S16 EXTERNAL TRANSFER — IMPLEMENTATION COMPLETION

Primary survey pair: AK_165 (1945-46) and AK_550 (1965), retained exactly as selected by the pre-outcome metadata-only procedure.
Primary scalar channel: MAG_DGRF in each survey, because the archive documents this as each survey's DGRF-corrected magnetic scalar. MAG_COMP/MAG_MERGE are not used for the primary test because archive documentation states those channels were tied/merged to statewide compilation grids.
Coordinates: supplied X/Y Alaska Albers coordinates in meters; both surveys use the same documented archive coordinate convention.
Frozen radii: 6, 8, 10, 12, 16, 20 km.
Quadrant rule: strict E/N quadrants relative to each center.
Quadratic operator: unweighted OLS F(E,N)=a+bE+cN+dE^2+eEN+fN^2 with E,N centered in km; H=[[2d,e],[e,2f]].
Signed-frame comparison: negative-curvature eigen-direction compared axially (modulo 180 degrees); determinant sign/topology reported separately.

Center-grid completion rule:
The original frozen S16 text did not specify a finite lattice for 'all mechanically eligible centers'. Before outcome-based selection or optimization, the primary implementation fixes a 1-km lattice anchored to the documented Alaska Albers coordinate origin over the common X/Y data bounding box. The 1-km choice is archive-native in scale (the metadata describe the Alaska Composite Grid as a one-kilometer grid). Every lattice center is then filtered only by the frozen >=20 observations + four-quadrant rule in BOTH surveys. No center is selected by magnetic outcome.

Important status:
Because the center lattice was not explicit in the original S16 freeze, this is an implementation completion, not a verbatim pre-specified parameter. The primary conclusion is therefore reported with this caveat. A 2-km and 4-km lattice diagnostic is included only as post-outcome robustness and is not used to choose a preferred result.
