Renormalized Local Observables and Stress–Energy

Renormalized local observables extract finite information from products of fields at the same point without choosing a preferred vacuum. The core workflow is always the same: characterize the universal Hadamard singularity, subtract it before coincidence, apply the required local differential operator, restore conservation, and state the finite curvature freedom. The state-dependent smooth remainder and the state-independent geometric terms must never be conflated.

Helpful background. Flat-Spacetime Quantum Field Theory provides quadratic field observables and normal ordering; Partial Differential Equations and Green Functions supplies distributional kernels and coincidence limits; Causality and Global Structure identifies the globally hyperbolic backgrounds used here; Stress–Energy and Conservation provides Hilbert variation and conservation; and Quantum Fields in Curved Spacetime supplies the volume map from Hadamard states to observables. The Riemann curvature tensor page develops the invariants entering local counterterms.

Enter this chapter

The chapter follows the scalar field $P=\Box-m^2-\xi R$ from its two-point function to quantities that can be compared between states or supplied to later semiclassical calculations. It assumes a smooth globally hyperbolic spacetime, a Hadamard state, and a declared subtraction scale and finite prescription. Boundaries, non-Hadamard states, and self-consistent metric response require additional analysis.

Read the pages in this order:

1. Point Splitting and Wick Squares subtracts a local parametrix and explains why the finite coincidence limit remains ambiguous.
2. Locally Covariant Wick Polynomials states the theorem restricting higher-power redefinitions to local curvature polynomials.
3. Renormalized Stress–Energy Tensor applies differential point splitting and restores exact conservation.
4. Renormalization Ambiguities and Gravitational Counterterms identifies finite stress-tensor shifts with gravitational coupling redefinitions.
5. Practical Stress-Tensor Calculations compares numerical methods and supplies an executable convergence benchmark.
6. Energy Conditions and Their Quantum Failure replaces classical pointwise positivity by state- and sampling-aware quantum bounds.

From a singular product to a local field

The two-point distribution $W_\omega(x,x')$ of a Hadamard state has a universal short-distance singular part $H_\ell(x,x')$. Their difference is smooth near $x=x'$, so

$$\langle\Phi^2(x)\rangle_{\omega,\ell} =\lim_{x'\to x}\left(W_\omega-H_\ell\right)(x,x')$$

exists. This is not merely normal ordering with respect to a reference vacuum: $H_\ell$ is constructed locally from the geometry and field equation. It therefore transforms naturally under admissible spacetime embeddings.

Finite does not mean unique. In four dimensions, the Wick square may be shifted by $c_1m^2+c_2R$. Higher powers have a triangular finite-renormalization structure. Local covariance, scaling, microlocal regularity, and the specified continuity and analyticity under background variations turn what might have been arbitrary functions into a finite list of curvature polynomials; the precise classification is Hollands and Wald 2001, §5.1, Theorem 5.1, and §5.2, PDF.

The proof architecture first compares two prescriptions and uses their common commutator to reduce the difference to lower Wick powers. In the original theorem, locality, analytic dependence on analytic metric families, and scaling then restrict the coefficient fields to finite curvature polynomials; continuity extends the result to smooth backgrounds. A counterexample to the claim that finiteness alone suffices is reference-vacuum normal ordering chosen independently on each spacetime: its coincidence limits can be finite while failing the compatibility under embeddings required of one natural field. Nor do covariance and conservation alone imply the finite polynomial classification; the scaling and regularity assumptions also matter.

A useful invariant is the difference between two Hadamard states in one prescription. The common parametrix and local finite terms cancel, leaving the coincidence limit of the smooth difference $W_\omega-W_{\omega'}$. This observation is both conceptual and computational: evaluating a smooth state difference directly can avoid subtracting two large renormalized numbers.

Stress–energy adds conservation

The stress tensor requires derivatives before coincidence. A covariant bidifferential operator acts on $W_\omega-H_\ell$, with primed indices parallel-transported to the unprimed point. Because a parametrix is not an exact bisolution, the provisional answer has a local divergence. A state-independent geometric correction restores $\nabla^a\langle T_{ab}\rangle=0$.

The conserved result still admits finite additions generated by local gravitational actions. In four dimensions these include the metric term, Einstein tensor, and two independent curvature-squared variations after the Euler identity is used. Their coefficients correspond to finite choices of cosmological, Newton, and higher-curvature couplings. Wald’s original point-splitting framework is developed in Wald 1977, pp. 1–19, with the conservation, trace, and uniqueness correction supplied by Wald 1978, p. 1477.

For a conformal field, conservation and local covariance leave a geometric trace anomaly. Its Euler and Weyl-squared coefficients carry different status from the scheme-dependent total derivative $\Box R$. The next chapter derives these coefficients using heat kernels; this chapter establishes why a stress-tensor prescription must state its scale and counterterm convention first.

Computation and physical interpretation

A numerical answer is not validated by finiteness alone. Mode sums, adiabatic subtraction, heat-kernel approximations, and numerical Green functions must be checked against at least one analytic or independent benchmark, multiple refinements, conservation, trace, boundary conditions, and state regularity. The chapter’s cylinder calculation uses Abel and Gaussian regulators to recover $L^2\rho=-\pi/6$ with quadratic convergence. It exposes wrong continuum subtractions by deliberate failure tests. Covariant point separation and its geometric subtraction architecture were established in Christensen 1976, pp. 2490–2501.

Quantum stress tensors also violate the pointwise weak and null energy conditions. Vacuum–two-particle interference or squeezing can make a local density negative while the global Hamiltonian remains bounded below. Quantum energy inequalities instead bound a smooth worldline average under stated field, state, trajectory, and sampling hypotheses. The absolute curved-spacetime result quoted here is specifically Fewster and Smith 2008, §3.1, Theorem 3.1, and §§3.2.2–3.3, not a universal theorem for every field or boundary condition.

What this chapter hands onward

At the end of the chapter, a well-posed local observable has five labels: field content, Hadamard state, background geometry, subtraction scale and scheme, and observer or tensor frame. State differences can be scheme independent even when absolute values are not. Conservation is an internal consistency condition, but it does not evolve the metric.

The anomaly chapter uses the same local counterterms to classify universal and scheme-dependent trace terms. The black-hole chapter uses state regularity and numerical point splitting near horizons. Semiclassical gravity begins only when a declared renormalized tensor is coupled to a metric equation and backreaction is solved.

For a first calculation, compare the Minkowski thermal state with the vacuum. The Wick-square page obtains $\Delta\langle\Phi^2\rangle=T^2/12$, and the stress-tensor page obtains $\rho=\pi^2T^4/30$ and $p=\rho/3$ for one real massless scalar. Both are finite state differences in a common prescription. Together they show why a finite temperature changes local observables without changing the universal short-distance subtraction.

Review the chapter

Two Hadamard states $\omega_1,\omega_2$ are evaluated with the same point-splitting prescription, and a colleague changes the prescription by adding a conserved local curvature tensor $C_{ab}$. Which of the absolute tensors and their difference change? What checks remain necessary before using the difference as a numerical result?

Solution

Each absolute expectation value shifts by the same $C_{ab}$, so their difference is unchanged. The common parametrix also cancels, leaving the bidifferential operator applied to the smooth two-point-function difference. Scheme independence does not by itself guarantee a correct computation: one must still verify mode or discretization convergence, conservation of the difference, the state and boundary conditions, tensor-frame regularity, and an analytic or independent benchmark. A shared normalization error in both states could otherwise survive, while a mismatch of subtraction prescriptions would invalidate the cancellation before any numerical comparison was attempted.

One practical check is to compute the smooth state difference directly and compare it with the subtraction of the two separately renormalized tensors. Agreement within the declared tolerance tests the implementation; disagreement points to inconsistent scales, counterterms, modes, or frames rather than to new state physics.

References

• Christensen, Stephen M. “Vacuum Expectation Value of the Stress Tensor in an Arbitrary Curved Background: The Covariant Point-Separation Method.” Physical Review D 14 (1976): 2490–2501. DOI.
• Décanini, Yves, and Antoine Folacci. “Hadamard Renormalization of the Stress-Energy Tensor for a Quantized Scalar Field in a General Spacetime of Arbitrary Dimension.” Physical Review D 78 (2008): 044025. DOI. Open PDF.
• Fewster, Christopher J., and Calvin J. Smith. “Absolute Quantum Energy Inequalities in Curved Spacetime.” Annales Henri Poincaré 9 (2008): 425–455. DOI. Open PDF.
• Hollands, Stefan, and Robert M. Wald. “Local Wick Polynomials and Time Ordered Products of Quantum Fields in Curved Spacetime.” Communications in Mathematical Physics 223 (2001): 289–326. DOI. Open PDF.
• Wald, Robert M. “The Back Reaction Effect in Particle Creation in Curved Spacetime.” Communications in Mathematical Physics 54 (1977): 1–19. DOI.
• Wald, Robert M. “Trace Anomaly of a Conformally Invariant Quantum Field in Curved Spacetime.” Physical Review D 17 (1978): 1477–1484. DOI.