Trumpet Geometry: A Cross-Disciplinary Overview

• Trumpet geometry is a class of structures defined by asymptotically cylindrical or flaring ends that control dynamic behavior in various physical and mathematical systems.
• In numerical relativity and JT gravity, trumpet configurations enable natural excision of black hole interiors and serve as key gluing interfaces in path integral formulations.
• In string theory and acoustic engineering, trumpet profiles inform T-dual background constructions and control wave propagation or shock suppression through precise area functions.

Searching arXiv for recent and foundational papers on “trumpet geometry” across the relevant subfields. “Trumpet geometry” denotes several domain-specific geometric structures that are not equivalent to one another but share a characteristic limiting profile: an asymptotically cylindrical or flaring end, or a variable-area passage whose narrowing or widening controls dynamics. In numerical relativity it refers to horizon-penetrating black-hole slices whose inner end is a finite-area limiting surface at infinite proper distance; in Jackiw–Teitelboim gravity it denotes hyperbolic trumpet and double-trumpet surfaces used in gluing formulas and spectral observables; in string theory it names the vector-gauged $SL(2,\mathbb{R})/U(1)$ background and related dual constructions; and in acoustics and compressible-flow engineering it describes geometries of ducts and nozzles whose area function governs wave steepening, reflection, radiation, or jet shaping (Thierfelder et al., 2010, Zhou, 13 Apr 2026, Giveon et al., 2013, Resch et al., 2016, Zhou et al., 2021).

1. Relativistic trumpet slices in Schwarzschild spacetime

In moving-puncture relativity, a Schwarzschild black hole evolved with the standard puncture gauges settles to a stationary, asymptotically cylindrical slice known as a trumpet. The numerical coordinates cover the exterior region from spatial infinity down to a finite areal radius $R_0$, but they do not continue to the second asymptotically flat end of the initial wormhole data. Instead, as the isotropic coordinate $r \to 0$, the slice approaches a cylindrical limit surface at finite areal radius $R_0\approx 1.3\,M$; the puncture behaves as a coordinate singularity with a regular exterior, and the unresolved interior is effectively removed by “natural excision” (Thierfelder et al., 2010).

The standard spherically symmetric line element used in this analysis is

$$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$

with areal radius $R=\sqrt{\gamma_{\theta\theta}}$. Near the puncture, stationary trumpet data satisfy the characteristic scalings

$$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$

and the coordinate-independent relation

$K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$

which matches the collapse end-state and analytic trumpet expansion to numerical accuracy (Thierfelder et al., 2010).

A distinct analytic family of horizon-penetrating, spatially isotropic Schwarzschild trumpets makes the cylindrical end explicit by prescribing the areal radius as

$R(r)=r+R_0,\qquad 0<R_0\le M.$

In that family,

$$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$

and the inner end has finite areal radius $R_0$0 but infinite proper radial distance. The special case $R_0$1 yields especially simple closed-form expressions, while the limit $R_0$2 removes the trumpet and reduces the coordinates to Painlevé–Gullstrand form (Dennison et al., 2014).

These constructions clarify a basic distinction often blurred in informal discussion. A trumpet slice is not a wormhole slice: wormhole data connect two asymptotically flat ends, whereas trumpet data terminate on a finite-area cylinder. Nor is the trumpet end a physical singularity. Its defining features are gauge-theoretic and geometric: finite limiting area, vanishing lapse, and divergent proper distance to the inner end (Dennison et al., 2014).

2. Dynamical formation and extensions in numerical relativity

Spherically symmetric stellar collapse with the same puncture gauges reaches the same late-time exterior trumpet geometry found in vacuum puncture evolutions. In the collapse studied in (Thierfelder et al., 2010), an apparent horizon forms at $R_0$3, and by $R_0$4 the exterior has become stationary; for the choice $R_0$5, the matter content remaining on the slice drops to atmosphere levels, $R_0$6, while the coordinate-independent diagnostic $R_0$7 falls below $R_0$8. The mechanism is the Gamma-driver shift: it stretches the radial coordinates so that matter falls inside the innermost grid point and becomes unresolved, while the exterior vacuum settles into stationary trumpet coordinates (Thierfelder et al., 2010).

This role of the shift is central. With $R_0$9, the innermost grid point acts like an outflow boundary for matter, the areal radius of that point grows rapidly to $r \to 0$0, and the effective areal resolution near the stellar surface drops by about an order of magnitude. The lapse simultaneously collapses through the singularity-avoidance of $r \to 0$1, freezing coordinate time in the strong-field region. A plausible implication is that the trumpet is best understood not as a property of vacuum data alone, but as the stationary gauge attractor for a broad class of black-hole-forming evolutions using puncture gauges (Thierfelder et al., 2010).

The trumpet concept extends well beyond static Schwarzschild. In Bondi accretion, transforming the fluid solution into trumpet coordinates yields regular expressions that extend through the horizon and into the black-hole interior up to the limiting surface. Two explicit cases are emphasized: the maximal trumpet with limiting areal radius $r \to 0$2, and an analytical trumpet with $r \to 0$3, for which

$r \to 0$4

making trumpet-coordinate Bondi flow a stationary code test in moving-puncture evolutions (Miller et al., 2016).

For rotating black holes, trumpet geometry can be characterized in a $r \to 0$5 formalism by three conditions: the lapse vanishes on a finite-area inner boundary, the proper distance to that boundary is infinite, and the exterior spatial geometry is regular. An analytic stationary family of Kerr trumpet slices realizes these properties. The trumpet surface is $r \to 0$6, its area is

$r \to 0$7

and the admissible parameter range is

$r \to 0$8

The same construction extends to Kerr–Newman–(A)dS black holes (Dennison et al., 2014).

Trumpet geometries also arise in higher dimensions and in cosmological settings. In Schwarzschild–Tangherlini spacetimes, special members of the maximal and generalized $r \to 0$9 families are trumpet slices, and $R_0\approx 1.3\,M$0 moving-puncture BSSN evolutions settle to the corresponding analytic trumpets (Dennison et al., 2010). In Schwarzschild–de Sitter spacetimes, static constant-mean-curvature trumpet slicings can be recast in a McVittie-like isotropic coordinate system. At large distances the metric asymptotes to a flat FLRW form with exponentially expanding scale factor, while as the comoving isotropic radius goes to zero it approaches a horizon-penetrating trumpet geometry rather than a second asymptotically de Sitter end (Dennison et al., 2017).

Hyperboloidal formulations sharpen the same geometric criteria. There the trumpet throat is a finite areal radius $R_0\approx 1.3\,M$1 inside the event horizon, with $R_0\approx 1.3\,M$2 and $R_0\approx 1.3\,M$3, so that the proper radial distance diverges while the slice still reaches future null infinity $R_0\approx 1.3\,M$4. In dynamical scalar-field collapse, the apparent horizon emerges at $R_0\approx 1.3\,M$5, and the hyperboloidal slices transition from regular initial data to trumpet geometry (Vañó-Viñuales, 2023).

Initial-data constructions exploit these properties directly. Trumpet initial data for boosted black holes in the moving-punctures approach start with trumpet geometry inside the horizons and reduce initial transients: the reported junk radiation is suppressed by as much as two orders of magnitude relative to comparable Bowen–York data (Slinker et al., 2018). More abstractly, non-constant-mean-curvature trumpet solutions of the vacuum constraint equations exist on manifolds with asymptotically Euclidean and asymptotically conformally cylindrical or periodic ends, placing trumpet geometry within the general conformal method for Einstein initial data (Leach, 2016). Spectral initial-data calculations likewise treat trumpet and wormhole punctures in a unified single-domain Galerkin–Collocation framework; for spinning trumpets, the limiting surface is deformed from a sphere to an oblate spheroid, with eccentricity tending to $R_0\approx 1.3\,M$6 at large spin (Clemente et al., 2017).

3. Trumpets and double trumpets in Jackiw–Teitelboim gravity

In JT gravity, a trumpet is a hyperbolic cylinder or half-infinite funnel with one asymptotic boundary and one geodesic boundary. Its Euclidean path integral furnishes a basic gluing kernel. In the normalization used in (Zolfi et al., 20 Feb 2025), the trumpet amplitude with asymptotic boundary length $R_0\approx 1.3\,M$7 and geodesic boundary length $R_0\approx 1.3\,M$8 is

$R_0\approx 1.3\,M$9

After the continuation $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$0, this kernel admits a tunneling interpretation: at fixed energy, it imposes $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$1, or equivalently $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$2, so the process can be read as converting an older black hole into a younger one while emitting a baby universe of length $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$3 (Zolfi et al., 20 Feb 2025).

The double trumpet is obtained by gluing two single trumpets along a common geodesic of length $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$4. In AdS$$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$5, it is the standard connected saddle for two-boundary observables, with Weil–Petersson measure $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$6. In the second-order formalism, the two-boundary path integral reduces to an integral over $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$7 of a product of “flaring boundary” factors, and in the spectral form factor the double trumpet gives the ramp contribution. With matter present, however, the moduli-space integral can diverge at small $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$8; suitable twisted boundary conditions render the ground-state energy positive and remove that divergence (Moitra et al., 2022).

The same double-trumpet structure reappears in random-state calculations. In the modular spectral form factor of a Haar random state, the large-bond-dimension replica answer is dominated by annular non-crossing permutations, and the paper explicitly interprets these as a discretized version of the double trumpet. The gluing modulus is mirrored by the number of connected loops $$ds^2 = -\alpha^2 dt^2 + \gamma_{rr}(dr + \beta^r dt)^2 + \gamma_{\theta\theta} d\Omega^2,$$9, the twist by the relative cyclic shift, and the resulting connected form factor develops a linear ramp, precisely as in the JT double-trumpet analysis (Cao et al., 2024).

Finite-cutoff JT gravity modifies the trumpet picture in a controlled way. The exact finite-cutoff disk amplitude can be written in an open-channel operator formulation using a rigid cosine kernel

$R=\sqrt{\gamma_{\theta\theta}}$0

together with a compact $R=\sqrt{\gamma_{\theta\theta}}$1-band $R=\sqrt{\gamma_{\theta\theta}}$2. The induced geodesic sector is therefore bandlimited, admits exact sampled representations, and the resulting finite-cutoff disk amplitude is a compact-support branch-difference observable rather than the ordinary thermal trace of a single lower-bounded $R=\sqrt{\gamma_{\theta\theta}}$3-independent Hamiltonian (Zhou, 13 Apr 2026).

A recurrent subtlety concerns how much of moduli space supports a given physical interpretation. For higher-genus or multi-boundary amplitudes, a naive use of the full Weil–Petersson volume introduces terms corresponding to differences of boundary lengths, not only sums of emitted baby-universe lengths. The Airy-limit ribbon-graph decomposition isolates the portion of moduli space whose exponential factors retain the tunneling interpretation based on sums of geodesic lengths (Zolfi et al., 20 Feb 2025).

4. String-theoretic cigar–trumpet backgrounds

In string theory, “trumpet” most commonly denotes the vectorially gauged $R=\sqrt{\gamma_{\theta\theta}}$4 supercoset. In the Euclidean near-horizon geometry of non-extremal NS5-branes, it describes the region beyond the singularity, while the axially gauged coset gives the cigar geometry outside the horizon. In cylindrical coordinates,

$R=\sqrt{\gamma_{\theta\theta}}$5

with $R=\sqrt{\gamma_{\theta\theta}}$6. The neck at $R=\sqrt{\gamma_{\theta\theta}}$7 is a non-compact cusp, $R=\sqrt{\gamma_{\theta\theta}}$8, and the geometry approaches a cylinder at $R=\sqrt{\gamma_{\theta\theta}}$9 (Giveon et al., 2013).

This trumpet is T-dual to the $$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$0 orbifold of the cigar SCFT. The relation is not merely geometric but exact at the level of worldsheet conformal field theory. The corresponding elliptic genera satisfy the “curious sum rule”

$$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$1

where the third term is the analytically continued negative-level minimal model associated with the Euclidean region between horizon and singularity (Giveon et al., 2013).

A heterotic generalization places the cigar and trumpet inside a one-parameter family of exact two-dimensional backgrounds obtained by relaxing isotropy in the generalized structure group of the duality-covariant construction. After Wick rotation, the trumpet endpoint $$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$2 is

$$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$3

while the cigar endpoint $$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$4 is

$$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$5

The heterotic gauge field strength vanishes at $$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$6, so the gauge sector decouples in the exact coset limits (Hassler et al., 2024).

A doubled-world-sheet derivation reaches the same pair from a different direction. Starting from the $$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$7 WZW parent with two abelian isometries, the gauging-and-reduction procedure permits only two consistent two-dimensional children: the trumpet,

$$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$8

and the cigar, obtained by gauging the other isometry. The two are related by Buscher T-duality, and the paper emphasizes that no other reduced conformal backgrounds arise in this $$\psi(r)\sim C\,r^{-1/2},\qquad \alpha(r)\sim a\,r^{1.09},\qquad K\to K_0\approx 0.30/M,$$9 example (Bakas et al., 2016).

A common misconception is that the trumpet is simply the cigar with a different coordinate chart. The data above show a stronger statement: cigar and trumpet are distinct gauged-coset or reduced backgrounds related by axial/vector gauging or T-duality, and their regularity properties differ sharply. The cigar has a smooth tip, whereas the trumpet has a cusp or strong-coupling singularity at its inner end (Giveon et al., 2013, Hassler et al., 2024).

5. Acoustic and compressible-flow meanings of trumpet geometry

In brass acoustics, trumpet geometry is governed by the area function $K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$0. The three-dimensional simulations in (Resch et al., 2016) show that the near-mouthpiece bore profile, especially the subtle widening in the first $K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$1 cm from the mouthpiece shank, strongly affects finite-amplitude wave propagation, harmonic enrichment, and radiated spectra. The B$K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$2 Barcelona BTR-200LQ trumpet was modeled with three computational geometries. A constant-bore approximation (“Geometry 1”) overestimated the radiated spectrum by $K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$3–$K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$4 dB, whereas a measured seven-point spline in the near-mouthpiece section (“Geometry 3”) matched experiment almost perfectly in the $K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$5–$K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$6 Hz band for $K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$7 and in the $K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$8–$K(\alpha)=0.300937-0.930916\,\alpha+O(\alpha^2),$9 Hz band for $R(r)=r+R_0,\qquad 0<R_0\le M.$0 (Resch et al., 2016).

The quantitative sensitivity is unusually strong. Increasing the near-mouthpiece radius by $R(r)=r+R_0,\qquad 0<R_0\le M.$1 mm shifts the radiated spectrum by $R(r)=r+R_0,\qquad 0<R_0\le M.$2 dB. In the simplified Geometry 1, the mouthpiece boundary radius was $R(r)=r+R_0,\qquad 0<R_0\le M.$3 that of the more accurate geometries, implying a $R(r)=r+R_0,\qquad 0<R_0\le M.$4 dB amplitude overestimate at the bell and yielding an observed $R(r)=r+R_0,\qquad 0<R_0\le M.$5–$R(r)=r+R_0,\qquad 0<R_0\le M.$6 dB discrepancy. The specific measured ratios reported in the paper—mouthpiece throat radius $R(r)=r+R_0,\qquad 0<R_0\le M.$7 the shank radius and bore radius at $R(r)=r+R_0,\qquad 0<R_0\le M.$8 cm downstream $R(r)=r+R_0,\qquad 0<R_0\le M.$9 the shank radius—quantify how rapidly the narrow region widens and why that widening moderates nonlinear wave steepening (Resch et al., 2016).

In laser-plasma accelerator target design, “trumpet” refers to a convex converging–diverging nozzle whose diverging wall is a circular arc with negative curvature radius $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$0. The governing geometric constraint is the circular-arc relation between throat radius $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$1, exit radius $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$2, diverging length $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$3, and exit half-angle $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$4, with optimization based on matching $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$5 to the Mach angle,

$$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$6

For $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$7 mm, $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$8 mm, $$\psi(r)=\sqrt{1+\frac{R_0}{r}},\qquad \alpha(r)=\frac{r}{r+R_0},$$9 mm, and $R_0$00, the study found $R_0$01 mm and $R_0$02 to be optimal (Zhou et al., 2021).

The physical effect is shock suppression. Bell nozzles focus compression waves and create strong density spikes; straight nozzles reduce that focusing but retain residual plateau modulations; trumpet nozzles push the shock diamonds upstream and behind the exit plane, producing flatter density plateaus at standoff distances $R_0$03–$R_0$04 mm. Plateau uniformity $R_0$05 was consistently lower for the trumpet than for the straight nozzle, for example $R_0$06 versus $R_0$07 for He at $R_0$08 mm and $R_0$09 versus $R_0$10 for Ar at $R_0$11 mm, at the cost of slightly thicker edges (Zhou et al., 2021).

These acoustic and nozzle results underscore a broader point. In both cases the word “trumpet” names not merely a gross outline but a precise area profile. The near-mouthpiece taper in a musical trumpet and the convex arc of a gas-jet nozzle are both first-order control parameters for wave or flow evolution, not secondary geometric refinements (Resch et al., 2016, Zhou et al., 2021).

6. Cross-disciplinary structure and recurrent distinctions

Across these literatures, the trumpet motif repeatedly combines a finite limiting cross section with nontrivial propagation toward or away from that limit. In relativity, the limiting sphere has finite area but lies at infinite proper distance, so the slice is regular outside while excluding the singularity (Dennison et al., 2014, Vañó-Viñuales, 2023). In JT gravity, the geodesic boundary has fixed finite length and serves as the gluing interface for amplitudes, ramps, and baby-universe processes (Moitra et al., 2022, Zolfi et al., 20 Feb 2025). In string theory, the trumpet is the T-dual or vector-gauged counterpart of the cigar, with a cusp or strong-coupling tip rather than a smooth cap (Giveon et al., 2013, Bakas et al., 2016). In acoustics and nozzle design, the decisive quantity is again a controlled radius or area function that changes reflection, nonlinearity, or shock structure (Resch et al., 2016, Zhou et al., 2021).

The major distinctions are equally important. Relativistic trumpets are gauge-dependent spatial slicings, not invariant characterizations of the spacetime manifold itself (Thierfelder et al., 2010). JT trumpets are hyperbolic surfaces used inside path integrals and gluing relations, and finite-cutoff JT replaces the usual asymptotic spectral picture by compact support and a branch-difference structure (Zhou, 13 Apr 2026). String-theoretic trumpets are exact or duality-related target-space backgrounds. Acoustic and nozzle trumpets are material or computational geometries whose performance depends sensitively on local shape. The shared label therefore indicates a family resemblance in limiting profile, not a single mathematical object.

This diversity explains why “trumpet geometry” has become a productive but context-sensitive term. In each field it isolates the geometry of a boundary or throat that governs the relevant dynamics: gauge settling and natural excision in black-hole numerics, gluing and spectral behavior in JT gravity, axial/vector duality in string theory, and nonlinear propagation or shock control in acoustics and compressible flows (Thierfelder et al., 2010, Cao et al., 2024, Hassler et al., 2024, Resch et al., 2016).