# NS-DCRP-13 — Supplier Trace Critical Lift, Finite-Family Anti-Diffusion, and Adjoint-Trace Bridge - date: 2026-08-16 - status: research proof checkpoint - canonical source: UTF-8 Markdown - canonical math delimiters: `$...$` and `$$...$$` - objective: bypass the DCRP-12 work-multiplicity obstruction by extracting a scale-uniform finite-family trace witness directly from the dissipation-boundary supplier atom. - no full Navier--Stokes regularity claim is made. - principal internal dependencies: - FCBP-02 filtered adjoint localization; - FCBP-04 heat-filter / moving-window architecture; - FCBP-05 combined observation hierarchy with selected adjoint-trace channel; - FCBP-06 CAR0--CAR3 / Native CAR Compiler; - MORP-01 through MORP-05; - DCRP-08 through DCRP-12. - external primary calibration: - Cheskidov--Dai, arXiv:1507.06611v6; - Cheskidov--Shvydkoy, arXiv:1102.1944v2. --- # 1. Executive result DCRP-12 showed that a fixed amount of global critical heat-work may be spread over arbitrarily many normalized cells, so $$ \boxed{ \text{fixed global work} \not\Rightarrow \text{uniform local work coefficient}. } \tag{1.1} $$ That multiplicity obstruction does **not** apply to the dissipation-boundary supplier endpoint itself. DCRP-08 established that at a dissipation-boundary shell $$ Q=Q(t), \qquad \Lambda=\lambda_Q, $$ one has $$ \boxed{ \|u_Q(t)\|_\infty \ge c_0\nu\Lambda. } \tag{1.2} $$ Define the critically rescaled shell $$ \boxed{ w(y) = \Lambda^{-1} u_Q \left( x_\ast+\Lambda^{-1}y, t \right). } \tag{1.3} $$ After choosing $$ x_\ast $$ at a point of almost maximal shell amplitude, $$ \boxed{ \|w\|_\infty \ge c_0\nu. } \tag{1.4} $$ The Fourier support of $$ w $$ lies in one fixed annulus independent of $$ Q. $$ Therefore Bernstein gives $$ \boxed{ \|\nabla w\|_\infty \le C_B \|w\|_\infty. } \tag{1.5} $$ This implies a uniform finite-family trace theorem: there exist universal constants $$ r_\ast>0, \qquad c_\ast>0, $$ a fixed nonnegative bump $$ \eta\in C_c^\infty(B_{r_\ast}), $$ and one of only six signed coordinate functionals $$ \boxed{ \mathcal L_{i,\sigma}(w) = \sigma \int \eta(y) w_i(y) \,dy, \qquad i\in\{1,2,3\}, \quad \sigma\in\{-1,+1\}, } \tag{1.6} $$ such that $$ \boxed{ \mathcal L_{i,\sigma}(w) \ge c_\ast\nu. } \tag{1.7} $$ Thus: $$ \boxed{ \textbf{ every dissipation-boundary supplier carries a fixed, scale-uniform, six-test local trace atom. } } \tag{1.8} $$ This is fundamentally different from DCRP-12's diffuse work measure. No number of distant work cells can make all six supplier-centered trace coefficients vanish. The witness is: - generated from the actual Navier--Stokes state; - located at the actual supplier scale; - located at an actual supplier center; - fixed-shape after normalization; - finite-dimensional; - quantitatively scale uniform. The same terminal bump can be propagated backward by the heat adjoint: $$ \boxed{ \phi_{i,\sigma}(\tau) = e^{-\tau\Delta} ( \sigma\eta e_i ) } \tag{1.9} $$ in backward-time notation, producing a canonical selected caloric-adjoint trace family. Therefore the old abstract CAR1 problem has a concrete solution **for the supplier state coordinate**: $$ \boxed{ \textbf{ supplier atom} \Longrightarrow \textbf{ uniform finite-family native trace separation}. } } \tag{1.10} $$ The remaining issue is no longer anti-diffusion. It is compiler compatibility: > Does the specific `selected adjoint trace` channel used by the FCBP/MORP finite-window audit admit this fixed terminal family, or a uniformly equivalent filtered version? If yes, then the supplier mechanism cannot lie in the exact combined-invisible kernel $$ O_W^T=0. $$ If no, the mismatch is now finite and explicit: it is a trace-family admissibility problem, not a diffuse-carrier problem. --- # 2. Supplier endpoint from the dissipation wavenumber For the Navier--Stokes dissipation wavenumber in the $$ r=\infty $$ form, $$ \Lambda(t) = \lambda_{Q(t)}, $$ Cheskidov--Dai / Cheskidov--Shvydkoy give the boundary estimate $$ \boxed{ \|u_{Q(t)}(t)\|_\infty \ge c_0\nu\Lambda(t) } \tag{2.1} $$ whenever $$ 1<\Lambda(t)<\infty. $$ DCRP-08 already converted this by Bernstein into $$ \boxed{ \Lambda \|u_Q\|_2^2 \ge c_1\nu^2. } \tag{2.2} $$ The present round uses the stronger pointwise form (2.1). --- # 3. Critical rescaling Let $$ x_\ast $$ satisfy $$ |u_Q(x_\ast,t)| \ge \frac34 \|u_Q(t)\|_\infty. $$ Define $$ \boxed{ w(y) = \Lambda^{-1} u_Q \left( x_\ast+\Lambda^{-1}y, t \right). } \tag{3.1} $$ Then $$ \boxed{ |w(0)| \ge \frac34c_0\nu. } \tag{3.2} $$ The Fourier support of $$ w $$ lies in a fixed annulus $$ \boxed{ \mathcal A = \{ \xi: c_-\le|\xi|\le c_+ \}, } \tag{3.3} $$ where $$ 00. $$ Status: $$ \boxed{ \textbf{PROVED}. } $$ --- # 6. NEW THEOREM — Finite-Family Supplier Trace Lift Choose a fixed function $$ \eta \in C_c^\infty(B_{r_\ast}(0)), $$ with $$ \eta\ge0 $$ and $$ \boxed{ \int \eta(y)\,dy = 1. } \tag{6.1} $$ For $$ i\in\{1,2,3\}, \qquad \sigma\in\{-1,+1\}, $$ define $$ \boxed{ \mathcal L_{i,\sigma}(w) = \sigma \int \eta(y) w_i(y) \,dy. } \tag{6.2} $$ ## Theorem 6.1 For every dissipation-boundary supplier shell, after the admissible critical rescaling and spatial re-centering above, $$ \boxed{ \max_{ 1\le i\le3, \ \sigma=\pm1 } \mathcal L_{i,\sigma}(w) \ge c_2\nu. } \tag{6.3} $$ ### Proof Use the selected pair $$ (i_\ast,\sigma_\ast) $$ from Section 5. Since $$ \eta\ge0, $$ has unit mass, and $$ \sigma_\ast w_{i_\ast}\ge c_2\nu $$ throughout its support, $$ \mathcal L_{i_\ast,\sigma_\ast}(w) \ge c_2\nu. $$ $$ \square $$ Status: $$ \boxed{ \textbf{PROVED}. } $$ --- # 7. Why this is a genuine anti-diffusion theorem The work multiplicity obstruction of DCRP-12 concerns a measure whose fixed total mass may be split into $$ N\to\infty $$ separate cells. Theorem 6.1 does not estimate a sum of work cells. It uses one dynamically selected supplier endpoint. Once the supplier center is chosen, one of six fixed local coordinate tests has a uniform lower bound. Therefore: $$ \boxed{ \text{supplier endpoint} \Longrightarrow \text{one local coefficient }\ge c\nu } \tag{7.1} $$ independently of: - number of other active cells; - total work multiplicity; - spatial distribution of the remaining solution; - pressure--flux cancellation elsewhere. Thus: $$ \boxed{ \textbf{ supplier-state trace visibility cannot be defeated by work fragmentation. } } \tag{7.2} $$ This bypasses rather than solves the global heat-work multiplicity problem. --- # 8. Physical-variable form Recall $$ w(y) = \Lambda^{-1} u_Q \left( x_\ast+\Lambda^{-1}y,t \right). $$ Then $$ \begin{aligned} \mathcal L_{i,\sigma}(w) &= \sigma \int \eta(y) \Lambda^{-1} u_{Q,i} \left( x_\ast+\Lambda^{-1}y,t \right) dy\\ &= \sigma \Lambda^2 \int \eta \left( \Lambda(x-x_\ast) \right) u_{Q,i}(x,t) dx. \end{aligned} $$ Hence Theorem 6.1 is equivalently $$ \boxed{ \max_{i,\sigma} \sigma \Lambda^2 \int \eta \left( \Lambda(x-x_\ast) \right) u_{Q,i}(x,t) dx \ge c_2\nu. } \tag{8.1} $$ The normalization $$ \Lambda^2 $$ is exactly the one dictated by the Navier--Stokes scaling of this local linear trace. --- # 9. Filtered-state interpretation Let $$ P_{\mathcal A} $$ denote the fixed unit-annulus Littlewood--Paley projector in normalized variables. The supplier shell is $$ w = P_{\mathcal A}v, $$ where $$ v $$ is the full normalized velocity state. Therefore: $$ \boxed{ \mathcal L_{i,\sigma}(w) = \sigma \int \eta \left( P_{\mathcal A}v \right)_i. } \tag{9.1} $$ This is a **filtered selected-time trace** of the actual normalized Navier--Stokes state. It is generated from the state by a fixed Fourier filter and a fixed local test. No dangerous/singular label is copied into the coordinate. Thus it passes the FCBP-06 Copied-Gate safety requirement. --- # 10. Native CAR1 interpretation FCBP-06 isolates CAR1 as the missing statement: $$ \boxed{ \operatorname{dist}_{\rm native} \ge a_0 \mu_{\rm dang} - \mathcal R^{extract}, } $$ with the requirement that the native geometry must be generated from Navier--Stokes rather than from a copied dangerous mark. Theorem 6.1 provides a concrete supplier-side separation: $$ \boxed{ \mathsf T_{\rm sup}(v) := \max_{i,\sigma} \mathcal L_{i,\sigma} ( P_{\mathcal A}v ) \ge c_2\nu. } \tag{10.1} $$ The quantity $$ \mathsf T_{\rm sup} $$ is: - state-generated; - scale normalized; - spatially re-rooted by an allowed symmetry; - finite-dimensional; - quantitatively uniform. Thus, for any native package norm that contains the six filtered trace coefficients as genuine components, $$ \boxed{ \operatorname{dist}_{\rm native} \ge c_3\nu } \tag{10.2} $$ away from the subspace where all six supplier traces vanish. Status: $$ \boxed{ \textbf{CAR1 PROVED FOR THIS CONCRETE SUPPLIER-TRACE SUBGEOMETRY}. } $$ This is not yet a theorem about the entire external admissible quotient $$ \Gamma_W. $$ That compiler identification remains explicit. --- # 11. Finite-dimensional anti-phantom advantage The old work-carrier branch had an effective number of cells $$ \mathfrak M_{\rm work} $$ that could diverge. The supplier trace vector is only: $$ \boxed{ \mathbf T_{\rm sup} = \left( \mathcal L_{1,+}, \mathcal L_{1,-}, \mathcal L_{2,+}, \mathcal L_{2,-}, \mathcal L_{3,+}, \mathcal L_{3,-} \right). } \tag{11.1} $$ Its dimension is fixed: $$ \boxed{ \dim \mathbf T_{\rm sup} = 6. } \tag{11.2} $$ Theorem 6.1 gives $$ \boxed{ \|\mathbf T_{\rm sup}\|_{\ell^\infty} \ge c_2\nu. } \tag{11.3} $$ Thus no moving-window multiplicity constant occurs at the extraction stage. This is precisely the geometry that FCBP-06's Native CAR Detector Compiler is designed to exploit once the trace vector is identified with an admissible detector/quotient component. --- # 12. Backward caloric adjoint family For each terminal test $$ \psi_{i,\sigma} = \sigma \eta e_i, $$ define the backward heat-adjoint family on normalized time $$ \tau\le0 $$ by $$ \boxed{ \Psi_{i,\sigma}(y,\tau) = e^{-\tau\Delta} \psi_{i,\sigma}(y), \qquad \tau\le0. } \tag{12.1} $$ Then $$ \boxed{ \partial_\tau \Psi_{i,\sigma} + \Delta \Psi_{i,\sigma} = 0, } \tag{12.2} $$ and $$ \boxed{ \Psi_{i,\sigma}(y,0) = \psi_{i,\sigma}(y). } \tag{12.3} $$ The terminal filtered trace is $$ \boxed{ \left< P_{\mathcal A}v(0), \Psi_{i,\sigma}(0) \right> = \mathcal L_{i,\sigma} ( P_{\mathcal A}v(0) ). } \tag{12.4} $$ Therefore one of this fixed six-element terminal adjoint family satisfies $$ \boxed{ \left< P_{\mathcal A}v(0), \Psi_{i,\sigma}(0) \right> \ge c_2\nu. } \tag{12.5} $$ This gives a canonical route from the supplier trace atom to a selected caloric-adjoint trace. --- # 13. Relation to FCBP filtered adjoint localization FCBP-02 already uses a backward filtered adjoint weight to cancel the principal localization residual. FCBP-04 / FCBP-05 explicitly retain a selected adjoint-trace channel in the combined observability hierarchy. Therefore the structure required by DCRP-13 is not foreign to the existing compiler. However the current corpus does not state, in one theorem, that the exact selected-adjoint family $$ \Psi_{i,\sigma} $$ from Section 12 is an admissible basis for $$ O_W^T. $$ The final identification must therefore be stated conditionally. --- # 14. Conditional theorem — direct collision with the trace-zero kernel ## Theorem 14.1 Assume the FCBP/MORP selected adjoint-trace channel $$ O_W^T $$ contains, after the standard scale/translation normalization, the six terminal caloric trace functionals generated by: $$ \psi_{i,\sigma} = \sigma\eta e_i. $$ Then every dissipation-boundary supplier state satisfies $$ \boxed{ O_W^T \ge c_T\nu } \tag{14.1} $$ for a universal: $$ c_T>0. $$ Consequently no such supplier state belongs to the exact trace-invisible kernel $$ \boxed{ O_W^T=0. } \tag{14.2} $$ ### Proof Theorem 6.1 gives one terminal coefficient at least $$ c_2\nu. $$ By the assumed channel inclusion, the trace observation norm dominates that coefficient up to a fixed normalization constant. $$ \square $$ Status: $$ \boxed{ \textbf{CONDITIONAL ONLY ON TRACE-FAMILY COMPILER ADMISSIBILITY}. } $$ No scale-uniform observability constant is otherwise needed for this direct finite-family branch. --- # 15. Corollary — supplier sequence cannot be combined-invisible if the trace family is admissible DCRP-08 proved that a hypothetical finite-time singularity requires a sequence $$ t_n\uparrow T $$ with: $$ \Lambda_n\to\infty $$ and supplier shells $$ Q_n=Q(t_n). $$ Under Theorem 14.1's compiler assumption, every normalized supplier state satisfies: $$ \boxed{ O_{W,n}^T \ge c_T\nu. } \tag{15.1} $$ Hence: $$ \boxed{ \textbf{ the supplier sequence cannot enter any combined-invisible branch that requires } O_W^T\to0. } \tag{15.2} $$ This directly attacks the FCBP-06 combined-invisible cascade survivor. Status: $$ \boxed{ \textbf{CONDITIONAL ON THE SAME TRACE-COMPILER IDENTIFICATION}. } $$ --- # 16. Why this is stronger than local positive energy alone A positive local $$ L^2 $$ mass statement gives: $$ \int_{B_R} |w|^2 \ge c\nu^2. $$ To convert that to a finite set of **linear** detector coefficients one would normally need a finite-dimensional approximation argument. The dissipation-boundary $$ L^\infty $$ lower bound is stronger. Because the field is band-limited, pointwise largeness persists on a fixed ball with a fixed coordinate sign. Therefore a fixed **six-element linear test family** already detects it. No compactness, singular-value decomposition, or increasing detector dimension is needed. --- # 17. The detector shape is not solution dependent The following are fixed once and for all: - the reference bump: $$ \eta; $$ - the six component/sign choices: $$ (i,\sigma); $$ - the unit-annulus filter: $$ P_{\mathcal A}. $$ The solution determines only: - the admissible spatial re-centering: $$ x_\ast; $$ - the admissible parabolic scale: $$ \Lambda^{-1}; $$ - which one of the six tests is positive. Thus the detector **family** is fixed and finite. This avoids the tautology: > choose the test to be the solution itself. No such solution-dependent shape is used. --- # 18. Rotational normalization If the MORP state normalization also allows spatial rotations, the six-test family can be reduced conceptually to one coordinate test after rotation. However no rotation is required. Keeping all six signed coordinate tests has two advantages: 1. it avoids a separate rotational selection theorem; 2. it makes finite-dimensionality explicit. Thus: $$ \boxed{ 6 } $$ is a safe universal detector count. --- # 19. Quantitative stability under approximate supplier threshold Suppose only: $$ \|u_Q\|_\infty \ge (c_0-\varepsilon)\nu\Lambda $$ with: $$ 0\le\varepsilon = - \left< F_q(s), \varphi_q(s) \right>. $$ Integrating: $$ \boxed{ \|u_q(t_1)\|_2^2 - \left< u_q(t_0), e^{\nu(t_1-t_0)\Delta} u_q(t_1) \right> = - \int_{t_0}^{t_1} \left< F_q(s), \varphi_q(s) \right> ds. } \tag{21.2} $$ This is an exact signed adjoint ancestry identity. --- # 22. NEW THEOREM — matched-adjoint nonlinear payment Let: $$ A_q(t) = \lambda_q^{1/2} \|u_q(t)\|_2. $$ Assume at: $$ t_1 $$ the supplier satisfies: $$ A_q(t_1) \ge a_0\nu. $$ Let: $$ K_0 = \|u(0)\|_2^2. $$ Choose: $$ \tau_q = \frac1{ c_h\nu\lambda_q^2 } \log \left( \frac{ 2\lambda_q^{1/2}K_0^{1/2} }{ a_0\nu } \right) $$ as in DCRP-09, and set: $$ t_0=t_1-\tau_q. $$ Then: $$ \boxed{ -\lambda_q \int_{t_0}^{t_1} \left< F_q(s), \varphi_q(s) \right> ds \ge \frac12 a_0^2\nu^2. } \tag{22.1} $$ ### Proof By shell heat decay: $$ \left\| e^{\nu\tau_q\Delta} u_q(t_1) \right\|_2 \le e^{-c_h\nu\lambda_q^2\tau_q} \|u_q(t_1)\|_2. $$ Hence: $$ \left| \left< u_q(t_0), e^{\nu\tau_q\Delta} u_q(t_1) \right> \right| \le K_0^{1/2} e^{-c_h\nu\lambda_q^2\tau_q} \|u_q(t_1)\|_2. $$ By definition of: $$ \tau_q, $$ the right side is at most: $$ \frac12 \|u_q(t_1)\|_2^2. $$ Equation (21.2) gives: $$ -\int_{t_0}^{t_1} \left< F_q, \varphi_q \right> ds \ge \frac12 \|u_q(t_1)\|_2^2. $$ Multiply by: $$ \lambda_q. $$ Since: $$ \lambda_q \|u_q(t_1)\|_2^2 = A_q(t_1)^2 \ge a_0^2\nu^2, $$ the result follows. $$ \square $$ Status: $$ \boxed{ \textbf{PROVED}. } $$ --- # 23. Significance of the matched-adjoint payment DCRP-09 proved an unsigned source-norm lower bound: $$ \lambda_q^{1/2} \int \|F_q\|_2 dt \gtrsim \nu. $$ Theorem 22.1 gives a stronger **signed dual pairing**: $$ \boxed{ -\lambda_q \int \left< F_q, e^{\nu(t_1-s)\Delta} u_q(t_1) \right> ds \gtrsim \nu^2. } \tag{23.1} $$ This pairing is: - exact; - actual-history; - scale critical; - sign definite after the heat-memory term is removed; - naturally adjoint. Thus the supplier branch generates both: $$ \boxed{ \text{terminal finite-family trace atom} } $$ and: $$ \boxed{ \text{signed matched-adjoint nonlinear payment}. } $$ This is much less compatible with an adjoint-trace-invisible minimal obstruction than a diffuse unsigned work measure. --- # 24. Why the matched adjoint is not yet an admissible detector by itself The terminal state: $$ u_q(t_1) $$ appears inside: $$ \varphi_q. $$ Therefore the matched adjoint shape is solution dependent. Using it directly as a detector would risk the same kind of tautological adaptivity that FCBP-06 warns against. For this reason: - Theorem 22.1 is retained as a native signed identity; - Theorem 6.1 is the actual finite-family detector extraction. The six-test trace lift removes the solution dependence from the detector family. A future compiler theorem may use Theorem 22.1 to prove that one of the six fixed adjoint channels inherits nonzero nonlinear payment, but that step is not claimed here. --- # 25. Updated Critical Lift status The original FCBP Critical Lift problem asked for a non-tautological, scale-uniform route from dangerous causal data to an auditable local observable. The DCRP chain has now produced: 1. hypothetical singularity: $$ \Longrightarrow $$ 2. arbitrarily high dissipation-boundary suppliers: $$ \Longrightarrow $$ 3. critical endpoint shell atom: $$ \Longrightarrow $$ 4. fixed local pointwise normalized amplitude: $$ \Longrightarrow $$ 5. finite six-test trace lower bound: $$ \boxed{ \max_{i,\sigma} \mathcal L_{i,\sigma} \ge c\nu. } $$ This is a genuine native, scale-uniform extraction statement. Therefore the supplier branch has solved the **geometry** of CAR1. What remains is a finite compiler question: $$ \boxed{ \textbf{ is the supplier trace family contained in, or uniformly controlled by, the already-declared }O_W^T\textbf{ adjoint-trace channel?} } \tag{25.1} $$ --- # 26. If the compiler answer is yes If: $$ O_W^T \gtrsim \max_{i,\sigma} \mathcal L_{i,\sigma}, $$ then: $$ \boxed{ O_W^T \ge c\nu } \tag{26.1} $$ on every supplier state. Since a hypothetical first singularity forces suppliers at arbitrarily high scales: $$ \boxed{ \liminf_{n\to\infty} O_{W,n}^T \ge c\nu. } \tag{26.2} $$ Thus the supplier branch cannot enter a moving-window combined-invisible cascade with: $$ O_{W,n}^{comb}\to0. $$ At that point the remaining global closure work would return to: - paid-side recurrence; - transition realization; - whether every hypothetical singular branch must pass through the supplier-normalized MORP minimal object. The diffuse-work multiplicity obstruction would no longer be relevant to supplier observability. --- # 27. If the compiler answer is no If the declared: $$ O_W^T $$ does **not** admit the six fixed caloric terminal traces, then the gap is now explicit. One must explain which of the following fails: 1. filtered velocity traces are not part of the trace state; 2. the allowed terminal adjoint family excludes fixed compact bumps; 3. the trace is defined only for a different tensor/source variable; 4. the filter class cannot include the fixed unit-annulus shell; 5. normalization loses the trace under actual return/re-root. Any such failure is a finite interface mismatch. It is no longer: $$ \boxed{ \text{unknown diffuse NS obstruction}. } $$ --- # 28. Relation to DCRP-12 DCRP-12 remains useful for the physical work ledger. Its result is: $$ \boxed{ \text{local PFET} \vee \text{paid backscatter} \vee \text{work escape}. } $$ DCRP-13 does not invalidate that theorem. It proves a different statement: $$ \boxed{ \text{work may diffuse, but the supplier state itself has a fixed trace atom}. } $$ Thus the two routes are complementary. ### Work route tracks **how the supplier is paid**. ### Trace route tracks **whether the supplier can be observationally invisible**. The trace route is immune to work-cell multiplicity. --- # 29. New exact frontier The previous frontier was: $$ \text{Quantitative Work Anti-Diffusion / Critical Lift}. $$ The supplier trace theorem bypasses the anti-diffusion half. The next exact target is now: $$ \boxed{ \textbf{ Supplier Trace / FCBP Adjoint-Channel Identification Lemma}. } $$ Desired statement: > After the standard supplier scale/translation normalization, the six fixed terminal filtered trace functionals > > $$ > \mathcal L_{i,\sigma} > $$ > > belong to the admissible selected-adjoint trace family defining > > $$ > O_W^T, > $$ > > or are uniformly dominated by that trace norm. > > Therefore: > > $$ > O_W^T > \ge > c\nu > $$ > > on every dissipation-boundary supplier. This is now a finite compiler theorem. No new PDE mechanism is required. --- # 30. Source ledger ## Cheskidov--Dai Alexey Cheskidov and Mimi Dai, *Regularity criteria for the 3D Navier-Stokes and MHD equations*, arXiv:1507.06611v6. Used for the dissipation-wavenumber architecture and the Navier--Stokes high-frequency boundary condition. ## Cheskidov--Shvydkoy Alexey Cheskidov and Roman Shvydkoy, *A unified approach to regularity problems for the 3D Navier-Stokes and Euler equations: the use of Kolmogorov's dissipation range*, arXiv:1102.1944v2. Contains the explicit boundary estimate: $$ \|u_{Q(t)}(t)\|_\infty \ge c_0\nu\Lambda(t) $$ on the active dissipation-wavenumber set. ## Internal FCBP-02 Already uses a backward filtered adjoint cutoff and proves cancellation of the principal localization residual. ## Internal FCBP-05 Declares the combined observation hierarchy: $$ O_W^{comb} = ( O_W^P, O_W^F, O_W^E, O_W^T ), $$ including the selected adjoint-trace channel. ## Internal FCBP-06 Factorizes Critical Lift into: $$ \mathrm{CAR0} \to \mathrm{CAR1} \to \mathrm{CAR2} \to \mathrm{CAR3}. $$ The supplier trace theorem provides a concrete scale-uniform native separation for a specific NS-generated state coordinate and respects the Copied-Gate prohibition. --- # 31. End state The key new theorem is: $$ \boxed{ \textbf{ Dissipation-boundary supplier} \Longrightarrow \textbf{ finite-family local trace atom}. } $$ Quantitatively: $$ \boxed{ \max_{ 1\le i\le3,\, \sigma=\pm1 } \sigma \int \eta(y) \left[ P_{\mathcal A}v(y) \right]_i dy \ge c\nu. } $$ The family has fixed dimension: $$ \boxed{ 6. } $$ No work multiplicity can dilute this coefficient. In addition, the supplier has a signed matched-adjoint nonlinear payment: $$ \boxed{ -\lambda_Q \int \left< F_Q(s), e^{\nu(t_Q-s)\Delta} u_Q(t_Q) \right> ds \ge c\nu^2. } $$ Thus the supplier mechanism is simultaneously: - state-visible; - trace-visible to a finite fixed family; - dynamically and nonlinearly generated. The next single frontier is: $$ \boxed{ \textbf{ Supplier Trace / FCBP Adjoint-Channel Identification Lemma}. } $$ If this finite compiler bridge holds, the dissipation-boundary supplier branch cannot be combined-invisible.