{"id":"064e6f9d-3eba-41ef-a856-2e2c11f35bc0","arxiv_id":"2505.13590","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"In Auriga simulations, boxy/peanut bulges form 1.1-1.6 Gyr after bar formation, always after a buckling event, and appear in 45% of barred galaxies at z=0, dropping to zero by z~1.","lead":"Using the Auriga cosmological zoom-in simulations, the authors track when boxy/peanut (b/p) bulges form and how common they are across cosmic time. The results show b/p bulges appear 1.1 to 1.6 billion years after their bars form, and that the b/p fraction declines from 45 percent today to zero by redshift one.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The b/p detection threshold d2Z/dr2 < -0.51, calibrated on the same galaxies it is used to validate, controls all headline timescales and fractions; Appendix D's admission that faint b/p signals remain visible when the code reports none makes this the pivotal assumption.","rationale":"The paper is a careful, well-presented study with public data and honest self-criticism (Appendix D). The central qualitative finding - that b/p bulges in Auriga form shortly after bar formation and are preceded by buckling - is supported by visual inspection and by independent parameters (A_buck, Theta_tilt), so it is not in serious doubt. The load-bearing fragility is the single calibration threshold in the automated detector, which determines not only whether a b/p is present at z=0 but also the formation time and the apparent dissolution/re-emergence. The authors' own Appendix D shows that a faint peanut signal survives when the code reports no detection, so 'dissolution' is at least partly a threshold artifact. Because the headline numbers (1.1-1.6 Gyr delay, 45% fraction, decline to zero by z~1) are all derived from this threshold, they should carry a sensitivity band. I therefore agree with the reader's CONDITIONAL verdict: the strongest claims are provisional until the threshold dependence is quantified. No change to the reader's verdict is needed, but the concrete test above should be run before the numbers are used for strong inferences.","tokens_in":27901,"tokens_out":6984,"duration_ms":63510,"concrete_test":"Run the Sec 2.3 pipeline on the public Auriga snapshots with the (d2Z/dr2)_min threshold varied from -0.35 to -0.65 in steps of 0.05, keeping the Butterworth smoothing and radial selection fixed. Record t_c_bp for each galaxy, the median delay, and the z=0 b/p fraction. If the median delay moves outside the quoted 1.1-1.6 Gyr range or the fraction shifts by more than ~10 percentage points, the headline numbers are threshold-dominated. This test also settles the dissolution claim: if lowering the threshold to -0.60 causes the Au10/Au17 'dissolved' intervals to register a peak, the dissolution is an artifact of the -0.51 cut.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2.3(v) defines b/p formation time t_c_bp by the first time (d2Z/dr2)_min drops below -0.51 and stays there for at least 5 snapshots. Footnote 3 states this threshold was determined by visually inspecting the same galaxies used to calibrate the method, and Appendix D concedes that during intervals where the code reports zero b/p strength, unsharp masks show a faint peanut signal near the ends of the bar. The threshold is therefore not a neutral detection limit but an ontological cut: it decides whether a b/p is present, when it forms, whether it dissolves, and whether it re-emerges. Every headline quantity - the median 1.6 Gyr delay (or 1.1 Gyr using visual weak times), the 45% z=0 fraction, the decline to zero by z~1, and the 'dissolution' episodes in Au10 and Au17 - is a direct function of this single number. If the threshold were slightly stricter or looser, the inferred formation times and fractions would shift; the paper provides no sensitivity analysis. The visual 'strong b/p' times (t_vs_bp) agree well with the code, which helps, but the b/p fraction curve in Fig. 6 is computed from the same threshold, and the visual method used for the thin line is not described quantitatively. The sample of 9 galaxies is small, but it is the threshold that makes the quantitative claims fragile, not Poisson statistics alone.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses the Auriga cosmological zoom-in simulations to study the formation and evolution of boxy/peanut (b/p) bulges. The authors develop an automated b/p detection method based on the second derivative of a normalized vertical-height profile, identify a sample of nine barred galaxies with b/p structures, and measure formation times relative to bar formation, b/p fractions as a function of redshift, b/p sizes, and the timing of buckling episodes. They report that b/p bulges form 1.1-1.6 Gyr after bar formation, always following a buckling event, that 45% of barred galaxies host a b/p at z=0 with the fraction declining to zero by z~1, and that b/p sizes are roughly half the bar length, with some galaxies showing b/p 'dissolution' and re-emergence.","tokens_in":28325,"tokens_out":3266,"duration_ms":29928,"significance":"If the central results hold, this is a valuable step in placing b/p bulge formation in a cosmological context, complementing earlier isolated-galaxy simulations and observational work. The paper is carefully structured and uses multiple independent diagnostics (visual inspection, Z-peak detection, A_buck, and Theta_tilt) that broadly agree, and the underlying Auriga data are public. The main quantitative claims, however, rest on a single detection threshold whose calibration is not independent of the measurements it produces, and the paper's own Appendix D concedes that faint b/p signals remain visible when the code reports no detection. These issues affect the headline formation times, the redshift evolution of the b/p fraction, and the interpretation of 'dissolution' events. The direction and qualitative conclusions are likely robust, but the reported numbers need additional sensitivity analysis before they can be accepted as quantitative predictions.","major_comments":[{"comment":"The detection threshold d^2Z/dr^2 < -0.51 is calibrated by visually inspecting the same nine galaxies whose formation times and fractions it is then used to measure (footnote 3), and Appendix D states that during intervals when the code reports zero b/p strength, unsharp masks show a faint peanut signal near the bar ends. Because this single threshold defines t_c_bp, the b/p fraction curve in Fig. 6, and the 'dissolution' episodes in Au10 and Au17, the paper should provide a sensitivity analysis, for example by varying the threshold or by using the visual weak detection times as an alternative criterion, before the quantitative claims (median 1.1-1.6 Gyr delay, 45% z=0 fraction, decline to zero at z~1) can be taken at face value.","section":"Sec. 2.3(v), footnote 3, and Appendix D"},{"comment":"The code-based b/p fraction is a direct function of the same -0.51 threshold, while the 'visual' b/p fraction shown as the thin blue line in Fig. 6 is not described quantitatively: there is no explicit protocol for what constitutes a visually detected b/p at each redshift, nor a consistency criterion between the visual and code counts. Without such a protocol, the claimed decline to zero at z~1 is not robust to the detection threshold, especially given the faint signals noted in Appendix D.","section":"Sec. 3.1 and Fig. 6"},{"comment":"The claim that b/p structures 'dissolve' (abstract, Sec. 3, Sec. 7) is stronger than the evidence supports. The authors themselves state in Appendix D that the unsharp masks show a faint peanut signal during the code's zero-strength intervals, which means the structure weakens below the detection threshold rather than demonstrably vanishing. This phrasing should be revised, or the strength criterion should be calibrated to the visually persistent feature, so that 'dissolution' is not an artifact of the threshold choice.","section":"Sec. 3 and Appendix D"}],"minor_comments":[{"comment":"The caption reads 'the formation time of the formation time of the bar'; the duplicate phrase should be corrected.","section":"Fig. 5 caption"},{"comment":"The quoted median delay of 1.1 Gyr for 'first appearance' is not clearly defined: Table 2 lists t_vw,1 only for Au13 and Au18 (and t_vw,2 for Au18), so the set of galaxies used to compute this median should be stated explicitly to avoid ambiguity.","section":"Sec. 6.1 and Fig. 5"},{"comment":"The text refers to 'IllsutrisTNG50' instead of 'IllustrisTNG50'.","section":"Sec. 3.1"},{"comment":"There is an inconsistency between the text of Sec. 5, which describes Au17's first buckling as spanning roughly 8.7-6 Gyr and the second as 3.8-0 Gyr, and the caption of Fig. 12, which quotes intervals of 8.09-7.62 Gyr and 2.79-0.34 Gyr; please reconcile these numbers.","section":"Sec. 5 and Fig. 12"}],"recommendation":"major_revision","confidential_remarks":"The paper is honest about its methodological limitations, which is to its credit, but the central quantitative claims currently hinge on a threshold calibrated on the same data it is used to measure. A sensitivity analysis and a quantitative description of the visual method used for the b/p fraction are necessary before the paper can be accepted. The scope of the paper is appropriate for MNRAS and the topic is timely."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a read if you work on bars or bulges. They track boxy/peanut bulges across time in the Auriga suite, something only isolated sims did carefully before. The main results: b/p bulges form 1.1–1.6 Gyr after the bar, always after a buckling episode; at z=0 about 45% of barred galaxies have a b/p, and the fraction declines to zero by z~1; b/p size is about half the bar length. They also document multiple buckling and cases where the b/p signal fades and then re-emerges.\n\nThe paper is careful. They use several independent diagnostics (visual inspection, the Z-profile curvature, Abuck, Theta_tilt), and the diagnostics mostly agree. The data are public. They compare against observed b/p fractions and isolated N-body results, and they flag where Auriga underproduces b/p relative to some observations. That is honest, useful work.\n\nThe soft spot is the automated detector. The b/p formation time is defined as when the minimum of d2Z/dr2 drops below -0.51. That threshold was picked by visually inspecting the same galaxies used to validate the method, and Appendix D concedes that during 'undetected' phases the unsharp masks still show a faint peanut near the ends of the bar. So the 'dissolution' and the exact delay time and the redshift evolution of the fraction are all hostage to one number, and there is no sensitivity test. A slightly different threshold would shift the timescales and the fraction curve. The sample is only nine galaxies, which is fine for this kind of study, but it makes the quantitative headline numbers provisional.\n\nThat said, the central claim that b/p formation in these simulations is tied to buckling and lags bar formation by roughly 1–2 Gyr is robust to reasonable threshold choices. The paper does not overclaim; it explicitly notes the ambiguity between a true dissolution and a signal falling below detectability. What it needs is a robustness analysis of the threshold and perhaps a more quantitative visual criterion for the weak b/p times.\n\nWho is this for? Anyone comparing cosmological simulations to observed b/p fractions or studying bar buckling. It is a legitimate subfield measurement, not a field reorganisation.\n\nI'd send it out for review. The threshold concern is real but fixable; the paper is worth refereeing, and a good referee could ask for the sensitivity test without demanding a rewrite.","headline":"First systematic account of b/p bulges in cosmological simulations, with a real caveat: the detection threshold controls every headline number and is calibrated on the same visual data used for validation.","tokens_in":28827,"tokens_out":1855,"would_cite":true,"duration_ms":18030,"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":"In the Auriga cosmological simulations, boxy/peanut bulges form 1.1–1.6 Gyr after their bar does, always after a buckling episode, and the b/p fraction among barred galaxies falls from 45% at $z=0$ to zero by $z\\sim1$.","keywords":["boxy/peanut bulges","galactic bars","buckling instability","cosmological zoom-in simulations","galaxy structure","redshift evolution","bar fraction","Auriga simulations"],"falsifier":"Re-run the peak-detection algorithm on the same nine galaxies with thresholds of, say, $-0.40$ and $-0.60$ and recompute formation times and the $z=0$ fraction: if the 1.1–1.6 Gyr delay or the 45 per cent fraction moves by more than the quoted uncertainties, the published timeline is a calibration artifact. A second test: a securely identified b/p bulge in a barred galaxy at $z>1$ would contradict the claim that the b/p fraction reaches zero by $z\\sim1$.","tokens_in":27660,"feed_emoji":"🥜","tokens_out":10501,"duration_ms":91858,"temperature":0.7,"pith_summary":"This paper asks when and how boxy/peanut (b/p) bulges appear in disc galaxies that form inside a cosmological context, rather than in the isolated galaxy models where these structures were first studied. Using 30 Milky-Way-mass haloes from the Auriga magneto-hydrodynamical zoom-in simulations, it develops an automatic measure of b/p strength from the vertical height profile of bar stars and traces that measure across cosmic time. It claims that b/p bulges form about 1.1–1.6 Gyr after a bar forms, that all nine b/p bulges in the sample follow a bar-buckling episode, and that the fraction of barred galaxies hosting a b/p falls from 45 per cent at $z=0$ to zero by $z\\sim1$. These claims matter because they turn a debated morphological detail into a dated dynamical event that both simulations and high-redshift surveys can agree to look for.","feed_headline":"Peanut bulges appear 1.1-1.6 Gyr after galactic bars","feed_subtitle":"In Auriga simulations, 45% of barred galaxies host them at z=0 and the fraction hits zero by z~1.","key_machinery":"The central object is the normalized vertical height profile $Z(r)=|z|_{\\mathrm{median}}(r)/|z|_{\\mathrm{median}}(0)$ built from bar stars, which measures how thick the stellar distribution is at each radius. A b/p bulge is registered as a local peak in $Z$, found automatically as the minimum of $\\mathrm{d}^2Z/\\mathrm{d}r^2$ dropping below $-0.51$ and staying there for at least five snapshots. The same peak's radial position defines the b/p size $R_{\\mathrm{bp}}$, and its height defines the b/p strength. Supporting diagnostics are the Fourier bar-strength profile $A_2(r)$, the buckling amplitude $A_{\\mathrm{buck}}$, and the meridional tilt angle $\\Theta_{\\mathrm{tilt}}$, whose peaks locate buckling episodes and tie the b/p peak to a physical instability.","core_discovery":"On the paper's own terms, the central discovery is that boxy/peanut bulges in cosmological simulations are not slow resonance-built structures but the visible aftermath of discrete buckling events. The peak-detection method finds nine b/p bulges in the 30 Auriga haloes; each one appears between 1.1 and 1.6 Gyr (median) after the bar forms, and in every case a visual break in the bar's mid-plane symmetry precedes the appearance. Two galaxies in the sample lose their b/p signal for a period and then regrow it after a later buckling, and one galaxy shows four buckling episodes. The paper also reports that b/p bulges extend to roughly half the bar length, with a median $R_{\\mathrm{bp}}/R_{\\mathrm{bar}}=0.48$ at $z=0$, and that the fraction of barred galaxies with a b/p drops from 45 per cent at $z=0$ to 20 per cent at $z=0.5$ and to zero at $z\\sim1$.","pith_inferences":["Because the $-0.51$ threshold was calibrated on the same nine galaxies it is applied to, the reported formation times and the 45 per cent fraction are partly circular; varying the threshold is a cheap check, and the paper's own appendix shows faint b/p signals may persist where the algorithm registers zero.","The universal buckling-first result may be a property of the Auriga galaxy formation model rather than of galaxies generally; the paper argues this model lacks massive central mass concentrations, so a simulation with a strong central concentration should produce a resonance-formed b/p without buckling.","The trend that b/p hosts have lower $\\sigma_z/\\sigma_r$ about 1 Gyr after bar formation suggests a quantitative threshold could separate future b/p hosts from non-hosts; that is a direct, testable extension to a larger sample of haloes.","If disc heating indeed suppresses buckling, then variations in feedback strength that make discs thinner should raise the b/p fraction toward the 69–80 per cent values reported by some observational studies."],"forward_implications":["The b/p fraction as a function of redshift becomes a clock for bar evolution: in these simulations the 45 per cent at $z=0$ falls to 20 per cent at $z=0.5$ and to zero by $z\\sim1$, so high-redshift surveys should see a steep drop.","A b/p bulge is not permanent once formed: bars can buckle, build a peanut, lose its detectable signal, and rebuild it after a later buckling, so single-snapshot classifications miss an important part of the population.","Because the delay between bar and b/p formation is only 1.1–1.6 Gyr, most b/p-bearing bars in the local Universe should be older than roughly 2 Gyr, and young bars should almost never show a b/p.","The b/p radius tracks, and is about half of, the bar radius, so measurements of one put a constraint on the other; the observed $R_{\\mathrm{bp}}/R_{\\mathrm{bar}}$ ratios agree with this simulated value."],"supporting_citations":[{"why":"It supplies the Auriga zoom-in haloes and the galaxy formation model that produce every galaxy analysed in this paper.","marker":"Grand et al. 2017"},{"why":"It identified six of the nine $z=0$ b/p galaxies in Auriga by visual inspection of unsharp masks, forming the seed of this sample.","marker":"Blázquez-Calero et al. 2020"},{"why":"It characterised b/p structures in Auriga at $z=0$ and reported a strong-b/p fraction near 30 per cent, the earlier estimate this paper extends.","marker":"Fragkoudi et al. 2020"},{"why":"It provides the comparison b/p fraction from a second cosmological simulation at $z=0$.","marker":"Anderson et al. 2024"},{"why":"It supplies the observed b/p fraction versus redshift, including the corrected detections used to quantify the decline toward $z\\sim1$.","marker":"Kruk et al. 2019"},{"why":"It gives the observed mass-dependent b/p fraction and the $R_{\\mathrm{box}}/R_{\\mathrm{bar}}$ size ratios that the simulated sizes are compared to.","marker":"Erwin & Debattista 2017"},{"why":"It established repeated buckling episodes and a roughly 1 Gyr bar-to-b/p delay in isolated $N$-body bars, the main dynamical comparison.","marker":"Martinez-Valpuesta et al. 2006"},{"why":"It showed resonance trapping can build b/p bulges without buckling, the alternative mechanism that the paper argues is absent in Auriga.","marker":"Sellwood & Gerhard 2020"},{"why":"It defines the meridional tilt angle and quantifies how disc thickness changes b/p strength and formation delay.","marker":"Ghosh et al. 2024"},{"why":"It introduced the $A_{\\mathrm{buck}}$ amplitude and linked gas content and central mass concentration to buckling suppression.","marker":"Debattista et al. 2006"}],"fun_headline_variants":["Peanut bulges arise from buckling in Auriga simulations","Cosmic peanut bulges form 1-2 Gyr after bar buckling","Auriga runs: peanut bulges in 45% of barred galaxies at z=0","Peanut bulges decay to zero by z~1 in cosmological runs"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the single threshold $\\mathrm{d}^2Z/\\mathrm{d}r^2 < -0.51$, chosen by visually inspecting the same galaxies, is what separates a real b/p bulge from a faint or transient one; because the paper itself shows faint b/p features remaining visible in unsharp masks when the code reports none, a different threshold would shift every formation time and fraction this paper reports.","fun_headline_variants_meta":{"raw":{"variants":["Peanut bulges arise from buckling in Auriga simulations","Cosmic peanut bulges form 1-2 Gyr after bar buckling","Auriga runs: peanut bulges in 45% of barred galaxies at z=0","Peanut bulges decay to zero by z~1 in cosmological runs"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000216,"raw_usage":{"total_tokens":1526,"prompt_tokens":1130,"completion_tokens":396,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":746,"completion_tokens_details":{"reasoning_tokens":314}},"tokens_in":746,"tokens_out":396,"duration_ms":3764,"temperature":1.0,"reasoning_tokens":314,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T20:13:05.175548+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the peak-detection algorithm on the same nine galaxies with thresholds of, say, $-0.40$ and $-0.60$ and recompute formation times and the $z=0$ fraction: if the 1.1–1.6 Gyr delay or the 45 per cent fraction moves by more than the quoted uncertainties, the published timeline is a calibration artifact. A second test: a securely identified b/p bulge in a barred galaxy at $z>1$ would contradict the claim that the b/p fraction reaches zero by $z\\sim1$.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It identified six of the nine $z=0$ b/p galaxies in Auriga by visual inspection of unsharp masks, forming the seed of this sample."}],"review_version":1}