{"id":"89c8a87f-7b26-4075-bf48-80576638e2f2","arxiv_id":"2411.16166","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Accounting for relaxation between stellar encounters reduces the surviving mass fraction of axion minihalos in the solar neighborhood to about 30%, down from earlier estimates of about 60%.","lead":"This paper calculates how much damage passing stars do to clumps of axion dark matter in our galaxy, using a more detailed model of repeated star encounters. It finds these clumps lose more mass than earlier work suggested, which could make axions easier to detect.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 30% survival fraction is driven almost entirely by the p=1 energy-combination rule for relaxed encounters, a parameter fit in a self-cited companion paper and applied with a step-function threshold at one dynamical time; if the true p is near 1.5-2, the result moves toward S2024's 58%.","rationale":"The reader's weakest_assumption correctly identifies the p=1 combination rule as the load-bearing element of the central claim. I re-examined the manuscript for other potential weak points: (i) the use of a single orbit per (M,z) grid point, (ii) not updating tdyn after each pass, (iii) the response function interpolation from Paper 1, and (iv) the assumed singular isothermal sphere mass distribution. None of these individually threatens the main comparison as much as the p rule. The paper itself flags the p=1 rule as 'conservative' and notes Paper 1 found p ≲ 1; the threshold at one tdyn is a simplification. Since the early disk passes (when minihalos are densest and most susceptible) fall on the p=1 side for typical orbits, the 30% result is essentially the p=1 result. Independent validation of p (or a scan over p) is therefore the decisive check. This does not change the reader's conditional verdict; it reinforces the requirement that the p=1 rule be independently confirmed before the specific 30% figure is treated as a firm prediction.","tokens_in":23147,"tokens_out":7228,"duration_ms":66346,"concrete_test":"Re-run the Monte Carlo pipeline with p as a continuous, sigmoid function of the ratio Δt/tdyn, e.g. p = 1 + 1/(1+exp((Δt/tdyn - 1)/w)) with w spanning 0.1-10, and also with p=1.5 and p=2 for all relaxed (Δt>tdyn) passes. If Msurv/Mori at ma=25 µeV, Mmin=10^2 Msun moves from 29.6% to >40% for p=1.5, the central quantitative claim is not robust. Additionally (or instead) run direct N-body simulations of an NFW minihalo with concentration c~10^3 hit by two successive impulses separated by Δt/tdyn ∈ {0.5, 1, 2, 5}, fit the p that reproduces the final bound mass, and check whether p≈1 for Δt/tdyn>1 or whether a transition region and p>1 persist.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Sec. V and Eq. (6) define the effective energy injection for multiple encounters. The hybrid method switches between p=2 (unrelaxed) and p=1 (relaxed) depending on whether the inter-disk-pass interval exceeds tdyn. Because minihalo tdyn is short at high redshift (tdyn ~ 80 Myr at z~10, from Eq. (9) with rho_vir=200 rho_crit(z)) while the orbital period in the singular isothermal sphere is ~250 Myr, the early passes, when the minihalo is most concentrated, fall on the p=1 side of the switch. The choice p=1 is not derived from first principles; it is a one-parameter fit from Paper 1 [47] by overlapping authors, quoted as 'p ≲ 1' and then rounded down to p=1 'to be conservative.' The paper applies a hard step at Δt=tdyn even though relaxation is a continuous process. Consequently the headline value Msurv/Mori=29.6% (Table II, ma=25 µeV, Mmin=10^2 Msun) is a direct function of that fitted p. If the true relaxed-combination exponent is p=1.5, or if complete relaxation requires ~3 tdyn (shifting some passes to p=2), the survival fraction moves substantially upward, toward S2024's 53.5% for the same parameters. The qualitative statement that multiple encounters are more destructive than S2024's linear addition would survive, but the quantitative 30%-vs-60% claim and the detection-inference conclusion would need re-scaling. This is the single load-bearing uncertainty; all other choices (tdyn not updated, Appendix G; discrete orbit sampling) are shown or argued to have sub-10% effects.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the disruption of QCD axion minihalos by stellar encounters in the Milky Way, extending the recent analysis of S2024. The authors Monte Carlo simulate minihalo orbits in a singular isothermal sphere potential, weight them by a pre-infall mass function from X2021, and track energy injections from passages through a two-component stellar disk. Their new element is a hybrid rule for summing energy injections from multiple encounters: p = 2 when consecutive encounters are separated by less than the minihalo dynamical time, and p = 1 when the minihalo has time to relax between encounters (Sec. V, Eqs. (6)-(9)). With this hybrid rule, they find Msurv/Mori ≈ 30% for ma = 25 µeV and Mmin = 10^2 M_sun, compared with ≈ 58% in S2024, and they argue that the extra mass in inter-minihalo space raises the local axion density and improves haloscope detection prospects. The p = 2 limit of their simulation reproduces S2024's results at the level of the published curves.","tokens_in":23553,"tokens_out":3554,"duration_ms":37565,"significance":"If the central quantitative claim holds, the paper makes a meaningful correction to estimates of the present-day minihalo mass function and to predictions for axion direct detection. The methodology is transparent: the orbit sampling is described in detail, the code and data are openly available, the comparison to S2024 in the p = 2 limit is shown, and the authors explicitly quantify a number of secondary effects (not updating the dynamical time, response-function differences, and Monte Carlo convergence). The key strength is that the paper isolates the one physical ingredient that changes the answer, namely how repeated energy injections combine when halos relax between encounters. The main limitation is that this ingredient is not independently derived or calibrated in the present work; it is a parameter fit from a companion paper with overlapping authorship, and the final survival fraction is highly sensitive to it.","major_comments":[{"comment":"The headline result, Msurv/Mori = 29.6% for ma = 25 µeV and Mmin = 10^2 M_sun in Table II, is driven almost entirely by the choice p = 1 for relaxed encounters. The value p = 1 is not derived here; it is taken from a parameter fit in Paper 1 (Ref. [47]), where the authors found p ≲ 1 and then rounded to p = 1 'to be conservative.' Since Eq. (8) makes the effective energy injection grow as the square of the sum of sqrt(E_frac,i), the difference between the hybrid result and S2024's p = 2 result is a direct consequence of this fitted exponent. If the true combination exponent for relaxed encounters is, say, p = 1.5, or if complete relaxation requires several dynamical times rather than one, the survival fraction will move substantially upward, toward S2024's 53.5% for the same parameters. The paper should either provide a first-principles derivation or direct N-body calibration of the relaxed-encounter combination rule, or systematically vary p and the relaxation threshold and show how the headline number changes. Without this, the quantitative 30%-vs-60% claim is not robustly established, even though the qualitative direction (more disruption than linear addition) is plausible.","section":"Sec. V, Eqs. (6)-(8) and Table II"},{"comment":"The hybrid method applies a sharp step: encounters separated by more than tdyn are combined with p = 1, and encounters separated by less than tdyn are combined with p = 2. Gravitational relaxation is a continuous process, and a halo that has only partially relaxed between encounters should presumably be treated with an intermediate combination rule. Because early disk passages, when the minihalo is most concentrated and tdyn is short, fall on the p = 1 side of the switch, this step could systematically overstate the amount of relaxation. The authors note in Sec. IX and Appendix G that updating tdyn changes the result by only a few percent, but that estimate does not test the sharpness of the p = 1/p = 2 transition itself. A concrete sensitivity test, for example using a smooth interpolation in Delta t/tdyn or a threshold at 2-3 tdyn, would directly address this concern.","section":"Sec. V, step-function threshold at tdyn"},{"comment":"The reported values of Msurv/Mori are quoted to one decimal place and the text says the Monte Carlo result is stable at the 0.1% level, but this stability is only with respect to re-drawing orbits for the same grid and model choices. The dominant uncertainty is model uncertainty in p, the relaxation threshold, and the assumed constant value (250 km/s)^2 for sigma_*^2 + v_mh^2 in Eq. (19). The paper would be considerably stronger if Table II or Fig. 6 included a propagation of these modeling uncertainties, rather than only Monte Carlo sampling noise. This is a presentation and completeness issue for the central claim.","section":"Sec. IX, Figs. 5-6 and Table II"}],"minor_comments":[{"comment":"The introduction states that the undisrupted mass function is derived in 'Sec. III' and then later refers to 'Sec. III' again for a different topic; the section numbering in the introduction should be corrected.","section":"Sec. I, organization paragraph"},{"comment":"The text 'In Sec. III, we derive the undisrupted mass function...' appears to duplicate the section label; the actual derivation appears in the second Sec. III (or should be renumbered).","section":"Sec. II, sentence after Eq. (B4)"},{"comment":"The overlap between the authors of this paper and Paper 1 is disclosed in a footnote, but the main text could more explicitly state that the p ≲ 1 result is from that companion paper and is not independently checked here.","section":"Sec. V, footnote 1"},{"comment":"There is a typo: 'assummed' should be 'assumed' in the sentence defining rc.","section":"Eq. (23) and surrounding text"},{"comment":"The phrase 'It is a good match to the analogous dashed gray curve in the top panel of Fig. 10 of 2024' should explicitly refer to S2024 rather than just '2024'.","section":"Sec. IX, text near Fig. 4"},{"comment":"The definition of E via Phi(robs) and the random variable R1 is clear, but the relation to the sampled energy distribution would benefit from one sentence explaining why the expression has the log term; currently it is only justified by the reference to Ref. [62].","section":"Sec. VI, Eq. (11)"}],"recommendation":"major_revision","confidential_remarks":"The paper is a reasonable incremental contribution that extends S2024 with a more detailed treatment of repeated encounters. The main result, however, is controlled by a combination rule that is parameter-fitted in a companion paper with overlapping authorship and is applied with a step-function relaxation criterion. I would ask the editor to ensure that the revised version either derives or independently calibrates this rule, or at minimum presents a careful sensitivity scan over p and the relaxation threshold. The paper's own appendix estimates that updating the dynamical time changes the answer by only a few percent, which is helpful, but it does not address the dominant uncertainty. This is fixable within the scope of the manuscript, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new thing here is the hybrid treatment of multiple stellar encounters: they simulate realistic orbits in a Milky Way potential, decide per gap whether the minihalo had time to relax, and combine energy injections with p=1 or p=2 accordingly. That is a real step beyond S2024's pure linear addition, and the machinery is described carefully enough to reproduce. The p=2 limit matches S2024, they checked convergence, and the code and data are public. Credit where due: this is a transparent, well-scoped simulation paper.\n\nThe soft spot is exactly where the headline comes from. The 30% versus ~60% survival difference is driven almost entirely by the switch to p=1 for relaxed encounters. That rule is not derived here; it is a fitted parameter from their own Paper 1, quoted as p ≲ 1 and then rounded down as a conservative choice. The threshold between relaxed and unrelaxed is a hard step at one dynamical time. The paper never shows what happens if the true p is 1.2 or 1.5, or if relaxation takes three dynamical times instead of one. The stress-test concern is legitimate: if either knob moves modestly, the 30% number slides back toward S2024's ~54%. The qualitative conclusion that multiple encounters are more destructive than linear addition would survive, but the quantitative claim would need rescaling.\n\nTwo smaller things. First, the detection-inference paragraph says the local axion density in minivoids increases, but they never compute that density; it is a plausible hand-wave, not a result. Second, there is a section-numbering typo in the introduction (Sec. III appears twice). Both are minor and fixable.\n\nWho is this for? People working on axion substructure and haloscope detection strategies. It deserves a serious referee. The referee should ask for a sensitivity scan over p and over the relaxation-time threshold, plus ideally an independent check of the p=1 rule against direct N-body simulations of relaxing minihalos. Conditional acceptance is the right verdict: the framework is sound, the headline number is not yet firm.","headline":"Solid Monte Carlo update on axion minihalo disruption, but the headline 30% survival fraction sits on a parameter from the authors' own prior paper and needs a sensitivity analysis before I'd trust the number.","tokens_in":24101,"tokens_out":2094,"would_cite":true,"duration_ms":22340,"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":"This paper argues that accounting for relaxation between successive stellar encounters makes stellar disruption of axion minihalos roughly twice as destructive as previously estimated, with only about 30% of the original minihalo mass…","keywords":["axion minihalos","stellar disruption","dark matter substructure","QCD axion","mass function","haloscope detection","dynamical relaxation","Milky Way stellar encounters"],"falsifier":"A controlled $N$-body experiment would settle it: take an NFW minihalo, apply two identical impulsive stellar kicks separated by waiting times of $0.5$, $1$, $2$, and $5$ dynamical times, and measure the final bound mass in each case. If the fully relaxed cases do not match the $p=1$ prediction, or if the transition from linear to nonlinear addition occurs at a different threshold than one dynamical time, the 30%-versus-58% contrast will not be reproduced.","tokens_in":22923,"feed_emoji":"⭐","tokens_out":14199,"duration_ms":111858,"temperature":0.7,"pith_summary":"This paper tries to establish that repeated stellar encounters in the Milky Way destroy axion minihalos more efficiently than earlier work concluded, because what matters is whether a minihalo has time to re-form between successive hits. The authors combine the energy injections from consecutive Galactic disk crossings with $p=1$ when the gap exceeds the minihalo's dynamical time and $p=2$ when it does not, a rule they call the hybrid method. For an axion mass of $25\\,\\mu\\mathrm{eV}$ and a minimum adiabatic-halo mass of $10^2\\,M_\\odot$, they find that only about 30% of the original minihalo mass survives at the solar radius, compared with about 58% under the earlier linear-addition method. If that is right, more axion dark matter sits in inter-minihalo space, the local axion density at Earth is higher, and haloscope searches become more promising than previously estimated. The paper also presents the disrupted mass function and checks that updating the dynamical time after each pass changes the result by only a few percent.","feed_headline":"Repeated star encounters cut axion minihalo survival to 30 percent","feed_subtitle":"Repeated stellar hits that allow reformation cut minihalo mass to ~30%, brightening direct dark-matter searches.","key_machinery":"The load-bearing object is the effective energy injection parameter $E_{\\mathrm{frac}}=\\Delta E/E_{\\mathrm{bind}}$ for a single stellar encounter, combined across many encounters by $E_{\\mathrm{frac,eff}}=\\left(\\sum_i E_{\\mathrm{frac},i}^{p/2}\\right)^{2/p}$. The exponent $p$ encodes the physical state of the halo: $p=2$ when encounters arrive so quickly that the minihalo cannot change between them, and $p=1$ when the minihalo fully relaxes in between, a value parameter-fitted in Paper 1 and adopted as a conservative choice. The decision is made by comparing the time between consecutive disk passages to the dynamical time $t_{\\mathrm{dyn}}=\\sqrt{3\\pi/(16G\\bar\\rho_{\\mathrm{vir}})}$; the hybrid method applies $p=1$ or $p=2$ separately to each consecutive pair of passages. Feeding the resulting effective $E_{\\mathrm{frac}}$ and the concentration parameter $c$ into the survival-fraction relation of Paper 1, over a Monte Carlo population of orbits evolved in the Galactic potential, produces the disrupted mass function and the 30% survival figure.","core_discovery":"The central claim is a mechanism and a number: when a minihalo is struck twice with enough time to relax in between, the second encounter acts on a looser object, so the two energy injections should be added nonlinearly ($p=1$) rather than linearly ($p=2$). Using Monte Carlo orbits drawn from a singular isothermal sphere so that the minihalo population is found at the solar neighborhood today, and summing the energy injected on every disk passage with the $p=1$/$p=2$ hybrid rule, the authors obtain a surviving mass fraction $M_{\\mathrm{surv}}/M_{\\mathrm{ori}}=30\\%$ for $m_a=25\\,\\mu\\mathrm{eV}$ and $M_{\\mathrm{min}}=10^2\\,M_\\odot$, against roughly 58% for linear addition. Their Table II gives 29-30% for axion masses $1.25,25,500\\,\\mu\\mathrm{eV}$ at that $M_{\\mathrm{min}}$, and 37-38% for $M_{\\mathrm{min}}=10^{-2}\\,M_\\odot$. Since Earth is more likely inside a minivoid than inside a minihalo, the extra disrupted mass raises the expected local axion density and strengthens haloscope detection prospects. The disrupted mass function is more suppressed than in the linear case and its peak shifts toward lower masses.","pith_inferences":["One consequence left implicit is that gravitational-lensing and pulsar-timing searches for small-scale dark-matter structure should expect a deficit of roughly $10^{-12}$ to $10^{-3}\\,M_\\odot$ halos relative to disruption-free predictions, turning the mass-retention ratio into an observable target.","The hybrid relaxation rule is independently testable in controlled $N$-body setups with two impulsive kicks separated by $0.5$ to $5$ dynamical times, which would calibrate where the $p=1$/$p=2$ transition actually sits.","The same relaxation logic should apply to other self-gravitating substructures that repeatedly cross stellar disks, so the factor-of-two reduction in retained mass could be a general feature of substructure evolution rather than an axion-specific one."],"forward_implications":["The stellar-disrupted minihalo mass function is roughly halved in area relative to the $p=2$ case, and its peak shifts toward lower masses, which directly changes predictions for any probe that counts compact dark-matter subhalos.","More axion mass ends up in inter-minihalo space; since Earth is more likely inside a minivoid than inside a minihalo, the expected local axion density at a haloscope is higher than previous estimates.","Haloscope detection prospects improve relative to the linear-addition estimate, with the same qualitative conclusion across axion masses $1.25$, $25$, and $500\\,\\mu\\mathrm{eV}$ and across the two assumed adiabatic-halo mass cutoffs.","Updating the dynamical time after each disk pass instead of holding it fixed changes $M_{\\mathrm{surv}}/M_{\\mathrm{ori}}$ by at most about 1-3%, so the main conclusion is stable against that simplification."],"supporting_citations":[{"why":"Provides the baseline linear-addition (p=2) estimate of ~58% retention and the disk-pass energy injection formula that this paper recombines with the hybrid rule.","marker":"[46]"},{"why":"Paper 1: supplies the single-encounter mass-loss response used to build the survival-fraction interpolation, and the parameter fit p ≈ 1 for fully relaxed encounters.","marker":"[47]"},{"why":"Supplies the general composition formula for multiple energy injections, Eq. (6), which the hybrid method switches between p=1 and p=2.","marker":"[60]"},{"why":"X2021: source of the modified Sheth-Tormen pre-infall minihalo mass function and the mass-concentration relation that seeds the simulation grid.","marker":"[36]"},{"why":"Provides the Monte Carlo orbit-sampling procedure used to draw minihalo orbits from the singular isothermal sphere distribution.","marker":"[62]"},{"why":"Standard source for the dynamical-time formula that sets the relaxation threshold between consecutive stellar encounters.","marker":"[61]"}],"fun_headline_variants":["Repeated star hits shred axion minihalos down to 30% mass","Axion minihalos lose 70% of mass to repeated stellar encounters","Stellar double hits cut axion minihalo survival to 30%","Multiple star encounters slash axion minihalo masses by 70%","Starry encounters strip axion minihalos to 30% survival"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that one dynamical time is enough for a minihalo to fully relax after a stellar encounter, so that a later encounter combines with the earlier one through the $p=1$ rule; if real relaxation needs several dynamical times, or if the relaxed rule is closer to $p=2$, the drop from about 58% to about 30% largely disappears.","fun_headline_variants_meta":{"raw":{"variants":["Repeated star hits shred axion minihalos down to 30% mass","Axion minihalos lose 70% of mass to repeated stellar encounters","Stellar double hits cut axion minihalo survival to 30%","Multiple star encounters slash axion minihalo masses by 70%","Starry encounters strip axion minihalos to 30% survival"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00076,"raw_usage":{"total_tokens":3408,"prompt_tokens":1008,"completion_tokens":2400,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":624,"completion_tokens_details":{"reasoning_tokens":2297}},"tokens_in":624,"tokens_out":2400,"duration_ms":15449,"temperature":1.0,"reasoning_tokens":2297,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:28:43.165948+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A controlled $N$-body experiment would settle it: take an NFW minihalo, apply two identical impulsive stellar kicks separated by waiting times of $0.5$, $1$, $2$, and $5$ dynamical times, and measure the final bound mass in each case. If the fully relaxed cases do not match the $p=1$ prediction, or if the transition from linear to nonlinear addition occurs at a different threshold than one dynamical time, the 30%-versus-58% contrast will not be reproduced.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the baseline linear-addition (p=2) estimate of ~58% retention and the disk-pass energy injection formula that this paper recombines with the hybrid rule."},{"cited_title":"DSouza and C","cited_arxiv_id":null,"evidence_quote":"Paper 1: supplies the single-encounter mass-loss response used to build the survival-fraction interpolation, and the parameter fit p ≈ 1 for fully relaxed encounters."},{"cited_title":"St¨ ucker, G","cited_arxiv_id":null,"evidence_quote":"Supplies the general composition formula for multiple energy injections, Eq. (6), which the hybrid method switches between p=1 and p=2."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"X2021: source of the modified Sheth-Tormen pre-infall minihalo mass function and the mass-concentration relation that seeds the simulation grid."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Monte Carlo orbit-sampling procedure used to draw minihalo orbits from the singular isothermal sphere distribution."},{"cited_title":"Binney and S","cited_arxiv_id":null,"evidence_quote":"Standard source for the dynamical-time formula that sets the relaxation threshold between consecutive stellar encounters."}],"review_version":1}