richness, low metallicity, low star formation efficiency, and blue color in LSB galaxies follows from the same post-merger state, so these traits should appear together.","LSB galaxies with bulges should generally have bulges as massive as those of HSB galaxies of the same stellar mass, but fainter, redder, and more diffuse; this is a direct prediction for resolved bulge studies.","Classical LSB galaxies are one end of a continuum of merger outcomes; selection by surface brightness alone does not isolate a separate formation channel.","In Romulus25, LSB galaxies make up roughly 60% of central galaxies in the $10^8$–$10^{10} \\, M_\\odot$ range, implying the merger-spin-up path is common rather than rare."],"supporting_citations":[{"why":"Showed in zoom-in NIHAO simulations that aligned, co-rotating, co-planar mergers produce larger, lower surface brightness galaxies; this paper extends that result to a large sample at higher masses.","marker":"A. Di Cintio et al. (2019)"},{"why":"Identified major-merger spin-up as the formation path of isolated ultra-diffuse galaxies in Romulus25; the current paper applies and tests the same mechanism for classical LSB galaxies.","marker":"A. C. Wright et al. (2021)"},{"why":"Presents the Romulus25 simulation and its calibrated subgrid star formation and feedback parameters from which the LSB sample is drawn.","marker":"M. Tremmel et al. (2017)"},{"why":"Supplies the definition of last major merger time based on virial radius overlap and the expected 25–30% spin-up from major mergers that the paper quantifies.","marker":"H. Hetznecker & A. Burkert (2006)"},{"why":"Established the conversion of orbital angular momentum to internal spin during major mergers, the physical basis for the spin-up mechanism.","marker":"M. Vitvitska et al. (2002)"},{"why":"Reported that the most extended LSB galaxies in the EAGLE simulations are products of spin-increasing major mergers, providing a simulation-side comparison for this result.","marker":"A. Kulier et al. (2020)"}],"fun_headline_variants":["Co-rotating mergers spin up galaxies into faintness","Aligned major mergers make low surface brightness galaxies","Gas-rich LSB galaxies born from co-rotating mergers","Faint galaxies emerge from aligned, high-spin mergers","Spin from co-rotating mergers dims galaxies"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper assumes the subgrid star formation and feedback prescriptions of Romulus25 (star formation efficiency 0.15, blastwave coupling 0.75, and density threshold $n > 0.2$ cm$^{-3}$) accurately reproduce how gas is consumed and how star formation spreads after mergers, even though the resolution limit may artificially boost star formation in low-density gas.","fun_headline_variants_meta":{"raw":{"variants":["Co-rotating mergers spin up galaxies into faintness","Aligned major mergers make low surface brightness galaxies","Gas-rich LSB galaxies born from co-rotating mergers","Faint galaxies emerge from aligned, high-spin mergers","Spin from co-rotating mergers dims galaxies"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000248,"raw_usage":{"total_tokens":1613,"prompt_tokens":1075,"completion_tokens":538,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":691,"completion_tokens_details":{"reasoning_tokens":459}},"tokens_in":691,"tokens_out":538,"duration_ms":6037,"temperature":1.0,"reasoning_tokens":459,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T12:59:15.373902+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run Romulus25 with a star formation model that prevents star formation in low-density gas (for example, by raising the density threshold) and check whether co-rotating major mergers still produce LSB remnants; if the spin-up-to-LSB connection disappears or weakens sharply, the central claim is an artifact of the subgrid recipe.","supporting_citations":[{"cited_title":"D., & Trayford , J","cited_arxiv_id":null,"evidence_quote":"Reported that the most extended LSB galaxies in the EAGLE simulations are products of spin-increasing major mergers, providing a simulation-side comparison for this result."}],"review_version":1}