@essrt/physics 1.14.0 → 1.16.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@essrt/physics",
-  "version": "1.14.0",
+  "version": "1.16.0",
   "description": "ESSRT (Expanding Solar System Resonance Theory) physics core \u2014 pure, injectable-constants, no I/O, no globals, no DOM. The complete model incl. fitted coefficients; every published version immutably ships one recorded model identity.",
   "essrt": {
     "modelVersion": "v11.0",

package/src/constants/generated.js

@@ -764,10 +764,10 @@ export const REFERENCE_DATA = Object.freeze({
     "sunModelTrueErrorDeg": 0.003,
     "sunTropicalYearDiffSeconds": 0.1,
     "sunSiderealYearDiffSeconds": 0.02,
-    "frameworkSunVsJplRmsArcsec": 0.95,
-    "meeusCh25SunVsJplRmsArcsec": 1.28,
-    "moonSeriesLonVsJplRmsArcsec": 2.92,
-    "moonSeriesLatVsJplRmsArcsec": 0.61,
+    "frameworkSunVsJplRmsArcsec": 0.93,
+    "meeusCh25SunVsJplRmsArcsec": 1.22,
+    "moonSeriesLonVsJplRmsArcsec": 2.84,
+    "moonSeriesLatVsJplRmsArcsec": 0.35,
     "mercuryNewtonianArcsecCy": 532,
     "mercuryObservedICRFArcsecCy": 575,
     "mercuryPark2017RateArcsecCy": 575.31,

package/src/eclipse/sun-planetary-completion.cjs

@@ -41,10 +41,21 @@
  *  (ii) the J2000 phase anchors ARG_L0 / PERI / D0 — epoch initial
  *       conditions ("anchored by design" class; the RATES are now
  *       framework-derived, injected);
- *  (iii) the 2lE term (+1.42″ sin 2lE) — DECLARED FITTED, not derived: a
- *       finder-annual artifact, measured chain-independent at E4/E5
- *       (re-fit noise −0.12″/−0.03″). Everything else in this file is
- *       derived. The Earth-around-EMB wobble is DERIVED: amplitude
+ *  (iii) RETIRED (FQ-7-Sun annual-channel attribution): the former
+ *       declared-fitted +1.42″ sin 2lE term. It was fitted against a
+ *       Sun-ALONE JPL comparison bridged by the leading nutation term
+ *       only, and the attribution measured it as the semiannual
+ *       nutation term −1.32″ sin(2F−2D+2Ω) ≡ −1.32″ sin 2lE in disguise
+ *       (2M residual 0.13″ under that bridge, 1.43″ ≡ the term's
+ *       negative under the fuller bridge). Nutation in longitude is a
+ *       frame rotation common to Sun AND Moon — this chain keeps both
+ *       bodies MEAN-of-date (mean obliquity, mean sidereal time; see
+ *       besselian.cjs), so a Sun-only nutation term is a frame
+ *       inconsistency for elongation. Removed: the nutation-free syzygy
+ *       fleet improves 3.877 → 3.756″ and the fuller-bridged Sun
+ *       comparison 2.344 → 2.123″ (fq7s-jpl-preview.mjs --no-2le;
+ *       fq7s-annual-channel.mjs). This file now carries ZERO fitted
+ *       constants. The Earth-around-EMB wobble is DERIVED: amplitude
  *       a_M·μ/AU with μ = 1/(1+M_E/M_M), injected by the model wiring
  *       from live package constants (6.4399″ at current constants), sign
  *       negative in the subtract convention.
@@ -175,9 +186,6 @@ const TERMS = [
   [[0, 0, 1, -1, 0, 0], [0, 0, -1, 0, 0, 0], -0.0353, -0.0379],
 ];
 
-/** The declared-fitted semiannual leftover (header (iii)): +1.42″ sin 2lE. */
-const FITTED_2LE_ARCSEC = 1.42;
-
 const D2R = Math.PI / 180;
 
 /**
@@ -222,8 +230,7 @@ function createSunPlanetaryCompletion({ embWobbleArcsec, carrierRatesDegPerCy })
       table += cA * Math.cos(th) + sA * Math.sin(th);
     }
     const D = (ARG_D0 + D1 * T) * D2R;
-    const arcsec = -table - embWobbleArcsec * Math.sin(D)
-      + FITTED_2LE_ARCSEC * Math.sin(2 * l[2]);
+    const arcsec = -table - embWobbleArcsec * Math.sin(D);
     return arcsec / 3600;
   }
   return { sunPlanetaryCompletionDeg };

package/src/moon/series-extension.cjs

@@ -80,6 +80,29 @@
  * fq7-r3-ship.mjs · fq7-r3-preview.mjs. The Ω carrier polynomial is an
  * IAU-class instrument literal — the same DECLARED-INPUT status as the
  * Delaunay carriers above (Ω ≡ Lp − F of the same catalog class).
+ *
+ * THE FQ-7 DUST ROUND (plan §12i) — 43 λ + 33 β derived Delaunay terms
+ * appended to LON_TERMS/LAT_TERMS below. The broadband-dust census
+ * (fq7-dust-census.mjs, secular-guarded v2: cubic detrend + T/T²
+ * quadrature absorbers on the head arguments, without which the
+ * doctrine-blocked T-modulated parameter class masquerades as fake
+ * 184-yr sideband families) ranked the post-R3 capturable content; the
+ * dense-JPL arbitration (fq7-dust-arbitration.mjs) measured EVERY
+ * trusted leader REAL at ratio 0.9–1.1; the labels bit-match MPP02
+ * main-problem catalog rows below the Meeus ~0.4″ truncation cutoff, in
+ * argument classes no earlier extraction catalog enumerated (odd kD,
+ * |kM| = 3, |kF| = 4/5). Main-problem ⇒ the D1 3-body lab derives them:
+ * one 120-yr run, λ and β columns, head fidelity 0.51%/0.07%,
+ * dt-halving drift 0.000″, lab/census ratio ∈ [0.7, 1.4] required
+ * (4 λ census rows the lab does NOT fully produce stay unshipped —
+ * derive or don't ship). Lab signs match the MPP02 catalog signs row
+ * for row (the frame-consistency control). JPL out-of-sample: official
+ * all-phase λ 2.922 → 2.840″, β 0.606 → 0.353″ — the β endpoint AT the
+ * R1-measured MPP02-vs-JPL floor (~0.33″); dense 2-day λ 2.321 →
+ * 2.219″, β 0.601 → 0.384″; syzygy fleet improves; the sign-flip
+ * control degrades every gate. Instruments: fq7-dust-census.mjs ·
+ * fq7-dust-arbitration.mjs · fq7-dust-extraction.mjs ·
+ * fq7-dust-preview.mjs.
  */
 
 'use strict';
@@ -155,6 +178,52 @@ const LON_TERMS = [
   [4, 0, -2, 2, -0.169],
   [4, -2, -2, 0, 0.161],
   [0, 0, 5, 0, 0.112],
+  // FQ-7 DUST ROUND — the argument classes no earlier catalog enumerated
+  // (odd kD, |kM| = 3, |kF| = 4); lab-derived, MPP02-catalog-certified,
+  // JPL-arbitrated (see header note):
+  [0, 0, 0, 4, 0.422],
+  [3, 0, 0, 0, 0.402],
+  [2, -3, 0, 0, 0.341],
+  [3, 1, -1, 0, 0.275],
+  [3, 0, 0, -2, -0.250],
+  [2, -3, -1, 0, 0.240],
+  [3, -1, -1, 0, -0.233],
+  [3, 1, 0, 0, 0.147],
+  [0, 3, 0, 0, -0.096],
+  [4, -1, -3, 0, 0.096],
+  [2, -1, 3, 0, 0.094],
+  [2, 0, -3, -2, -0.093],
+  [0, 0, 1, 4, 0.092],
+  [6, -1, -2, 0, 0.091],
+  [1, 1, 2, 0, 0.090],
+  [4, 0, 0, 2, -0.084],
+  [3, -1, -2, 0, -0.082],
+  [0, 0, 1, -4, -0.081],
+  [2, 0, 4, 0, 0.078],
+  [2, 0, 0, -4, -0.078],
+  [0, 1, 0, -2, 0.074],
+  [2, -1, 1, 2, -0.073],
+  [2, -1, -1, -2, -0.073],
+  [6, -1, -1, 0, 0.072],
+  [2, 1, -1, 2, 0.068],
+  [2, 0, -5, 0, 0.067],
+  [2, 1, 0, 2, 0.066],
+  [4, 0, 1, -2, -0.065],
+  [2, -2, 2, 0, 0.065],
+  [2, -2, 0, -2, 0.065],
+  [3, -1, 0, 0, 0.063],
+  [0, 2, 2, 0, -0.062],
+  [5, 0, -2, 0, -0.059],
+  [0, 1, -2, -2, 0.050],
+  [4, 1, 1, 0, -0.050],
+  [0, 1, -4, 0, -0.048],
+  [0, 1, 2, 2, 0.045],
+  [1, 0, 1, 2, 0.044],
+  [3, 0, -3, 0, -0.043],
+  [1, 0, 3, 0, -0.042],
+  [1, 2, 0, 0, -0.041],
+  [0, 3, -1, 0, -0.038],
+  [1, -2, 0, 0, 0.037],
 ];
 
 /** Latitude tail: [kD, kM, kMp, kF, sin″] — 30 derived terms ≥ 0.1″
@@ -190,6 +259,41 @@ const LAT_TERMS = [
   [1, 0, -2, -1, -0.1093],
   [2, 2, -1, 1, -0.1044],
   [1, 1, 1, 1, 0.1014],
+  // FQ-7 DUST ROUND — lab-derived, MPP02-catalog-certified,
+  // JPL-arbitrated (see header note):
+  [3, 0, 0, -1, -0.352],
+  [3, 0, -1, -1, -0.305],
+  [4, -1, -2, 1, 0.214],
+  [3, 0, -1, 1, -0.206],
+  [0, 2, -1, 1, -0.094],
+  [6, 0, -1, -1, 0.094],
+  [0, 0, 4, -1, 0.091],
+  [2, -1, 0, -3, 0.087],
+  [6, 0, -2, -1, 0.081],
+  [1, 0, -2, 1, -0.078],
+  [0, 1, -3, -1, -0.075],
+  [2, 1, -2, -1, 0.073],
+  [2, -2, 1, -1, 0.069],
+  [2, 0, -2, 3, -0.065],
+  [1, 0, 2, 1, -0.064],
+  [2, -1, -2, 1, -0.064],
+  [2, 1, 2, -1, -0.063],
+  [4, 0, 0, -3, 0.062],
+  [0, 1, 3, 1, -0.062],
+  [6, 0, -2, 1, 0.060],
+  [2, -2, 1, 1, 0.059],
+  [2, 1, 0, -3, -0.059],
+  [4, -1, 1, -1, 0.057],
+  [0, 2, 0, 1, -0.055],
+  [0, 2, 1, 1, -0.054],
+  [2, -1, -3, -1, 0.054],
+  [4, -1, -2, -1, 0.052],
+  [4, 1, -1, 1, -0.051],
+  [1, 1, -1, 1, -0.051],
+  [2, 1, -2, 1, 0.049],
+  [3, 0, -2, 1, -0.048],
+  [4, 0, 2, -1, 0.048],
+  [6, 0, -1, 1, 0.042],
 ];
 
 /** Planetary mean-longitude polynomials, degrees per Julian century TT