Harmonic Coarse Spaces

• Harmonic coarse spaces are constructions that use local harmonic constraints to distill large-scale information into robust, coarse-level representations for both computational and geometric settings.
• In numerical Helmholtz solvers, these spaces employ locally Helmholtz-harmonic modes and SPD spectral selection to maintain near wavenumber-independent iteration counts.
• In geometric analysis, the method promotes unstable coarse maps to canonical harmonic representatives under stability conditions, ensuring analytical rigidity.

Harmonic coarse spaces are coarse-level constructions built from local harmonicity conditions, with the term appearing in two technically distinct settings in recent arXiv literature. In numerical domain decomposition for the heterogeneous Helmholtz equation, a harmonic coarse space is a coarse space whose basis functions are locally Helmholtz-harmonic and are selected by a self-adjoint positive-definite spectral criterion (Dolean et al., 2 Sep 2025). In geometric analysis of harmonic maps to Hadamard targets, the phrase “Harmonic Coarse Spaces” organizes a viewpoint in which coarse Lipschitz or quasi-isometric data admit canonical harmonic representatives under a stability condition, thereby turning coarse classes into analytically rigid objects (Riestenberg et al., 14 Nov 2025). The common thread is that coarse-scale information is filtered through harmonicity to obtain robust or canonical global representatives.

1. Terminological scope and governing ideas

In the Helmholtz setting, the guiding idea is to build coarse basis functions that are locally Helmholtz-harmonic—i.e., they exactly satisfy the homogeneous heterogeneous Helmholtz equation in each subdomain—and then assemble them globally with partition-of-unity weights (Dolean et al., 2 Sep 2025). The resulting coarse correction is used in two-level domain decomposition methods, especially ORAS-based preconditioners, to improve robustness with respect to the wavenumber in heterogeneous media.

In the geometric-analysis setting, the central idea is different in implementation but similar in spirit. A coarse Lipschitz map from a complete manifold $X$ to a proper Hadamard space $Y$ is required to satisfy a stability condition, expressed as strict coarse subharmonicity under averaging, and this condition implies the existence of a harmonic map at bounded distance (Riestenberg et al., 14 Nov 2025). Under additional assumptions, that harmonic representative is unique. This suggests a broader conceptual use of the phrase: coarse-geometric data are promoted to harmonic objects when a suitable positivity or drift condition is available.

A plausible implication is that “harmonic coarse space” functions as a unifying label for constructions in which nonlocal or large-scale data are compressed into harmonic representatives. In the computational literature, the output is a finite-dimensional coarse space for iterative solvers; in the geometric literature, the output is a bounded-distance harmonic representative for a coarse map.

2. Harmonic coarse spaces for the heterogeneous Helmholtz equation

The computational setting considered in (Dolean et al., 2 Sep 2025) is the heterogeneous Helmholtz PDE

$$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$

with boundary conditions including the impedance condition

$$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$

and possibly Dirichlet data on $\Gamma_D=\partial\Omega\setminus\Gamma_R$. A typical choice is $Z(x)=\sqrt{a(x)\,m(x)}$. In the constant-coefficient case, $a\equiv 1$, $m\equiv 1$, this reduces to the standard Helmholtz equation $-\Delta u-k^2u=f$ with $k=\omega$ (Dolean et al., 2 Sep 2025).

With a simplicial mesh $Y$0 and Lagrange elements, the discrete problem is

$Y$1

where $Y$2 is indefinite and frequency-dependent. The overview emphasizes that accuracy at high $Y$3 requires mesh resolution scaling with $Y$4 because of the pollution effect, and notes the common practice of fixing points per wavelength, for example $Y$5–$Y$6 in $Y$7, with similar or slightly lower values in $Y$8 when higher-order elements are used (Dolean et al., 2 Sep 2025).

Within two-level ORAS, the one-level preconditioner is built from overlapping subdomains and local Robin problems,

$Y$9

while the two-level correction incorporates a coarse basis matrix $$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$0 through

$$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$1

An additive form is also given: $$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$2 In this setting, harmonic coarse spaces are designed to provide the matrix $$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$3 in a manner aligned with the wave physics (Dolean et al., 2 Sep 2025).

The significance of this construction is explicitly framed in terms of wavenumber robustness. Restricting the search space to local Helmholtz solutions ensures the coarse basis captures the propagative content of the PDE rather than extraneous volumetric components, while the spectral selection identifies the modes that are least well handled by purely local iteration (Dolean et al., 2 Sep 2025).

3. Local Helmholtz-harmonic subspaces and spectral selection

On a subdomain $$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$4, the local Helmholtz-harmonic subspace is defined by

$$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$5

that is, the space of heterogeneous Helmholtz solutions on $$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$6 with the original boundary conditions applied on $$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$7 (Dolean et al., 2 Sep 2025). Equivalently, these functions satisfy the local homogeneous PDE

$$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$8

The positive Helmholtz energy is introduced as the SPD form

$$-\nabla\cdot\big(a(x)\nabla u\big) - \omega^2\,m(x)\,u = f \quad \text{in }\Omega,$$9

Using the partition-of-unity operator $$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$0, the harmonic spectral problem is then: find $$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$1 such that

$$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$2

The overview states that this problem is self-adjoint and coercive and that all eigenvalues are real and positive (Dolean et al., 2 Sep 2025). The interpretation given there is that the eigenproblem selects locally Helmholtz-harmonic modes whose energy changes least under partition-of-unity weighting.

In discretized form, Helmholtz-harmonicity can be imposed by a constraint, leading to the saddle-point eigenproblem

$$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$3

where $$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$4 is the local Dirichlet Helmholtz matrix, $$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$5 is its restriction to interior rows, $$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$6 and $$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$7 are the local matrices associated with $$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$8, and $$a(x)\,\partial_n u + i\,\omega\,Z(x)\,u = 0 \quad \text{on }\Gamma_R,$$9 is a Lagrange multiplier enforcing Helmholtz-harmonicity (Dolean et al., 2 Sep 2025).

The retained eigenvectors are those with $\Gamma_D=\partial\Omega\setminus\Gamma_R$0, where $\Gamma_D=\partial\Omega\setminus\Gamma_R$1 is a user-defined threshold, and the global coarse basis is assembled by

$\Gamma_D=\partial\Omega\setminus\Gamma_R$2

Each resulting basis vector is globally supported through partition-of-unity weighting, locally Helmholtz-harmonic on its subdomain, and chosen by an SPD spectral criterion (Dolean et al., 2 Sep 2025).

4. Extended harmonic coarse spaces, oversampling, and implementation

An alternative construction in (Dolean et al., 2 Sep 2025) is the extended harmonic coarse space, described as an algebraic variant with oversampling. For each $\Gamma_D=\partial\Omega\setminus\Gamma_R$3, an enlarged subdomain $\Gamma_D=\partial\Omega\setminus\Gamma_R$4 is formed by adding at least one layer of mesh elements across interfaces, with restriction $\Gamma_D=\partial\Omega\setminus\Gamma_R$5 and matrix $\Gamma_D=\partial\Omega\setminus\Gamma_R$6. The harmonic projector is

$\Gamma_D=\partial\Omega\setminus\Gamma_R$7

where $\Gamma_D=\partial\Omega\setminus\Gamma_R$8 is the local Robin inverse on $\Gamma_D=\partial\Omega\setminus\Gamma_R$9. The local eigenproblem becomes

$Z(x)=\sqrt{a(x)\,m(x)}$0

The overview again states that this problem is SPD and that its eigenvalues are real and positive; moreover, only matrix-vector products with $Z(x)=\sqrt{a(x)\,m(x)}$1 are needed, so explicit assembly is unnecessary (Dolean et al., 2 Sep 2025).

Retaining the modes with $Z(x)=\sqrt{a(x)\,m(x)}$2, the basis is assembled as

$Z(x)=\sqrt{a(x)\,m(x)}$3

By construction, these basis functions belong to the global finite element space and correspond to local discrete Helmholtz solutions in $Z(x)=\sqrt{a(x)\,m(x)}$4 (Dolean et al., 2 Sep 2025). Oversampling is described as improving robustness in a manner akin to MS-GFEM.

The implementation pathway is laid out explicitly. One partitions $Z(x)=\sqrt{a(x)\,m(x)}$5 into $Z(x)=\sqrt{a(x)\,m(x)}$6 subdomains, creates overlap by adding layers of elements, assembles the local Robin matrices $Z(x)=\sqrt{a(x)\,m(x)}$7, builds $Z(x)=\sqrt{a(x)\,m(x)}$8, $Z(x)=\sqrt{a(x)\,m(x)}$9, $a\equiv 1$0, and $a\equiv 1$1, defines the partition-of-unity weights $a\equiv 1$2, solves the local SPD eigenproblems, assembles $a\equiv 1$3, forms and factors the coarse matrix $a\equiv 1$4, and finally applies right-preconditioned GMRES with the two-level ORAS preconditioner (Dolean et al., 2 Sep 2025). The same source notes that wider overlaps improve convergence and that a partition-of-unity variant with vanishing derivative at the interface slightly improves results.

For parameter selection, the overview reports empirically effective thresholds:

• Harmonic: $a\equiv 1$5.
• Extended harmonic: $a\equiv 1$6–$a\equiv 1$7.

It also emphasizes that the number of independent Helmholtz-harmonic modes that can be glued across interfaces is bounded by the number of interface degrees of freedom $a\equiv 1$8, which scales like $a\equiv 1$9, even though the local eigenproblems themselves involve $m\equiv 1$0 degrees of freedom (Dolean et al., 2 Sep 2025). This compactness is a central practical distinction from volumetric spectral spaces.

5. Robustness, comparison with other coarse spaces, and numerical trade-offs

The overview identifies three reasons for the performance of harmonic coarse spaces in Helmholtz solvers (Dolean et al., 2 Sep 2025). First, the construction is physics-informed, because the candidate space consists of local Helmholtz solutions. Second, the spectral machinery is stable: unlike DtN or indefinite Helmholtz spectral problems, harmonic coarse space eigenproblems are SPD with real spectra and therefore admit simple selection criteria. Third, empirical robustness is observed across several challenging scenarios.

The reported numerical evidence covers $m\equiv 1$1 homogeneous and heterogeneous squares, medical imaging with PML, the $m\equiv 1$2 COBRA cavity, and a $m\equiv 1$3 crustal geomodel (Dolean et al., 2 Sep 2025). Across these cases, harmonic and extended harmonic coarse spaces are described as delivering near wavenumber-independent iteration counts when $m\equiv 1$4 is increased appropriately, with coarse space dimensions growing roughly with interface degrees of freedom. Extended harmonic is reported as consistently among the most robust choices, particularly in $m\equiv 1$5.

The comparison with GenEO-type spectral coarse spaces is explicit. $m\equiv 1$6-GenEO is said to be robust for elliptic problems but to deteriorate as $m\equiv 1$7 grows because of mismatch with Helmholtz physics. Hk-GenEO replaces the right-hand SPD operator with one closer to Helmholtz, improving robustness relative to $m\equiv 1$8-GenEO, but often producing larger coarse spaces and, with Robin conditions, complex eigenvalues that are selected by $m\equiv 1$9 (Dolean et al., 2 Sep 2025). By contrast, harmonic coarse spaces are restricted to the Helmholtz-harmonic subspace $-\Delta u-k^2u=f$0, making them smaller and more physically relevant than full-volume GenEO spaces.

The main trade-off identified is between setup cost and iterative robustness. Increasing $-\Delta u-k^2u=f$1 adds modes and reduces iteration counts until diminishing returns, while the global coarse operator factorization eventually dominates setup time (Dolean et al., 2 Sep 2025). The overview therefore recommends choosing $-\Delta u-k^2u=f$2 only large enough for iteration counts to plateau. It also recommends using $-\Delta u-k^2u=f$3 for internal Robin transmission and in building the positive Helmholtz operators, at least two layers of overlap, and adequate points per wavelength so that pollution error does not mask preconditioner performance (Dolean et al., 2 Sep 2025).

6. Harmonic representatives of coarse maps in Hadamard targets

A separate but mathematically related use of the “Harmonic Coarse Spaces” theme appears in (Riestenberg et al., 14 Nov 2025). There, $-\Delta u-k^2u=f$4 is a proper Hadamard space, meaning a complete CAT(0) metric space with compact closed balls. CAT(0) geometry implies unique geodesic segments between pairs of points, convexity of distance functions and Busemann functions, and comparison properties including Reshetnyak’s four-point subembedding and the Ptolemy inequality

$-\Delta u-k^2u=f$5

These structures are used in harmonic map estimates (Riestenberg et al., 14 Nov 2025).

A map $-\Delta u-k^2u=f$6 is coarse Lipschitz if there exist $-\Delta u-k^2u=f$7 such that

$-\Delta u-k^2u=f$8

or equivalently

$-\Delta u-k^2u=f$9

The domain $k=\omega$0 is assumed to be a complete Riemannian $k=\omega$1-manifold with

$k=\omega$2

which yields heat kernel and Green’s function estimates, mean value inequalities, and Gaussian decay (Riestenberg et al., 14 Nov 2025).

For harmonic maps $k=\omega$3 in the Korevaar–Schoen sense, the energy density $k=\omega$4 satisfies the weak differential inequalities

$k=\omega$5

The paper then introduces stability as a sufficient criterion for a coarse Lipschitz map to lie within bounded distance of a harmonic map (Riestenberg et al., 14 Nov 2025). One formulation is the heat-evolution condition: there exists $k=\omega$6 such that for all $k=\omega$7 and $k=\omega$8,

$k=\omega$9

Another uses harmonic probability measures supported in balls of radius $Y$00, yielding coarse strict subharmonicity at scale $Y$01. Stability is coarse: if $Y$02 is stable and $Y$03 is uniformly bounded, then $Y$04 is also stable, possibly at a changed scale (Riestenberg et al., 14 Nov 2025).

The main existence theorem states that if $Y$05 is a complete $Y$06-manifold with $Y$07, $Y$08 is a proper Hadamard space, and $Y$09 is an $Y$10-coarse Lipschitz map stable at scale $Y$11, then there exists a harmonic map $Y$12 and a constant $Y$13 such that

$Y$14

(Riestenberg et al., 14 Nov 2025). In symmetric-space targets of noncompact type, uniqueness also holds under cocompactness assumptions on $Y$15: among harmonic maps at bounded distance from a stable $Y$16, there is a unique representative (Riestenberg et al., 14 Nov 2025).

In this framework, the “harmonic coarse space” perspective is that a coarse class $Y$17 with positive drift admits a canonical harmonic representative $Y$18 at bounded distance (Riestenberg et al., 14 Nov 2025). The same paper uses this to generalize the Schoen–Li–Wang conjecture beyond rank $Y$19, to define a universal Hitchin component $Y$20 for $Y$21, and to identify that component both with quasi-symmetric positive maps $Y$22 and with harmonic maps $Y$23 (Riestenberg et al., 14 Nov 2025).

7. Conceptual synthesis, limitations, and open directions

The two literatures differ in objects and methods, but each uses harmonicity to control coarse-scale structure. In (Dolean et al., 2 Sep 2025), local Helmholtz-harmonicity and SPD spectral selection generate compact coarse corrections for two-level ORAS. In (Riestenberg et al., 14 Nov 2025), coarse strict subharmonicity and positive drift select bounded-distance harmonic representatives of coarse maps. This suggests a common abstract pattern: large-scale information is not used directly, but is filtered through harmonic constraints that isolate the globally relevant modes.

The limitations are also explicit. In the geometric theory, $Y$24 must be proper CAT(0) for compactness limits; if $Y$25 splits as $Y$26, no map is stable; stability implies $Y$27; and non-NPC targets fall outside the theory, with extensions to buildings requiring more work (Riestenberg et al., 14 Nov 2025). In the Helmholtz theory, excessively small $Y$28 yields insufficient coarse spaces, excessively large $Y$29 inflates coarse-solve cost, poor impedance choices or minimal overlap degrade performance, and inadequate points per wavelength can obscure the benefits of the preconditioner (Dolean et al., 2 Sep 2025).

The open problems also differ by setting. For harmonic maps, (Riestenberg et al., 14 Nov 2025) asks whether positive drift is necessary and sufficient for bounded-distance harmonic representatives, whether the theory extends to non-proper targets such as Euclidean buildings, whether $Y$30 can be made sharp and algorithmic, and whether the finite non-transversality mechanism extends from full positivity to Guichard–Wienhard $Y$31-positivity. For Helmholtz solvers, (Dolean et al., 2 Sep 2025) frames the practical problem as identifying the best algorithms and numerical strategies for benchmark problems modelled by the Helmholtz equation, with the effectiveness of harmonic and related spectral coarse spaces depending on the specific problem and numerical configuration.

Taken together, these developments establish harmonic coarse spaces as a technically precise notion rather than a single construction. In computational wave propagation, they are coarse basis spaces built from locally Helmholtz-harmonic modes and selected by SPD spectral criteria (Dolean et al., 2 Sep 2025). In nonlinear geometric analysis, they designate a viewpoint in which coarse maps with positive drift acquire canonical harmonic representatives, with applications to rigidity and higher Teichmüller theory (Riestenberg et al., 14 Nov 2025).