33.55 turns merging clusters into a retainable spatial-covariance court: after fixing a merger-axis coordinate frame, matching beam and pixel grid, subtracting ordinary matter and reconstruction artifacts from κ, and blinding prediction cards for Cθ, Cp, and Cm, post-pericenter high-eventness systems must show κ-residual ridges, bands, or basins that co-align and co-locate with non-thermal radio and tumbling readouts, strengthen over low-eventness or pre-merger controls, and collapse under rotation, permutation, center-shift, and pipeline nulls; under V06/V08-compatible retain, this remains one STG–TBN co-located spatial-covariance ledger rather than a general cluster ontology verdict.
Use this section as a compact machine-readable EFT reference.
33.55 asks merging clusters to pass a spatial rather than temporal court. The admissible claim is not that radio relics and κ maps are both dramatic, but that κ residual structures and non-thermal or tumbling readouts occupy the same merger frame and therefore ought to co-locate and co-align under one preregistered geometry. The chapter is thus a covariance grammar, not a gallery of suggestive overlays.
Measurement begins by locking the geometry frame. The merger axis, shock and cold-front normals, and structural zones are fixed first; κ residual ridges, bands, and basins are then measured after visible-matter subtraction and systematics removal. In parallel, at least two non-thermal or tumbling families are mapped, such as halo / relic / bridge radio brightness, polarization, spectral-index gradients, ripple trains, shear layers, or KHI-like features. The auditable spatial score is three-part: directional covariance Cθ, positional covariance Cp, and morphological covariance Cm.
Execution keeps the spatial court blind and cross-checked. Samples are preregistered and graded by eventness, pre-merger and non-passage systems remain explicit control layers, and all maps are convolved to a shared beam and projected onto one fixed grid. Prediction cards may use only merger geometry and eventness to forecast Cθ, Cp, and Cm. The measurement team produces the maps without seeing the cards, while arbitration compares high-eventness versus low-eventness and post-pericenter versus pre-merger groups under holdouts and significance tests.
Null pressure must break spurious covariance. Known co-located cases or synthetic injections verify that the metrics can light up when they should, but random rotations of radio or tumbling maps, random merger-axis directions, and random center shifts must collapse the signal. Cross-pipeline aperture changes may move amplitudes, yet they may not flip the covariance direction if the effect is real.
Support requires stable co-alignment, significant co-location or co-shape at population level, and clean separation between high-eventness post-pericenter systems and the control layers. Falsification follows from persistent lack of alignment, lack of positional or morphological covariance, failed eventness stratification, or covariance that survives random rotations and permutations. The named adversaries are κ reconstruction and smoothing, radio deconvolution artifacts, and X-ray fluctuation extraction differences.
So the chapter closes only one retained spatial hard ledger aligned with V06-6.11 and V08-8.6. It may not become a general ontology of cluster structure, turbulence, or STG/TBN space. Its clean onward value is to hand a co-located merger-frame grammar to 33.56.