Heat and Wave kernel expansions for stationary spacetimes

This paper generalizes classical spectral geometry to stationary spacetimes by investigating the relationship between wave-trace expansions, heat kernel coefficients, and zeta function residues, specifically deriving a complex second wave-trace coefficient that reduces to the standard scalar curvature term in the ultrastatic limit.

The Gist

Imagine the universe not as a static stage, but as a flowing river. In physics, we often study how waves (like sound or light) ripple through this river. Usually, the river is chaotic and changing, making it impossible to predict the waves' behavior easily.

However, this paper focuses on a special kind of river: a stationary spacetime. Think of this as a river that flows in a perfectly steady, unchanging pattern. Even though the water is moving, the shape of the flow looks exactly the same if you take a snapshot today or a snapshot a year from now. In physics, this steady flow is guided by something called a "Killing field"—a kind of invisible, unchanging clockwork that keeps the universe's rhythm consistent.

The authors, Alexander Strohmaier and the late Steve Zelditch, are trying to solve a massive puzzle: How do we count and describe the "notes" (frequencies) that this river can sing?

Here is a breakdown of their work using simple analogies:

1. The Musical Instrument Analogy

In standard geometry (like studying a drum), we look at the drumhead and ask: "What are the frequencies it can vibrate at?" This is called spectral geometry.

• The Drum: A Riemannian manifold (a curved surface).
• The Vibration: The Laplacian operator (the math that describes how waves move).
• The Notes: The eigenvalues (the specific frequencies).

This paper asks: "What happens if our drum is actually a spacetime (space + time) that is flowing steadily?"

• The New Instrument: A stationary spacetime.
• The New Vibration: The Wave Equation (which includes time).
• The New Notes: The frequencies of time-translations.

The authors show that even though this is a 4D (or higher) spacetime problem, it behaves mathematically very much like the simpler 3D drum problem, but with some extra "twists" caused by the flow of time.

2. The "Wave-Trace" and the Echo

Imagine shouting into a cave. The sound bounces off the walls and comes back to you. If you analyze the echo, you can learn about the shape of the cave.

• The Shout: A wave traveling through spacetime.
• The Echo: The Wave-Trace. This is a mathematical tool that sums up all the possible vibrations of the universe at a specific moment.

The authors are interested in what happens right at the very beginning of the echo (time = 0). In mathematics, this moment is "singular" (it blows up), but if you zoom in, you see a pattern. It's like looking at a fractal; the pattern repeats itself in layers.

3. The "Heat Kernel" and the Recipe Book

To understand these complex echoes, the authors use a trick involving Heat.

• Imagine heating up a metal plate. The heat spreads out over time.
• The Heat Kernel is a mathematical recipe that tells you exactly how the heat spreads from one point to another.
• In this paper, the authors show that the "notes" of the spacetime (the wave-trace) are deeply connected to how "heat" would spread in this flowing universe.

They discovered that the first few terms of this "heat recipe" tell us the most important things about the universe's shape.

• The First Term: Tells us the total "volume" of the universe (how big the cave is).
• The Second Term (The Big Discovery): This is the main result of the paper. It tells us about the curvature and the twist of the spacetime.

4. The "Frame-Dragging" Twist

In a simple, static universe (like a calm pond), the math is straightforward. But in a stationary universe (like a spinning black hole or a rotating star), space itself gets "dragged" along with the rotation. This is called frame-dragging.

The authors calculated the second term of their expansion.

• In a calm pond: This term is just about the curvature (how bumpy the surface is).
• In a spinning river: This term gets complicated. It includes the curvature plus the effects of the spin (the "shift vector") and how the flow of time stretches or squeezes space (the "lapse function").

They found a complex formula that acts like a universal translator. If you plug in a simple, non-spinning universe, the formula simplifies to the classic result everyone knows. But if you plug in a spinning universe, it reveals the hidden geometry of that spin.

5. Why Does This Matter?

You might ask, "Why do we need such a complicated formula?"

1. Universal Laws: It proves that the rules of spectral geometry (which we know well for static shapes) can be extended to dynamic, flowing universes. It unifies the math of "still" space and "moving" space.
2. Black Holes and Stars: Real astrophysical objects like rotating stars or the regions far away from black holes are "stationary." This math helps physicists understand the quantum vibrations (particles) in these extreme environments.
3. The "Fingerprint" of the Universe: Just as a fingerprint identifies a person, the "wave-trace invariants" (the coefficients they calculated) identify the geometry of the spacetime. If you know these numbers, you know the shape and flow of the universe, even if you can't see it directly.

Summary

Think of the universe as a giant, flowing musical instrument.

• Old Math: Could only describe the notes of a stationary drum.
• This Paper: Describes the notes of a spinning, flowing drum.
• The Result: They wrote down the exact recipe (the second coefficient) to calculate the "sound" of this spinning drum. This recipe accounts for the curvature of space, the flow of time, and the "drag" caused by rotation.

It's a bridge between the abstract world of pure math and the physical reality of our dynamic, rotating universe.

Technical Summary

1. Problem Statement and Context

The paper addresses the spectral geometry of the wave equation on stationary spacetimes. While classical spectral geometry focuses on the Laplacian on Riemannian manifolds, this work generalizes these concepts to Lorentzian manifolds possessing a complete timelike Killing vector field $Z$.

• The Operator: The generator of time-translations for the wave equation (or Klein-Gordon equation) on such spacetimes acts as a Hamiltonian $H = i\mathcal{L}_Z$. When the spacetime is ultrastatic (product of time and a Riemannian manifold), $H$ reduces to the square root of the Laplacian.
• The Challenge: For general stationary spacetimes (which may include rotation or "frame-dragging" effects), the spectral analysis is more complex. The authors aim to compute wave-trace invariants near time zero ($t=0$). These invariants are crucial because they:
  1. Determine the Weyl law (asymptotic distribution of eigenvalues).
  2. Provide coefficients for the heat kernel expansion.
  3. Relate to the poles of the spectral zeta function.
• Specific Gap: Previous work ([SZ18]) established the leading term (Weyl law) and related it to the geometry of null geodesics. This paper seeks to compute the second non-zero term in the wave-trace expansion and express it in terms of local geometric invariants, bridging the gap between Lorentzian spectral theory and classical Riemannian heat kernel coefficients.

2. Methodology

The authors employ a combination of microlocal analysis, Hadamard expansions, and invariant theory.

A. Geometric Setup

The spacetime $(M, g)$ is assumed to be globally hyperbolic, stationary, and spatially compact.

• Metric Decomposition: The metric is analyzed in two ways:
  1. Cauchy Surface View: $M \cong \mathbb{R} \times \Sigma$. The metric is written using the Lapse function $N$ and Shift vector $w$:
     $$\Phi^*g = -N^2 dt^2 + h_{jk}(dx^j + w^j dt)(dx^k + w^k dt)$$
  2. Principal Bundle View: $M$ is viewed as an $\mathbb{R}$-principal bundle over the space of Killing orbits $K$. This provides a more invariant description involving the norm of the Killing field $|Z| = \sqrt{-g(Z,Z)}$.

B. Hadamard Expansion

The core computational tool is the Hadamard expansion of the fundamental solution (Green's function) $G$ of the wave operator $\square_g + W$.

• The singularity of $G$ near the diagonal is expanded using Riesz distributions $R_\beta$ and smooth Hadamard coefficients $V_k$.
• The authors specialize this expansion to the stationary setting, exploiting the symmetry under the Killing flow to show that coefficients depend only on the time difference $t - t'$.

C. Wave-Trace Computation

The wave trace is defined as the distribution $\text{tr}(e^{-itH})$.

• The authors relate the global trace to a local wave-trace $Q_y(s)$ integrated over a Cauchy surface $\Sigma$.
• $Q_y(s)$ is derived from the normal derivatives of the Green's function on the diagonal: $Q_y(s) = (\partial_{\nu, y} G_s - \partial_{\nu, y'} G_s)|_{y'=y}$.
• By substituting the Hadamard expansion into this expression, they perform an asymptotic analysis as $s \to 0$ to extract the coefficients of the homogeneous distributions $\mu_\alpha(s)$.

D. Gauge Invariance

A significant technical hurdle is that the local expression for the second coefficient $c_2(x)$ appears to depend on the choice of the Cauchy surface $\Sigma$ (specifically via the unit normal $\nu$).

• The authors utilize the Pohozaev-Schoen identity (adapted to Lorentzian manifolds) and conformal transformation properties of the Ricci tensor to prove that the integral of this coefficient over $\Sigma$ is actually a spectral invariant, independent of the choice of $\Sigma$.

3. Key Contributions and Results

A. The Wave-Trace Expansion

The wave trace admits an expansion at $t=0$ of the form:
$$\text{tr}(e^{-itH}) \sim c_0 \mu_{n-1}(t) + c_1 \mu_{n-2}(t) + c_2 \mu_{n-3}(t) + \dots$$
where $\mu_\alpha$ are homogeneous distributions.

Main Theorem (Theorem 1.3):

1. Vanishing of $c_1$: The coefficient $c_1$ is identically zero.
2. Explicit Formula for $c_2$: The coefficient $c_2$ is given by the integral over a Cauchy surface $\Sigma$:
   $$c_2 = \int_\Sigma c_2(x) \, d\text{Vol}_h(x)$$
   where the local density $c_2(x)$ is:
   $$\begin{aligned} \tilde{c}_2(x) = &\left(\frac{1}{6}\text{scal}(x) - W(x)\right) N(x)|Z|^{-n+2} \\ &- \frac{n-2}{6} \text{Ric}(Z, Z) N(x)|Z|^{-n} \\ &+ \frac{n(n-2)}{12} N(x) g(\nabla_Z Z, \nabla_Z Z) |Z|^{-n-2} \\ &+ \frac{1}{3} \text{Ric}(\nu, Z) |Z|^{-n+2} + \frac{n-2}{3} |Z|^{-n} g(\nabla^2_Z Z, \nu) \end{aligned}$$
   Here, $\text{scal}$ is the scalar curvature, $\text{Ric}$ is the Ricci tensor, $\nabla$ is the covariant derivative, and $\nu$ is the future-pointing unit normal to $\Sigma$.

B. Connection to Heat Kernel and Ultrastatic Limits

• Ultrastatic Case: When $N=1$ and $w=0$ (ultrastatic), the complex formula simplifies drastically to the classical second heat kernel coefficient:
  $$\tilde{c}_2(x) = \frac{1}{6}\text{scal}(x) - W(x)$$
  This confirms the consistency of the general theory with classical Riemannian spectral geometry.
• Heat Trace: The result implies an asymptotic expansion for the heat trace $\text{tr}(e^{-tH^2})$ as $t \to 0^+$, allowing for the meromorphic continuation of the spectral zeta function.

C. Example: Rotating Sphere

The authors apply their formula to a rotating round sphere (Example 1.4).

• They explicitly compute the spectrum of the wave operator.
• They verify that the integral of the derived $c_2$ matches the asymptotic expansion of the spectral sum, demonstrating the formula's validity in a non-trivial stationary (rotating) setting.

4. Significance

1. Generalization of Spectral Geometry: This work successfully extends the machinery of heat kernel coefficients and Weyl laws from Riemannian manifolds to the broader class of stationary spacetimes. It establishes that the "second coefficient" in the wave trace is a robust spectral invariant even in the presence of frame-dragging (shift vector $w \neq 0$).
2. Physical Relevance: In General Relativity, spectral properties describe the dispersion and decay of quantum fields on curved backgrounds. These coefficients provide precise quantitative data about how the geometry (curvature, rotation) influences the density of states and the short-time behavior of quantum propagators.
3. Resolution of Gauge Dependence: The paper resolves the apparent non-invariance of the local coefficient $c_2(x)$ under changes of the Cauchy surface. By proving the integral is invariant, it solidifies the geometric interpretation of these spectral invariants as properties of the spacetime itself, not the foliation.
4. Foundation for Future Work: The techniques developed (combining Hadamard expansions with principal bundle geometry) provide a framework for studying more complex spacetimes, potentially including non-compact cases or those with horizons, via scattering theory.

In summary, the paper provides a rigorous derivation of the second wave-trace invariant for stationary spacetimes, linking it explicitly to local curvature and Killing field geometry, and demonstrating its reduction to classical results in the ultrastatic limit.