diff --git a/models/GADM-0.5.md b/models/GADM-0.5.md
index 9a8f217..27f5eb7 100644
--- a/models/GADM-0.5.md
+++ b/models/GADM-0.5.md
@@ -294,3 +294,65 @@ output. It points to non-AI factors that GADM lacks: industrial structure,
 the energy mix of material processing, offshoring. With the task-data
 φ_r, the decoupling reading holds. The EU fit remains the warning case, if
 future AI use shifts toward automating physical production.
+
+## 7. Grounding of the ecological classes (ASSEMBLY-WP-0018-T02)
+
+### 7.1 Values
+
+- **Starting levels.** Each boundary is placed in its zone of increasing
+  risk, with the boundary at 0.4 and the high-risk line at 0:
+
+  | Boundary | Value now | Boundary / high-risk line | Level |
+  |---|---|---|---|
+  | Climate (CO₂) | 423 ppm | 350 / 450 | 0.11 |
+  | Forest cover | 60 % | 75 / 54 % | 0.12 |
+  | Biosphere (HANPP) | 30 % | 10 / 20 % | 0 |
+  | Ocean aragonite saturation | 81 % of preindustrial | 80 % | ≈ 0.38 |
+  | Ozone, aerosols | safe | | ≈ 0.6 |
+
+  Freshwater, nitrogen, phosphorus and novel entities have no values here;
+  they are taken as deep in the zone. That gives **renewables 0.12 and sinks
+  0.30**, close to the priors (0.15, 0.33). Secondary sources: the Planetary
+  Health Check PDF is not machine-readable.
+- **Compute load.** Data centers used 415 TWh in 2024, about 1.5 % of world
+  electricity, and about 945 TWh are projected for 2030 (IEA *Energy and AI*).
+  GADM-0.2's `e2` = 0.03 gave compute 8–17 % of the ecological draw. The
+  grounded **e2 = 0.0015** gives about 1 % at the calibrated present.
+- **Physical load (structural fix).** The physical load now follows *rival
+  capacity*, 1 − (1 − K_phys)(1 − φ_r K_Ω), so organisms that automate
+  physical production draw on the classes like physical capital. Before, only
+  K_phys did, and the overstated compute term had hidden that gap.
+- **Materials.** Copper reserves rose from 680 Mt (2013) to 880–980 Mt (2025)
+  while about 22 Mt were mined each year. Reserve additions outpace use, which
+  supports replenishment at rest.
+
+### 7.2 What changes
+
+| Reading | Before grounding | After |
+|---|---|---|
+| class most hit by organism growth (§ 2) | sinks > renewables > materials | none: direct draw about 0 (compute small, rival automation small at φ_r 0.15) |
+| generative regime's sinks breach (§ 2) | t 29–83 | none within 100 years; with strict material depletion, materials bind at about t 33 |
+| decoupled decades with demand split, generative B2 / B3 / B4 (§ 4) | 0.69 / 0.71 / 0.25 | 0.78 / 0.69 / 0.38 |
+| φ_r fitted to dematerialization, EU / DE (§ 5) | 0.66 / 0.16 | **0.79 / 0.57** |
+| decoupled decades at the DE fit | 0.55 / 0.72 / 0.24 (φ_r 0.15) | 0.15 / 0.45 / 0.19 (φ_r 0.57) |
+
+### 7.3 Readings
+
+1. **The ecological case against the generative regime rested on an
+   overstated compute draw.** With grounded compute energy, the regime
+   breaches no class floor within the horizon when materials are replenished.
+   The T02 grounding test (organism growth binds energy and sinks before
+   materials) is moot: the direct draw is about 1 % of the total.
+2. **Renewables are already below their floor in every scenario.** This does
+   not depend on organisms.
+3. **The conflict in BR-011 sharpens.** Task data give φ_r ≤ 0.25. Both
+   dematerialization series now give 0.57–0.79. At the task-data value the
+   model dematerializes far faster than observed (EU −4.6 % per year, against
+   −1.85 % observed). Two explanations remain open:
+   - the model lacks something that keeps output coupled to materials, for
+     example rebound effects or the physical footprint of digital services;
+   - the 2010–2024 data reflect pre-AI structural factors, not an AI-era
+     parameter.
+
+   Decoupling holds at the task-data φ_r and becomes partial or disappears at
+   the dematerialization fits.
diff --git a/research/BreakRegister.md b/research/BreakRegister.md
index b59538f..2e3fc93 100644
--- a/research/BreakRegister.md
+++ b/research/BreakRegister.md
@@ -23,4 +23,4 @@ that part of the band.
 | BR-008 | 2026-10-13 (WP-0011-T05) | Which levers bound domination (BR-007 follow-up) | closed model: host rights levers → coupled model: only host access rules on foreign operators and the home's levers reach the foreign-run channels | measured channels (Pd, Pmem, Plin, Pi) are operated from the US for EU and AR hosts | **structural, implemented:** cross-border coupling (`world.py`); observed DMA effect on search ≈ 0 (at most a 0.006/yr) | update 2026-10-13 (WP-0016-T01): a host lineage access rule does reach imported-lineage concentration (EU Plin 0.91 → 0.19 at a 0.15); portability needs the domestic lineage market's concentration (CB3) |
 | BR-009 | 2026-10-13 (WP-0012) | Sentiment dynamics of the legitimacy loop | Σ relaxes toward 0.5 + gain × Δ(perceived benefit), so from the measured US Σ 0.285 it rises → measured US sentiment fell 2021–2026 (0.405 → 0.285) while adoption grew fast | the neutral reference 0.5 is an assumption; US adoption also shows no braking by low sentiment (all loop-on candidates fail the US match) | **resolved 2026-10-13 (WP-0017-T01):** parameter `sig_ref`; calibrated runs anchor Σ at its measured level, and adoption and legislation respond to changes relative to it. Default (None = 0.5) keeps earlier results. R1–R4 and H1/H2 verdicts unchanged; the matching loop weights become ≈ 0.5 | resolved; `sig_adopt` not yet identified |
 | BR-010 | 2026-10-13 (WP-0007-T01) | Non-rival growth decouples welfare from throughput (core hypothesis, GADM-0.1 § 14) | output split rival / non-rival (GADM-0.5-dev): no genuine decoupling in any anchor; the generative regime's ecological cost unchanged | GADM has no demand composition: non-rival output adds to, but never replaces, rival consumption | **resolved 2026-10-13 (WP-0007-T04):** two-good demand split; the generative regime decouples in 45–71 % of decades at B2–B3, and its sinks breach moves 15–50 years later; band-dependent (none at B1) | resolved; calibration to observed dematerialization open (WP-0018) |
-| BR-011 | 2026-10-13 (WP-0018-T01) | Genuine decoupling in the generative regime (BR-010 resolution) | holds for φ_r ≤ 0.2 (task-data prior; DE dematerialization fit 0.15) → disappears at φ_r 0.66 (EU dematerialization fit) | the reading depends on the rival share of organism output, which the 2010–2024 data constrain only weakly (pre-generative-AI) | none structural; empirical: watch L4 task shares and E5 material intensity after 2024. Update (WP-0018-T03): task data give φ_r 0.05–0.25, so the EU fit points to non-AI factors; the reading holds at the task-data φ_r | open, leaning toward 'holds' |
+| BR-011 | 2026-10-13 (WP-0018-T01) | Genuine decoupling in the generative regime (BR-010 resolution) | holds for φ_r ≤ 0.2 (task-data prior; DE dematerialization fit 0.15) → disappears at φ_r 0.66 (EU dematerialization fit) | the reading depends on the rival share of organism output, which the 2010–2024 data constrain only weakly (pre-generative-AI) | none structural; empirical: watch L4 task shares and E5 material intensity after 2024. Update (WP-0018-T03): task data give φ_r 0.05–0.25. Update (WP-0018-T02, compute load grounded, physical load on rival capacity): dematerialization fits rise to EU 0.79 / DE 0.57, so decoupling is partial or absent at the fits and holds only at the task-data φ_r. Candidate missing structure: material rebound, physical footprint of digital services | open, conflict sharpened |
diff --git a/simulations/gadm02/gadm05.py b/simulations/gadm02/gadm05.py
index c0ce58b..0838c81 100644
--- a/simulations/gadm02/gadm05.py
+++ b/simulations/gadm02/gadm05.py
@@ -28,7 +28,12 @@ import gadm04 as g4
 MODEL_VERSION = "GADM-0.5-dev"
 STOCKS = g4.STOCKS
 INIT = g4.INIT
-PARAMS = {**g4.PARAMS, "phi_r": 0.15, "xn": 0.6}
+PARAMS = {**g4.PARAMS, "phi_r": 0.15, "xn": 0.6,
+          # WP-0018-T02: compute's ecological draw. Data centers used 415 TWh in 2024, about 1.5 % of
+          # global electricity (IEA Energy and AI, 2025), well under 1 % of energy use and emissions.
+          # GADM-0.2's e2 = 0.03 gives compute 8-17 % of the draw; e2 = 0.0015 gives about 1 % at the
+          # calibrated present (K_Omega 0.12).
+          "e2": 0.0015}
 
 
 def simulate(params: dict | None = None, init: dict | None = None, **kw) -> list[dict]:
@@ -60,7 +65,14 @@ def throughput(row: dict, p: dict | None = None) -> float:
 # regenerate; recycling (circ, circular material use rate) slows depletion.
 ECO = ("Er", "Ed", "Es")
 STOCKS_ECO = g4.STOCKS + ECO
-ECO_PARAMS = {"eco": 0.0, "Er0": 0.15, "Ed0": 0.45, "Es0": 0.33,
+# WP-0018-T02 grounding of starting levels (position in the zone of increasing risk, boundary = 0.4,
+# high-risk line = 0): CO2 423 ppm (350 / 450) 0.11; forest cover 60 % (75 / 54 %) 0.12; HANPP 30 %
+# (10 / 20 %) 0; ocean aragonite 81 % of preindustrial (boundary 80 %) about 0.38; ozone, aerosols safe
+# about 0.6. Freshwater, nitrogen, phosphorus and novel entities without values: deep in the zone.
+# Renewables (land, biosphere, freshwater) about 0.12; sinks about 0.30. Materials replenishment:
+# copper reserves 680 Mt (2013) -> 880-980 Mt (2025) with about 22 Mt/year mined, so additions
+# outpace use ("rest" is supported).
+ECO_PARAMS = {"eco": 0.0, "Er0": 0.12, "Ed0": 0.45, "Es0": 0.30,
               "share_k": (0.4, 0.3, 0.3), "share_o": (0.3, 0.2, 0.5),
               "lowc": 0.40, "circ": 0.12, "rr": None, "rs": None, "rm": 0.0, "eco_floor": 0.20}
 # rm: material replenishment (reserve growth through discovery, substitution and recycling technology).
@@ -72,7 +84,11 @@ def eco_E(s: dict) -> float:
 
 
 def _loads(s: dict, a: dict, p: dict) -> tuple[float, float]:
-    return s["Kp"] * a["dem"] * p["thr"], s["Ko"]
+    """(physical load, compute load). WP-0018-T02: the physical load follows *rival capacity*
+    1 - (1 - K_phys)(1 - phi_r K_Omega), so organisms that automate physical production draw on the
+    classes like physical capital (before, only K_phys did; the overstated compute term hid that)."""
+    k_r = 1 - (1 - s["Kp"]) * (1 - p.get("phi_r", 1.0) * s["Ko"])
+    return k_r * a["dem"] * p["thr"], s["Ko"]
 
 
 def _draws(s: dict, a: dict, p: dict, q: dict) -> tuple[float, float, float]:
diff --git a/simulations/gadm02/results/demand.md b/simulations/gadm02/results/demand.md
index 004a252..23c9b9e 100644
--- a/simulations/gadm02/results/demand.md
+++ b/simulations/gadm02/results/demand.md
@@ -7,38 +7,38 @@ years 0–10 (observed EU dematerialization ≈ −1.6 %/year).
 
 | Scenario | Anchor | Case | VFAC t100 | rival demand share Cr t100 | intensity trend | decoupled windows | sinks breach | Es t100 |
 |---|---|---|---|---|---|---|---|---|
-| Baseline (current trends) | B1 | no demand split | 0.397 | 1.00 | +0.0 % | 0.00 | — | 0.31 |
-| Baseline (current trends) | B1 | split (λ 0.05, ε 2) | 0.402 | 0.91 | +0.0 % | 0.00 | — | 0.33 |
-| Baseline (current trends) | B1 | split, strong (λ 0.1, ε 4) | 0.411 | 0.81 | +0.0 % | 0.46 | — | 0.36 |
-| Baseline (current trends) | B2 | no demand split | 0.343 | 1.00 | -0.1 % | 0.00 | — | 0.33 |
-| Baseline (current trends) | B2 | split (λ 0.05, ε 2) | 0.353 | 0.67 | -0.1 % | 0.00 | — | 0.36 |
-| Baseline (current trends) | B2 | split, strong (λ 0.1, ε 4) | 0.374 | 0.05 | -0.1 % | 0.00 | — | 0.41 |
-| Baseline (current trends) | B3 | no demand split | 0.331 | 1.00 | -0.2 % | 0.18 | — | 0.27 |
-| Baseline (current trends) | B3 | split (λ 0.05, ε 2) | 0.344 | 0.45 | -0.3 % | 0.26 | — | 0.31 |
-| Baseline (current trends) | B3 | split, strong (λ 0.1, ε 4) | 0.354 | 0.05 | -0.3 % | 0.08 | — | 0.33 |
-| Baseline (current trends) | B4 | no demand split | 0.303 | 1.00 | -0.3 % | 0.41 | — | 0.21 |
-| Baseline (current trends) | B4 | split (λ 0.05, ε 2) | 0.313 | 0.29 | -0.3 % | 0.46 | — | 0.23 |
-| Baseline (current trends) | B4 | split, strong (λ 0.1, ε 4) | 0.318 | 0.05 | -0.4 % | 0.14 | — | 0.24 |
-| Generative regime | B1 | no demand split | 0.608 | 1.00 | +0.0 % | 0.00 | t 83 | 0.19 |
-| Generative regime | B1 | split (λ 0.05, ε 2) | 0.614 | 0.88 | +0.0 % | 0.00 | — | 0.21 |
-| Generative regime | B1 | split, strong (λ 0.1, ε 4) | 0.625 | 0.72 | +0.0 % | 0.21 | — | 0.24 |
-| Generative regime | B2 | no demand split | 0.585 | 1.00 | -0.1 % | 0.00 | t 49 | 0.17 |
-| Generative regime | B2 | split (λ 0.05, ε 2) | 0.602 | 0.67 | -0.1 % | 0.69 | — | 0.20 |
-| Generative regime | B2 | split, strong (λ 0.1, ε 4) | 0.645 | 0.05 | -0.1 % | 0.90 | — | 0.30 |
-| Generative regime | B3 | no demand split | 0.567 | 1.00 | -0.3 % | 0.00 | t 37 | 0.15 |
-| Generative regime | B3 | split (λ 0.05, ε 2) | 0.591 | 0.47 | -0.3 % | 0.71 | t 70 | 0.20 |
-| Generative regime | B3 | split, strong (λ 0.1, ε 4) | 0.618 | 0.05 | -0.4 % | 0.73 | — | 0.25 |
-| Generative regime | B4 | no demand split | 0.513 | 1.00 | -0.4 % | 0.00 | t 29 | 0.14 |
-| Generative regime | B4 | split (λ 0.05, ε 2) | 0.541 | 0.21 | -0.4 % | 0.25 | t 42 | 0.19 |
-| Generative regime | B4 | split, strong (λ 0.1, ε 4) | 0.551 | 0.05 | -0.5 % | 0.34 | — | 0.21 |
+| Baseline (current trends) | B1 | no demand split | 0.394 | 1.00 | -0.1 % | 0.05 | — | 0.30 |
+| Baseline (current trends) | B1 | split (λ 0.05, ε 2) | 0.400 | 0.91 | -0.1 % | 0.30 | — | 0.32 |
+| Baseline (current trends) | B1 | split, strong (λ 0.1, ε 4) | 0.410 | 0.81 | -0.1 % | 0.52 | — | 0.35 |
+| Baseline (current trends) | B2 | no demand split | 0.344 | 1.00 | -0.6 % | 0.04 | — | 0.36 |
+| Baseline (current trends) | B2 | split (λ 0.05, ε 2) | 0.355 | 0.68 | -0.6 % | 0.04 | — | 0.41 |
+| Baseline (current trends) | B2 | split, strong (λ 0.1, ε 4) | 0.378 | 0.06 | -0.6 % | 0.04 | — | 0.49 |
+| Baseline (current trends) | B3 | no demand split | 0.338 | 1.00 | -1.4 % | 0.20 | — | 0.37 |
+| Baseline (current trends) | B3 | split (λ 0.05, ε 2) | 0.353 | 0.47 | -1.4 % | 0.28 | — | 0.44 |
+| Baseline (current trends) | B3 | split, strong (λ 0.1, ε 4) | 0.367 | 0.05 | -1.5 % | 0.30 | — | 0.51 |
+| Baseline (current trends) | B4 | no demand split | 0.316 | 1.00 | -1.8 % | 0.42 | — | 0.37 |
+| Baseline (current trends) | B4 | split (λ 0.05, ε 2) | 0.331 | 0.33 | -1.8 % | 0.46 | — | 0.45 |
+| Baseline (current trends) | B4 | split, strong (λ 0.1, ε 4) | 0.339 | 0.05 | -1.9 % | 0.47 | — | 0.50 |
+| Generative regime | B1 | no demand split | 0.608 | 1.00 | -0.1 % | 0.00 | — | 0.21 |
+| Generative regime | B1 | split (λ 0.05, ε 2) | 0.615 | 0.89 | -0.1 % | 0.00 | — | 0.23 |
+| Generative regime | B1 | split, strong (λ 0.1, ε 4) | 0.627 | 0.73 | -0.1 % | 0.61 | — | 0.26 |
+| Generative regime | B2 | no demand split | 0.593 | 1.00 | -0.6 % | 0.00 | — | 0.21 |
+| Generative regime | B2 | split (λ 0.05, ε 2) | 0.613 | 0.70 | -0.6 % | 0.78 | — | 0.27 |
+| Generative regime | B2 | split, strong (λ 0.1, ε 4) | 0.672 | 0.05 | -0.7 % | 0.99 | — | 0.50 |
+| Generative regime | B3 | no demand split | 0.579 | 1.00 | -1.5 % | 0.00 | — | 0.22 |
+| Generative regime | B3 | split (λ 0.05, ε 2) | 0.609 | 0.55 | -1.5 % | 0.69 | — | 0.32 |
+| Generative regime | B3 | split, strong (λ 0.1, ε 4) | 0.656 | 0.05 | -1.6 % | 0.78 | — | 0.51 |
+| Generative regime | B4 | no demand split | 0.529 | 1.00 | -1.9 % | 0.00 | — | 0.23 |
+| Generative regime | B4 | split (λ 0.05, ε 2) | 0.566 | 0.34 | -2.0 % | 0.38 | — | 0.37 |
+| Generative regime | B4 | split, strong (λ 0.1, ε 4) | 0.593 | 0.05 | -2.1 % | 0.40 | — | 0.50 |
 
 ## Readings and robustness class
 
 | Reading | B1 | B2 | B3 | B4 | class |
 |---|---|---|---|---|---|
-| decoupled windows, split − none: Baseline (current trends) | +0.000 | +0.000 | +0.080 | +0.050 | QUALITATIVE BREAK |
-| ΔVFAC split − none: Baseline (current trends) | +0.006 | +0.010 | +0.012 | +0.010 | quantitative shift |
-| ΔEs t100 split − none: Baseline (current trends) | +0.016 | +0.032 | +0.032 | +0.022 | quantitative shift |
-| decoupled windows, split − none: Generative regime | +0.000 | +0.690 | +0.710 | +0.250 | QUALITATIVE BREAK |
-| ΔVFAC split − none: Generative regime | +0.006 | +0.017 | +0.025 | +0.028 | quantitative shift |
-| ΔEs t100 split − none: Generative regime | +0.016 | +0.036 | +0.047 | +0.051 | quantitative shift |
+| decoupled windows, split − none: Baseline (current trends) | +0.250 | +0.000 | +0.080 | +0.040 | QUALITATIVE BREAK |
+| ΔVFAC split − none: Baseline (current trends) | +0.006 | +0.011 | +0.014 | +0.014 | quantitative shift |
+| ΔEs t100 split − none: Baseline (current trends) | +0.018 | +0.045 | +0.066 | +0.079 | quantitative shift |
+| decoupled windows, split − none: Generative regime | +0.000 | +0.780 | +0.690 | +0.380 | QUALITATIVE BREAK |
+| ΔVFAC split − none: Generative regime | +0.007 | +0.020 | +0.030 | +0.037 | quantitative shift |
+| ΔEs t100 split − none: Generative regime | +0.021 | +0.061 | +0.099 | +0.141 | quantitative shift |
diff --git a/simulations/gadm02/results/dematerialization.md b/simulations/gadm02/results/dematerialization.md
index 6a6ea9b..c248b56 100644
--- a/simulations/gadm02/results/dematerialization.md
+++ b/simulations/gadm02/results/dematerialization.md
@@ -10,30 +10,30 @@ start (Calibration v0.5) with ecological classes. Total throughput per output is
 
 | Jurisdiction | observed | model, no split (φ 0.2) | model, split ε 2 (φ 0.2) | φ_r reproducing the observation (split ε 2) |
 |---|---|---|---|---|
-| EU | -1.85 %/y | -2.43 %/y | -2.91 %/y | 0.66 |
-| DE | -2.73 %/y | -2.23 %/y | -2.66 %/y | 0.16 |
+| EU | -1.85 %/y | -4.55 %/y | -5.24 %/y | 0.79 |
+| DE | -2.73 %/y | -3.98 %/y | -4.59 %/y | 0.57 |
 
 ## T04 readings at the prior and at the EU-fitted φ_r (generative regime, material replenishment at rest)
 
 | φ_r | Anchor | decoupled decades no split → split | sinks breach no split → split | VFAC no split → split |
 |---|---|---|---|---|
-| 0.2 | B1 | 0.00 → 0.00 | t 83 → none | 0.608 → 0.614 |
-| 0.2 | B2 | 0.00 → 0.55 | t 49 → none | 0.585 → 0.601 |
-| 0.2 | B3 | 0.00 → 0.70 | t 37 → t 62 | 0.567 → 0.589 |
-| 0.2 | B4 | 0.00 → 0.24 | t 29 → t 40 | 0.512 → 0.538 |
-| 0.16 | B1 | 0.00 → 0.00 | t 83 → none | 0.608 → 0.614 |
-| 0.16 | B2 | 0.00 → 0.65 | t 49 → none | 0.585 → 0.602 |
-| 0.16 | B3 | 0.00 → 0.71 | t 37 → t 68 | 0.567 → 0.591 |
-| 0.16 | B4 | 0.00 → 0.25 | t 29 → t 42 | 0.512 → 0.540 |
-| 0.66 | B1 | 0.00 → 0.00 | t 83 → t 100 | 0.607 → 0.609 |
-| 0.66 | B2 | 0.00 → 0.00 | t 49 → t 58 | 0.583 → 0.589 |
-| 0.66 | B3 | 0.00 → 0.00 | t 37 → t 43 | 0.564 → 0.572 |
-| 0.66 | B4 | 0.00 → 0.02 | t 29 → t 33 | 0.509 → 0.518 |
+| 0.2 | B1 | 0.00 → 0.00 | none → none | 0.607 → 0.614 |
+| 0.2 | B2 | 0.00 → 0.77 | none → none | 0.592 → 0.611 |
+| 0.2 | B3 | 0.00 → 0.68 | none → none | 0.579 → 0.606 |
+| 0.2 | B4 | 0.00 → 0.36 | none → none | 0.528 → 0.562 |
+| 0.57 | B1 | 0.00 → 0.00 | none → none | 0.606 → 0.609 |
+| 0.57 | B2 | 0.00 → 0.15 | none → none | 0.589 → 0.597 |
+| 0.57 | B3 | 0.00 → 0.45 | none → none | 0.573 → 0.586 |
+| 0.57 | B4 | 0.00 → 0.19 | none → none | 0.520 → 0.537 |
+| 0.79 | B1 | 0.00 → 0.00 | none → none | 0.605 → 0.606 |
+| 0.79 | B2 | 0.00 → 0.00 | none → none | 0.586 → 0.590 |
+| 0.79 | B3 | 0.00 → 0.02 | none → none | 0.570 → 0.576 |
+| 0.79 | B4 | 0.00 → 0.02 | none → none | 0.515 → 0.523 |
 
 ## Readings and robustness class
 
 | Reading | B1 | B2 | B3 | B4 | class |
 |---|---|---|---|---|---|
-| decoupled decades with split, φ_r 0.16 | +0.00 | +0.65 | +0.71 | +0.25 | QUALITATIVE BREAK |
-| decoupled decades with split, φ_r 0.2 | +0.00 | +0.55 | +0.70 | +0.24 | QUALITATIVE BREAK |
-| decoupled decades with split, φ_r 0.66 | +0.00 | +0.00 | +0.00 | +0.02 | QUALITATIVE BREAK |
+| decoupled decades with split, φ_r 0.2 | +0.00 | +0.77 | +0.68 | +0.36 | QUALITATIVE BREAK |
+| decoupled decades with split, φ_r 0.57 | +0.00 | +0.15 | +0.45 | +0.19 | QUALITATIVE BREAK |
+| decoupled decades with split, φ_r 0.79 | +0.00 | +0.00 | +0.02 | +0.02 | QUALITATIVE BREAK |
diff --git a/simulations/gadm02/results/eco_classes.md b/simulations/gadm02/results/eco_classes.md
index cc55a44..8be4847 100644
--- a/simulations/gadm02/results/eco_classes.md
+++ b/simulations/gadm02/results/eco_classes.md
@@ -8,35 +8,35 @@ below the floor. t = 100 levels.
 
 | Scenario | Anchor | Er / Ed / Es t100 | renewables below floor | materials below floor | sinks below floor | lowest at t100 |
 |---|---|---|---|---|---|---|
-| Baseline (current trends) | B1 | 0.15 / 0.08 / 0.31 | below at start | t 46 | — | materials |
-| Baseline (current trends) | B2 | 0.19 / 0.11 / 0.33 | below at start | t 45 | — | materials |
-| Baseline (current trends) | B3 | 0.17 / 0.09 / 0.27 | below at start | t 43 | — | materials |
-| Baseline (current trends) | B4 | 0.13 / 0.07 / 0.21 | below at start | t 37 | — | materials |
-| Generative regime | B1 | 0.08 / 0.02 / 0.19 | below at start | t 30 | t 83 | materials |
-| Generative regime | B2 | 0.08 / 0.01 / 0.17 | below at start | t 28 | t 49 | materials |
-| Generative regime | B3 | 0.08 / 0.01 / 0.15 | below at start | t 26 | t 37 | materials |
-| Generative regime | B4 | 0.08 / 0.01 / 0.14 | below at start | t 25 | t 29 | materials |
-| Attention oligopoly | B1 | 0.24 / 0.16 / 0.42 | below at start | t 67 | — | materials |
-| Attention oligopoly | B2 | 0.24 / 0.16 / 0.39 | below at start | t 64 | — | materials |
-| Attention oligopoly | B3 | 0.20 / 0.13 / 0.31 | below at start | t 54 | — | materials |
-| Attention oligopoly | B4 | 0.15 / 0.09 / 0.22 | below at start | t 42 | — | materials |
+| Baseline (current trends) | B1 | 0.13 / 0.09 / 0.30 | below at start | t 50 | — | materials |
+| Baseline (current trends) | B2 | 0.17 / 0.14 / 0.36 | below at start | t 59 | — | materials |
+| Baseline (current trends) | B3 | 0.18 / 0.16 / 0.37 | below at start | t 70 | — | materials |
+| Baseline (current trends) | B4 | 0.18 / 0.16 / 0.37 | below at start | t 75 | — | materials |
+| Generative regime | B1 | 0.07 / 0.02 / 0.21 | below at start | t 33 | — | materials |
+| Generative regime | B2 | 0.08 / 0.03 / 0.21 | below at start | t 33 | — | materials |
+| Generative regime | B3 | 0.08 / 0.03 / 0.22 | below at start | t 34 | — | materials |
+| Generative regime | B4 | 0.09 / 0.04 / 0.23 | below at start | t 35 | — | materials |
+| Attention oligopoly | B1 | 0.20 / 0.18 / 0.40 | below at start | t 77 | — | materials |
+| Attention oligopoly | B2 | 0.21 / 0.20 / 0.41 | below at start | t 97 | — | materials |
+| Attention oligopoly | B3 | 0.21 / 0.20 / 0.41 | below at start | t 100 | — | materials |
+| Attention oligopoly | B4 | 0.20 / 0.20 / 0.40 | below at start | t 97 | — | materials |
 
 ## Materials replenishment: at rest today
 
 | Scenario | Anchor | Er / Ed / Es t100 | renewables below floor | materials below floor | sinks below floor | lowest at t100 |
 |---|---|---|---|---|---|---|
-| Baseline (current trends) | B1 | 0.15 / 0.46 / 0.31 | below at start | — | — | renewables |
-| Baseline (current trends) | B2 | 0.19 / 0.53 / 0.33 | below at start | — | — | renewables |
-| Baseline (current trends) | B3 | 0.17 / 0.49 / 0.27 | below at start | — | — | renewables |
-| Baseline (current trends) | B4 | 0.13 / 0.44 / 0.21 | below at start | — | — | renewables |
-| Generative regime | B1 | 0.08 / 0.30 / 0.19 | below at start | — | t 83 | renewables |
-| Generative regime | B2 | 0.08 / 0.29 / 0.17 | below at start | — | t 49 | renewables |
-| Generative regime | B3 | 0.08 / 0.28 / 0.15 | below at start | — | t 37 | renewables |
-| Generative regime | B4 | 0.08 / 0.29 / 0.14 | below at start | — | t 29 | renewables |
-| Attention oligopoly | B1 | 0.24 / 0.61 / 0.42 | below at start | — | — | renewables |
-| Attention oligopoly | B2 | 0.24 / 0.60 / 0.39 | below at start | — | — | renewables |
-| Attention oligopoly | B3 | 0.20 / 0.55 / 0.31 | below at start | — | — | renewables |
-| Attention oligopoly | B4 | 0.15 / 0.47 / 0.22 | below at start | — | — | renewables |
+| Baseline (current trends) | B1 | 0.13 / 0.47 / 0.30 | below at start | — | — | renewables |
+| Baseline (current trends) | B2 | 0.17 / 0.57 / 0.36 | below at start | — | — | renewables |
+| Baseline (current trends) | B3 | 0.18 / 0.58 / 0.37 | below at start | — | — | renewables |
+| Baseline (current trends) | B4 | 0.18 / 0.58 / 0.37 | below at start | — | — | renewables |
+| Generative regime | B1 | 0.07 / 0.32 / 0.21 | below at start | — | — | renewables |
+| Generative regime | B2 | 0.08 / 0.33 / 0.21 | below at start | — | — | renewables |
+| Generative regime | B3 | 0.08 / 0.34 / 0.22 | below at start | — | — | renewables |
+| Generative regime | B4 | 0.09 / 0.37 / 0.23 | below at start | — | — | renewables |
+| Attention oligopoly | B1 | 0.20 / 0.61 / 0.40 | below at start | — | — | renewables |
+| Attention oligopoly | B2 | 0.21 / 0.64 / 0.41 | below at start | — | — | renewables |
+| Attention oligopoly | B3 | 0.21 / 0.63 / 0.41 | below at start | — | — | renewables |
+| Attention oligopoly | B4 | 0.20 / 0.62 / 0.40 | below at start | — | — | renewables |
 
 ## Organism growth: class draw attributable to adoption speed (baseline, B4 − B1)
 
@@ -44,5 +44,5 @@ Relative change of each class between t 0 and t 50, B4 minus B1 (negative = orga
 
 | Materials replenishment | renewables | materials | sinks | order (most hit first) |
 |---|---|---|---|---|
-| none | -0.032 | -0.018 | -0.036 | sinks > renewables > materials |
-| at rest | -0.032 | -0.018 | -0.036 | sinks > renewables > materials |
+| none | -0.000 | +0.000 | -0.001 | sinks > renewables > materials |
+| at rest | -0.000 | +0.000 | -0.001 | sinks > renewables > materials |
diff --git a/simulations/gadm02/results/selection.md b/simulations/gadm02/results/selection.md
index 8abec61..85abf08 100644
--- a/simulations/gadm02/results/selection.md
+++ b/simulations/gadm02/results/selection.md
@@ -6,34 +6,34 @@ Shock = {'Pd': 0.25, 'Pm': 0.15} added to the starting state. t = 100.
 
 | Scenario | Anchor | VFAC fixed → endogenous | σ fixed → endogenous | capture t0 → t100 (fixed weights at t100) | max endogenous capture change/yr | shock VFAC loss fixed / endogenous |
 |---|---|---|---|---|---|---|
-| Baseline (current trends) | B1 | 0.433 → 0.434 | 0.428 → 0.433 | 0.42 → 0.44 (0.43) | 0.0019 | -0.005 / -0.005 |
-| Baseline (current trends) | B2 | 0.378 → 0.378 | 0.372 → 0.389 | 0.42 → 0.54 (0.45) | 0.0019 | -0.001 / -0.001 |
-| Baseline (current trends) | B3 | 0.368 → 0.367 | 0.333 → 0.342 | 0.42 → 0.58 (0.48) | 0.0021 | -0.000 / -0.000 |
-| Baseline (current trends) | B4 | 0.340 → 0.340 | 0.292 → 0.294 | 0.42 → 0.63 (0.53) | 0.0024 | +0.000 / +0.000 |
-| Attention oligopoly | B1 | 0.348 → 0.348 | 0.360 → 0.379 | 0.42 → 0.53 (0.42) | 0.0017 | +0.001 / +0.000 |
-| Attention oligopoly | B2 | 0.320 → 0.320 | 0.314 → 0.324 | 0.42 → 0.56 (0.44) | 0.0019 | -0.000 / -0.000 |
-| Attention oligopoly | B3 | 0.314 → 0.314 | 0.268 → 0.262 | 0.42 → 0.60 (0.48) | 0.0025 | -0.000 / -0.000 |
-| Attention oligopoly | B4 | 0.290 → 0.290 | 0.216 → 0.201 | 0.42 → 0.65 (0.53) | 0.0033 | +0.000 / -0.000 |
-| Generative regime, concentrated lineages | B1 | 0.529 → 0.530 | 0.530 → 0.465 | 0.42 → 0.17 (0.44) | 0.0041 | -0.001 / -0.000 |
-| Generative regime, concentrated lineages | B2 | 0.477 → 0.478 | 0.479 → 0.474 | 0.42 → 0.38 (0.47) | 0.0036 | -0.004 / -0.003 |
-| Generative regime, concentrated lineages | B3 | 0.431 → 0.431 | 0.441 → 0.454 | 0.42 → 0.47 (0.51) | 0.0031 | -0.000 / -0.000 |
-| Generative regime, concentrated lineages | B4 | 0.396 → 0.396 | 0.416 → 0.424 | 0.42 → 0.51 (0.55) | 0.0027 | -0.000 / +0.000 |
-| Generative regime | B1 | 0.687 → 0.686 | 0.661 → 0.691 | 0.42 → 0.23 (0.47) | 0.0044 | -0.000 / -0.000 |
-| Generative regime | B2 | 0.663 → 0.662 | 0.613 → 0.653 | 0.42 → 0.29 (0.50) | 0.0043 | -0.000 / -0.001 |
-| Generative regime | B3 | 0.644 → 0.642 | 0.581 → 0.619 | 0.42 → 0.34 (0.53) | 0.0040 | -0.000 / -0.001 |
-| Generative regime | B4 | 0.582 → 0.581 | 0.507 → 0.522 | 0.42 → 0.39 (0.55) | 0.0038 | -0.000 / -0.000 |
+| Baseline (current trends) | B1 | 0.435 → 0.435 | 0.428 → 0.433 | 0.42 → 0.44 (0.43) | 0.0019 | -0.005 / -0.005 |
+| Baseline (current trends) | B2 | 0.382 → 0.382 | 0.372 → 0.389 | 0.42 → 0.54 (0.45) | 0.0019 | -0.001 / -0.001 |
+| Baseline (current trends) | B3 | 0.375 → 0.375 | 0.333 → 0.342 | 0.42 → 0.58 (0.48) | 0.0021 | -0.000 / -0.000 |
+| Baseline (current trends) | B4 | 0.351 → 0.351 | 0.292 → 0.294 | 0.42 → 0.63 (0.53) | 0.0024 | +0.000 / +0.000 |
+| Attention oligopoly | B1 | 0.349 → 0.349 | 0.360 → 0.379 | 0.42 → 0.53 (0.42) | 0.0017 | +0.001 / +0.000 |
+| Attention oligopoly | B2 | 0.323 → 0.323 | 0.314 → 0.324 | 0.42 → 0.56 (0.44) | 0.0019 | -0.000 / -0.000 |
+| Attention oligopoly | B3 | 0.320 → 0.320 | 0.268 → 0.262 | 0.42 → 0.60 (0.48) | 0.0025 | -0.000 / -0.000 |
+| Attention oligopoly | B4 | 0.300 → 0.300 | 0.216 → 0.201 | 0.42 → 0.65 (0.53) | 0.0033 | +0.000 / -0.000 |
+| Generative regime, concentrated lineages | B1 | 0.532 → 0.532 | 0.530 → 0.465 | 0.42 → 0.18 (0.44) | 0.0041 | -0.001 / -0.000 |
+| Generative regime, concentrated lineages | B2 | 0.483 → 0.483 | 0.479 → 0.474 | 0.42 → 0.38 (0.47) | 0.0036 | -0.003 / -0.003 |
+| Generative regime, concentrated lineages | B3 | 0.441 → 0.441 | 0.441 → 0.454 | 0.42 → 0.47 (0.51) | 0.0031 | -0.000 / -0.000 |
+| Generative regime, concentrated lineages | B4 | 0.410 → 0.410 | 0.416 → 0.424 | 0.42 → 0.51 (0.55) | 0.0027 | +0.000 / +0.000 |
+| Generative regime | B1 | 0.694 → 0.693 | 0.661 → 0.691 | 0.42 → 0.23 (0.47) | 0.0044 | -0.000 / -0.000 |
+| Generative regime | B2 | 0.675 → 0.674 | 0.613 → 0.653 | 0.42 → 0.29 (0.50) | 0.0043 | -0.000 / -0.001 |
+| Generative regime | B3 | 0.659 → 0.657 | 0.580 → 0.618 | 0.42 → 0.34 (0.53) | 0.0040 | -0.000 / -0.001 |
+| Generative regime | B4 | 0.599 → 0.599 | 0.506 → 0.521 | 0.42 → 0.39 (0.55) | 0.0038 | -0.000 / -0.000 |
 
 ## Readings and robustness class
 
 | Reading | B1 | B2 | B3 | B4 | class |
 |---|---|---|---|---|---|
-| endogenous capture at t100 (endogenous − fixed weights): Baseline (current trends) | +0.010 | +0.097 | +0.100 | +0.099 | quantitative shift |
+| endogenous capture at t100 (endogenous − fixed weights): Baseline (current trends) | +0.011 | +0.097 | +0.100 | +0.099 | quantitative shift |
 | ΔVFAC endogenous − fixed: Baseline (current trends) | +0.000 | -0.000 | -0.000 | -0.000 | QUALITATIVE BREAK |
 | endogenous capture at t100 (endogenous − fixed weights): Attention oligopoly | +0.108 | +0.122 | +0.126 | +0.128 | robust |
 | ΔVFAC endogenous − fixed: Attention oligopoly | +0.000 | +0.000 | +0.000 | +0.001 | quantitative shift |
-| endogenous capture at t100 (endogenous − fixed weights): Generative regime, concentrated lineages | -0.265 | -0.095 | -0.033 | -0.039 | quantitative shift |
-| ΔVFAC endogenous − fixed: Generative regime, concentrated lineages | +0.001 | +0.001 | -0.000 | -0.000 | QUALITATIVE BREAK |
-| endogenous capture at t100 (endogenous − fixed weights): Generative regime | -0.240 | -0.212 | -0.191 | -0.164 | robust |
+| endogenous capture at t100 (endogenous − fixed weights): Generative regime, concentrated lineages | -0.263 | -0.091 | -0.032 | -0.039 | quantitative shift |
+| ΔVFAC endogenous − fixed: Generative regime, concentrated lineages | +0.001 | +0.000 | -0.000 | -0.000 | QUALITATIVE BREAK |
+| endogenous capture at t100 (endogenous − fixed weights): Generative regime | -0.239 | -0.210 | -0.188 | -0.161 | robust |
 | ΔVFAC endogenous − fixed: Generative regime | -0.001 | -0.002 | -0.002 | -0.001 | quantitative shift |
 | shock loss difference (endogenous − fixed), generative regime | -0.000 | -0.000 | -0.000 | -0.000 | quantitative shift |
 
diff --git a/workplans/ASSEMBLY-WP-0018-gadm05-grounding.md b/workplans/ASSEMBLY-WP-0018-gadm05-grounding.md
index a124c79..571a755 100644
--- a/workplans/ASSEMBLY-WP-0018-gadm05-grounding.md
+++ b/workplans/ASSEMBLY-WP-0018-gadm05-grounding.md
@@ -49,12 +49,17 @@ decoupling and sinks-breach readings survive?
 
 ```task
 id: ASSEMBLY-WP-0018-T02
-status: progress
+status: done
 priority: high
 state_hub_task_id: "3691dd92-6ba0-51f2-8a95-6c569cbe1aab"
 ```
 
-**Started 2026-10-13, interrupted (search limit):** one control variable so
+**Done 2026-10-14:** starting levels from control variables (renewables
+0.12, sinks 0.30), compute load grounded (IEA: e2 0.03 → 0.0015), physical load
+on rival capacity (structural fix), material replenishment supported (copper
+reserves). The generative regime's sinks breach disappears; the BR-011
+conflict sharpens (fits EU 0.79 / DE 0.57 against task data ≤ 0.25).
+`models/GADM-0.5.md` § 7. Earlier note: **Started 2026-10-13, interrupted (search limit):** one control variable so
 far: atmospheric CO₂ 423 ppm, boundary 350, high-risk line 450 (PHC 2025, via
 a secondary source). That puts the sink class 73 % into the zone of increasing
 risk. Next: the full PHC 2025 table (PIK executive summary PDF) for all 13