Interactive companion to Paper H's joint fit of ω Cen's central dark mass. Shows the mass–extension (M, a) posterior region for four verdict configurations (tracer profile × assumed distance), the 16-row ln K sensitivity ladder those configurations sit on, and the G3 formation-channel buildability overlay - all at the paper's fiducial bracket.
Embargoed to the fiducial bracket - 90% regions only
Every number on this page is the paper's fiducial-bracket, 90%-credibility output (HPD90 / PL90). This tool never asserts the tension is resolved - Paper H's own conclusion is that the joint data cannot yet decide between a compact and an extended central dark mass. The ln K rows with an error-model dependence are labelled "error-model sensitivity rows," matching the paper's own label, not "profile-leg" or any other shorthand. Full method and caveats: Paper H, full text · PDF · AXI methods note.
Verdict configuration
Posterior plane - M vs. extension parameter a (fiducial bracket, 90%)
M (M☉, x-axis) vs. a (dimensionless extension parameter, y-axis, as defined on the paper's shipped result grid)
G3 buildability overlay - consistency check only, never a likelihood term
557
Buildable, pessimistic remnant IMF
153
Buildable, optimistic remnant IMF
140
Not buildable (either IMF)
111
Literature silent
Bañares-Hernández et al. 2025 (arXiv:2408.00939, A&A 693, A104) favour an extended central mass of 2–3×10⁵ M☉ at a 1.5–2.2 pc scale, emulating a cluster of heavy stellar remnants - alongside their independently reported 3σ IMBH upper limit of 6,000 M☉ from pulsar timing. This overlay is a formation-physics consistency check on candidate (M, a) regions; per the paper, it never enters any likelihood or prior.
What this tool shows
Paper H ran a joint fit of ω Cen's central dark mass against four tracer-profile / assumed-distance configurations (Plummer or αβγ tracer profile, at 5,200 pc or 5,494 pc), reporting each configuration's 90% HPD (highest posterior density, r_a marginalised) and 90% profile-likelihood (r_a profiled) region in the (M, a) plane, and a 16-row ln K sensitivity ladder spanning the same four configurations plus knot-prior and error-model variants. This tool renders that shipped output directly (fH_posterior_plane.json, fH_lnk_ladder.json, fH_g3_overlay.json) - it recomputes nothing.
Reading the ln K ladder
Rows marked error-model sensitivity row use a profiled error-model leg (use_profile: true); this is the paper's own row label, kept verbatim here. Rows without that tag use the mandatory no-profile-leg companion setup. The ladder's spread across knot-prior variants (τ=0.5, 1.0, 2.0) is itself part of the paper's reported sensitivity - not a single point estimate.
What this tool deliberately does not do
Does not assert the mass tension is resolved. Paper H's conclusion is that current data cannot decide between a compact and an extended central dark mass at the fiducial bracket.
Does not plot the G3 buildability overlay spatially on the (M, a) plane - the shipped data gives verdict counts and one favoured-region summary in different units (mass and parsec scale), not a full outline on this plane, so it is shown as a text panel instead of a fabricated spatial overlay.
Does not quote any single-sentence "consistency statement" from the paper outside its full context - see the paper page for that language in place.