The manuscript (paper/manuscript.qmd) now carries the full arc measurement → depth → stress as its spine, but Parts 2 (Depth, #sec:depth) and 3 (Stress + attribution, #sec:stress) are written as literature-grounded framework sections with no result figures (by design, this pass).
Deliverable: once the modules land (uq_depth multi-band inversion, uq_density/#N1, uq_stress_depth/#5, uq_attribution/#7), replace the framework prose with real, truth-known synthetic results:
- a δV_S/V_S(z) + δρ/ρ(z) depth-profile figure with propagated C_m(z);
- an effective-stress/strain(z) figure showing the aleatoric⊕epistemic budget;
- a depth–source attribution chart (shares A_k(z) with credible intervals + coupling slice).
Then rebuild the PDF (part of #N3). Keep the best-practice/deviation tables (tab:bp-measure, tab:bp-depth, tab:bp-stress) already added.
The manuscript (
paper/manuscript.qmd) now carries the full arc measurement → depth → stress as its spine, but Parts 2 (Depth,#sec:depth) and 3 (Stress + attribution,#sec:stress) are written as literature-grounded framework sections with no result figures (by design, this pass).Deliverable: once the modules land (
uq_depthmulti-band inversion,uq_density/#N1,uq_stress_depth/#5,uq_attribution/#7), replace the framework prose with real, truth-known synthetic results:Then rebuild the PDF (part of #N3). Keep the best-practice/deviation tables (
tab:bp-measure,tab:bp-depth,tab:bp-stress) already added.