At most two smooth holomorphic fibrations
over hyperbolic curves
Abstract
We prove that a connected compact complex surface admits at most two holomorphic submersions with connected fibres onto compact curves of genus at least two, up to isomorphism of the bases. In particular, there is no surface with three distinct Kodaira fibrations, answering the existence question posed by Catanese (Jpn. J. Math., 2017). The result also includes isotrivial fibrations and requires no projectivity hypothesis. The proof relates the ramification of a pair of fibrations to an obstruction to a third. Equality of pulled-back hyperbolic metrics forces each reduced ramification component to be smooth and unramified over both bases. Its neighbourhood carries a cyclic action, and its normal bundle has negative degree. A third fibration would produce a holomorphic splitting whose cyclic average is tangent to the ramification curve. When all three fibrations are tangent there, the splittings are meromorphic; averaging and a constant linear combination cancel their principal parts. In either case the resulting regular foliation has the ramification curve as a compact leaf, contradicting the degree-zero condition imposed by its normal connection. The unramified case follows from factorization through a finite product cover.
Note. This paper was generated entirely by AI, including MiMo, using an automated research pipeline developed by Chenghua Liu and Hanyu Li.
1 Introduction
A Kodaira fibration is a smooth, non-isotrivial holomorphic family of curves over a compact curve: its fibres vary in complex structure, although the underlying map is a differentiable fibre bundle. Kodaira’s construction [6] and the double Kodaira fibrations studied by Catanese and Rollenske [4] show that a single compact complex surface can support two such families. Catanese [3, Question 10] asked whether three distinct Kodaira fibrations can occur on the same surface. Salter [8, Question 3.3] asked a related question about smooth surface bundle structures on complex surfaces, and more generally on four-manifolds with nonzero signature.
There are restrictions on a possible third fibration and exact counts for particular surfaces. Chen [5, Theorems 1.1–1.2] proved that the Atiyah–Kodaira manifold considered there, and regular finite covers of products of hyperbolic surfaces, have exactly two surface bundle structures up to -fibrewise diffeomorphism. Llosa Isenrich and Py [7, Theorem 2] proved that three distinct Kodaira fibrations on a compact complex surface would induce a finite-index image in the product of the three base fundamental groups. That theorem restricts such a configuration but does not exclude it. We exclude it by studying the ramification curves of the pairwise maps to products.
Throughout, a complex surface is a connected complex manifold of complex dimension two, and a curve is a connected compact Riemann surface. A curve is hyperbolic if its genus is at least two. A fibration is a surjective holomorphic map with connected fibres; it is smooth if it is a submersion. Two fibrations are equivalent if for an isomorphism . No automorphism of is included in this equivalence relation.
Theorem 1.1.
A compact complex surface admits at most two equivalence classes of smooth fibrations onto hyperbolic curves.
The bound is attained by the projections of a product of two hyperbolic curves, and also by double Kodaira fibrations [4].
Corollary 1.2.
A compact complex surface admits at most two inequivalent Kodaira fibrations.
Proof.
The proof has two geometric steps. First, the joint map of two inequivalent smooth fibrations is finite. On the normalization of any ramification component, the two restricted maps have the same ramification divisor. Uniqueness of their pulled-back hyperbolic metrics then identifies the local ramification image with the graph of a hyperbolic isometry. We deduce that the reduced ramification curves are smooth and pairwise disjoint. A component with ramification index has a cyclic neighbourhood model and satisfies
These statements concern ramification curves in the source. They do not assert that the entire branch divisor in the product is smooth: different points of the source may lie over the same branch value.
Second, a third fibration would force a regular holomorphic foliation tangent to . Schwarz–Pick rigidity shows that its tangency with the first two fibrations occurs either everywhere on or nowhere on . In the latter case, cyclic averaging of a holomorphic splitting produces the required foliation. In the former, two meromorphic splittings have principal coefficients whose ratio is constant on . Averaging removes the intermediate Laurent terms, and an affine combination cancels the remaining pole. A compact leaf of a regular holomorphic foliation has a degree-zero normal bundle, giving the contradiction. This degree obstruction is classical in the theory of holomorphic connections [2, 1]; the construction of the splitting from the three fibrations is the step needed here. Appendix A gives an alternative calculation of the common-tangency obstruction. If the original pair is unramified, an elementary factorization argument on a finite product cover completes the proof.
The theorem counts holomorphic submersions for the given complex structure on . It is not a bound on arbitrary smooth surface bundle structures on four-manifolds, whose number can be arbitrarily large [8, 9]. Hyperbolicity of the bases and smoothness of each fibration are essential to the argument. In particular, smoothness must be checked separately for the two projections of a branched cover; Remark 2.5 explains this distinction for a construction appearing in the literature.
2 Ramification and hyperbolic metrics
For a smooth compact curve in a complex surface , write for its normal bundle, so that . We first establish the finiteness of a pair map and recall the degree obstruction used later.
Lemma 2.1.
If and are inequivalent smooth fibrations from a compact complex surface, then is finite and surjective.
Proof.
The fibres of are homologous, so the topological degree of the restriction of to a fibre of is independent of the fibre. A holomorphic map of compact curves has degree zero exactly when it is constant. Thus, if is constant on one fibre of , it is constant on every fibre. Local holomorphic sections of then give a holomorphic factorization . The map is nonconstant. Over a regular value of , a fibre of is the disjoint union of fibres of . Its connectedness forces , contrary to inequivalence.
Consequently is nonconstant on every fibre of . Each such fibre is a smooth connected, hence irreducible, compact curve, so no curve can be contracted by . Its fibres are therefore zero-dimensional compact analytic sets, hence finite. The map is proper, so it is finite. Its image is a closed analytic subset of dimension two in the connected surface , and is therefore the whole product. ∎
A holomorphic connection on a line bundle over a curve is a -linear sheaf map satisfying for local holomorphic functions and sections . The following form of the connection obstruction will suffice; compare [2] and [1, Theorem 7.1 and Corollary 7.3].
Lemma 2.2.
A holomorphic line bundle on a compact curve that admits a holomorphic connection has degree zero. In particular, if a regular holomorphic foliation on a neighbourhood of a smooth compact curve has as a leaf, then .
Proof.
Choose a nonzero meromorphic section of the line bundle and write . If , where is a holomorphic frame and is a holomorphic unit, then
with holomorphic. Thus the poles of have residues equal to the orders of the zeros and poles of . The residue theorem gives
For the second assertion, choose foliation charts with and leaves given by . On a connected overlap the transverse coordinates satisfy , with and . The normal transition functions are constant. They define a flat holomorphic connection on , namely the normal Bott connection. The first assertion applies. ∎
For a hyperbolic curve , let denote its metric of curvature . Pullback by an unramified covering gives the curvature metric on the covering curve. We also need uniqueness when two pullbacks have the same degeneracy divisor.
Lemma 2.3.
Let and be nonconstant maps of curves, with . If their ramification divisors on coincide, then . Near every point of , the image of is contained in the graph of a holomorphic local hyperbolic isometry from to .
Proof.
In a local coordinate on , write the two metrics as and . At a point with common ramification order , both densities are times a smooth positive function. Hence
is a smooth real function on all of , independent of the coordinate. Away from the common zeros, the curvature equation gives
where is the Euclidean Laplacian in . Both sides extend smoothly across the zeros. If and are the pulled-back area forms, this equation is . Multiplication by and integration by parts yield
The left-hand side is nonpositive, so is constant. Since is positive away from finitely many points, the curvature equation then gives .
Where is locally invertible, is a holomorphic local hyperbolic isometry. Lift small target charts to the unit disc. Such an isometry agrees locally with the disc automorphism having the same value and differential: isometries with the same first-order data agree along geodesics from that point. For a neighbourhood of a ramification point of , choose a nearby unramified point and the corresponding disc automorphism . The lifted maps satisfy on a nonempty open set, hence on the whole neighbourhood by the identity theorem. Descending to smaller target charts proves the graph assertion at the ramification point as well. ∎
For two inequivalent smooth fibrations and , define
The wedge is a nonzero holomorphic section of , where denotes the canonical line bundle. Thus is an effective divisor, possibly empty. Its support is the tangency locus of the two fibre foliations.
Proposition 2.4.
Let and be inequivalent smooth fibrations from a compact complex surface to hyperbolic curves. The reduced support of is a disjoint union of smooth compact curves , and and are unramified coverings. If has coefficient in , then and
| (1) |
There is a neighbourhood of on which and lift to holomorphic retractions . Moreover, admits a cyclic action of order fixing pointwise and preserving and . Locally along , suitable coordinates give
| (2) |
Here the same coordinate on is used for both retractions.
Proof.
Let be the normalization map of an irreducible component of . The cotangent maps induced by and are nowhere-zero maps of line bundles into . Their images coincide along , so they differ by a nowhere-zero holomorphic bundle isomorphism. Composing with shows that and are related by the same isomorphism. If either restricted map were constant, both would be constant, contradicting Lemma 2.1. They are therefore nonconstant and have identical ramification divisors.
By Lemma 2.3, each local branch of the ramification image is a graph of a local hyperbolic isometry. Fix a point of the ramification divisor. All source branches through give the same graph germ: its value is , and its derivative is the unique scalar relating the nonzero covectors and in local target coordinates. To obtain the derivative even when the source branch is singular, take a limit from points of its normalization where the restricted differentials are nonzero. A local hyperbolic isometry is determined by its value and differential, so the graph germs agree.
We now apply this conclusion to the finite germ of at . Choose disjoint source neighbourhoods around the finitely many points over , and shrink a target bidisc so that its inverse image is contained in their union. The component representing the germ at is finite over the bidisc, has only over its centre, and may be chosen connected. After further shrinking, its branch locus is contained in the single graph just identified. This assertion concerns only this source representative; branch images from other points over the centre need not agree with it.
Straighten the graph to . Away from its inverse image, the representative is an unramified cover of , where is a disc and . This cover is connected: a proper analytic divisor in a connected complex manifold does not disconnect it, since paths can be perturbed off its real-codimension-two strata. The fundamental group of is , so each connected finite cover is isomorphic to
The isomorphism is holomorphic because both covering maps are local biholomorphisms. It extends across the divisor by uniqueness of the normal finite extension: both extensions are obtained by integral closure of the target structure sheaf in the same covering field. Since the source is smooth, it is normal. Thus this is also the local form of the original finite map. Its reduced ramification curve is smooth and maps isomorphically to the local branch graph. In particular, different ramification components cannot meet, and both restrictions to each component are unramified. The local index is the coefficient of that component in plus one.
Fix a component . Choose a tubular neighbourhood which deformation retracts onto . The induced image of under is the image of under . The covering-space lifting criterion for the unramified covering therefore gives a lift with . This lift is holomorphic because the covering is locally biholomorphic. The same argument gives a holomorphic lift of , also restricting to . The pair is locally a cyclic cover branched over the diagonal; in local coordinates it consequently has the form , .
If is a change of coordinate on , the corresponding source coordinates satisfy the exact identities
| (3) |
Writing gives . The normal coordinate changes therefore have th powers equal to the tangent coordinate changes on , proving . Taking degrees gives (1); here because covers a hyperbolic curve.
To glue the cyclic action, factor the second identity in (3) as
On each sufficiently small connected overlap, a holomorphic th root of gives for a constant th root of unity . Every transition thus commutes with . The coordinate actions glue as germs along , preserve , and have order . For completeness, these germs can be realized on an invariant open neighbourhood. Compactness gives representatives of their finitely many powers on a common neighbourhood . Choose a smaller neighbourhood of such that all their images lie in and all group composition identities hold on . The union of these images of is invariant, because composing any two representatives on gives another power. Restricting to this union yields the asserted action on an actual neighbourhood. ∎
Remark 2.5 (Checking both projections).
A branched cover which gives one smooth fibration need not give a second. For a cyclic cover branched along a smooth graph , the local map is
The first projection is a submersion. The differential of the second is , which vanishes at a ramification point with .
This distinction matters for the branch-cover construction discussed in [4, Remark 2.6], where the branch divisor is étale over one factor but ramified over the other. In the intermediate genus-six family of [10, pp. 186–187], a genus-nine curve has degree-two maps to a genus-three curve. After an unramified base change , the family is a double cover of branched along the two resulting graphs. Hurwitz’s formula gives
This ramification persists under the unramified base change, and the local calculation above shows that the projection of this intermediate family to is not a submersion. The construction therefore does not supply two smooth projections. This observation concerns the intermediate family, not Zaal’s subsequent construction of a family of genus-three curves.
3 Excluding a third fibration
Suppose that , , are pairwise inequivalent smooth fibrations onto hyperbolic curves. Put . We first show that a third fibration cannot be tangent to just part of a component of .
Lemma 3.1.
Let be a component of . Either , or is a component of all three divisors. In the latter case its ramification index is the same for all three pairs.
Proof.
Suppose meets a component of at . By Proposition 2.4, and are unramified coverings, and hence pull back the base metrics to . The covectors and are proportional at , so the two rank-one Hermitian forms and are proportional. Their common nonzero restriction to makes the proportionality factor one.
Since is an unramified covering, the restriction of this identity to says that attains equality in Schwarz–Pick at . In particular it is nonconstant. The equality case of Schwarz–Pick, applied to its lift between universal covering discs, implies that is an unramified covering.
Set , , and let be the index of for the pair . The bundle map on given by
vanishes on , and therefore induces a holomorphic map . The bundle of such maps has degree
A line bundle of negative degree has no nonzero holomorphic section, so . Hence , and proportionality with gives as well. The same reasoning, with indices and interchanged, applies if initially meets . Distinct components of each are disjoint by Proposition 2.4.
In the common-component case, (1) applied to each pair states that . As , all three indices are equal. ∎
We will average splittings of a retraction over a cyclic action preserving . A splitting is a bundle map satisfying ; it may also be meromorphic. If generates an action of order , define
| (4) |
All terms have domain , since . Each term is again a splitting, so is an invariant splitting. The average preserves holomorphicity and, for meromorphic splittings, the bound on pole order along .
Lemma 3.2.
The disjoint alternative in Lemma 3.1 is impossible.
Proof.
Use the retraction and cyclic action associated to by Proposition 2.4. Since , the line bundle is transverse to along . After shrinking to an invariant neighbourhood, the restriction of to is an isomorphism. Its inverse gives a holomorphic splitting .
At any , the differential of has eigenvalue on and eigenvalue on the vertical line . The invariant splitting therefore has image at . Its image on is a holomorphic line subbundle of . A line subbundle is integrable: if is a local nonvanishing generator, then . It thus defines a regular holomorphic foliation with compact leaf . Lemma 2.2 gives , contradicting (1). ∎
In the common-component case, the fibre directions instead give meromorphic splittings. The cyclic action leaves only one possible negative Laurent power, which can be cancelled by an affine combination.
Lemma 3.3.
The common-component alternative in Lemma 3.1 is impossible.
Proof.
Lift the three maps on a tubular neighbourhood of to retractions with , as in the proof of Proposition 2.4. Shrink so that no other component of any meets it. Let be the common ramification index. Use the cyclic action for , and local coordinates with
The derivative has order exactly along , and . Consequently
| (5) |
On changing coordinates as in (3), comparison of the coefficients of in gives , where denotes the coefficient in the new chart. Thus the coefficients define a holomorphic function on the compact curve , and this function is a constant . Moreover,
The common index for is , so .
Away from , the line bundles and define splittings of . They extend meromorphically across , with poles of order at most . In the chosen coordinates,
The cyclic action preserves and , so it fixes . Average by (4).
Under the pushforward by , a vertical monomial is multiplied by . An invariant vertical Laurent series can therefore contain this monomial only when . Among the integers from through , the only such integer is . The leading monomial in survives averaging, and hence
Since is a global constant different from , the bundle map
| (6) |
is globally defined. Its pole cancels in every chart, so it is holomorphic across . The coefficients in this affine combination sum to one, giving . It is invariant and has image along ; equivalently, its local expression is .
The image of defines a regular holomorphic foliation with compact leaf . Lemma 2.2 contradicts . ∎
It remains to treat an unramified pair. The following holomorphic factorization argument is sufficient here; Chen [5, Theorem 1.2] treats surface bundle structures on regular finite covers of products in the stronger topological setting.
Lemma 3.4.
Let be a finite unramified covering, with . Suppose the induced factor maps have connected fibres. Every fibration onto a hyperbolic curve is then equivalent to or .
Proof.
Write and identify with a finite-index subgroup . Set
Each has finite index in , and . Let be the corresponding unramified curve coverings. Covering-space theory gives a finite unramified cover over . It is holomorphic because it is locally a lift through a biholomorphic covering chart.
Consider . The restrictions have a common topological degree. If this degree is zero, every is constant and depends only on . If the degree is positive, differentiation in a vector of gives a holomorphic section of on . Its degree is , so that section vanishes. This holds for every and every such vector, hence is independent of the first factor. In either case, factors through one projection.
Since is a local biholomorphism, this factorization implies that vanishes on for one . The map is therefore constant on every connected fibre of . Local holomorphic sections of give for a holomorphic map . The map is nonconstant, and connectedness of the fibres of forces , as in Lemma 2.1. Thus and are equivalent. ∎
Proof of Theorem 1.1.
References
- [1] M. Abate, F. Bracci, T. Suwa, and F. Tovena, Localization of Atiyah classes, Rev. Mat. Iberoam. 29 (2013), no. 2, 547–578. doi:10.4171/RMI/730.
- [2] M. F. Atiyah, Complex analytic connections in fibre bundles, Trans. Amer. Math. Soc. 85 (1957), 181–207. doi:10.1090/S0002-9947-1957-0086359-5.
- [3] F. Catanese, Kodaira fibrations and beyond: methods for moduli theory, Jpn. J. Math. 12 (2017), no. 2, 91–174. doi:10.1007/s11537-017-1569-x.
- [4] F. Catanese and S. Rollenske, Double Kodaira fibrations, J. Reine Angew. Math. 628 (2009), 205–233. arXiv:math/0611428.
- [5] L. Chen, The number of fiberings of a surface bundle over a surface, Algebr. Geom. Topol. 18 (2018), no. 4, 2245–2263. doi:10.2140/agt.2018.18.2245.
- [6] K. Kodaira, A certain type of irregular algebraic surfaces, J. Analyse Math. 19 (1967), 207–215. doi:10.1007/BF02788717.
- [7] C. Llosa Isenrich and P. Py, Mapping class groups, multiple Kodaira fibrations, and CAT(0) spaces, Math. Ann. 380 (2021), no. 1–2, 449–485. doi:10.1007/s00208-020-02125-y.
- [8] N. Salter, Surface bundles over surfaces with arbitrarily many fiberings, Geom. Topol. 19 (2015), no. 5, 2901–2923. doi:10.2140/gt.2015.19.2901.
- [9] N. Salter, Correction to the article “Surface bundles over surfaces with arbitrarily many fiberings”, Geom. Topol. 25 (2021), no. 5, 2707–2711. doi:10.2140/gt.2021.25.2707.
- [10] C. Zaal, Explicit complete curves in the moduli space of curves of genus three, Geom. Dedicata 56 (1995), no. 2, 185–196. doi:10.1007/BF01267642.
Appendix A The common-tangency obstruction in coordinates
We give a second proof of Lemma 3.3, starting after the constant has been obtained. It produces a holomorphic connection directly on , rather than constructing a foliation and using its normal connection.
In a chart write
For an overlap with coordinate change , equation (3) gives
Let be the coefficients of in the new chart. Compare the coefficients of in
For , the first correction to the leading term of has order , so these terms contribute nothing at order . Terms with also contribute nothing. Thus
or equivalently
| (7) |
The root-of-unity factor in the change of normal coordinate has no effect on this identity, since it is raised to the power .
Define a holomorphic function in each chart by
Equation (7) becomes
This is the compatibility condition for the local connections
Indeed, , and the Leibniz rule gives precisely the displayed transformation law. The local connections therefore glue to a holomorphic connection on . Lemma 2.2 would imply , whereas . This proves the required contradiction.