Rank One Spectral Local Systems

• Rank one spectral local systems are exact constructions that collapse complex matrix data to scalar invariants across diverse mathematical settings.
• They appear in settings such as twistor deformations on Kähler manifolds, Aomoto complexes for arrangement complements, and ℓ-adic deformation spaces, each offering concrete computational frameworks.
• In Floer theory and spectral Fukaya categories, the rank one specialization produces clear scalar holonomy and bordism corrections that contrast sharply with higher-rank phenomena.

“Rank one spectral local systems” is not a single standardized term across the current literature. In one precise modern sense, it denotes maps $\xi:L\to BGL_1(R)$ used to twist objects of a spectral Fukaya category (Porcelli et al., 25 Sep 2025). Closely related but distinct rank-one theories appear in twistor deformation of character varieties on compact Kähler manifolds (Saito, 2013), in Aomoto-complex calculations for arrangement complements (Saito, 2018), in deformation spaces of rank one $\ell$-adic local systems in positive characteristic (Esnault et al., 2019), in Deligne pairings for families of curves with flat relative connections (Montplet et al., 2015), and in Lagrangian Floer theory, where rank one is the scalar specialization of a higher-rank formalism (Konstantinov, 2017). This suggests an umbrella usage: rank-one local systems become “spectral” when their variation is organized by twistor parameters, cohomology jump loci, Floer-theoretic curvature, or bordism-valued open–closed operations.

1. Exact meanings and principal settings

The literature isolates several exact models rather than a single universal definition.

Setting	Rank-one object	Structural datum
Spectral Fukaya theory	$\xi:L\to BGL_1(R)$	Thom spectrum $R$, open–closed map (Porcelli et al., 25 Sep 2025)
Compact Kähler geometry	$$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$	Twistor deformation by $z^{-1}\theta+z\bar\theta$ (Saito, 2013)
Arrangement complements	Rank $1$ local system on $U=Y\setminus X$	Aomoto complex, Milnor monodromy eigenspaces (Saito, 2018)
Positive-characteristic $\ell$-adic theory	$$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$	Quasilinear jump loci, Hard Lefschetz (Esnault et al., 2019)
Families of curves	Flat relative line bundles	Deligne pairing and intersection connection (Montplet et al., 2015)
Equivariant orbit geometry	$\ell$0-equivariant rank $\ell$1 local systems on $\ell$2-orbits	Bruhat $\ell$3-order (Carmassi, 2021)

The phrase is therefore best treated as a comparative concept. In some papers it is literal and formal, as in the spectral Fukaya category (Porcelli et al., 25 Sep 2025). In others it is implicit: the local systems are rank one, and the “spectral” component arises from monodromy eigenspaces, character varieties, or deformation-theoretic support loci (Saito, 2013, Saito, 2018, Esnault et al., 2019).

A recurrent theme is that rank one collapses matrix-valued structures to scalar data. This collapse is explicit in Floer theory, where endomorphism-valued curvature in higher rank becomes a scalar weighted Maslov–2 count in rank one (Konstantinov, 2017). It also appears in character-variety geometry, where the moduli space reduces to an abelian torus of characters (Saito, 2013).

2. Character varieties, families, and twistor deformation

For a compact Kähler manifold $\ell$4, rank one $\ell$5-local systems are identified with characters

$\ell$6

On the identity component, after choosing generators $\ell$7 of $\ell$8, this becomes $\ell$9, and there is a canonical morphism

$\xi:L\to BGL_1(R)$0

sending a local system to the underlying holomorphic line bundle (Saito, 2013). The kernel is described explicitly by

$\xi:L\to BGL_1(R)$1

with $\xi:L\to BGL_1(R)$2 (Saito, 2013). Each fiber of $\xi:L\to BGL_1(R)$3 contains a unique unitary local system (Saito, 2013).

The corresponding twistor deformation is equally explicit. If $\xi:L\to BGL_1(R)$4 is a Higgs line bundle and $\xi:L\to BGL_1(R)$5 is the associated unitary local system, then the smooth twistor module of rank one is represented by the family of connections

$\xi:L\to BGL_1(R)$6

and the induced family of local systems is

$\xi:L\to BGL_1(R)$7

(Saito, 2013). In rank one, the spectral datum is therefore just the holomorphic $\xi:L\to BGL_1(R)$8-form $\xi:L\to BGL_1(R)$9, and the deformation law is governed by exponentials of its periods together with those of $R$0.

For smooth families of projective curves $R$1, the same rank-one character data are organized by flat relative connections. A relative flat line bundle determines a classifying map

$R$2

and locally the corresponding monodromy characters satisfy

$R$3

(Montplet et al., 2015). The variation of the family is measured by the Gauss–Manin invariant $R$4, and this enters directly into the geometry of Deligne pairings.

The principal output is a canonical and functorial intersection connection on the Deligne pairing. For holomorphic line bundles $R$5 with flat relative connections, the connection on $R$6 has curvature

$R$7

(Montplet et al., 2015). In the case of trivial fibrations, the paper also shows that the Deligne isomorphism is flat with respect to the connections constructed there, and uses this to produce a meromorphic connection on the hyperholomorphic line bundle over the twistor space of rank one flat connections on a Riemann surface (Montplet et al., 2015).

3. Cohomological support loci and computation

For hyperplane arrangements, rank one local systems are encoded by logarithmic residues. If $R$8 and $R$9, a rank-one local system $$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$0 is determined by local monodromies $$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$1 satisfying

$$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$2

The associated Aomoto complex is

$$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$3

(Saito, 2018). Classical comparison theorems require nonresonance along every dense edge, but for $$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$4 the paper proves that the canonical map

$$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$5

is bijective for $$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$6 under a more delicate set of geometric conditions, even when the usual condition on $$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$7 fails (Saito, 2018). This is used to compute monodromy eigenspaces of Milnor fibers, for example proving

$$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$8

for the reflection arrangement of type $$H^1(X,\mathbf C^*)=\operatorname{Hom}(H_1(X,\mathbf Z),\mathbf C^*)$$9 (Saito, 2018).

In positive characteristic, the deformation space of rank one $z^{-1}\theta+z\bar\theta$0-adic local systems is the formal character space

$z^{-1}\theta+z\bar\theta$1

with $z^{-1}\theta+z\bar\theta$2 a finitely generated free $z^{-1}\theta+z\bar\theta$3-module (Esnault et al., 2019). For an arithmetic complex $z^{-1}\theta+z\bar\theta$4, the cohomology jump loci are

$z^{-1}\theta+z\bar\theta$5

The main theorem shows that Frobenius-stable closed subsets of this deformation space are quasilinear, meaning finite unions of torsion translates of formal Lie subgroups (Esnault et al., 2019). As a consequence, the jump loci $z^{-1}\theta+z\bar\theta$6 are quasilinear, Hard Lefschetz holds for rank one $z^{-1}\theta+z\bar\theta$7-local systems on smooth projective varieties, and one obtains generic vanishing statements such as

$z^{-1}\theta+z\bar\theta$8

for abelian varieties (Esnault et al., 2019).

These results make the rank-one situation unusually rigid. In both the arrangement and $z^{-1}\theta+z\bar\theta$9-adic settings, the cohomology of a moving rank one local system is controlled by explicit algebraic loci rather than by arbitrary variation. This suggests that “spectral” in rank one often means the geometry of cohomology support rather than higher-rank eigenspace decomposition.

4. Rank one as the scalar specialization in Floer theory

In monotone Lagrangian Floer theory over $1$0, a local system is a functor

$1$1

and for a transverse pair $1$2 with local systems $1$3, the Floer cochain group is

$1$4

(Konstantinov, 2017). The higher-rank obstruction is the endomorphism-valued section

$1$5

which is independent of auxiliary choices and is a parallel section of $1$6 (Konstantinov, 2017).

The rank-one specialization is immediate and decisive. If $1$7 has rank $1$8, then $1$9, each $U=Y\setminus X$0 is multiplication by the scalar holonomy around $U=Y\setminus X$1, and

$U=Y\setminus X$2

The obstruction equation

$U=Y\setminus X$3

reduces, for $U=Y\setminus X$4, to the scalar equality

$U=Y\setminus X$5

(Konstantinov, 2017). In self-Floer rank one there is no additional matrix constraint, because every endomorphism is already scalar.

The paper makes the contrast with higher rank concrete on the Chiang Lagrangian $U=Y\setminus X$6. Through each point there pass exactly three Maslov–2 discs, and for a rank-one representation one has $U=Y\setminus X$7, so $U=Y\setminus X$8. Consequently, rank-one local systems cannot kill the obstruction on $U=Y\setminus X$9, whereas the rank-2 irreducible representation $\ell$0 over $\ell$1 satisfies

$\ell$2

and yields nonzero Floer cohomology (Konstantinov, 2017). The rank-one theory is therefore the scalar limit of a genuinely richer endomorphism-valued framework.

This scalar limit is exactly what disappears when one passes to higher rank: noncentral curvature, noncommuting holonomy contributions, and fiberwise centralizer conditions. The paper does not diagonalize $\ell$3, but it explicitly identifies rank one as the case in which the entire latent spectral structure collapses to a single scalar value (Konstantinov, 2017).

5. Spectral Fukaya categories and the open–closed map

The most literal current use of the phrase occurs in the spectral Fukaya category attached to a graded Liouville domain $\ell$4 and a commutative tangential pair $\ell$5. The pair determines a Thom spectrum $\ell$6, and the paper incorporates rank one spectral local systems as maps

$\ell$7

on closed exact Lagrangians $\ell$8 whose tangent bundles lift compatibly to $\ell$9 (Porcelli et al., 25 Sep 2025). Here $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$0 is the space of units of the commutative ring spectrum $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$1, and $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$2 classifies invertible $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$3-module twists.

The twist is implemented geometrically. From $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$4 one constructs coherent twisting data on Floer moduli spaces, then forms twisted moduli spaces as zero-sets of Thom-space-valued maps. The resulting enlarged category has objects $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$5, and its morphism groups are defined by bordism groups of the twisted flow categories (Porcelli et al., 25 Sep 2025). This is the spectral analogue of adding a rank one local system to an exact brane, but the coefficient object is now an invertible $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$6-module rather than a $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$7-dimensional vector space.

The open–closed map detects the new phenomenon. If $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$8 denotes the untwisted bordism class and $$G(\overline{\mathbf Q}_\ell)=\operatorname{Hom}_{\mathrm{cont}}(T,\overline{\mathbf Q}_\ell^\times)$$9 the class defined using the twisted unit and twisted open–closed module, then

$\ell$00

in $\ell$01, where $\ell$02 is represented by the composition

$\ell$03

(Porcelli et al., 25 Sep 2025). The correction is governed by the stable Hopf map $\ell$04, and the resulting unit is multiplicatively two-torsion (Porcelli et al., 25 Sep 2025).

This is precisely the point at which the modern expression “rank one spectral local systems” becomes formal rather than metaphorical. In the classical exact Fukaya category over $\ell$05, the image of a Lagrangian under the open–closed map is independent of the local system; in the spectral setting it changes by the universal $\ell$06-correction (Porcelli et al., 25 Sep 2025). The paper therefore identifies a genuinely new bordism-theoretic effect that only appears after passing from ordinary rings to ring spectra.

6. Equivariant and combinatorial rank-one theories

Another exact but different use of rank one local systems appears on $\ell$07-orbits in a Hermitian symmetric variety $\ell$08. Here

$\ell$09

where $\ell$10 denotes isomorphism classes of $\ell$11-equivariant $\ell$12-local systems of rank $\ell$13 on $\ell$14 (Carmassi, 2021). Such a local system is equivalent to a character of the component group of a stabilizer: $\ell$15

The paper studies the Bruhat $\ell$16-order on $\ell$17, a refinement of orbit closure order. On the subset $\ell$18 of trivial local systems, this order coincides exactly with the ordinary Bruhat order (Carmassi, 2021). New behavior enters only through nontrivial rank-one local systems, and the structure depends strongly on type. In the simply connected simply laced case, either all local systems are trivial, or exactly the maximum-rank orbits admit one nontrivial rank-one local system, and then the Hasse diagram of $\ell$19 has exactly two connected components: all trivial local systems, and the nontrivial local systems on maximum-rank orbits (Carmassi, 2021).

In type $\ell$20, nontrivial local systems again occur exactly on maximum-rank orbits, but comparability between trivial and nontrivial sectors is more intricate (Carmassi, 2021). In type $\ell$21, the rank-one local systems are much richer: if $\ell$22 contains $\ell$23 long roots $\ell$24, then the orbit $\ell$25 carries exactly $\ell$26 nonisomorphic rank $\ell$27 local systems, parametrized by sign sequences (Carmassi, 2021).

This branch of the subject is not “spectral” in the twistor, Floer, or bordism sense. It is instead combinatorial and equivariant. Even so, it fits the broader umbrella because the local systems are rank one and are organized by a rigid global structure, here a closure-like partial order rather than a deformation or support theory. A common misconception is that rank-one local systems are always classified only by ordinary characters; this example shows that their geometry can be refined by orbit combinatorics and stabilizer component groups (Carmassi, 2021).

Taken together, these theories show that rank one spectral local systems are best understood as a family of exact constructions rather than a single doctrine. In Kähler geometry they are characters with explicit twistor deformation; in families of curves they acquire canonical Deligne-pairing connections; in arrangement theory and positive-characteristic arithmetic they are controlled by cohomology jump loci; in Floer theory they are the scalar specialization of higher-rank curvature; and in the spectral Fukaya category they become genuine $\ell$28-valued twists whose open–closed image is modified by $\ell$29 (Saito, 2013, Montplet et al., 2015, Saito, 2018, Esnault et al., 2019, Konstantinov, 2017, Porcelli et al., 25 Sep 2025). The unifying feature is rank one; the “spectral” content depends on which structure—twistor parameter, monodromy eigenspace, support locus, endomorphism-valued curvature, or bordism class—is being used to organize it.