PERSEUS libcbm vs GCBM cross-state intercomparison (CONUS) — LCMS-anchored corrected scenario sweep

v1.3.0 headline correction. v1.2.0 published a CONUS scenario sweep using literature-calibrated regional disturbance rates that overestimated the actual schedule by approximately 5.6x in Minnesota. v1.3.0 supersedes the v1.2.0 scenarios with an LCMS-anchored corrected sweep computed via the --from-bundle mode of cbm_conus/tools/raster_to_sit_events.py. Under empirically observed disturbance rates, scenario policy levers produce CONUS-wide year-50 effects within ~25 TgC of BAU (HIST 0, RH +17, RH_CC +2, WARM_HIST -24). This is approximately 270x smaller than the inventory stratification effect (B1.1 vs B1.3 FIA EXPNS = +6,760 TgC at n=39), confirming that methodological uncertainty axes dominate scenario uncertainty in this generation of CBM-CFS3 carbon models at empirical disturbance rates. What was wrong in v1.2.0. The v1.2.0 BAU was a mix: 6 of 48 states (GA, ID, IN, ME, MN, WA) had LCMS-calibrated bundles; the other 42 had no bundle events and BAU was a no-disturbance counterfactual. The literature-rate HIST/RH/RH_CC/WARM_HIST variants overestimated disturbance ~5.6x, producing a spurious 15,000 TgC BAU-to-HIST swing. What v1.3.0 contains. 192 corrected scenario trajectories (HIST/RH/RH_CC/WARM_HIST across 48 states, bundled as one tarball). Corrected CONUS, regional, and per-state scenario rollup CSVs. All v1.0 through v1.2 contents remain in this concept DOI lineage. What is unchanged. The qualitative methods finding (inventory stratification dominates cross-model uncertainty) and the n=48 BAU baseline trajectory are unchanged. The regional DOM pool fingerprint at v1.0 stands. Five refinement axes (updated). Axis 1 inventory stratification: published, +6,760 TgC. Axis 2 HWP storage: live for ME. Axis 3 disturbance risk: corrected here to small at empirical rates (~25 TgC CONUS-wide scenario spread). Axis 4 landowner dynamics: pending. Axis 5 economics + UQ: NPV layer flagged as not production-ready and masked in the live explorer. Companion materials: methods note at holoros.github.io/perseus-forest-intelligence/methods/inventory-stratification/ and American Forests comparison dashboard at holoros.github.io/perseus-forest-intelligence/methods/af-comparison/.