OCS Research · Preprints
The Omega Centauri Research Papers
Eleven companion preprints (Swanson 2026) develop and test a single idea: that the thermodynamics of computation, rather than the urge to expand, predicts where the oldest technological civilizations end up, and that Omega Centauri (NGC 5139) is the most accessible place to look. Papers F–H and the AXI note extend the observational case: a joint accretion bound, an X-ray source census, and a pre-registered analysis of the contested central dark mass with its methods companion. Read each abstract below, or open the full paper.
Paper A · The hypothesisThe Macro Transcension Hypothesis: Spinning Black Holes in Dense Stellar Clusters as Thermodynamic Attractors for Advanced Civilizations, with Omega Centauri as an Observational Test Bed
AbstractMost proposed resolutions of the Fermi paradox assume that long-lived technological civilizations either expand outward, perish, or deliberately hide. We develop a fourth alternative, the Macro Transcension Hypothesis (MTH): that civilizations which optimize for long-term computation are driven by thermodynamics, rather than preference, toward a specific class of astrophysical environment, namely rapidly spinning massive black holes embedded in dense, old stellar systems, and that the migration and its endpoint are both electromagnetically quiet. The MTH extends the transcension hypothesis of Smart (2012) from planet-scale "inner space" to macroscopic black-hole infrastructure, and differs from the aestivation hypothesis of Sandberg, Armstrong & Ćirković (2016) in requiring no waiting strategy: the relevant free-energy and entropy-disposal advantages are available now. We quantify the case in four steps: (i) thin-disk accretion onto a Kerr black hole releases 5.7–30 per cent of rest-mass energy across the spins disk accretion can reach, rising to 42 per cent only in the formal a★ → 1 limit that radiative spin equilibrium forbids, versus 0.5 per cent for complete fusion of stellar material (0.7 per cent for hydrogen alone), while magnetically arrested disks extract additional spin energy at effective efficiencies exceeding 100 per cent of accreted rest mass; (ii) by the generalized second law, an event horizon is a thermodynamically ideal entropy sink, and a worked delivery budget (Section 4.2) shows the realized erasure cost is of order the CMB-limited Landauer floor (a gain of ~0.05 on the conservative loss budget) if the carrier modes are thermally occupied by the 2.7 K sky, and lands a factor of ~103 below it in a fed flow (~107 in the unfuelled ambient medium) only if the carrier modes are kept in the vacuum state (Neff ≪ 1; T ≪ 6 mK at 90 MHz, T ≪ 0.3 μK at 4.3 kHz) from the emitter to the horizon, which requires a sky-occluding aperture and a path not thermalized by plasma at Te ≫ 2.7 K; (iii) the Bekenstein–Hawking entropy of a ~2×104 M☉ black hole corresponds to ~6×1085 bits, or ≈3×1085 at the Thorne spin the architecture ends at, exceeding any material archive; and (iv) per unit of harvested mass, this architecture outperforms complete fusion of the same fuel by a factor of ~11–60, outperforms Dyson-type stellar harvesting without star lifting by ~115–600 in lifetime energy yield (~11–60 against a star-lifting economy), and exceeds a solar Dyson swarm by a factor of ~109 (6.6×108 at 2×104 M☉) in instantaneous Eddington-limited power. We then identify Omega Centauri (NGC 5139), a ~4×106 M☉, ~12-Gyr-old stripped dwarf-galaxy nucleus hosting the nearest strong candidate intermediate-mass black hole, as the most observationally accessible system satisfying the MTH selection criteria, and present a falsification framework built on six instrument-matched tests: accretion-luminosity limits (JWST, ATCA), mid-infrared waste-heat limits (JWST), a gravitational-wave signature inconsistent with a single point mass, an EMRI point-mass measurement, and an EMRI spin measurement (the latter three via LISA, contingent on a compact-object inspiral occurring in-band during the mission), plus neutrino burst searches (KM3NeT). Current data, including the unresolved tension between a ≥8,200 M☉ kinematic lower bound (Häberle et al. 2024) and a ≲6,000 M☉ pulsar-timing upper bound (Bañares-Hernández et al. 2025) and the complete electromagnetic silence of the central object (Mahida et al. 2026; Chen et al. 2025), are consistent with both the MTH and the more parsimonious gas-starvation null hypothesis; we state explicitly which forthcoming observations would discriminate between them, and which would falsify the MTH outright. The hypothesis is offered in the falsificationist tradition of Sandberg, Armstrong & Ćirković (2016) and Dvali & Osmanov (2023): a speculative but physically grounded working model whose value lies in the concrete observational program it motivates.
Full paper → Paper B · The reviewInward Resolutions of the Fermi Paradox: A Critical Review of Migration Down Thermodynamic Gradients
AbstractMost catalogued resolutions of the Fermi paradox modify one of three things: the abundance of technological life, its longevity, or its visibility. A smaller family modifies its direction: these are the inward-migration hypotheses, which hold that mature technological intelligence migrates down thermodynamic gradients, toward denser, serially faster, colder and more computationally efficient configurations of matter, and that the observed silence of the sky is the external appearance of that migration. The family comprises six distinguishable proposals, built on a two-decade precursor lineage from Dyson’s eternal-computation bound to Bradbury’s Matrioshka-brain engineering: the migration hypothesis of Ćirković and Bradbury, Smart’s transcension hypothesis, Vidal’s stellivore interpretation, the aestivation hypothesis of Sandberg, Armstrong and Ćirković, black-hole computing proposals from Inoue and Yokoo through Dvali and Osmanov, and the recent Macro Transcension Hypothesis. They share one load-bearing premise: that the thermodynamics of computation, rather than expansion, reproduction, or communication, is the correct lens for predicting the behaviour of the oldest intelligence. The family has not previously been reviewed as such, nor its mutual inconsistencies catalogued or its falsifiability graded. We reconstruct the family tree and its intellectual debts; restate the unifying physics (Landauer’s principle, the Margolus–Levitin bound, Bekenstein–Hawking entropy, and the temperature hierarchy of available entropy sinks) with explicit numbers; construct a comparative matrix of assumptions, energy logics, predicted observables, standing objections, and observational status for each member; and grade each against six falsifiability criteria. Against the sociological solutions that predict the same silence from different premises (zoo, dark-forest, and sustainability), the inward family’s distinguishing virtue is residue: thermodynamic optimization leaves dynamical and high-energy traces that fear and ethics do not. Open problems are stated as research questions: goal stability over 108-year horizons, migration economics under Bostrom-type opportunity costs, the incomplete-compliance gap, and population-level consistency with grabby-aliens selection effects. We close with the instrument-matched tests feasible or scheduled for each hypothesis over 2026–2040, with the cold-infrared and LISA legs named where neither applies, anchored by the first dedicated globular-cluster technosignature surveys and the multi-messenger campaign proposed for Omega Centauri.
Full paper → Paper C · The observational campaignA Multi-Messenger Technosignature and Anomaly-Detection Campaign for Omega Centauri
AbstractOmega Centauri (NGC 5139), the most massive Galactic globular cluster and the probable stripped nucleus of an accreted dwarf galaxy, presents a unique conjunction of observational circumstances: the strongest current candidate for an intermediate-mass black hole (IMBH) in the Galaxy, anchored by seven stars moving above the local escape velocity (Häberle et al. 2024); a formal inconsistency, a factor 1.37 apart, between the kinematic lower bound (≥8,200 M☉) and a pulsar-timing upper bound (≲6,000 M☉; Bañares-Hernández et al. 2025); no compact-accretion signature at the deepest radio and infrared limits yet placed on this cluster (its GeV emission being attributed to the millisecond-pulsar ensemble), and among the deepest placed on any globular-cluster core (Mahida et al. 2026; Chen et al. 2025); and a southern declination optimal for the newest southern-hemisphere facilities. To our knowledge no dedicated technosignature search of ω Cen has been reported at any wavelength. We present a coordinated, hypothesis-agnostic, multi-messenger campaign of eight instrument-matched programs (enumerated in Section 1) spanning infrared imaging, radio timing and SETI, astrometry, gravitational waves, neutrinos, gamma rays, the optical time domain, and archival channels, addressing conventional astrophysics (IMBH reality, mass, and spin; cluster dynamics) and technosignature hypotheses with the same data. For each program we state quantitative sensitivities, time requests, decision thresholds, and explicit falsification criteria, including negative results: with the acceleration estimator calibrated to the proper-motion precision the oMEGACat catalogue itself reports, direct astrometric acceleration detection stands at 0.21σ for a star at the 0.08 pc outer edge of the fast-star region, and stays below 1σ there before ~2040; at the seven stars' measured projected radii the forecast for 2×104 M☉ is 0.02–5.3σ per star (0.06–13σ at 4.9×104 M☉), an upper bound because projected radius bounds the acceleration from above, with only the innermost star above 1σ at the lower mass and its measured acceleration (0.011 ± 0.013 mas yr−2) already below the 0.068 predicted. The decision-grade astrometry therefore routes through photocentric-wander and reference-frame measurements instead. Total cost is ≲ US$10M over 2026–2040, of which the cash lines are dominated by archival analysis and commensal observing, though the facility request includes ~650 h of dedicated MeerKAT time, 65 h of JWST and 50 h of CTAO. A per-mille mass and spin measurement is available from LISA if an inspiral is caught in band, at a target-specific probability of ~3×10−7 over a four-year mission; the probable LISA deliverable is instead a resolved-source count for the cluster’s remnant population. Every null result constrains conventional astrophysics, and no anomaly claim advances without confirmation from at least two independent messengers; under realistic outcomes the campaign adjudicates the astrophysical hypotheses, while the technosignature hypothesis is constrained only along specific low-probability branches.
Full paper → Paper D · The economicsThe Economics of Inward Migration: Relocation versus Densification for Computation-Maximizing Civilizations
AbstractThe inward-migration resolutions of the Fermi paradox hold that computation-optimizing civilizations relocate to thermodynamically privileged environments, with rapidly spinning intermediate-mass black holes (IMBHs) in dense old clusters as the strongest candidate destination. Bostrom (2003) priced the opportunity cost of delayed expansion and Bennett et al. (2019) showed that dormancy confers no thermodynamic advantage, but relocation itself, the abandonment of accumulated local infrastructure for a transit of 105–106 years toward a large deferred payoff, is one for which we are not aware of a prior pricing. This paper treats that trade as a decision problem over a common utility (discounted integrated computation) with three strategies: stay-and-densify (Matrioshka-style local engineering), migrate (beamed-sail relocation to the nearest suitable IMBH cluster), and seed-and-stay (a self-replicating seed dispatched while densification continues at home). Using payoff kernels assembled from established physics, we derive closed-form crossover conditions. The central result is a threshold on the effective discount-plus-hazard rate: migration dominates densification whenever ρ + λ < ln(G ps)/τ, provided G ps > 1, where G is the destination computation-rate multiplier (103–109 on power alone, from an ADIOS-suppressed floor to full Eddington capacity, depending on fuel imports), ps the transit survival probability, and τ the transit-plus-construction time. At fiducial parameters the threshold is ρ + λ ≲ 2×10−4 yr−1: any lineage whose combined discount-plus-hazard half-life exceeds roughly 3,500 years should migrate. Under additive utility, symmetric valuation, and a seed fidelity within the tolerance (1 − f)(G ps)x−1 of the migrant’s, with x = (ρ + λ)/ρ* (about 4.5×10−5 at x = 0.5, rising to unity at the threshold and falling to 2×10−9 as ρ + λ goes to zero), the seed-and-stay hybrid dominates both pure strategies throughout the migration-favourable region and in a band extending modestly beyond the threshold (to ρ + λ ≈ 1.7 ρ* at a physical seed cost f = 10−6), because seed mass is a negligible fraction of local output; beyond that band the exponential discount on the deferred payoff extinguishes the seed's value. Embedding the decision rule in a mixed population quantifies how much inward migration thins the expected loud population, and yields a new population-level residue: the predicted sky ratio of Matrioshka-type infrared sources to quiet-cluster systems, for which archival infrared nulls supply the joint-likelihood frame for a limit conditional on an assumed residue-to-formation ratio and maintenance lifetimes. All results are conditional on the optimization premise shared by the hypothesis family, on symmetric valuation of drifted successors, and on power-only destination accounting; the contribution is the pricing structure, not the premise.
Full paper → Paper E · The engineering and the adjudicationEngineered Intermediate-Mass Black Hole Systems: Infrastructure Constraints, Observable Residue, and a Multi-Messenger Adjudication Framework
AbstractThe framework developed here cannot adjudicate an engineered black-hole system (Heng) on any electromagnetic null. Its mid-infrared channel is one-sided, so its evidence saturates at ln fd = −0.69 however deep the photometry goes, and without a detection it can never favor Heng; the cap depends on the dormancy prior (ln fd = −2.3 at fd = 0.1). Rapidly spinning intermediate-mass black holes (IMBHs) in dense, old clusters are proposed destinations for computation-optimizing civilizations. We ask whether infrastructure can persist around a 2×104 M☉ Kerr IMBH in the ω Cen core. A flyby Monte Carlo gives an allowed envelope from ~102 gravitational radii to the stripping radius (~4×103 AU). Unbound flybys never set the boundary; bound-remnant diffusion does, with a floor of a few 107 yr when a remnant is present. With outflow suppression engineered away, ambient fueling supports up to 7.6×105 L☉ after a thin-disc spin-up and ≈103–4×103 L☉ after a magnetically arrested (MAD) spin-up, which also leaves a★ ≈ 0.035–0.07, within 1σ of the natural-capture null and below the T5 trigger a★ < 0.1. Bayes factors against an explicit null menu score three forward-modeled observables: a waste-heat floor, MAD regulation, and EMRI dephasing. On current data and forecast JWST depths the mid-infrared channel returns ln K = −0.47 and the engineered-mass channel −0.03, a total of −0.50. It registers the 1967 pulsar discovery as a strong anomaly (ln K = +7.9), collapses it once the resolving hypothesis joins the menu, and gives no false positive on two mundane cases; one pre-registered criterion fails and is reported. Results are conditional on the optimization premise and on unrefereed companion-paper inputs.
Full paper → Paper F · The accretion limitA Radio and X-ray Bound on Bondi-fed Accretion onto the Candidate Intermediate-Mass Black Hole in Omega Centauri, with an Infrared Consistency Check
AbstractThree deep non-detections constrain accretion onto the candidate intermediate-mass black hole (IMBH) in Omega Centauri: 170 hours of ATCA radio imaging at 1.1 μJy rms (Mahida et al. 2026), JWST imaging with no accretion-like point source (Chen et al. 2026), and a 291-ks Chandra limit LX ≤ 1.6×1030 erg s−1 (Haggard et al. 2013). We build a forward model from the Bondi radiative efficiency εB ≡ Lbol/(ṀBc2), the mass and eight shared nuisance parameters, and report the 95 per cent credible upper limit ε95 for three accretion-flow families. For a radiatively inefficient flow, ε95 = 3.0×10−10 at 8,200 M☉ and 1.1×10−11 at 4×104 M☉. These limits are conditional on the fundamental plane; without the radio leg they relax to 1.4×10−8 and 8.1×10−10. The infrared leg never binds, so this is a radio-plus-X-ray bound with an infrared consistency check. Against a natural outflow-suppressed hot flow at a 47 Tucanae-like gas density, the data reject 85, 77 and 99 per cent of the parameter space at 8,200 M☉, the pulsar-timing cap of 6,000 M☉ and 4×104 M☉ (28, 20 and 76 per cent without radio). At one tenth of that density, just below the 0.03–0.08 cm−3 wind-fed estimate, the figures fall to 57, 47 and 94 per cent (4, 2 and 32). The verdict applies the lowest branch of an efficiency fit three to four decades below that branch’s upper edge, and a pulsar dispersion-measure determination of the density is foreground-limited until the sight-line scatter is modelled down. Block-by-block analysis of the radio campaign excludes 100 μJy, 2-hr flares recurring faster than every 31 hr, and leaves 1 mJy half-hour flares unconstrained beyond 3.4 days.
Full paper → Paper G · The X-ray censusA Residual Census of Unidentified X-ray Sources in Omega Centauri after the Modern Pulsar Catalogue
AbstractThe Chandra astrometric frame of Omega Centauri sits +0.42″ in declination from the radio timing frame, which is comparable to the 0.36″ to 0.49″ tier-1 match radii. Registering it takes tier-1 matches of the 233-source catalogue of Henleywillis et al. (2018) against the 19 pulsars of Colom i Bernadich et al. (2026) from three of eight to five of eight. Five registered associations lie 0.02″ to 0.34″ from their pulsars, with chance-coincidence probabilities below 2.3×10−3 under each of three estimators, reported separately. They are also the anchors that fix the registration, so they are not independent of it. MSP H is not an anchor: its timing position lies 0.21″ from the X-ray source h1 of Zhao & Heinke (2023) (two-coordinate frame transfer, Section 5.3) and 1.83″ from Henleywillis et al. (2018) source 14c, whose position matches the thermal source h2 to within 0.1″. A 47 Tucanae control run first recovers 18 of 21 known counterparts. From the 39 unidentified sources inside a colour-flux region defined by the confirmed counterparts (15 to 39 under anchor leave-one-out) we publish a follow-up list ranked under a rule fixed before the ranking was computed. The residual of unidentified sources above the flux limit, against a log N–log S background, is −1.9 (−70.5, +52.6), consistent with zero. It sets the scale of what a follow-up can find and is not a detection.
Full paper → Paper H · The mass tensionThe Contested Central Dark Mass of Omega Centauri: A Pre-registered Joint Analysis of Fast Stars, Pulsar Timing, and the Limits of the Proper-Motion Dispersion Profile
AbstractThe data cannot yet decide between a compact and an extended central dark mass in Omega Centauri. Fast stars require a dark mass of at least 8,200 M☉ (Häberle et al. 2024), while joint kinematic and pulsar-timing modeling favors an extended ~2–3×105 M☉ component with a point-mass cap of 6,000 M☉ (Bañares-Hernández et al. 2025). We report a pre-registered joint likelihood analysis of the fast stars, the 19-pulsar TRAPUM timing set (Colom i Bernadich et al. 2026) as one-sided bounds, and the oMEGACat dispersion and anisotropy profiles (Häberle et al. 2025). The dispersion profile rejects every smooth spherical non-rotating model, with statistical errors ten to seventy times below its systematic budget, so a pre-set gate retires it. The pulsar accelerations span 3.0 nats. Pulsar jerks cannot discriminate at 19 pulsars: with the published nearest-neighbour jerk constant corrected (ξ = 3.4596, not 3.04), the floor is a 1-stable process, so discrimination needs an indicative census of order 102 timed pulsars (simulated at 10−21 m s−3), and at 19 pulsars the verdict is flat from 10−22 to 10−19 m s−3. The fast stars and pulsars jointly prefer a compact solution (Mdark ≈ 2.0–2.5×104 M☉ at the point-mass limit, ln K = +10 to +12), quotable only as a calibrated 90 per cent consistency statement (injection coverage 0.8825 extended, 0.865 point). It holds at the fiducial mass-to-light bracket, under a Bahcall–Wolf tracer cusp (at the shallower published slope the all-legs injection check fails, coverage 0.660; no leg-dropped run exists), under the TRAPUM pulsar reduction (the earlier reduction reverses the sign of ln K), and for the speed-tail family (a hard-truncated isothermal tail lowers ln K by 0.17–0.39 nats across the four leg-dropped configurations). The profile-leg tension is unresolved.
Full paper → AXI note · Methods companion to Paper HThe Component-Differential Residual in the Omega Centauri Proper-Motion Dispersion Profile: A Pre-Registered Campaign That Cannot Identify Its Mechanism
AbstractA fit of the oMEGACat proper-motion dispersion profile of ω Cen under a shared smooth discrepancy in the mean leaves a component-differential residual: the model under-predicts the sky-radial dispersion and over-predicts the sky-tangential one, coherently, at every radius and every reading of the grid (Swanson 2026f). This note records a campaign pre-registered to identify the mechanism behind that residual (Swanson 2026c), run to its stopping rule, and closed on its own pre-registered outcome that no mechanism can be named. Four hypotheses were specified in advance with separating discriminants: anisotropy structure beyond one Osipkov-Merritt scale, flattening with azimuthal averaging, unrelaxed accretion-origin substructure, and component-dependent measurement systematics. Stage 0 could run two of the four on the frozen radial table, and both discriminants failed at all twelve residual-evaluation points; the flattening and substructure hypotheses were not tested. Stage 1 fitted M1, a component-resolved discrepancy model that decomposes the shared spline into a shared term δS and a differential term δD on the same four knots. M1 clears the per-component adequacy test the shared model failed, at all 36 readings, with a minimum p of 0.0301 over 144 tests. Two other gates go against it. Cross-axis stability fails on one of five pre-committed statements, through a discrete switch between the differential and the white systematic term that moves with the δD prior width; and the budget check fires on both of its scoring conventions, first at the 201.15″ bin (≈199″ interpolated), by up to a factor of 3.8, on the as-coded |δD|-versus-band convention, or first at the 142.43″ bin (≈137″ interpolated), by up to a factor of 7.5 over all 36 readings, on the convention that follows the equipartition band from which the budget was derived. The calibrated-region boundary itself, a separate threshold on δD alone, is crossed earlier, near 134″ interpolated. Under the pre-registered consequence clauses this makes decision criterion 4, "the campaign cannot identify the mechanism", the only quotable outcome. The measured object that survives is the δD amplitude, 0.16 to 0.45 km s−1 over 36 readings, positive everywhere and inside the calibrated region everywhere, with a median split of 0.29 to 0.33 km s−1 inside the knot span against the 0.27 to 0.31 km s−1 that stage 0 measured independently. Escalation to an axisymmetric model is forbidden by the budget gate's own clause. Paper H is untouched throughout.
Full paper → XI note · Erratum companion to Paper HAn Erratum to the Jerk Noise-Floor Constant in Prager et al. (2017) and the works that adopt it
AbstractPulsar spin-frequency second derivatives, jerks, probe cluster gravitational potentials and central masses against a stochastic floor of nearest-neighbour stellar encounters, whose characteristic amplitude is ȧ0 = (2πξ/3) G⟨m⟩σn, with n the stellar number density, σ the velocity dispersion, and ⟨m⟩ the mean stellar mass. Prager et al. (2017) quote the dimensionless constant ξ ≃ 3.04, and Abbate et al. (2018, 2019) adopt it. The double integral that defines ξ evaluates in closed form to √(2π) [1 + ln(2+√3)/(2√3)] = 3.4595806610: five evaluations of that one integral (a closed-form reduction, adaptive quadrature, symbolic evaluation, antiderivative differentiation, and high-precision quadrature) plus two independently coded Monte Carlo simulations of the underlying stellar field that never evaluate the integral, at 15.8σ and 14.6σ respectively, both reject 3.04; Section 3 states the common-mode caveat on the first five. Because ȧ0 is linear in ξ, the printed constant understates the jerk floor by 12.1 per cent, inflating the apparent significance of jerk-based central-mass inferences.
Full paper → Paper I · Gas-density provenance · v0.11 working draftThe Gas-Density Provenance of Radio Intermediate-Mass Black-Hole Limits in Globular Clusters: An Audit and a Recompute
AbstractRadio searches for intermediate-mass black holes (IMBHs) in Galactic globular clusters convert a flux-density limit into a mass limit through an assumed intracluster gas density, and the field-standard value is 0.2 cm−3. This paper audits where that value comes from and recomputes the limits with a gas prior for each cluster. The audit traces the 0.2 to four millisecond-pulsar dispersion measures in M15 (Freire et al. 2001), which Strader et al. (2012) adopted as a prior and Tremou et al. (2018) applied to all 50 clusters, with their assumed density n = 0.2 cm−3 corresponding to an electron density of half that value. Later users cite the 47 Tuc measurement (Abbate et al. 2018; Wrobel and Nyland 2020) alongside the root papers. The recomputation uses a measured prior for 47 Tuc, a Zhang et al. (2025) prior for five clusters and a one-sided upper-envelope prior for the remaining 44; Section 4.3 gives the amendment and original-scope result. In all three headline scenarios no cluster has a recomputed 95th-percentile limit below 103 M⊙, against 12 of 50 for the printed 3σ point limits, which are evaluations at the assumed density rather than percentiles; the tighter [T3/10, T3] variant leaves M15 alone below the threshold, at 537 M⊙, under an envelope above the published M15 upper limit. A pulsar-timing control on 47 Tuc failed in its pre-registered form, which removes the pipeline route on M15 and leaves the radio route as the operative constraint, with the X-ray route as corroboration.
Full paper → Paper J · Gaia ejecta test · v0.8 working draftA Gaia Test of Published and Preprint Predictions for Stars Ejected from Milky Way Globular Clusters
AbstractTwo models predict populations of stars ejected from Milky Way globular clusters by encounters with stellar-mass compact objects: Evans et al. (2025) predict the population in aggregate with Gaia detectability applied, and Cabrera and Rodriguez (2023) publish per-cluster sky and proper-motion credible regions for the whole-lifetime ejecta. We register a blinded count test of both predictions in Gaia DR3 and, under the pre-registered switch rule, in DR4. The test asks whether the pre-declared predicted regions hold the predicted excess over a field background estimated from rotated control regions. A blinded power gate, evaluated on control-region counts and template-derived quantities before any signal-region count is opened, treats a gate stop as a reportable outcome. The first result is a bound that needs no signal model. Summed over clusters, the squared signal-to-background ratio is at most 1302/(2.0×106) = 8.4×10−3, so the expected Z0 under the alternative is at most 0.09, and at the registered Z0 ≥ 3 threshold the design has 0.18 per cent power against a 0.13 per cent null tail, for any distribution of the roughly 130 predicted Arm C stars across clusters. The second result is a defect in the Cabrera and Rodriguez archive. Its per-cluster ejecta tables store a literal sentinel in every position and velocity field of the recent 500 Myr window, which the pipeline’s own transform maps to a single sky position that the mock sky exclusion removes, so the preregistered Arm C validation cannot be scored on the archive as published. Independently, the registered information floor asks for 118 detectable stars in a single cluster at the median background of the 148-row archive record against an arm-wide predicted total of about 130. Arm E could not be run either. Its template model was never implemented, and its one testable region, the NGC 7099 box, fails the registered control-validity rule through the same absolute fdrop defect that affects Arm C. No power result exists for Arm E, and no signal-region count has been computed for either arm. The Methods section records the registered design with each number traced to its source line, and the deviations table records both Arm C findings in the eleven-field form.
Full paper →