LCK metrics on complex spaces with quotient singularities
George-Ionut Ionita, Ovidiu Preda

TL;DR
This paper extends the concept of locally conformally Kaehler (LCK) metrics to complex spaces with quotient singularities, establishing conditions for their existence and behavior under blow-ups.
Contribution
It introduces LCK metrics on singular complex spaces, characterizes their existence on quotient singularities, and shows stability under blow-ups.
Findings
LCK metrics exist on quotient singularities if the universal cover admits a compatible Kähler metric.
Blow-ups at points of LCK complex spaces preserve the LCK property.
Abstract
In this article we introduce a generalization of locally conformally Kaehler metrics from complex manifolds to complex analytic spaces with singularities and study which properties of locally conformally Kaehler manifolds still hold in this new setting. We prove that if a complex analytic space has only quotient singularities, then it admits a locally conformally Kaehler metric if and only if its universal cover admits a Kaehler metric such that the deck automorphisms act by homotheties of the Kaehler metric. We also prove that the blow-up at a point of a LCK complex space is also LCK.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
LCK metrics on complex spaces with quotient singularities
George-Ionuţ Ioniţă
Department of Mathematics and Computer Science
”Politehnica” University Bucharest
313 Splaiul Independenţei, Bucharest 060042, Romania
and
Ovidiu Preda
Institute of Mathematics of the Romanian Academy
P.O. Box 1-764, Bucharest 014700, Romania
Abstract.
In this article we introduce a generalization of locally conformally Kähler metrics from complex manifolds to complex analytic spaces with singularities and study which properties of locally conformally Kähler manifolds still hold in this new setting. We prove that if a complex analytic space has only quotient singularities, then it admits a locally conformally Kähler metric if and only if its universal cover admits a Kähler metric such that the deck automorphisms act by homotheties of the Kähler metric. We also prove that the blow-up at a point of an LCK complex space is also LCK.
2010 Mathematics Subject Classification:
32C15; 53C55
Ovidiu Preda was supported by a grant of Ministry of Research and Innovation, CNCS - UEFISCDI, project number PN-III-P1-1.1-PD-2016-0182, within PNCDI III
1. Introduction
A Kähler manifold is a complex manifold admitting a (1,1)-form which is positive definite and -closed. This form is called Kähler form, or by an abuse of language which is unlikely to cause any confusion, Kähler metric, since it corresponds to a hermitian metric. By Dolbeault’s lemma, a Kähler form can be written locally , where is a strictly plurisubharmonic function, called Kähler potential. Hence, is completely determined by a family of Kähler potentials , which verify the compatibility condition on the open subset where both are defined. Grauert [3] and Moishezon [5] used this equivalent definition to extend the notion of Kähler metrics to complex spaces with singularities.
A locally conformally Kähler manifold is a complex manifold admitting a (1,1)-form such that every point has a neighborhood and there is a smooth function such that is Kähler. By using local Kähler potentials and compatibility conditions, the definition of locally conformally Kähler metrics can be extended to complex spaces with singularities in the same manner as Grauert and Moishezon did for the Kähler case.
A well known characterization of locally conformally Kähler manifolds (LCK for short) is the following result: a complex manifold admits an LCK metric if and only if its universal cover admits a Kähler metric such that the deck automorphisms act on by Kähler homotheties.
Therefore, it is natural to ask wether or not the equivalent definition for LCK remains valid for complex analytic spaces. The main result of this article solves this affirmatively in the case of complex analytic spaces with quotient singularities:
*Theorem 1.1: *** Let be a complex analytic space which has only quotient singularities.
*Then, admits an LCK metric if and only if its universal cover admits a Kähler metric such that deck automorphisms act on by homotheties of the Kähler metric. *
We also generalize to complex spaces a classical theorem regarding LCK manifolds. We prove the following result:
**Theorem 1.2: *** The blow-up at a point of an LCK complex space is also LCK. *
In section 4 of this paper, we give some examples of LCK singular complex spaces which do not admit Kähler metrics. They are obtained as a quotient of an LCK manifold by a finite group of automorphisms with fixed points. In the last section, we make some remarks and propose an open problem, for further study of LCK complex spaces.
2. Preliminaries
In this section, we collect the notions, definitions, and results that we need for the main theorem.
**Definition 2.1: ** Let be a complex space. A Kähler metric on is a collection , where is an open covering of , is a strongly plurisubharmonic function on , such that on each nonempty intersection we have the following compatibility condition:
[TABLE]
where is a holomorphic function on .
If is a complex manifold, such a collection defines indeed a Kähler form on , given locally, on each set , by .
If the collection verifies the open covering and the compatibility conditions from the Kähler metric definition above, but each function is only assumed to be plurisubharmonic, and strictly plurisubharmonic on the complement of an analytic subset of positive codimension in , then is called a weakly Kähler metric on .
For a short presentation of locally conformally Kähler (LCK) manifolds, one may read the survey [7] by Ornea and Verbitsky. There are more equivalent definitions of LCK manifolds:
**Definition 2.2: ** A complex manifold is called LCK if it verifies one of the following equivalent conditions:
- (1)
There exists a -form on such that for every , there exists an open neighborhood of and a smooth function such that is a Kähler form on ; 2. (2)
has a Hermitian metric such that , where is a closed 1-form on , called the Lee form; 3. (3)
The universal cover of has a Kähler metric such that the deck transform group acts on by Kähler homotheties; 4. (4)
admits an oriented, flat, real line bundle and an -valued (1,1)-form which is Kähler with respect to .
Using the first of these equivalent definitions, we can generalize LCK metrics to singular complex spaces, in the following way:
**Definition 2.3: ** Let be a complex space. An LCK metric on is a collection , where is an open covering of , is a strongly plurisubharmonic function on , and is a smooth function, such that we have the following compatibility condition:
[TABLE]
on each nonempty intersection .
As in the Kähler case, such a collection will be called weakly LCK if we require every function to be only plurisubharmonic, and strictly plurisubharmonic on the complement of an analytic subset of positive codimension in .
Two Kähler (or LCK) metrics are considered to be equal if they determine the same (1,1)-form on the regular locus of the complex space.
If is a Kähler space, , and is a smooth function, we denote by the metric . We say that an automorphism acts by homotheties of the Kähler metric if , where .
**Definition 2.4: ** If is a complex space, then a point is called quotient singularity if there exists a finite subgroup of automorphisms of such that the germs and are biholomorphic.
The folowing result about the properties of quotient singularities is a particular case of H. Cartan’s [2, Théorème 1]:
*Theorem 2.5: *** Let be a complex manifold and a finite group of automorphisms of .
*Then, the quotient space with the sheaf induced by the canonical projection is a normal space. *
The next theorem, by Bierstone and Milman [1, Theorem 13.4], is the fundamental result on global desingularization of complex spaces.
**Theorem 2.6: *** Any complex space admits a desingularization such that is the composition of a locally finite sequence of blow-ups with smooth centers and is a divisor with normal crossings in . *
In this theorem locally finite sequence of blow-ups means that on every compact subset, all but finitely many blow-ups are trivial.
The following theorem of Kollár, which combines [4, Lemma 7.2] and [4, Theorem 7.5], gives a sufficient condition under which the fundamental group of a normal space and the fundamental group of a desingularization of it, are isomorphic.
*Theorem 2.7: *** Let be a normal space which has only quotient singularities and a resolution of singularities.
*Then, the induced homomorphism is an isomorphism. *
Of course, taking into account 2, the assumption of normality in Kollár’s theorem is superfluous, but we kept the original statement. We mention that a different proof for 2 was given by Verbitsky [9, Theorem 4.1].
3. The main results
3.1. A characterization theorem for LCK complex spaces
Proof of 1.
Firstly, we prove the direct implication, so we know by hypothesis that admits an LCK metric. We denote by the universal cover of and we consider a resolution of singularities for . These two maps induce a resolution of singularities and a cover such that the following diagram commutes:
[TABLE]
Now, denote by the LCK metric on . Then, with the notations: , , and for every , we have that is a weakly LCK metric on . Since is a covering of , if we define , , and , we obtain that is a weakly LCK metric on . Denote by its induced Lee form. Since is the universal cover of a complex space with only quotient singularities, it also has only quotient singularities, and by 2 it is also normal. Hence, by 2, is simply connected, which further implies that is exact: there exists such that .
Next, we may assume that the sets of the LCK structure on are connected and sufficiently small such that for each , is a disjoint union of open sets in , each of them biholomorphic to . Then, for each , is a union of pairwise disjoint open connected sets, each of them biholomorphic to , and we denote ; for every and , we have that . Also, for every and every , is a biholomorphism and on , we have: . Hence, there exists a constant such that on . Now, it is not difficult to verify that
[TABLE]
is a weakly Kähler metric on .
Furthermore, since are by construction constant on the fibers of , they descend to , where we denote them by . Moreover, since and is a local biholomorphism, they are strictly plurisubharmonic, hence the family
[TABLE]
is a Kähler metric on .
From this point forward, the proof is similar to the one for manifolds, with the necessary adaptations. Knowing that is constant on the fibers of , we deduce that descends to a function on . Also, since on , it follows that descends to a function on . Hence, we have on .
Consider . By the commutativity of the diagram, we also have , hence it is invariant to the action of . That being so, taking into account that on , we get that the 1-form defined on is invariant to the action of . Since on , there exists such that on . By the continuity of and the connectedness and density of , we deduce that on . Next, we want to see how acts on the Kähler metric. For each , there is exactly one such that , thus:
[TABLE]
[TABLE]
which consequently gives , ending the proof for the direct implication.
Now, in order to prove the reversed implication, we suppose that the universal cover of the complex space , has a Kähler metric such that acts on by Kähler homotheties. Hence, we have the character morphism given by . On we consider the following equivalence relation: if there exists such that and . Then, is a line bundle which is trivial, since there exists an open cover of and a choice of transition maps which are all positive. Given a section on which is non-zero at every point, we obtain a section for the line bundle on , which is also trivial. Hence, we may consider that is a function with values in . For any , by the construction of the line bundle , we get
[TABLE]
hence is deck-invariant. There exists a real function such that . Also, we may consider that , where for every , the family is made of connected open sets which are projected by biholomorphically on . Then, for every , we choose an arbitrary and denote , and . With these notations, it is now easy to verify that is an LCK metric on .
3.2. The blow-up at a point of an LCK complex space
A classical result, by Tricerri [8] and Vuletescu [10], says that the blow-up at a point of an LCK manifold is also an LCK manifold. In the next lines we show that this result can be easily generalized to complex spaces.
Proof of 1.
Let be a complex space with the LCK metric , and . We may assume that has a neighborhood such that , and for all . We may also assume that is sufficiently small such that it can be embedded as a closed complex subspace in the unit ball , and such that extends to a strictly plurisubharmonic function on (we keep the same notation for the extended function).
Now, denote by the blow-up of in , and by the projection. Then, . Using the technique from [10], one can construct a (1,1)-form as the curvature of a line bundle on , such that has the following properties: it is negative definite along (i.e. for every and every non-zero vector , where is the complex structure on ), it is negative semi-definite at points of (i.e. for every and every ), and it is zero outside a compact subset of .
Then, for a sufficiently small constant , the (1,1)-form
[TABLE]
is positive definite. It is also -closed, since is the curvature form of a line bundle. Hence, it is a Kähler form on . By Dolbeault’s lemma, it can be represented in the form of the generalized definition of Kähler metrics, as . With the notations and , we have that is a Kähler metric on , the blow-up of at . Since outside , the strictly plurisubharmonic functions and determine the same Kähler metric on . For this reason, by glueing and in the natural way, with the notation , the compatibility condition holds on , for any and any , because . Hence,
[TABLE]
is an LCK metric on , the blow-up at of .
4. Examples
In this section we give examples of LCK complex spaces which do not admit Kähler metrics. They are obtained as the quotient of an LCK (non-Kähler) manifold by a finite group of automorphisms which have fixed points.
**Example 4.1: ** Quotients of Hopf manifolds of dimension at least 3.
Consider , , and the matrix . Denote . Then, the Hopf manifold is an LCK manifold which does not admit Kähler metrics. Also, take the matrix and denote . Define the function by . By the results in [2, Section 4], is a singular complex space biholomorphic to . Consider the function on . Then, is a Kähler form on which induces a Kähler metric on . Denote by the Kähler metric on obtained via the biholomorphism . Also, denote . It is not difficult to verify that acts by homotheties of the Kähler metric . Finally, by theorem 1, which for the converse implication is true (with the same proof) for any cover, not only the universal cover, is LCK. However, does not admit Kähler metrics, since it contains as a closed complex subspace which is a 2-dimensional Hopf mainfold.
**Example 4.2: ** Quotients of compact LCK surfaces.
Let be a compact LCK (non-Kähler) manifold which has a finite cyclic group for which the metric is invariant, and such that the fixed point locus of is a finite set. For every point , there exists a neighborhood in such that on , , where is smooth and is strictly plurisubharmonic. If the metric on can also be written , then leads to , hence , which further implies . Therefore, we may assume that is -invariant. Moreover, by taking the pull-back of the metric by all the elements of and then taking the average metric, we may assume that is also -invariant, hence both and descend to functions on the singular space .
Consequently, the LCK metric descends to , which is an LCK metric on . However, the functions and may not be smooth at the singular points of . We may assume that the projection on of every fixed point of has a small neighborhood in which intersects only one of the sets . We can modify both and on this small neighborhood, to make them smooth, with the modified still strictly plurisubharmonic, thus obtaining a modified metric on which is still LCK.
Now, if we assume that has a Kähler metric, then its pull-back to is a (1,1)-form on which is Kähler on the complement of the set of fixed points of , and it can be modified at those points to obtain a Kähler metric on the whole , yielding a contradiction. Hence, the singular complex space does not admit Kähler metrics.
5. Remarks
Mumford [6] proved that if is a normal point of a 2-dimensional algebraic variety, then is locally simply connected around if and only if is a regular point of . Thus, the regular locus of a simply connected normal complex space is not, in general, simply connected. For this reason, the proof of our theorem cannot be modified to use normalizations instead of desingularizations, which would have been better, since the method would have worked for any complex space. But it is worth remarking that even if for our proof the additional assumption on the type of singularities is essential, the theorem might be true for the general case of singular complex spaces. Thus, we propose the following problem:
**Problem 5.1: ** Prove that a complex space admits an LCK metric if and only if its universal cover admits a Kähler metric such that deck automorphisms act on by homotheties of the Kähler metric, or find a counterexample to this statement.
We also want to point out that the Hopf surface (2-dimensional) and the Inoue surface do not have “enough” automorphisms to be used for examples like 4, and the quotient of any of these surfaces by a finite group of automorphisms is again smooth and in the same class as the initial surface. Also, since all our concrete examples are quotients of LCK, non-Kähler manifolds, it would be interesting to find a different way to construct an example of LCK complex space which does not admit Kähler metrics.
Acknowledgment
Ovidiu Preda is grateful to Professor Liviu Ornea for guiding his learning of geometry which led to the problem studied in this article. Both authors are thankful to Alexandra Otiman for repeatedly taking the time to answer their questions.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] Bierstone E.; Milman P.D.: Canonical desingularization in characteristic zero by blowing up the maximum strata of a local invariant , Invent. Math. 128 no. 2, 207–302, (1997).
- 2[2] Cartan H.: Quotient d’une variété analytique par un groupe discret d’automor-phismes , Séminaire Henri Cartan, tome 6 (1953–1954), p. 1–13.
- 3[3] Grauert H.: Über Modifikationen und exzeptionelle analytische Mengen , Math. Ann. 146 , 331–368, (1962).
- 4[4] Kollár J.: Shafarevich maps and plurigenera of algebraic varieties , Invent. Math. 113 , 177–215, (1993).
- 5[5] Moishezon B.G.: Singular Kählerian spaces , Proceedings of the International Conference on Manifolds and Related Topics in Topology (Tokyo 1973), Tokyo, 343–351, (1975).
- 6[6] Mumford D.: The topology of normal singularities of an algebraic surface and a criterion for simplicity , Publications mathématiques de l’I.H.É.S., tome 9, 5–22, (1961).
- 7[7] Ornea L.; Verbitsky M.: A report on locally conformally Kähler manifolds . Harmonic Maps and Differential Geometry 135–149, Contemporary Mathematics, Volume: 542 , Amer. Math. Soc., Providence (2011).
- 8[8] Tricerri F.: Some examples of locally conformal Kähler manifolds , Rend. Sem. Mat. Torino 40 , no.1, 81–92, (1982).
