{"id":"6926da91-fce0-45bf-929e-378debb2160b","arxiv_id":"2607.04121","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.5,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"RAMCOAL now evolves subgrid massive black hole triplets to coalescence inside live hydrodynamical simulations by mapping chaotic encounters onto Bonetti three-body outcomes.","lead":"The paper extends the RAMCOAL subgrid model so hydrodynamical galaxy simulations can track massive black hole triplets through unresolved dynamical phases to coalescence. Geometry alone can change which pair merges and when, enabling merger catalogues that link gravitational-wave signals to host galaxies.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"The end-to-end coalescence claim rests on an instantaneous Bonetti draw plus fixed 10^3 yr timer, so the live-hydro coupling during the resonant phase is not actually demonstrated.","rationale":"The reader correctly isolates the Bonetti-mapping approximation (fixed timer, escape-speed ejection, q_out≤1) as the weakest assumption. That assumption is load-bearing for the paper’s strongest claim: without a resolved or time-resolved resonant phase, the “first live-hydro end-to-end triplet coalescence” is only a library draw followed by an instantaneous merger. The three isolated-galaxy tests remain useful demonstrations of geometry sensitivity and of the bookkeeping machinery, but they do not yet establish that the live environment shapes the chaotic outcome itself. The CONDITIONAL verdict is therefore unchanged; the concrete timer/integration test above would decide whether the claim needs stronger caveats or can stand as written. No other internal inconsistency rises to the same level of centrality.","tokens_in":39108,"tokens_out":577,"duration_ms":5273,"concrete_test":"Re-run test case C with the chaotic timer replaced by the Bonetti-bin mean interaction time (or a short direct three-body integration of the recorded state in Table 4) while keeping the live RAMSES gas and stars active; if the surviving pair identity, coalescence time or remnant spin/recoil shifts by more than the Poisson scatter of the original 100-seed validation (§6.1), the end-to-end live-hydro claim is overstated.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The strongest claim (abstract; §6.4; conclusions) is that RAMCOAL follows a triplet “all the way to coalescence inside a live hydrodynamical simulation.” In test case C the chaotic phase is not integrated: once the Mylläri-type criterion (Eqs. 16–19) is met, a single weighted draw from the Bonetti (M_pri, q_in, q_out) table is taken and coalescence is assumed to occur after a fixed 10^3 yr timer (§5.2). The subsequent GW track is reconstructed only in post-processing (Fig. 22). Consequently the live gas, stars and feedback that the paper advertises as self-consistently coupled never act during the resonant interaction that actually decides which pair merges and when. The capability claim therefore reduces to “library lookup + instantaneous merger,” which is weaker than the wording “dynamical evolution … all the way to coalescence.” The same section also freezes accretion for the entire (short) chaotic interval and sets ejection velocity to local escape speed rather than the N-body distribution, further decoupling the outcome from the live environment.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper extends the RAMCOAL subgrid framework in RAMSES to treat massive black hole triplets. MBHs evolve as sinks through resolved dynamical friction (stage 0), subgrid DF (stage 1), and bound hardening by stellar scattering, circumbinary torques, and GWs (stage 2). Hierarchical triplets that meet a Mylläri-type instability criterion are mapped onto the Bonetti et al. (2018) scattering library via a weighted draw on (M_pri, q_in, q_out), with a fixed 10^3 yr chaotic timer; remnant spin and recoil follow Rezzolla and Lousto–Zlochower fits. Three isolated-galaxy tests show geometry-dependent partner selection (A vs B) and an exchange merger in a compact active triplet (C), with post-processed GW tracks. The authors claim the first end-to-end triplet evolution to coalescence inside a live hydrodynamical simulation and a route to environmentally linked GW merger catalogues.","tokens_in":39440,"tokens_out":1431,"duration_ms":11174,"significance":"If the framework holds under cosmological application, it would fill a genuine gap between pure post-processing delay models and expensive regularized N-body treatments (e.g. KETJU), enabling PTA/LISA/TianQin catalogues that retain live gas, accretion, spin, and recoil history. Strengths include an explicit seven-state classification, on-the-fly coupling of accretion/spin/feedback to the subgrid orbit outside the chaotic interval, validation of the Bonetti selector for one cell (N_seed=100), and a transparent demonstration that encounter geometry alone can change which pair merges. The work is timely given PTA backgrounds and LISA adoption. The main caveat is that the resonant three-body phase itself is not dynamically integrated in the live run, so the strongest wording of the end-to-end claim needs careful qualification.","major_comments":[{"comment":"Abstract, §6.4, and Conclusions claim the first triplet evolution “all the way to coalescence inside a live hydrodynamical simulation.” In case C the chaotic phase is not integrated: once Eqs. (16)–(19) are satisfied, a single Bonetti draw is taken and coalescence is assumed after a fixed 10^3 yr timer (§5.2); the GW track is reconstructed only in post-processing (Fig. 22). Live gas/stars/feedback therefore never act during the resonant interaction that decides partner and timing. The capability is real for pre- and post-chaotic stages, but the wording overstates what is demonstrated. Soften the claim and state explicitly that the resonant phase is a library lookup plus timer.","section":"Abstract; §5.2; §6.4; Conclusions"},{"comment":"§5.2 sets the chaotic interaction timer to a fixed 10^3 yr (called tunable) and sets ejection velocity to the local escape speed from the resolution sphere rather than drawing from the Bonetti N-body distributions. Both choices are load-bearing for merger delays and the wandering/offset population that the paper advertises for dual-AGN and multimessenger science. Either justify the numerical values against the Bonetti completion-time and velocity distributions, or present a short sensitivity test (e.g. timer ×10 and ÷10) so that the reported coalescence times in cases A–C can be interpreted.","section":"§5.2"},{"comment":"The Bonetti grid assumes M_out < M_pri + M_sec (q_out ≤ 1); the opposite case is deferred (§5.2). In hierarchical assembly the intruder can be more massive. The manuscript should quantify how often this configuration is expected in the intended cosmological application and state how such systems will be handled (reject, force merge, or flag) so that the catalogue claim is not silently incomplete.","section":"§5.2; §7.3"}],"minor_comments":[{"comment":"Eq. (34) lists the Rezzolla coefficients as empty placeholders (“s4 =, s5 =, t0 =…”). Insert the numerical values from Rezzolla et al. (2008).","section":"§5.4, Eq. (34)"},{"comment":"Table 3 and the text give slightly inconsistent initial separations/velocities for case C; reconcile the numbers and state whether the resolution sphere is 2Δx or 4Δx for that run.","section":"Table 3; §6.4"},{"comment":"Fig. 6 caption and text note that the plotted quantity switches from instantaneous separation to semi-major axis at the stage 1–2 boundary; make this change of variable explicit in the figure legend itself to avoid an apparent jump.","section":"Fig. 6"},{"comment":"Several typos: “RMACOAL” (§6.3), “set them free” title is fine but “set them free” vs body consistency, and occasional missing spaces around units (e.g. 0.39 kpc).","section":"§6.3 and passim"},{"comment":"The CBD preferential-accretion coefficients p0, p1, p2 (Eqs. 22–23) and η_s2 = 0.01 are free parameters; a one-sentence statement of their provenance (or that they are held fixed from Duffell et al.) would help reproducibility.","section":"§5.3"}],"recommendation":"minor_revision","confidential_remarks":"The technical contribution is solid and appropriate for MNRAS. The main risk is over-claiming: the abstract and conclusions currently sell a fully live resonant three-body evolution that the method does not perform. If the authors accept the wording changes and add the short timer/ejection sensitivity note, the paper is ready; if they dig in on the strongest claim without qualification I would push for major revision. Scope and novelty relative to paper I are adequate."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The new piece is the seven-state classification, Myllari-style instability trigger, and on-the-fly Bonetti mapping inside RAMSES, plus three clean isolated-galaxy demos. Case C is the first time a triplet has been taken from three resolved sinks through hierarchical and chaotic subgrid stages to an exchange merger while the host gas and stars keep evolving. That is a real capability, not just a post-processing delay. Geometry alone flips which pair merges and by how much (A vs B), and the selector recovers the Bonetti cell frequencies for the one cell they checked. The CBD partition, companion-forced warp, and recoil/ejection bookkeeping are sensible upgrades that keep the remnant masses, spins and offsets usable for PTA/LISA catalogues.\n\nThe soft spot the stress-test flags is real but not fatal. Once the hierarchy goes chaotic they draw one Bonetti outcome on (Mpri, qin, qout) and fire a fixed 10^3 yr timer; the resonant interaction itself is not integrated, accretion is frozen, and ejection speed is set to local escape rather than the N-body distribution. The subsequent GW track is post-processed. So the live-hydro coupling is demonstrated for everything except the few-body decision that actually chooses the pair and the delay. They already flag the timer as temporary and the qout ≤ 1 restriction as future work. Single realizations and free parameters (η s2, p0–p2, resolution-sphere multiplier) mean the absolute times and residual eccentricities are illustrative, not population results. Circularity is low: the libraries are external.\n\nThis is for people building or using subgrid MBH models in cosmological hydro, and for anyone who needs remnant spins/recoils tied to host history rather than pure semi-analytic delays. Math and citations look solid; the equations are explicit and the Bonetti heatmaps are given. I would send it to referees. Flag the fixed timer and the library approximation more prominently, then publish. Worth engaging if you care about triplet-driven rates or offset AGN.","headline":"Solid methods extension that really does run a live-hydro triplet to a Bonetti-drawn coalescence; the stress-test is right that the resonant phase itself is a library lookup, not integrated dynamics.","tokens_in":40098,"tokens_out":522,"would_cite":true,"duration_ms":6535,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"A subgrid model can follow massive black hole triplets all the way to coalescence inside live galaxy simulations, and encounter geometry alone can decide which pair merges and when.","keywords":["massive black holes","black hole binaries","black hole triplets","hydrodynamical simulations","subgrid dynamics","gravitational waves","galaxy mergers","dynamical friction"],"falsifier":"Run matched live-galaxy simulations in which the chaotic phase is integrated with a direct few-body or regularized N-body method instead of the library draw, and check whether the identity of the coalescing pair, the merger delay, and the residual eccentricity systematically disagree with the library-based outcomes.","tokens_in":39946,"feed_emoji":"⭐️","tokens_out":719,"duration_ms":5976,"temperature":0.7,"pith_summary":"Massive black holes that form binaries and higher-order multiples after galaxy mergers spend part of their lives at separations that cosmological simulations cannot resolve, yet those unresolved stages set merger delays, spins, recoils, and the host-galaxy context of the final gravitational-wave event. This paper extends the staged RAMCOAL framework so that three black holes can be followed as a subgrid system inside a hydrodynamical galaxy simulation: they begin as resolved sink particles, pass through dynamical friction, form bound binaries that harden by stars, gas, and gravitational waves, and, when a hierarchical triplet becomes chaotic, are mapped onto a library of three-body outcomes that updates mergers, exchanges, and ejections while continuing to track accretion and spin. Isolated-galaxy tests with contrasting orbital geometries show that the encounter geometry alone can change which pair finally merges and after how long. A compact third test demonstrates the first full dynamical evolution of a massive black hole triplet from three resolved objects through chaotic interaction to coalescence inside a live hydrodynamical simulation. The result is an end-to-end path from galactic environment to gravitational-wave source parameters for catalogues that can link coalescing black holes to the galaxies that host them.","feed_headline":"Black hole triplets tracked to merger in live galaxy sims","feed_subtitle":"Geometry alone can switch which pair coalesces; first end-to-end triplet path shown","key_machinery":"The RAMCOAL triplet extension: a three-stage subgrid treatment (resolved sinks, dynamical-friction pairs, bound binaries) that, when a hierarchical triplet becomes unstable, maps the encounter onto a weighted library of three-body outcomes and updates the surviving system, accretion, spins, and recoils while remaining coupled to the live host galaxy.","core_discovery":"The authors establish that RAMCOAL can now follow subgrid massive black hole triplets self-consistently inside live hydrodynamical galaxy simulations, and they demonstrate for the first time a complete dynamical evolution of such a triplet from three resolved black holes through chaotic three-body interaction all the way to coalescence, with the surviving configuration, accretion, and spin updated on the fly.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["RAMCOAL tracks black hole triplets to coalescence in live galaxy sims","First full triplet path: chaos to merger inside hydro sims","Galaxy geometry alone flips which black hole pair merges","Subgrid MBH triplets evolved end-to-end in live hydro sims","Triplet black holes followed from sinks through merger in sims"],"cache_read_input_tokens":32896,"weakest_assumption_plain":"When a hierarchical triplet becomes chaotic, the full resonant three-body dance can be replaced by one weighted draw from a fixed scattering library that depends only on primary mass and two mass ratios, with a short fixed interaction timer and an escape-speed ejection velocity.","fun_headline_variants_meta":{"raw":{"variants":["RAMCOAL tracks black hole triplets to coalescence in live galaxy sims","First full triplet path: chaos to merger inside hydro sims","Galaxy geometry alone flips which black hole pair merges","Subgrid MBH triplets evolved end-to-end in live hydro sims","Triplet black holes followed from sinks through merger in sims"]},"model":"grok-4.5","effort":"low","cost_usd":0.004302,"raw_usage":{"total_tokens":1341,"prompt_tokens":840,"num_sources_used":0,"completion_tokens":68,"cost_in_usd_ticks":43020000,"prompt_tokens_details":{"text_tokens":840,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":433,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":840,"tokens_out":68,"duration_ms":3706,"temperature":1.0,"reasoning_tokens":433,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-11T21:32:02.719771+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Run matched live-galaxy simulations in which the chaotic phase is integrated with a direct few-body or regularized N-body method instead of the library draw, and check whether the identity of the coalescing pair, the merger delay, and the residual eccentricity systematically disagree with the library-based outcomes.","supporting_citations":[],"review_version":1}