This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

\diagramstyle

[labelstyle=]

para-Kähler-Einstein 4-manifolds and
non-integrable twistor distributions

Gil Bor, Omid Makhmali, and Paweł Nurowski
Gil Bor

Centro de Intevetigación en Matemáticas (CIMAT), Guanajuato, Mexico

Email address: [email protected]


Omid Makhmali

Center for Theoretical Physics, PAS, Al. Lotników 32/46, 02-668 Warszawa, Poland

Email address: [email protected]


Paweł Nurowski

Center for Theoretical Physics, PAS, Al. Lotników 32/46, 02-668 Warszawa, Poland

Email address: [email protected]
Abstract.

We study the local geometry of 4-manifolds equipped with a para-Kähler-Einstein (pKE) metric, a special type of split-signature pseudo-Riemannian metric, and their associated twistor distribution, a rank 2 distribution on the 5-dimensional total space of the circle bundle of self-dual null 2-planes. For pKE metrics with non-zero scalar curvature this twistor distribution has exactly two integral leaves and is ‘maximally non-integrable’ on their complement, a so-called (2,3,5)-distribution. Our main result establishes a simple correspondence between the anti-self-dual Weyl tensor of a pKE metric with non-zero scalar curvature and the Cartan quartic of the associated twistor distribution. This will be followed by a discussion of this correspondence for general split-signature metrics which is shown to be much more involved. We use Cartan’s method of equivalence to produce a large number of explicit examples of pKE metrics with non-zero scalar curvature whose anti-self-dual Weyl tensor have special real Petrov type. In the case of real Petrov type D,D, we obtain a complete local classification. Combined with the main result, this produces twistor distributions whose Cartan quartic has the same algebraic type as the Petrov type of the constructed pKE metrics. In a similar manner, one can obtain twistor distributions with Cartan quartic of arbitrary algebraic type. As a byproduct of our pKE examples we naturally obtain para-Sasaki-Einstein metrics in five dimensions. Furthermore, we study various Cartan geometries naturally associated to certain classes of pKE 4-dimensional metrics. We observe that in some geometrically distinguished cases the corresponding Cartan connections satisfy the Yang-Mills equations. We then provide explicit examples of such Yang-Mills Cartan connections.

Key words and phrases:
Para-complex structure, para-Kähler structure, Einstein metric, Cartan geometry, Cartan reduction, Petrov type, (2,3,5)-distribution, conformal structure
1991 Mathematics Subject Classification:
53C10, (53C15, 58A30, 58A15, 53C29)
00footnotetext: Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

1. Introduction and main results

The main purpose of this article is to give a detailed treatment of para-Kähler-Einstein structures in dimension 4 and examine their relation to maximally non-integrable distributions of rank 2 in dimension 5, referred to as (2,3,5)-distributions.

Our initial motivation for this article is twofold. Firstly, it is an extension of the observation made in [BHLN18] where, inspired by the rolling problem of Riemannian surfaces [BH93], a notion of projective rolling was defined which gives rise to (2,3,5)-distributions. Consequently, it was observed that the (2,3,5)-distributions whose algebra of infinitesimal symmetries is maximal i.e. the split real form of 𝔤2,\mathfrak{g}_{2}, can be obtained from such construction with a direct link to the homogeneous para-Kähler-Einstein metric on 𝐒𝐋3()/𝐆𝐋2()\mathbf{SL}_{3}(\mathbb{R})/\mathbf{GL}_{2}(\mathbb{R}) referred to as the dancing metric. We point out that para-Hermitian structures and their variations naturally appear in various geometric settings since they were first defined in [Lib52]. The reader may consult [CFG96] for a survey.

Our second motivation comes from the twistorial construction of rank 2 distributions from conformal structures of split signature in dimension four, referred to as twistor distributions, described in [AN14]. In open subsets where the self-dual Weyl curvature is non-zero such distributions, which are naturally induced on the 𝕊1\mathbb{S}^{1}-bundle of self-dual null planes, are (2,3,5). We will discuss that, in general, the fundamental invariant of twistor distributions, referred to as the Cartan quartic, depend on the fourth jet of the components of the Weyl curvature of the conformal structure. This poses a basic question: whether the Cartan quartic of twistor distributions can have any root type? In this article we answer this question affirmatively. As a by-product of our construction one naturally obtains explicit examples of para-Sasaki-Einstein metrics.

Before proceeding further, a brief definition of the geometric structures appearing in this article is in order. As will be defined in § 2.1, an almost para-complex structure on a manifold, M,M, is defined as an endomorphism K:TMTMK:\mathrm{T}M\to\mathrm{T}M satisfying K2=IdTMK^{2}=\mathrm{Id}_{\mathrm{T}M} whose ±1\pm 1-eigenspaces have rank 2. As a result, unlike almost-complex structures, the eigenspaces of KK split each tangent space of MM into two transversal distributions. The integrability of these distributions induces a para-complex structure on MM which results in two transversal foliations of M.M. Similarly, an almost-para-Hermitian, para-Hermitian and para-Kähler structure can be defined in terms of KK and a pseudo-Riemannian metric of split signature satisfying certain compatibility condition, as explained in § 2.2.1 and § 2.3.1.

The first objective of this article, presented in § 2, is to give a unifying treatment of almost para-Hermitian and para-Kähler structures in dimension four via Cartan’s method of equivalence and analyze the curvature decompositions in each case. To our knowledge such presentation has been absent in the literature.

Our main topic of interest, treated in § 3, is para-Kähler-Einstein (pKE) metrics, defined as para-Kähler structures for which the metric is Einstein i.e. its trace-free Ricci tensor is zero. Because of our interest in non-integrable twistor distributions, we restrict our considerations to only those pKE metrics with non-zero scalar curvature. In § 3.1 it is shown that such metrics define a Cartan geometry of type (𝐒𝐋3(),𝐆𝐋2()).(\mathbf{SL}_{3}(\mathbb{R}),\mathbf{GL}_{2}(\mathbb{R})). In § 3.2 we investigate five classes of pKE metrics for which the root type, or the Petrov type, of the anti-self-dual Weyl tensor Weyl{Weyl}^{-} (see § 2.2.4), is non-generic and real. Let us mention briefly that by root type, or the Petrov type, of Weyl{Weyl}^{-} we refer to the multiplicity pattern of the roots of the 4th order polynomial obtained from the representation of Weyl{Weyl}^{-} as a binary quartic at each point. For these five classes of pKE metrics we carry out the Cartan reduction procedure case by case. This enables us to find all homogeneous models of pKE metrics in dimension 4. Moreover, we find explicit examples of pKE metrics with non-zero scalar curvature in every special real root type of Weyl{Weyl}^{-}. In particular, Theorem 3.12 gives explicit examples of pKE structures of Petrov type IIII, Theorem 3.14 gives examples for Petrov type IIIIII, and Theorem 3.16 gives examples of Petrov type NN. The pKE structures of real Petrov type DD are described in Theorem 3.10. This Petrov type is particularly interesting, since we found all possible such pKE metrics with non-zero scalar curvature. The method we have used to derive these examples is known as Cartan’s reduction method which is followed by integrating the reduced structure equations. Moreover, Cartan’s reduction method combined with Cartan-Kähler analysis is used to obtain the local generality of pKE metrics for each Petrov type in the real analytic case, as presented in Table 1. We point out that § 3.2 serves as an example of how effective Cartan’s method of equivalence is in studying geometric structures.

In § 4 we discuss the relationship between pKE metrics and (2,3,5)-distributions. More precisely, in § 4.1 we briefly review the basic facts about (2,3,5)-distributions, including the Cartan connection, the Cartan quartic and associated conformal structure of split signature. In § 4.2.1, we recall the fact that for any pKE structure, and more generally, any indefinite conformal structure in dimension four, a rank 2 distribution, referred to as a twistor distribution, is naturally induced on its 5-dimensional space of null self-dual planes, 𝒩+,\mathcal{N}_{+}, and that another rank 2 twistor distribution is induced on the space of null anti-self-dual planes, 𝒩;\mathcal{N}_{-}; this was originally observed in [AN14]. Moreover, the distribution induced on 𝒩+\mathcal{N}_{+} or 𝒩\mathcal{N}_{-} is a (2,3,5) distribution at a point if and only if the self-dual Weyl tensor, Weyl+,Weyl^{+}, or the anti-self-dual Weyl tensor, Weyl,Weyl^{-}, is non-zero at that point, respectively. After necessary coframe adaptations, which are performed at the beginning of § 4.2.1, we state our prime result on the surprising proportionality of the two quartics, the Cartan quartic of the (2,3,5) twistor distribution on 𝒩+\mathcal{N}_{+}, and the anti-self-dual Weyl quartic of the pKE metric, as in the following theorem.

Theorem 4.9.

Given a pKE metric for which the scalar curvature is non-zero, the Cartan quartic for the non-integrable twistor distribution on 𝒩+\mathcal{N}_{+} is a non-zero multiple of the quartic representation of the anti-self-dual Weyl curvature WeylWeyl^{-}. In particular, the Cartan quartic of the twistor distribution and the anti-self-dual Weyl curvature of the metric have the same root type.

In Remark 4.10 we explain the natural identification that underlies this theorem. As a result of Theorem 4.9 and our examples of pKE metrics of non-generic Petrov types in § 3, we obtain a large class of explicit examples of twistor distributions for each special real algebraic type of the Cartan quartic. Additionally, one obtains that the associated (3,2)(3,2) signature conformal structure on an open subset of 𝒩+\mathcal{N}_{+} has an Einstein representative whose conformal holonomy is a subgroup of 𝐒𝐋3()𝐆2𝐒𝐎4,3;\mathbf{SL}_{3}(\mathbb{R})\subset\mathbf{G}_{2}^{*}\subset\mathbf{SO}_{4,3}; this was also observed in [SW17]. The main purpose of § 4.3 is to show why the coincidence of the root types of the quartics, explained in Theorem 4.9, is remarkable. This is done by obtaining the Cartan quartic for the twistor distribution on 𝒩\mathcal{N}_{-}. It is shown that the coefficients of the Cartan quartic on 𝒩\mathcal{N}_{-} depend on the 4th jet of WeylWeyl^{-} and there is no obvious relation between the algebraic types of these quartics.

Finally, we also mention that, starting in § 3.1, a number of Cartan geometries are introduced which are naturally associated with pKE metrics in dimension 4. Since these Cartan geometries live on principal bundles over 4-dimensional manifolds with split-signature conformal metrics, one can study the vacuum Yang-Mills equations for the corresponding Cartan connections. As far as we know few papers are concerned with such studies. Here an honorable exception is a paper by S. Merkulov [Mer84], who established in 1984 that the vacuum Yang-Mills equations for the Cartan normal conformal connection of a 4-dimensional conformal structure (M,[g])(M,[g]), are equivalent to the vanishing of its Bach tensor. Also in this vein is the work [NS03], where in particular, the vacuum Yang-Mills equations for Cartan connections associated with 3-dimensional parabolic geometries of type (𝐒𝐋3(),𝐏12),(\mathbf{SL}_{3}(\mathbb{R}),\mathbf{P}_{12}), and (𝐒𝐔2,1,𝐏12)(\mathbf{SU}_{2,1},\mathbf{P}_{12}) were considered, in which 𝐏12\mathbf{P}_{12} denotes the Borel subgroup. It turns out that Cartan geometries of various type appear in the process of Cartan reduction performed on a given Cartan geometry. It is also clear that reduction of a Cartan geometry results in a principal bundle over the same base. Therefore, if the vacuum Yang-Mills equations can be defined for a certain Cartan geometry, they can also be defined for all the Cartan geometries obtained from the reduction procedure. We take this approach in § 3.1 and in the subsequent sections, where we reduce the initial (𝐒𝐋3(),𝐆𝐋2())(\mathbf{SL}_{3}(\mathbb{R}),\mathbf{GL}_{2}(\mathbb{R}))-type Cartan geometry to Cartan geometries of various types depending on the Petrov types. In particular, we find in Proposition 3.3, as a consequence of Theorem 3.2, that all pKE 4-dimensional structures for which the Einstein constant is equal to -3 satisfy vacuum Yang-Mills equations for the 𝔰𝔩3()\mathfrak{sl}_{3}(\mathbb{R})-valued Cartan connection of the associated (𝐒𝐋3(),𝐆𝐋2())(\mathbf{SL}_{3}(\mathbb{R}),\mathbf{GL}_{2}(\mathbb{R}))-type Cartan geometry. Similarly, in Theorem 3.12 we give examples of pKE metrics satisfying the vacuum Yang-Mills equations for the Cartan connection \mathcal{B} of a Cartan geometry of type (𝐒𝐎2,2,𝐓2)(\mathbf{SO}_{2,2},\mathbf{T}^{2}), which can be obtained for pKE metrics of real special Petrov type.

The EDS calculations mentioned in the text are carried out using the Cartan package in Maple written by Jeanne Clelland and an exterior differential package for Mathematica written by Sotirios Bonanos.

Conventions

In this article we will be working in the real smooth category. Since our results are of local nature, the manifolds can be taken to be the maximal open sets over which the assumptions made in each statement is valid.

The manifold MM is always 4 dimensional equipped with a metric gg of split signature. The 1-forms θ1,θ2,θ3,θ4\theta^{1},\theta^{2},\theta^{3},\theta^{4} represent a coframe on MM with respect to which g=2θ1θ3+2θ2θ4g=2\theta^{1}\theta^{3}+2\theta^{2}\theta^{4} where it is understood that the terms such as θ1θ3\theta^{1}\theta^{3} represent symmetric tensor product of the 1-forms θ1\theta^{1} and θ3\theta^{3} i.e.

θaθb=12(θaθb+θbθa)\theta^{a}\theta^{b}=\textstyle{\frac{1}{2}(\theta^{a}\otimes\theta^{b}+\theta^{b}\otimes\theta^{a})}

Given an nn-dimensional manifold, N,N, equipped with a coframe {β1,,βn},\{\beta^{1},\dots,\beta^{n}\}, the corresponding set of frame will be expressed as {β1,,βn}\{\frac{\partial}{\partial\beta^{1}},\dots,\frac{\partial}{\partial\beta^{n}}\} i.e. βbβa=δba.\textstyle{\frac{\partial}{\partial\beta^{b}}}\im\beta^{a}=\delta^{a}_{~{}b}. Given a function F:N,F:N\to\mathbb{R}, the so-called coframe derivatives of F,F, denoted by Fi:N,F_{i}:N\to\mathbb{R}, are defined as

(1.1) Fi=βidF.F_{i}=\textstyle{\frac{\partial}{\partial\beta^{i}}}\im\mathrm{d}F.

When a set of 1-forms on NN is introduced as I={γ1,,γk},I=\{\gamma^{1},\dots,\gamma^{k}\}, it represents the ideal that is algebraically generated by the 1-forms γ1,,γkTN,\gamma^{1},\dots,\gamma^{k}\in\mathrm{T}^{*}N, and is called a Pfaffian system. A Pfaffian system is called integrable if it satisfies the Frobenius condition, dII,\mathrm{d}I\subset I, where d\mathrm{d} is the exterior derivative. The integral manifolds of an integrable Pfaffian system are called its leaves. Locally, around a generic point xN,x\in N, the leaves of an integrable Pfaffian system induce a smooth foliation which enables one to consider the quotient space of its leaves, referred to as the leaf space of I.I. Since our treatment is local, we can always work in sufficiently small neighborhoods which allows one to define the leaf space of an integrable Pfaffian system.

Acknowledgments

This project benefited from the Simons Semester Symmetry and Geometric Structures at IMPAN in 2018. GB was supported by Conacyt grant A1-S-45886. OM would like to thank Kael Dixon and Niky Kamran for their interest and helpful conversations. OM acknowledges the partial support of the grant GA19-06357S from the Czech Science Foundation, GAČR. The research leading to these results has also received funding from the Norwegian Financial Mechanism 2014-2021 with project registration number 2019/34/H/ST1/00636.

2. Almost para-Hermitian and para-Kähler structures

The goal of this section is to fix notation, give necessary definitions and recall some facts that will be needed in subsequent sections. More precisely, in § 2.1 we recall some basic facts about pseudo-Riemannian metrics in dimension four. The notion of almost para-Hermitian structure and the decomposition of its curvature into irreducible components with respect to the action of its structure group is defined in § 2.2. Furthermore, in § 2.2 we define the so-called Petrov type of the Weyl curvature. In § 2.3 we define para-Kähler structures in terms of additional integrability and compatibility conditions imposed on an almost para-Hermitian structure. We derive their structure equations, curvature decomposition and give a local coordinate expression in terms of a potential function. We end the section by giving examples of pKE structures in terms of potential functions which, as will be shown in § 3.2, correspond to the only homogeneous models with non-zero scalar curvature.

2.1. Rudiments of indefinite pseudo-Riemannian metrics in dimension 4

In this section we briefly recall the decomposition of the space of 2-forms into self-dual and anti-self-dual 2-forms using the Hodge star operator. As a result, two 5-dimensional circle bundles of self-dual null planes and anti-self-dual null planes is obtained for any indefinite pseudo-Riemannian metric in dimension four. Subsequently, we recall the structure equations of split signature metrics and their curvature decomposition.

2.1.1. The Hodge star operator.

From now on let (M,g)(M,g) be a 4-dimensional real oriented manifold equipped with a split signature metric gg. Locally, we can always find a real coframe (θ1,θ2,θ3,θ4)=(α1,α2,α¯,1α¯)2(\theta^{1},\theta^{2},\theta^{3},\theta^{4})=(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}) in which the metric takes the form:

(2.1) g=gabθaθb=2θ1θ3+2θ2θ4=2α1α¯+12α2α¯.2g=g_{ab}\theta^{a}\theta^{b}=2\theta^{1}\theta^{3}+2\theta^{2}\theta^{4}=2\alpha^{1}\bar{\alpha}{}^{1}+2\alpha^{2}\bar{\alpha}{}^{2}.

We denote by (θ1,θ2,θ3,θ4)=(α1,α2,α¯1,α¯2)(\textstyle{\frac{\partial}{\partial\theta^{1}},\frac{\partial}{\partial\theta^{2}},\frac{\partial}{\partial\theta^{3}},\frac{\partial}{\partial\theta^{4}}})=(\textstyle{\frac{\partial}{\partial\alpha^{1}},\frac{\partial}{\partial\alpha^{2}},\frac{\partial}{\partial\bar{\alpha}^{1}},\frac{\partial}{\partial\bar{\alpha}^{2}}}) the dual frame. The coframe (θ1,θ2,θ3,θ4)(\theta^{1},\theta^{2},\theta^{3},\theta^{4}) is null i.e. the metric gg has constant coefficients gabg_{ab}, with g13=g31=g24=g42=1g_{13}=g_{31}=g_{24}=g_{42}=1 as the only non-vanishing ones, which implies nullity, g(θa,θb)=0g(\frac{\partial}{\partial\theta^{a}},\frac{\partial}{\partial\theta^{b}})=0 for a=1,,4a=1,\dots,4. Note that in this paper the coframe/frame on MM will be denoted by two notations (θ)(\theta) or (α,α¯)(\alpha,\bar{\alpha}{}), and by (θ)(\frac{\partial}{\partial\theta}) or (α,α¯)(\frac{\partial}{\partial\alpha},\frac{\partial}{\partial\bar{\alpha}}), respectively. The reason for this redundancy will be made clear in the next sections, when sometimes one, and sometimes the other notation will be more convenient.

We assume that the coframe (α1,α2,α¯,1α¯)2(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}) is positively oriented, therefore, the volume form 𝐯𝐨𝐥g\mathbf{vol}_{g} on MM can be expressed as

𝐯𝐨𝐥g=α1α2α¯1α¯.2\mathbf{vol}_{g}=\alpha^{1}\wedge\alpha^{2}\wedge\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{2}.

This enables one to define the Hodge star operator i.e. a linear map :Λ2TMΛ2TM*:\Lambda^{2}\mathrm{T}^{*}M\to\Lambda^{2}\mathrm{T}^{*}M such that

ω(X,Y)𝐯𝐨𝐥g=ωXY*\omega(X,Y)\mathbf{vol}_{g}=\omega\wedge X^{\flat}\wedge Y^{\flat}

Here XX^{\flat} is a 1-form associated to a vector field XX such that YX=g(X,Y)Y\im X^{\flat}=g(X,Y) for all vector fields YY on M.M.

It is easy to see that

2=idΛ2TM,*^{2}=\operatorname{id}_{\Lambda^{2}\mathrm{T}^{*}M},

and therefore the operator * splits Λ2TM\Lambda^{2}\mathrm{T}^{*}M into the direct sum

Λ2TM=Λ+2Λ2\Lambda^{2}\mathrm{T}^{*}M=\Lambda^{2}_{+}\oplus\Lambda^{2}_{-}

of its eigenspaces Λ±2\Lambda^{2}_{\pm}, corresponding to its respective ±1\pm 1 eigenvalues. In what follows we will frequently use the basis (σ±1,σ±2,σ±3)(\sigma^{1}_{\pm},\sigma^{2}_{\pm},\sigma^{3}_{\pm}) of the ±1\pm 1-eignespaces of *, expressed in terms of the null coframe (α1,α2,α¯,1α¯)2(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}) as

(2.2a) σ+1\displaystyle\sigma^{1}_{+} =α1α2,σ+2=α¯1α¯,2σ+3=α1α¯+1α2α¯2Λ+2\displaystyle=\alpha^{1}\wedge\alpha^{2},\quad\sigma^{2}_{+}=\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{2},\quad\sigma^{3}_{+}=\alpha^{1}\wedge\bar{\alpha}{}^{1}+\alpha^{2}\wedge\bar{\alpha}{}^{2}\quad\in\Lambda^{2}_{+}
(2.2b) σ1\displaystyle\sigma^{1}_{-} =α1α¯,2σ2=α¯1α2,σ3=α1α¯1α2α¯2Λ2\displaystyle=\alpha^{1}\wedge\bar{\alpha}{}^{2},\quad\sigma^{2}_{-}=\bar{\alpha}{}^{1}\wedge\alpha^{2},\quad\sigma^{3}_{-}=\alpha^{1}\wedge\bar{\alpha}{}^{1}-\alpha^{2}\wedge\bar{\alpha}{}^{2}\quad\in\Lambda^{2}_{-}

As a result, in terms of (α1,α2,α¯1,α¯2),(\textstyle{\frac{\partial}{\partial\alpha^{1}},\frac{\partial}{\partial\alpha^{2}},\frac{\partial}{\partial\bar{\alpha}^{1}},\frac{\partial}{\partial\bar{\alpha}^{2}}}), a basis for the self-dual and anti-self-dual bivectors in Λ2TM=Λ2,+Λ2,\Lambda^{2}\mathrm{TM}=\Lambda^{2,+}\oplus\Lambda^{2,-} can be expressed as

σ1+=α1α2,σ2+=α¯1α¯2,σ3+=α1α¯1+α2α¯2Λ2,+,\displaystyle\textstyle{\sigma^{+}_{1}=\frac{\partial}{\partial\alpha^{1}}\wedge\frac{\partial}{\partial\alpha^{2}},\quad\sigma^{+}_{2}=\frac{\partial}{\partial\bar{\alpha}^{1}}\wedge\frac{\partial}{\partial\bar{\alpha}^{2}},\quad\sigma^{+}_{3}=\frac{\partial}{\partial\alpha^{1}}\wedge\frac{\partial}{\partial\bar{\alpha}^{1}}+\frac{\partial}{\partial\alpha^{2}}\wedge\frac{\partial}{\partial\bar{\alpha}^{2}}\quad\in\Lambda^{2,+},}
σ1=α1α¯2,σ2=α2α¯1,σ3=α1α¯1α2α¯2Λ2,.\displaystyle\textstyle{\sigma^{-}_{1}=\frac{\partial}{\partial\alpha^{1}}\wedge\frac{\partial}{\partial\bar{\alpha}^{2}},\quad\sigma^{-}_{2}=\frac{\partial}{\partial\alpha^{2}}\wedge\frac{\partial}{\partial\bar{\alpha}^{1}},\quad\sigma^{-}_{3}=\frac{\partial}{\partial\alpha^{1}}\wedge\frac{\partial}{\partial\bar{\alpha}^{1}}-\frac{\partial}{\partial\alpha^{2}}\wedge\frac{\partial}{\partial\bar{\alpha}^{2}}\quad\in\Lambda^{2,-}.}

At each point pp of MM the bivector α1α2\frac{\partial}{\partial\alpha^{1}}\wedge\frac{\partial}{\partial\alpha^{2}} defines a null plane N+=Span{α1,α2}N_{+}=\mathrm{Span}\{\frac{\partial}{\partial\alpha^{1}},\frac{\partial}{\partial\alpha^{2}}\}. Recall that a plane N+N_{+} is null if g(X,X)=0g(X,X)=0 for all XN+X\in N_{+}. At every point pMp\in M we then have the space 𝒩p\mathcal{N}_{p} of all null planes at pp.

Let us consider a pair (a,Np)(a,N_{p}) where a𝐒𝐎2,2a\in\mathbf{SO}_{2,2} and Np𝒩p.N_{p}\in\mathcal{N}_{p}. The group 𝐒𝐎2,2\mathbf{SO}_{2,2} acts naturally on the space 𝒩p\mathcal{N}_{p} via:

(a,Np)aNp={aX|XNp}.(a,N_{p})\,\to\,a\cdot N_{p}=\{aX\,|\,X\in N_{p}\}.

This action decomposes 𝒩p\mathcal{N}_{p} into two orbits

𝒩p=𝒩p+𝒩p.\mathcal{N}_{p}=\mathcal{N}_{p+}\sqcup\mathcal{N}_{p-}.

Each of these orbits is diffeomorphic to a circle 𝒩p±𝕊1\mathcal{N}_{p\pm}\cong\mathbb{S}^{1}. Take Np+𝒩p+N_{p+}\in\mathcal{N}_{p+} and assume Np+=Span{α1|p,α2|p}.N_{p+}=\mathrm{Span}\{\frac{\partial}{\partial\alpha^{1}}|_{p},\frac{\partial}{\partial\alpha^{2}}|_{p}\}. Note that its defining bivector α1α2|p\frac{\partial}{\partial\alpha^{1}}{\,{\wedge}\;}\frac{\partial}{\partial\alpha^{2}}|_{p} is self-dual. Since the action of 𝐒𝐎2,2\mathbf{SO}_{2,2} does not change self-duality of null planes, the orbit 𝒩p+\mathcal{N}_{p+} is called the space of self-dual null planes at p.p. Consequently the orbit 𝒩p\mathcal{N}_{p-} is comprised of null planes defined by anti-self-dual bivectors and is therefore called the space of anti-self-dual null planes at p.p.

More explicitly, at pMp\in M the set of self-dual and anti-self-dual null planes can be parametrized, respectively, by λ,μ{}\lambda,\mu\in\mathbb{R}\cup\{\infty\} in the following way

(2.3) 𝒩p+=Ker{θ1+μθ4,θ2μθ3},𝒩p=Ker{θ2λθ1,θ3+λθ4}.\mathcal{N}_{p+}=\mathrm{Ker}\{\theta^{1}+\mu\theta^{4},\theta^{2}-\mu\theta^{3}\},\quad\mathcal{N}_{p-}=\mathrm{Ker}\{\theta^{2}-\lambda\theta^{1},\theta^{3}+\lambda\theta^{4}\}.

This parametrization will be used in § 4. The bundles 𝒩±:=𝑝𝒩p±\mathcal{N}_{\pm}:=\underset{p}{\bigcup}\mathcal{N}_{p\pm} equipped with the projections

(2.4) ν+:𝒩+M,ν:𝒩M,\nu_{+}\colon\mathcal{N}_{+}\to M,\qquad\nu_{-}\colon\mathcal{N}_{-}\to M,

where (ν±)1(p)=𝒩p±,(\nu_{\pm})^{-1}(p)=\mathcal{N}_{p_{\pm}}, at pM,p\in M, are referred to by various names, including circle twistor bundles ([AN14]) or bundles of real α\alpha-planes and β\beta-planes [AG96]. In what follows we will frequently refer to the circle bundles 𝒩+\mathcal{N}_{+} and 𝒩\mathcal{N}_{-} as the bundle of self-dual and anti-self-dual null planes, respectively.

2.1.2. Structure equations.

The null coframe (θ1,θ2,θ3,θ4)(\theta^{1},\theta^{2},\theta^{3},\theta^{4}) uniquely defines the Levi-Civita connection 1-forms Γba\Gamma^{a}_{~{}b} via the first structure equations:

dθa+Γbaθb=0\displaystyle{\rm d}\theta^{a}+\Gamma^{a}_{~{}b}\wedge\theta^{b}=0 (torsionfreeness),\displaystyle\mathrm{(torsionfreeness)},
gacΓbc+gbcΓac=0\displaystyle g_{ac}\Gamma^{c}_{~{}b}+g_{bc}\Gamma^{c}_{~{}a}=0 (metricity).\displaystyle\mathrm{(metricity)}.

As a result, the Riemann curvature of the metric gg, given by the 𝔰𝔬2,2\mathfrak{so}_{2,2}-valued 2-form 12Rbcdaθcθd,\tfrac{1}{2}R^{a}_{~{}bcd}\theta^{c}\wedge\theta^{d}, is defined via the second structure equations

(2.5) dΓba+ΓcaΓbc=12Rbcdaθcθd.{\rm d}\Gamma^{a}_{~{}b}+\Gamma^{a}_{~{}c}\wedge\Gamma^{c}_{~{}b}=\tfrac{1}{2}R^{a}_{~{}bcd}\theta^{c}\wedge\theta^{d}.

Via the action of 𝐒𝐎2,2,\mathbf{SO}_{2,2}, the Riemann curvature decomposes into the components known as the traceless Ricci tensor, Weyl curvature and the scalar curvature. The Ricci tensor is defined as Rab=RacbcR_{ab}=R^{c}_{~{}acb} and the scalar curvature is R=RabgabR=R_{ab}g^{ab}, where gabgbc=δca.g^{ab}g_{bc}=\delta^{a}_{~{}c}. The trace-free part of the Ricci tensor is defined as Rab=Rab14Rgab\overset{\circ}{R}_{ab}=R_{ab}-\tfrac{1}{4}Rg_{ab}. Defining the Schouten tensor as

Pab=12Rab112Rgab,{\mbox{\sf P}}_{ab}=\tfrac{1}{2}R_{ab}-\tfrac{1}{12}Rg_{ab},

the Weyl tensor is expressed as

(2.6) Cbcda=Rbcda+gadPcbgacPdb+gbcPdagbdPca.C^{a}_{~{}bcd}=R^{a}_{~{}bcd}+g_{ad}{\mbox{\sf P}}_{cb}-g_{ac}{\mbox{\sf P}}_{db}+g_{bc}{\mbox{\sf P}}_{da}-g_{bd}{\mbox{\sf P}}_{ca}.

Solving the metricity condition for the first structure equations, it follows that the connection 1-forms Γba\Gamma^{a}_{~{}b} can be expressed as

(2.7) Γba=([cc|cc]Γ11Γ210Γ41Γ12Γ22Γ4100Γ14Γ11Γ12Γ140Γ21Γ22).\Gamma^{a}_{~{}b}=\begin{pmatrix}[cc|cc]\Gamma^{1}_{~{}1}&\Gamma^{1}_{~{}2}&0&\Gamma^{1}_{~{}4}\\ \Gamma^{2}_{~{}1}&\Gamma^{2}_{~{}2}&-\Gamma^{1}_{~{}4}&0\\ \cline{1-4}\cr 0&-\Gamma^{4}_{~{}1}&-\Gamma^{1}_{~{}1}&-\Gamma^{2}_{~{}1}\\ \Gamma^{4}_{~{}1}&0&-\Gamma^{1}_{~{}2}&-\Gamma^{2}_{~{}2}\end{pmatrix}.

Consequently, the torsion-free condition yields

(2.8) dα1=Γ11α1Γ21α2Γ41α¯2\displaystyle{\rm d}\alpha^{1}=-\Gamma^{1}_{~{}1}\wedge\alpha^{1}-\Gamma^{1}_{~{}2}\wedge\alpha^{2}-\Gamma^{1}_{~{}4}\wedge\bar{\alpha}{}^{2}
dα2=Γ12α1Γ22α2+Γ41α¯1\displaystyle{\rm d}\alpha^{2}=-\Gamma^{2}_{~{}1}\wedge\alpha^{1}-\Gamma^{2}_{~{}2}\wedge\alpha^{2}+\Gamma^{1}_{~{}4}\wedge\bar{\alpha}{}^{1}
dα¯=1Γ14α2+Γ11α¯+1Γ12α¯2\displaystyle{\rm d}\bar{\alpha}{}^{1}=\Gamma^{4}_{~{}1}\wedge\alpha^{2}+\Gamma^{1}_{~{}1}\wedge\bar{\alpha}{}^{1}+\Gamma^{2}_{~{}1}\wedge\bar{\alpha}{}^{2}
dα¯=2Γ14α1+Γ21α¯+1Γ22α¯.2\displaystyle{\rm d}\bar{\alpha}{}^{2}=-\Gamma^{4}_{~{}1}\wedge\alpha^{1}+\Gamma^{1}_{~{}2}\wedge\bar{\alpha}{}^{1}+\Gamma^{2}_{~{}2}\wedge\bar{\alpha}{}^{2}.

Passing to the second structure equations, one notes that due to the symmetries of the Riemann tensor, Rabcd=R[ab][cd]=RcdabR_{abcd}=R_{[ab][cd]}=R_{cdab}, setting Rab=cdgaegbfRefcdR^{ab}{}_{cd}=g^{ae}g^{bf}R_{efcd}, one obtains a linear map given by

Riemann:Λ2TMΛ2TM,Riemann(θaθb)=12Rabθccdθd.Riemann:\Lambda^{2}\mathrm{T}^{*}M\to\Lambda^{2}\mathrm{T}^{*}M,\quad\quad Riemann(\theta^{a}\wedge\theta^{b})=\tfrac{1}{2}R^{ab}{}_{cd}\theta^{c}\wedge\theta^{d}.

Since Λ2TM=Λ+2Λ2\Lambda^{2}\mathrm{T}^{*}M=\Lambda^{2}_{+}\oplus\Lambda^{2}_{-}, the matrix form of this map can be expressed as

(2.9) Riemann=([c||c]Weyl++112RidΛ+2RicciRicciWeyl+112RidΛ2).Riemann=\begin{pmatrix}[c||c]Weyl^{+}+\tfrac{1}{12}R\operatorname{id}_{\Lambda^{2}_{+}}&\overset{\circ}{Ricci}\\ \cline{1-2}\cr\cline{1-2}\cr\overset{\circ}{Ricci}{}^{*}&Weyl^{-}+\tfrac{1}{12}R\operatorname{id}_{\Lambda^{2}_{-}}\end{pmatrix}.

Here Weyl+Weyl^{+} and WeylWeyl^{-} are traceless 3×33\times 3 matrices, and Ricci\overset{\circ}{Ricci}{}^{*} is a 3×33\times 3 matrix related to the 3×33\times 3 matrix of trace-free Ricci tensor, Ricci,\overset{\circ}{Ricci}, via Ricci=(HRicciH1)T\overset{\circ}{Ricci}{}^{*}=(H\overset{\circ}{Ricci}H^{-1})^{T}, where H=(001020100)H=\tiny{\begin{pmatrix}0&0&-1\\ 0&2&0\\ -1&0&0\end{pmatrix}}. The matrices HWeyl+,HWeylHWeyl^{+},HWeyl^{-} are symmetric and their components will be denoted by (Ψ0,Ψ1,Ψ2,Ψ3,Ψ4)(\Psi_{0}{}^{\prime},\Psi_{1}{}^{\prime},\Psi_{2}{}^{\prime},\Psi_{3}{}^{\prime},\Psi_{4}{}^{\prime}) and (Ψ0,Ψ1,Ψ2,Ψ3,Ψ4)(\Psi_{0},\Psi_{1},\Psi_{2},\Psi_{3},\Psi_{4}) respectively. Moreover, let us denote the the 9-components of Ricci\overset{\circ}{Ricci} by (P11,P12,P22,P14,P13P24,P23,P33,P34,P44).({\mbox{\sf P}}_{11},{\mbox{\sf P}}_{12},{\mbox{\sf P}}_{22},{\mbox{\sf P}}_{14},{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24},{\mbox{\sf P}}_{23},{\mbox{\sf P}}_{33},{\mbox{\sf P}}_{34},{\mbox{\sf P}}_{44}). It follows that the scalar curvature can be written as R=12(P13+P24)R=12({\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24}), as given in (2.21). Since the Ricci tensor RabR_{ab} and the Schouten tensor Pab{\mbox{\sf P}}_{ab} are linearly related, we will be using the Schouten tensor in the sequel. As a result, the second structure equations (2.5) read

(2.10) 12d\displaystyle\tfrac{1}{2}{\rm d} (Γ11+Γ22)+Γ41Γ14=\displaystyle(\Gamma^{1}_{~{}1}+\Gamma^{2}_{~{}2})+\Gamma^{1}_{~{}4}\wedge\Gamma^{4}_{~{}1}=
Ψ3σ+1Ψ1σ+212(2Ψ2P13P24)σ+3+P14σ1P23σ2+12(P13P24)σ3\displaystyle-\Psi_{3}^{\prime}\sigma^{1}_{+}-\Psi_{1}^{\prime}\sigma^{2}_{+}-\tfrac{1}{2}(2\Psi_{2}^{\prime}-{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24})\sigma^{3}_{+}+{\mbox{\sf P}}_{14}\sigma^{1}_{-}-{\mbox{\sf P}}_{23}\sigma^{2}_{-}+\tfrac{1}{2}({\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24})\sigma^{3}_{-}
d\displaystyle{\rm d} Γ14+Γ14(Γ11+Γ22)=\displaystyle\Gamma^{4}_{~{}1}+\Gamma^{4}_{~{}1}\wedge(\Gamma^{1}_{~{}1}+\Gamma^{2}_{~{}2})=
Ψ4σ+1(Ψ2+P13+P24)σ+2Ψ3σ+3P11σ1P22σ2+P12σ3\displaystyle-\Psi_{4}^{\prime}\sigma^{1}_{+}-(\Psi_{2}^{\prime}+{\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24})\sigma^{2}_{+}-\Psi_{3}^{\prime}\sigma^{3}_{+}-{\mbox{\sf P}}_{11}\sigma^{1}_{-}-{\mbox{\sf P}}_{22}\sigma^{2}_{-}+{\mbox{\sf P}}_{12}\sigma^{3}_{-}
d\displaystyle{\rm d} Γ41+(Γ11+Γ22)Γ41=\displaystyle\Gamma^{1}_{~{}4}+(\Gamma^{1}_{~{}1}+\Gamma^{2}_{~{}2})\wedge\Gamma^{1}_{~{}4}=
(Ψ2+P13+P24)σ+1+Ψ0σ+2+Ψ1σ+3+P44σ1+P33σ2+P34σ3,\displaystyle(\Psi_{2}^{\prime}+{\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24})\sigma^{1}_{+}+\Psi_{0}^{\prime}\sigma^{2}_{+}+\Psi_{1}^{\prime}\sigma^{3}_{+}+{\mbox{\sf P}}_{44}\sigma^{1}_{-}+{\mbox{\sf P}}_{33}\sigma^{2}_{-}+{\mbox{\sf P}}_{34}\sigma^{3}_{-},

with analogous equations for the ‘unprimed’ objects:

(2.11) 12d\displaystyle\tfrac{1}{2}{\rm d} (Γ11Γ22)+Γ21Γ12=\displaystyle(\Gamma^{1}_{~{}1}-\Gamma^{2}_{~{}2})+\Gamma^{1}_{~{}2}\wedge\Gamma^{2}_{~{}1}=
Ψ1σ1+Ψ3σ212(2Ψ2P13P24)σ3+P12σ+1P34σ+2+12(P13P24)σ+3\displaystyle\Psi_{1}\sigma^{1}_{-}+\Psi_{3}\sigma^{2}_{-}-\tfrac{1}{2}(2\Psi_{2}-{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24})\sigma^{3}_{-}+{\mbox{\sf P}}_{12}\sigma^{1}_{+}-{\mbox{\sf P}}_{34}\sigma^{2}_{+}+\tfrac{1}{2}({\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24})\sigma^{3}_{+}
d\displaystyle{\rm d} Γ12+Γ12(Γ11Γ22)=\displaystyle\Gamma^{2}_{~{}1}+\Gamma^{2}_{~{}1}\wedge(\Gamma^{1}_{~{}1}-\Gamma^{2}_{~{}2})=
Ψ0σ1(Ψ2+P13+P24)σ2+Ψ1σ3P11σ+1P44σ+2+P14σ+3\displaystyle-\Psi_{0}\sigma^{1}_{-}-(\Psi_{2}+{\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24})\sigma^{2}_{-}+\Psi_{1}\sigma^{3}_{-}-{\mbox{\sf P}}_{11}\sigma^{1}_{+}-{\mbox{\sf P}}_{44}\sigma^{2}_{+}+{\mbox{\sf P}}_{14}\sigma^{3}_{+}
d\displaystyle{\rm d} Γ21+(Γ11Γ22)Γ21=\displaystyle\Gamma^{1}_{~{}2}+(\Gamma^{1}_{~{}1}-\Gamma^{2}_{~{}2})\wedge\Gamma^{1}_{~{}2}=
(Ψ2+P13+P24)σ1+Ψ4σ2Ψ3σ3+P22σ+1+P33σ+2+P23σ+3.\displaystyle(\Psi_{2}+{\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24})\sigma^{1}_{-}+\Psi_{4}\sigma^{2}_{-}-\Psi_{3}\sigma^{3}_{-}+{\mbox{\sf P}}_{22}\sigma^{1}_{+}+{\mbox{\sf P}}_{33}\sigma^{2}_{+}+{\mbox{\sf P}}_{23}\sigma^{3}_{+}.

Here we used the respective basis (σ±1,σ±2,σ±3)(\sigma^{1}_{\pm},\sigma^{2}_{\pm},\sigma^{3}_{\pm}) of Λ±2\Lambda^{2}_{\pm}, as defined in (2.2).

Remark 2.1.

The usual way of employing the system of equations (2.8), (2.10)-(2.11), is to think about (θ1,θ2,θ3,θ4)(\theta^{1},\theta^{2},\theta^{3},\theta^{4}) as a given coframe on MM, and to use the equations (2.8), (2.10)-(2.11) to uniquely determine the Levi-Civita connection forms Γji\Gamma^{i}_{~{}j}, and consequently the curvature RabcdR^{a}{}_{bcd} of gg, in terms of this chosen coframe. Alternatively, in the language of GG-structures, one observes that θi\theta^{i}’s are ambiguous up to an action of 𝐒𝐎2,2\mathbf{SO}_{2,2} since they were chosen so that (2.1) is satisfied. One says that 𝐒𝐎2,2\mathbf{SO}_{2,2} is the structure group of the pseudo-Riemannian structure. As a result, one can define a principal 𝐒𝐎2,2\mathbf{SO}_{2,2}-bundle π:M\pi\colon\mathcal{F}\to M, as the bundle of all null coframes with respect to which (2.1) holds. In this language the θi\theta^{i}’s give rise to a lifted null coframe at each point of \mathcal{F} and the Γji\Gamma^{i}_{j}’s mimic the Maurer-Cartan forms of 𝔰𝔬2,2\mathfrak{so}_{2,2}; they are uniquely defined on \mathcal{F} as a result of the torsion-free condition. Moreover, these 1-forms, together with θi\theta^{i}s, form a basis of 1-forms at every point of \mathcal{F}. Hence, one obtains a unique coframe at each point of \mathcal{F}, consisting of 1-forms (θi,Γi)j(\theta^{i},\Gamma^{i}{}_{j}), which is transformed equivariantly in each fiber of \mathcal{F} and satisfy the equations (2.8), (2.10)-(2.11) everywhere on \mathcal{F}. We refer to [Gar89, Olv95] for an overview of this exterior differential system (EDS) viewpoint.

2.2. Almost para-Hermitian metrics

In this section we define almost para-complex structures and almost para-Hermitian metrics. We obtain the structure equations and curvature decomposition. Using the curvature decomposition, we recall the well-known Petrov classification of such structures.

2.2.1. Definitions

An almost para-Hermitian structure (M,g,K)(M,g,K) on a 4-dimensional manifold MM with a metric gg of signature (+,+,,)(+,+,-,-) is defined in terms of an endomorphism

K:TMTM,K:\mathrm{T}M\to\mathrm{T}M,

such that

K2=idTM,(Kparacomplex),\hskip 128.0374ptK^{2}=\operatorname{id}_{\mathrm{T}M},\qquad\qquad(K~{}\mathrm{paracomplex}),

whose ±1\pm 1-eigenvalues have rank 2 and, additionally, satisfies the compatibility condition

g(KX,KY)=g(X,Y),X,YTM,(Kmetriccompatible).g(KX,KY)=-g(X,Y),\qquad\forall X,Y\in\mathrm{T}M,\qquad(K~{}\mathrm{metric~{}compatible}).

An almost para-Hermitian structure (M,g,K)(M,g,K) distinguishes a pair of rank 2 distributions \mathscr{H} and ¯\bar{\mathscr{H}} on MM defined as the ±1\pm 1-eigenspaces of KK i.e.

(2.12) =(K+idTM)TM,and¯=(KidTM)TM.{\mathscr{H}}=(K+\operatorname{id}_{\mathrm{T}M})\mathrm{T}M,\quad\mathrm{and}\quad\bar{\mathscr{H}}=(K-\operatorname{id}_{\mathrm{T}M})\mathrm{T}M.

It follows that

TM=¯.\mathrm{T}M={\mathscr{H}}\oplus\bar{\mathscr{H}}.

Moreover \mathscr{H} and ¯\bar{\mathscr{H}} are null with respect to gg and must belong to the same orbit in the space 𝒩=𝒩+𝒩{\mathcal{N}}={\mathcal{N}}_{+}\sqcup{\mathcal{N}}_{-} of all null planes.

An almost para-Hermitian structure (M,g,K)(M,g,K) is called half-para-Hermitian if precisely one of \mathscr{H} or ¯\bar{\mathscr{H}} are integrable i.e. either [,][{\mathscr{H}},{\mathscr{H}}]\subset{\mathscr{H}} or [¯,¯]¯.[\bar{\mathscr{H}},\bar{\mathscr{H}}]\subset\bar{\mathscr{H}}. If \mathscr{H} and ¯\bar{\mathscr{H}} are both integrable i.e.

[,],and[¯,¯]¯,[{\mathscr{H}},{\mathscr{H}}]\subset{\mathscr{H}},\quad\mathrm{and}\quad[\bar{\mathscr{H}},\bar{\mathscr{H}}]\subset\bar{\mathscr{H}},

then the almost-para-Hermitian structure (M,g,K)(M,g,K) is called para-Hermitian.

An almost para-Hermitian structure (M,g,K)(M,g,K) defines a para-Kähler 2-form

(2.13) ρ(X,Y):=g(KX,Y).\rho(X,Y):=g(KX,Y).

The fact that ρ\rho is skew symmetric, ρ(X,Y)=ρ(Y,X)\rho(X,Y)=-\rho(Y,X), follows from the algebraic properties of KK.

An almost para-Hermitian structure (M,g,K)(M,g,K) is called almost para-Kähler if and only if the 2-form ρ\rho is closed i.e.

dρ=0.{\rm d}\rho=0.

An almost para-Hermitian structure (M,g,K)(M,g,K) is para-Kähler if it is para-Hermitian and almost para-Kähler i.e. \mathscr{H} and ¯\bar{\mathscr{H}} are integrable and ρ\rho is closed.

2.2.2. Almost para-Hermitian structure in an adapted frame.

A coframe (α1,α2,α¯,1α¯)2(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}) on a 4-dimensional manifold MM is adapted to an almost para-Hermitian structure (M,g,K)(M,g,K) if and only if

(2.14) g=2α1α¯+12α2α¯2\displaystyle g=2\alpha^{1}\bar{\alpha}{}^{1}+2\alpha^{2}\bar{\alpha}{}^{2}
K=α1α1+α2α2α¯1α¯1α¯2α¯2.\displaystyle K=\textstyle{\alpha^{1}\otimes\frac{\partial}{\partial\alpha^{1}}+\alpha^{2}\otimes\frac{\partial}{\partial\alpha^{2}}-\bar{\alpha}{}^{1}\otimes\frac{\partial}{\partial\bar{\alpha}^{1}}-\bar{\alpha}{}^{2}\otimes\frac{\partial}{\partial\bar{\alpha}^{2}}}.

It follows that in such adapted coframes

(2.15) ρ=α1α¯+1α2α¯,2\rho=\alpha^{1}\wedge\bar{\alpha}{}^{1}+\alpha^{2}\wedge\bar{\alpha}{}^{2},

At every point of a 4-dimensional almost para-Hermitian manifold (M,g,K)(M,g,K) the stabilizer 𝐇𝐆𝐋4()\mathbf{H}\subset\mathbf{GL}_{4}(\mathbb{R}) of the pair (g,K)(g,K), i.e.

𝐇={U𝐆𝐋4():g(UX,UY)=g(X,Y)&K(UX)=UK(X)},\mathbf{H}=\{U\in\mathbf{GL}_{4}(\mathbb{R})\,:\,g(UX,UY)=g(X,Y)\,\,\&\,\,K(UX)=UK(X)\},

satisfies

𝐇𝐆𝐋2()𝐒𝐎2,2.\mathbf{H}\cong\mathbf{GL}_{2}(\mathbb{R})\subset\mathbf{SO}_{2,2}.

Expressing 𝐇\mathbf{H} in the coframe (θ1,(\theta^{1}, θ2,θ3,θ4)=(α1,α2,α¯,1α¯)2\theta^{2},\theta^{3},\theta^{4})=(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}) as in (2.14), provides the 4-dimensional reducible representation

T:𝐇𝐆𝐋4()T:\mathbf{H}\to\mathbf{GL}_{4}(\mathbb{R})

of 𝐇\mathbf{H} given by

(2.16) T(U)=(A00(AT)1)withA=(a11a12a21a22)𝐆𝐋2().T(U)=\begin{pmatrix}A&0\\ 0&(A^{T})^{-1}\end{pmatrix}\quad\mathrm{with}\quad A=\begin{pmatrix}a_{11}&a_{12}\\ a_{21}&a_{22}\end{pmatrix}\in\mathbf{GL}_{2}(\mathbb{R}).

As a result the geometry arising from the pair (g,K)(g,K) reduces the structure group of MM from 𝐆𝐋4()\mathbf{GL}_{4}(\mathbb{R}) to 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) via representation T.T. The 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) irreducible decomposition of 4\mathbb{R}^{4} as a 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R})-module is 4=2(2)\mathbb{R}^{4}=\mathbb{R}^{2}\oplus(\mathbb{R}^{2})^{*}. It reflects the splitting of TM\mathrm{T}M, into TM=¯\mathrm{T}M=\mathscr{H}\oplus\bar{\mathscr{H}}.

Proposition 2.2.

Every almost-para-Hermitian structure (M,g,K)(M,g,K) on a 4-dimensional manifold locally admits an adapted coframe. If (θa)=(α1,α2,α¯,1(\theta^{a})=(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1}, α¯)2\bar{\alpha}{}^{2}) is a coframe adapted to (M,g,K)(M,g,K) then the most general adapted coframe is given by

(2.17) θ~a=T(U)baθb,\tilde{\theta}^{a}=T(U)^{a}_{~{}b}\theta^{b},

where the 4×44\times 4 matrices T(U)=(T(U)a)bT(U)=(T(U)^{a}{}_{b}) are as in (2.16).

2.2.3. 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) invariant curvature decomposition

Any coframe adapted to (M,g,K)(M,g,K) is in particular a null coframe, as in (2.1). Thus to analyze the properties of (M,g,K)(M,g,K) we can use the structure equations (2.8), (2.10)-(2.11). The stabilizer 𝐇𝐆𝐋2()\mathbf{H}\cong\mathbf{GL}_{2}(\mathbb{R}) of the pair (g,K)(g,K) is therefore the structure group of the almost para-Hermitian structure (M,g,K)(M,g,K). It acts, via the representation TT, on any adapted coframe (θa)(\theta^{a}) as in (2.17). The induced transformation of the Levi-Civita connection (2.7) and its curvature is given by

(2.18a) θaθ~=aT(U)baθb,\displaystyle\theta^{a}\to\tilde{\theta}{}^{a}=T(U)^{a}_{~{}b}\theta^{b},
(2.18b) ΓbaΓ~=baT(U)caΓdcT(U)1dbdT(U)caT(U)1c,b\displaystyle\Gamma^{a}_{~{}b}\to\tilde{\Gamma}{}^{a}_{~{}b}=T(U)^{a}_{~{}c}\Gamma^{c}_{~{}d}T(U)^{-1d}{}_{b}-dT(U)^{a}_{~{}c}T(U)^{-1c}{}_{b},
(2.18c) RbcdaR~=bcdaT(U)eaRfgheT(U)1fTb(U)1gTc(U)1h.d\displaystyle R^{a}_{~{}bcd}\to\tilde{R}{}^{a}_{~{}bcd}=T(U)^{a}_{~{}e}R^{e}_{~{}fgh}T(U)^{-1f}{}_{b}T(U)^{-1g}{}_{c}T(U)^{-1h}{}_{d}.

The transformations (2.18c) gives the action of 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) on the 20-dimensional vector space of the curvature tensors RbcdaR^{a}_{~{}bcd}. Using this action one can decompose the curvature tensor into its indecomposable components. First we define 10 vector spaces defined in terms of the curvature components (Pab,Ψμ,Ψμ)({\mbox{\sf P}}_{ab},\Psi_{\mu},\Psi_{\mu}^{\prime}), a,b=1,2,3,4a,b=1,2,3,4, μ=0,1,2,3,4\mu=0,1,2,3,4 as

(2.19) Ric13=\displaystyle Ric^{3}_{1}= {ΠABs.t.ΠAB=(P11P12P12P22)}\displaystyle\{\Pi_{AB}~{}\mathrm{s.t.}~{}\Pi_{AB}=\begin{pmatrix}{\mbox{\sf P}}_{11}&{\mbox{\sf P}}_{12}\\ {\mbox{\sf P}}_{12}&{\mbox{\sf P}}_{22}\end{pmatrix}\}
Ric23=\displaystyle Ric^{3}_{2}= {Π¯ABs.t.Π¯AB=(P33P34P34P44)}\displaystyle\{\bar{\Pi}_{AB}~{}\mathrm{s.t.}~{}\bar{\Pi}_{AB}=\begin{pmatrix}{\mbox{\sf P}}_{33}&{\mbox{\sf P}}_{34}\\ {\mbox{\sf P}}_{34}&{\mbox{\sf P}}_{44}\end{pmatrix}\}
Ric33=\displaystyle Ric^{3}_{3}= {PBAs.t.PBA=(P13P242P232P14P13+P24)}\displaystyle\{P^{A}_{~{}B}~{}\mathrm{s.t.}~{}P^{A}_{~{}B}=\begin{pmatrix}{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24}&2{\mbox{\sf P}}_{23}\\ 2{\mbox{\sf P}}_{14}&-{\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24}\end{pmatrix}\}
Scal1=\displaystyle Scal^{1}= {P13+P24}\displaystyle\{{\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24}\}
Weyl15=\displaystyle Weyl^{5}_{1}= {WABCD=W(ABCD)s.t\displaystyle\{W_{ABCD}=W_{(ABCD)}~{}\mathrm{s.t}
W0000=Ψ0,W0001=Ψ1,W0011=Ψ2,W0111=Ψ3,W1111=Ψ4,}\displaystyle W_{0000}=\Psi_{0},~{}W_{0001}=\Psi_{1},~{}W_{0011}=\Psi_{2},~{}W_{0111}=\Psi_{3},~{}W_{1111}=\Psi_{4},\}
Weyl11=\displaystyle Weyl^{1}_{1}= {Ψ0},Weyl21={Ψ1},Weyl31={Ψ2},Weyl41={Ψ3},Weyl51={Ψ4}.\displaystyle\{\Psi_{0}^{\prime}\},\quad Weyl^{1}_{2}=\{\Psi_{1}^{\prime}\},\quad Weyl^{1}_{3}=\{\Psi_{2}^{\prime}\},\quad Weyl^{1}_{4}=\{\Psi_{3}^{\prime}\},\quad Weyl^{1}_{5}=\{\Psi_{4}^{\prime}\}.

Here the (spinorial) indices A,B,C,D=0,1A,B,C,D=0,1, and the equations WABCD=W(ABCD)W_{ABCD}=W_{(ABCD)} mean that WABCDW_{ABCD} is totally symmetric in indices A,B,C,DA,B,C,D. The notation for the spaces RicjiRic^{i}_{j} and WeyljiWeyl^{i}_{j} is such that the upper index indicates the dimension of each space, and the lower index enumerates spaces of the same dimension. In particular the spaces Ric13Ric^{3}_{1} and Ric23Ric^{3}_{2} have dimensions 3 as spaces of symmetric 2×22\times 2 matrices, Ric33Ric^{3}_{3} has dimension 3 as the space of traceless 2×22\times 2 matrices, and Weyl15Weyl^{5}_{1} has dimension 5 as the space of symmetric tensors of degree 4 in dimension 2.

Proposition 2.3.

The 𝐆𝐋2()𝐒𝐎2,2\mathbf{GL}_{2}(\mathbb{R})\subset\mathbf{SO}_{2,2} invariant decomposition of the 20-dimensional curvature space, Riemann20,Riemann^{20}, of an almost para-Hermitian structure (M,g,K)(M,g,K) in dimension 4 is

(2.20) Riemann20=\displaystyle Riemann^{20}= Ric13Ric23Ric33traceless Ricci\displaystyle\underbrace{Ric^{3}_{1}\oplus Ric^{3}_{2}\oplus Ric^{3}_{3}}_{\text{traceless Ricci}}\oplus
Scal1Ricci scalar\displaystyle\underbrace{Scal^{1}}_{\text{Ricci scalar}}\oplus
Weyl11Weyl21Weyl31Weyl41Weyl51self-dual Weyl\displaystyle\underbrace{Weyl^{1}_{1}\oplus Weyl^{1}_{2}\oplus Weyl^{1}_{3}\oplus Weyl^{1}_{4}\oplus Weyl^{1}_{5}}_{\text{self-dual Weyl}}\oplus
Weyl15.anti-self-dual Weyl\displaystyle\underbrace{Weyl^{5}_{1}.}_{\text{anti-self-dual Weyl}}
Proof.

This decompositions can be obtained similar to the decomposition of Riemann20Riemann^{20} into Weyl±Weyl^{\pm}, Ricci\overset{\circ}{Ricci} and RR via the 𝐒𝐎2,2\mathbf{SO}_{2,2} invariant decomposition of Λ2TM\Lambda^{2}\mathrm{T}^{*}M. In this case one decomposes Λ2TM\Lambda^{2}\mathrm{T}^{*}M using the 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) invariant decomposition of the tangent space

TM=¯,\mathrm{T}M={\mathscr{H}}\oplus\bar{\mathscr{H}},

which is possible for any almost para-Hermitian manifold (M,g,K)(M,g,K). The associated decomposition of the cotangent bundle Λ1TM\Lambda^{1}\mathrm{T}^{*}M is given by

Λ1TM=Λ(1,0)Λ(0,1),\Lambda^{1}\mathrm{T}^{*}M=\Lambda^{(1,0)}\oplus\Lambda^{(0,1)},

where

Λ(1,0)\displaystyle\Lambda^{(1,0)} ={ωΛ1TM|X¯ω=0,X¯¯}\displaystyle=\{~{}\omega\in\Lambda^{1}\mathrm{T}^{*}M~{}|~{}\bar{X}\im\omega=0,~{}\forall\bar{X}\in\bar{\mathscr{H}}~{}\}
Λ(0,1)\displaystyle\Lambda^{(0,1)} ={ω¯Λ1TM|Xω¯=0,X}.\displaystyle=\{~{}\bar{\omega}\in\Lambda^{1}\mathrm{T}^{*}M~{}|~{}X\im\bar{\omega}=0,~{}\forall X\in{\mathscr{H}}~{}\}.

As a result, Λ2TM\Lambda^{2}\mathrm{T}^{*}M is decomposed into

Λ2TM=Λ(2,0)Λ(1,1)Λ(0,2).\Lambda^{2}\mathrm{T}^{*}M=\Lambda^{(2,0)}\oplus\Lambda^{(1,1)}\oplus\Lambda^{(0,2)}.

It turns out that Λ(2,0)\Lambda^{(2,0)} and Λ(0,2)\Lambda^{(0,2)} are 1-dimensional, and Λ(1,1)\Lambda^{(1,1)} has dimension 4. Choosing an adapted coframe (α1,α2,α¯,1α¯)2(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}) we can write a basis for these spaces in terms of the self-dual and anti-self-dual 2-forms, (σ±1,σ±2,σ±3),(\sigma^{1}_{\pm},\sigma^{2}_{\pm},\sigma^{3}_{\pm}), in (2.2), as follows.

Λ(2,0)Λ+2\displaystyle\Lambda^{(2,0)}\cap\Lambda^{2}_{+} =Span{σ+1}=Λ(2,0),\displaystyle=\mathrm{Span}\{\sigma^{1}_{+}\}=\Lambda^{(2,0)},
Λ(0,2)Λ+2\displaystyle\Lambda^{(0,2)}\cap\Lambda^{2}_{+} =Span{σ+2}=Λ(0,2)\displaystyle=\mathrm{Span}\{\sigma^{2}_{+}\}=\Lambda^{(0,2)}
Λ(1,1)Λ+2\displaystyle\Lambda^{(1,1)}\cap\Lambda^{2}_{+} =Span{σ+3},\displaystyle=\mathrm{Span}\{\sigma^{3}_{+}\},
Λ(1,1)Λ2\displaystyle\Lambda^{(1,1)}\cap\Lambda^{2}_{-} =Span{σ1,σ2,σ3}=Λ2,\displaystyle=\mathrm{Span}\{\sigma^{1}_{-},\sigma^{2}_{-},\sigma^{3}_{-}\}=\Lambda^{2}_{-},

This gives a natural decomposition of Λ+2\Lambda^{2}_{+} into 1-dimensional 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) invariant subspaces

Λ+2=Span{σ+1}Span{σ+3}Span{σ+2}.\Lambda^{2}_{+}=\mathrm{Span}\{\sigma^{1}_{+}\}\oplus\mathrm{Span}\{\sigma^{3}_{+}\}\oplus\mathrm{Span}\{\sigma^{2}_{+}\}.

Using this we can further decompose the map RiemannRiemann from (2.9) as

(2.21) Riemann=\displaystyle Riemann= 112Rid6×6+([c|c|c||ccc]Ψ22Ψ3Ψ4P222P12P11Ψ12Ψ2Ψ3P23P13P24P14Ψ02Ψ1Ψ2P332P34P44P442P14P11Ψ22Ψ1Ψ0P34P13P24P12Ψ32Ψ2Ψ1P332P23P22Ψ42Ψ3Ψ2).\displaystyle\tfrac{1}{12}R\operatorname{id}_{6\times 6}+\begin{pmatrix}[c|c|c||ccc]\Psi_{2}^{\prime}&-2\Psi_{3}^{\prime}&\Psi_{4}^{\prime}&{\mbox{\sf P}}_{22}&2{\mbox{\sf P}}_{12}&{\mbox{\sf P}}_{11}\\ \cline{1-6}\cr\Psi_{1}^{\prime}&-2\Psi_{2}^{\prime}&\Psi_{3}^{\prime}&{\mbox{\sf P}}_{23}&{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24}&-{\mbox{\sf P}}_{14}\\ \cline{1-6}\cr\Psi_{0}^{\prime}&-2\Psi_{1}^{\prime}&\Psi_{2}^{\prime}&{\mbox{\sf P}}_{33}&-2{\mbox{\sf P}}_{34}&{\mbox{\sf P}}_{44}\\ \cline{1-6}\cr\cline{1-6}\cr{\mbox{\sf P}}_{44}&2{\mbox{\sf P}}_{14}&{\mbox{\sf P}}_{11}&\Psi_{2}&2\Psi_{1}&\Psi_{0}\\ {\mbox{\sf P}}_{34}&{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24}&-{\mbox{\sf P}}_{12}&-\Psi_{3}&-2\Psi_{2}&-\Psi_{1}\\ {\mbox{\sf P}}_{33}&-2{\mbox{\sf P}}_{23}&{\mbox{\sf P}}_{22}&\Psi_{4}&2\Psi_{3}&\Psi_{2}\end{pmatrix}.

Comparing this with the decomposition (2.9) one obtains

  • Weyl+Weyl^{+} gets decomposed into five 1-dimensional 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) invariant subspaces denoted by Weyl11,,Weyl51Weyl^{1}_{1},\dots,Weyl^{1}_{5} which correspond to the components Ψ0,,Ψ4\Psi^{\prime}_{0},\dots,\Psi^{\prime}_{4} in (2.21) respectively.

  • Ricci\overset{\circ}{Ricci} is decomposed into three invariant subspaces, Ric3,1Ric3,2Ric33Ric^{3}{}_{1},Ric^{3}{}_{2},Ric^{3}{}_{3} which correspond to the rows (P22,P12,P11)({\mbox{\sf P}}_{22},{\mbox{\sf P}}_{12},{\mbox{\sf P}}_{11}), (P23,P13P24,P14)({\mbox{\sf P}}_{23},{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24},{\mbox{\sf P}}_{14}), and (P33,({\mbox{\sf P}}_{33}, P34,P44){\mbox{\sf P}}_{34},{\mbox{\sf P}}_{44}) in (2.21) respectively.

  • WeylWeyl^{-} remains indecomposable with its 5-dimensional representation Weyl15Weyl^{5}_{1} whose components are (Ψ0,Ψ1,Ψ2,Ψ3,Ψ4)(\Psi_{0},\Psi_{1},\Psi_{2},\Psi_{3},\Psi_{4}).

  • The Ricci scalar R=12(P13+P24)R=12({\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24}) is proportional to the trace of RiemannRiemann and gives the 1-dimensional invariant subspace Scal1Scal^{1}.

As a result one obtains the decompositions (2.20). ∎

It is straightforward to find the explicit action of the 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) group on the indecomposable components of the curvature in (2.19).

Proposition 2.4.

The curvature components Ric13Ric^{3}_{1}, Ric23Ric^{3}_{2}, Ric33Ric^{3}_{3} and Weyl15Weyl^{5}_{1} are ‘tensorial’ with respect to the action of 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) i.e. for U𝐆𝐋2()𝐒𝐎2,2U\in\mathbf{GL}_{2}(\mathbb{R})\subset\mathbf{SO}_{2,2} given by (2.16) if an adapted coframe (θa)(\theta^{a}) is transformed by

θaθ~=aT(U)baθb,\theta^{a}\to\tilde{\theta}{}^{a}=T(U)^{a}_{~{}b}\theta^{b},

then the transformation law for the curvature components ΠAB\Pi_{AB}, Π¯AB\bar{\Pi}_{AB} and PBAP^{A}_{~{}B} in (2.19) is

(2.22a) ΠABΠ~=ABΠCDAtCAtDA,B\displaystyle\Pi_{AB}\to\tilde{\Pi}{}_{AB}=\Pi_{CD}A^{tC}{}_{A}A^{tD}{}_{B},
(2.22b) Π¯ABΠ¯~=ABΠ¯CDA1CA1DA,B\displaystyle\bar{\Pi}_{AB}\to\tilde{\bar{\Pi}}{}_{AB}=\bar{\Pi}_{CD}A^{-1C}{}_{A}A^{-1D}{}_{B},
(2.22c) PBAP~BA=ACAPDCA1D,B\displaystyle P^{A}_{~{}B}\to\tilde{P}^{A}_{~{}B}=A^{A}_{~{}C}P^{C}_{~{}D}A^{-1D}{}_{B},
(2.22d) WABCDW~=ABCDWEFGHA1EA1FAA1GBA1HC.D\displaystyle W_{ABCD}\to\tilde{W}{}_{ABCD}=W_{EFGH}A^{-1E}{}_{A}A^{-1F}{}_{B}A^{-1G}{}_{C}A^{-1H}{}_{D}.

The curvature scalars Ψ0\Psi_{0}^{\prime}, Ψ1\Psi_{1}^{\prime}, Ψ2\Psi_{2}^{\prime}, Ψ3\Psi_{3}^{\prime}, Ψ4\Psi_{4}^{\prime}, P14+P23{\mbox{\sf P}}_{14}+{\mbox{\sf P}}_{23}, are weighted scalars and transform to

(2.23) Ψ0\displaystyle\Psi_{0}^{\prime} Ψ~=0(detA)2Ψ0,\displaystyle\to\tilde{\Psi}{}_{0}^{\prime}=(\det A)^{2}~{}\Psi_{0}^{\prime},\quad Ψ4\displaystyle\Psi_{4}^{\prime} Ψ~=4(detA)2Ψ4,\displaystyle\to\tilde{\Psi}{}_{4}^{\prime}=(\det A)^{-2}~{}\Psi_{4}^{\prime},
Ψ1\displaystyle\Psi_{1}^{\prime} Ψ~=1(detA)Ψ1,\displaystyle\to\tilde{\Psi}{}_{1}^{\prime}=(\det A)~{}\Psi_{1}^{\prime},\quad Ψ3\displaystyle\Psi_{3}^{\prime} Ψ~=3(detA)1Ψ3,\displaystyle\to\tilde{\Psi}{}_{3}^{\prime}=(\det A)^{-1}~{}\Psi_{3}^{\prime},
Ψ2\displaystyle\Psi_{2}^{\prime} Ψ~=2Ψ2,\displaystyle\to\tilde{\Psi}{}_{2}^{\prime}=\Psi_{2}^{\prime},\quad P13+P24\displaystyle{\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24} P~+13P~=24P13+P24.\displaystyle\to\tilde{{\mbox{\sf P}}}{}_{13}+\tilde{{\mbox{\sf P}}}{}_{24}={\mbox{\sf P}}_{13}+{\mbox{\sf P}}_{24}.
Corollary 2.5.

Every almost para-Hermitian structure (M,g,K)(M,g,K) in dimension 4 possesses two scalar invariants which are the scalar curvature of the metric g,g, given by R=12(P12+P34)R=12({\mbox{\sf P}}_{12}+{\mbox{\sf P}}_{34}), and Ψ2,\Psi_{2}^{\prime}, arising from the self-dual Weyl tensor of the metric. Moreover the vanishing of each of the 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) densities, Ψ0,,Ψ4\Psi_{0}^{\prime},\dots,\Psi_{4}^{\prime}, as well as each of the 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) tensors, ΠAB\Pi_{AB}, Π¯AB\bar{\Pi}_{AB} and PBA,P^{A}_{~{}B}, is an invariant property of almost para-Hermitian structures.

2.2.4. Cartan-Penrose-Petrov classification of the Weyl tensor

One of the basic pointwise invariants of 4-dimensional metrics of split signature is the so-called Petrov type of its self-dual and anti-self-dual Weyl curvatures. To define it, note that the transformation law (2.22d) shows the action of the structure group 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) on the anti-self-dual Weyl tensors, Weyl15,Weyl^{5}_{1}, as an 5-dimensional representation. This representation is isomorphic with the standard representation of 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) on Sym4(2)\mathrm{Sym}^{4}(\mathbb{R}^{2})^{*} i.e. the degree 4 homogeneous polynomials in two variables. Using (2.19), the quartic polynomial is given by

W(ξ)\displaystyle W(\xi) =WABCDξAξBξCξD\displaystyle=W_{ABCD}\xi^{A}\xi^{B}\xi^{C}\xi^{D}
=Ψ4(ξ1)4+4Ψ3(ξ1)3(ξ0)+6Ψ2(ξ1)2(ξ0)2+4Ψ1(ξ1)(ξ0)3+Ψ0(ξ0)4,\displaystyle=\Psi_{4}(\xi^{1})^{4}+4\Psi_{3}(\xi^{1})^{3}(\xi^{0})+6\Psi_{2}(\xi^{1})^{2}(\xi^{0})^{2}+4\Psi_{1}(\xi^{1})(\xi^{0})^{3}+\Psi_{0}(\xi^{0})^{4},

where ξ=(ξ0,ξ1).\xi=(\xi^{0},\xi^{1}). It turns out that ξ\xi can serve as a homogeneous coordinate for the circle bundle of anti-self-dual planes, 𝒩.\mathcal{N}_{-}. More precisely, using the Weyl curvature (2.6), define Cabcd=gadCbcdd,C_{abcd}=g_{ad}C^{d}_{~{}bcd}, which can be used to define the multilinear map

𝐖:=Cabcd(θaθb)(θcθd):Sym2(Λ2TM)𝒩+(M).\mathbf{W}:=C_{abcd}(\theta^{a}{\,{\wedge}\;}\theta^{b})\circ(\theta^{c}{\,{\wedge}\;}\theta^{d}):\mathrm{Sym}^{2}(\Lambda^{2}\mathrm{T}M)\rightarrow{\mathcal{N}^{*}_{+}}^{\infty}(M).

Restricting to anti-self-dual null planes, 𝒩Λ2TM\mathcal{N}_{-}\subset\Lambda^{2}\mathrm{T}M, as in (2.3), one can define the quartic polynomial

(2.24) W(λ)\displaystyle W(\lambda) =𝐖(θ1+λθ2,θ4λθ3,θ1+λθ2,θ4λθ3)\displaystyle\textstyle{=\mathbf{W}(\frac{\partial}{\partial\theta^{1}}+\lambda\frac{\partial}{\partial\theta^{2}},\frac{\partial}{\partial\theta^{4}}-\lambda\frac{\partial}{\partial\theta^{3}},\frac{\partial}{\partial\theta^{1}}+\lambda\frac{\partial}{\partial\theta^{2}},\frac{\partial}{\partial\theta^{4}}-\lambda\frac{\partial}{\partial\theta^{3}})}
=Ψ4λ4+4Ψ3λ3+6Ψ2λ2+4Ψ1λ+Ψ0\displaystyle=\Psi_{4}\lambda^{4}+4\Psi_{3}\lambda^{3}+6\Psi_{2}\lambda^{2}+4\Psi_{1}\lambda+\Psi_{0}

where

Ψ4=C2323,Ψ3=C1323,Ψ2=C1423,Ψ1=C2414,Ψ0=C1414,\Psi_{4}=C_{2323},\quad\Psi_{3}=C_{1323},\quad\Psi_{2}=-C_{1423},\quad\Psi_{1}=C_{2414},\quad\Psi_{0}=C_{1414},

which establishes the relation λ=ξ1ξ0\lambda=\frac{\xi^{1}}{\xi^{0}} between the parameters. The quartic (2.24) is a representation of the anti-self-dual Weyl curvature of the metric g.g.

Similarly, restricting to self-dual null planes, 𝒩+Λ2TM\mathcal{N}_{+}\subset\Lambda^{2}\mathrm{T}M, and using the affine parametrization (2.3), one can define the quartic polynomial

(2.25) W(μ)\displaystyle W^{\prime}(\mu) =𝐖(θ4μθ1,θ3+μθ2,θ4μθ1,θ3+μθ2)\displaystyle\textstyle{=\mathbf{W}(\frac{\partial}{\partial\theta^{4}}-\mu\frac{\partial}{\partial\theta^{1}},\frac{\partial}{\partial\theta^{3}}+\mu\frac{\partial}{\partial\theta^{2}},\frac{\partial}{\partial\theta^{4}}-\mu\frac{\partial}{\partial\theta^{1}},\frac{\partial}{\partial\theta^{3}}+\mu\frac{\partial}{\partial\theta^{2}})}
=Ψ4μ4+4Ψ3μ3+6Ψ2μ2+4Ψ1μ+Ψ0\displaystyle=\Psi^{\prime}_{4}\mu^{4}+4\Psi^{\prime}_{3}\mu^{3}+6\Psi^{\prime}_{2}\mu^{2}+4\Psi^{\prime}_{1}\mu+\Psi^{\prime}_{0}

where

Ψ4=C1212,Ψ3=C1213,Ψ2=C1234,Ψ1=C1334,Ψ0=C3434,\Psi^{\prime}_{4}=C_{1212},\quad\Psi^{\prime}_{3}=C_{1213},\quad\Psi^{\prime}_{2}=C_{1234},\quad\Psi^{\prime}_{1}=C_{1334},\quad\Psi^{\prime}_{0}=C_{3434},

The quartic W(μ)W^{\prime}(\mu) is a representation of the self-dual Weyl curvature of gg whose coefficients transform according to (2.23).

The Petrov type at each point is the root type of the quartics W(λ)W(\lambda) and W(μ)W^{\prime}(\mu) at that point, since multiplicity pattern of the roots is invariant under the induced action of the structure group 𝐒𝐎2,2\mathbf{SO}_{2,2}. Note that since the coefficients of the quartics are real and transform under the action of 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}), the root type is closed under complex conjugation. As a result, there are 10 root types for each of the quartics W(λ)W(\lambda) and W(μ).W^{\prime}(\mu). Following the tradition in General Relativity, where the metric has Lorentzian signature, root types are grouped into the six Petrov types, denoted by GG, IIII, IIIIII, NN, DD and OO. In the case of metrics of split signature, due to different reality conditions, one obtains a finer classification of Petrov types given by

  1. (1)

    type GrG^{r}: 4 real simple roots.

  2. (2)

    type GcG^{c}: 2 real simple roots and 2 complex conjugate roots.

  3. (3)

    type GccG^{cc}: 2 pairs of complex conjugate roots..

  4. (4)

    type IIrII^{r}: 1 double real root, 2 simple real roots.

  5. (5)

    type IIcII^{c}: 1 double real root, 2 complex conjugate roots.

  6. (6)

    type IIIIII: 1 triple real root and 1 simple real root.

  7. (7)

    type DrD^{r}: 2 double real roots.

  8. (8)

    type DcD^{c}: 2 double complex conjugate roots.

  9. (9)

    type NN: 1 quadruple real root.

  10. (10)

    type OO: when all the coefficients of the quartic are zero.

The letter GG stands for general type since, generically, the Petrov type of a quartic is GG. If the quartic is non-zero, then the 9 root types are listed in Figure  1, which shows the self-conjugate pattern of roots in each type. The horizontal line represents the real line and conjugation of roots is given by reflection with respect to the horizontal line.

Refer to caption
Figure 1. Root types of a non-zero quartic with real coefficients.

In what follows the root types Gr,IIr,III,Dr,N,OG^{r},II^{r},III,D^{r},N,O will be referred to as the special real Petrov types. It is clear from our discussion that since the parameters λ\lambda and μ\mu in the quartics W(λ)W(\lambda) and W(μ),W^{\prime}(\mu), parametrize 𝒩\mathcal{N}_{-} and 𝒩+,\mathcal{N}_{+}, respectively, a choice of real root for these quartics determine a choice of an anti-self-dual and self-dual null plane. This enables one to consider anti-self-dual or self-dual null planes that correspond to a real root of the quartics W(λ)W(\lambda) or W(μ).W^{\prime}(\mu).

2.3. Para-Kähler (pK) metrics

In this section the para-Kähler condition is used to reduce the structure equations of an almost para-Hermitian structure. After deriving their structure equations and curvature decomposition, we show that para-Kähler structures can be described in terms of a potential function, using which we give two examples of para-Kähler-Einstein metrics. These two examples turn out to be homogeneous as will be explained in § 3.2.

2.3.1. PK structures in an adapted coframe.

A special feature of every almost para-Hermitian geometry (M,g,K)(M,g,K) is that in addition to the (weighted) tensorial invariants arising from the curvature of the Levi-Civita connection, it has invariants of lower order referred to as the intrinsic torsion. These are defined in terms of the (Grey-Harvella type) decomposition of the covariant derivative of the 2-form ρ\rho (2.13) with respect to 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}). Two of these (relative) invariants are of particular interest in our setting. We will express them in terms of the Levi-Civita connection 1-forms Γa.b\Gamma^{a}{}_{b}. Using (2.18b), one obtains that the transformation of the connection 1-forms Γ41\Gamma^{1}_{~{}4} and Γ14\Gamma^{4}_{~{}1} does not involve the inhomogeneous terms d(T(U))T(U)1{\rm d}(T(U))T(U)^{-1}, which leads to the following proposition.

Proposition 2.6.

Under the gauge transformation (2.18) of adapted coframes (θ1,θ2,θ3,θ4)(\theta^{1},\theta^{2},\theta^{3},\theta^{4}) for an almost para-Hermitian structure, the connection 1-forms Γ41\Gamma^{1}_{~{}4} and Γ14\Gamma^{4}_{~{}1} transform as

Γ41Γ~41=(detA)Γ41,\displaystyle\Gamma^{1}_{~{}4}\to\tilde{\Gamma}^{1}_{~{}4}=(\det A)~{}\Gamma^{1}_{~{}4},\qquad Γ14Γ~14=(detA)1Γ14\displaystyle\Gamma^{4}_{~{}1}\to\tilde{\Gamma}^{4}_{~{}1}=(\det A)^{-1}~{}\Gamma^{4}_{~{}1}

where A𝐆𝐋2()A\in\mathbf{GL}_{2}(\mathbb{R}). As a result, the vanishing of each of the connection 1-forms Γ41\Gamma^{1}_{~{}4} and Γ14\Gamma^{4}_{~{}1} is an invariant property of an almost para-Hermitian structure.

We have the following proposition.

Proposition 2.7.

An almost para-Hermitian structure (M,g,K)(M,g,K) is para-Kähler if and only if

Γ41=0andΓ14=0,\Gamma^{1}_{~{}4}=0\quad\mathrm{and}\quad\Gamma^{4}_{~{}1}=0,

in one (and therefore any) adapted coframe. As a result, the Levi-Civita connection form of gg is reduced to

Γba=(Γ00ΓT),withΓ𝔤𝔩2()Λ1TM𝐄𝐧𝐝(2)Λ1TM,\Gamma^{a}_{~{}b}=\begin{pmatrix}\Gamma&0\\ 0&-\Gamma^{T}\end{pmatrix},\quad\mathrm{with}\quad\Gamma\in\mathfrak{gl}_{2}(\mathbb{R})\otimes\Lambda^{1}\mathrm{T}^{*}M\equiv{\bf End}(\mathbb{R}^{2})\otimes\Lambda^{1}\mathrm{T}^{*}M,

in any adapted coframe.

Proof.

By Frobenius theorem the integrability of \mathscr{H} and ¯\bar{\mathscr{H}} in an adapted coframe is equivalent to dα1α1α2=0=dα2α1α2{\rm d}\alpha^{1}\wedge\alpha^{1}\wedge\alpha^{2}=0={\rm d}\alpha^{2}\wedge\alpha^{1}\wedge\alpha^{2} and dα¯1α¯1α¯=20=dα¯2α¯1α¯2{\rm d}\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{2}=0={\rm d}\bar{\alpha}{}^{2}\wedge\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{2} respectively. Using the first structure equations (2.8), it follows that the simultaneous integrability of \mathscr{H} and ¯\bar{\mathscr{H}} implies

(2.26) dΓ14α1α¯1α¯=20,dΓ14α2α¯1α¯=20,\displaystyle{\rm d}\Gamma^{4}_{~{}1}\wedge\alpha^{1}\wedge\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{2}=0,\quad\quad{\rm d}\Gamma^{4}_{~{}1}\wedge\alpha^{2}\wedge\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{2}=0,\quad
dΓ41α¯1α1α2=0,dΓ41α¯2α1α2=0.\displaystyle{\rm d}\Gamma^{1}_{~{}4}\wedge\bar{\alpha}{}^{1}\wedge\alpha^{1}\wedge\alpha^{2}=0,\quad\quad{\rm d}\Gamma^{1}_{~{}4}\wedge\bar{\alpha}{}^{2}\wedge\alpha^{1}\wedge\alpha^{2}=0.

On the other hand, the almost Kähler condition dρ=0{\rm d}\rho=0, when written in an adapted coframe reads

d(α1α2+α¯1α¯)2=0.{\rm d}(\alpha^{1}\wedge\alpha^{2}+\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{2})=0.

Using the first structure equations (2.8) it follows that this condition is equivalent to

(2.27) Γ41α¯1α¯+2Γ14α1α2=0.\Gamma^{1}_{~{}4}\wedge\bar{\alpha}{}^{1}\wedge\bar{\alpha}{}^{2}+\Gamma^{4}_{~{}1}\wedge\alpha^{1}\wedge\alpha^{2}=0.

It follows from (2.26) and (2.27) that Γ4=10\Gamma^{4}{}_{1}=0 and Γ1=40\Gamma^{1}{}_{4}=0, as claimed. ∎

Proposition 2.7 leads to the following “para” analogue of the well-known fact that the holonomy of Riemannian 4-manifolds which are Kähler is a subgroup of 𝐔2\mathbf{U}_{2}.

Corollary 2.8.

For any 4-dimensional para-Kähler structure (M,g,K)(M,g,K) the pseudo-Riemannian holonomy of the metric gg is reduced from 𝐒𝐎2,2\mathbf{SO}_{2,2} to 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) via the representation TT in (2.17). This holonomy reduction is equivalent to the property that KK is parallel with respect to the Levi-Civita connection \nabla of g.g.

Remark 2.9.

The corollary above is a consequence of the so-called holonomy principle in pseudo-Riemannian geometry which establishes a one to one correspondence between the space of parallel sections of tensor bundles and the invariant vectors in TxM\mathrm{T}_{x}M under the action of the holonomy group Holx\mathrm{Hol}_{x} at each point xMx\in M. We refer to [BBI97] for further discussion of the holonomy group of pseudo-Riemannian metrics of split signature.

Let us also point out that in the spirit of Remark 2.1, Proposition 2.7 implies that the bundle of adapted null frames for para-Kähler structures is a principal 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R})-bundle 8M\mathcal{F}^{8}\to M obtained from reducing the 𝐒𝐎2,2\mathbf{SO}_{2,2}-bundle M,\mathcal{F}\to M, with the property that the reduced first order structure equations given by (2.8), (2.10)-(2.11) have no intrinsic torsion.

2.3.2. Curvature of pK geometry and pK-Einstein (pKE) condition

In this section we discuss the curvature of para-Kähler structures.

Proposition 2.10.

The curvature Riemann20Riemann^{20} of every 4-dimensional para-Kähler structure (M,g,K)(M,g,K) can be decomposed as

Riemann20=Ric23(Scal1Weyl31)Weyl15,Riemann^{20}=Ric^{3}_{2}\oplus\Big{(}Scal^{1}\cong Weyl^{1}_{3}\Big{)}\oplus Weyl^{5}_{1},

which, compared to (2.19), means Ric13=Ric33=Weyl11=Weyl21=Weyl41=Weyl51=0Ric^{3}_{1}=Ric^{3}_{3}=Weyl^{1}_{1}=Weyl^{1}_{2}=Weyl^{1}_{4}=Weyl^{1}_{5}=0.

More explicitly, the curvature operator RiemannRiemann in (2.21), expressed in term of the basis of 2-forms (σ±i)(\sigma^{i}_{\pm}) in (2.2), is given by

Riemann=\displaystyle Riemann= Ψ2id6×6+([c|c|c||ccc]Ψ200002Ψ20P23P13P24P1400Ψ202P14Ψ22Ψ1Ψ00P13P240Ψ32Ψ2Ψ12P23Ψ42Ψ3Ψ2).\displaystyle-\Psi_{2}^{\prime}\operatorname{id}_{6\times 6}+\begin{pmatrix}[c|c|c||ccc]\Psi_{2}^{\prime}&0&0&&0&\\ \cline{1-6}\cr 0&-2\Psi_{2}^{\prime}&0&{\mbox{\sf P}}_{23}&{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24}&-{\mbox{\sf P}}_{14}\\ \cline{1-6}\cr 0&0&\Psi_{2}^{\prime}&&0&\\ \cline{1-6}\cr\cline{1-6}\cr&2{\mbox{\sf P}}_{14}&&\Psi_{2}&2\Psi_{1}&\Psi_{0}\\ 0&{\mbox{\sf P}}_{13}-{\mbox{\sf P}}_{24}&0&-\Psi_{3}&-2\Psi_{2}&-\Psi_{1}\\ &-2{\mbox{\sf P}}_{23}&&\Psi_{4}&2\Psi_{3}&\Psi_{2}\end{pmatrix}.
Proof.

The proof follows from a simple substitution of Γ4=10\Gamma^{4}{}_{1}=0 and Γ1=40\Gamma^{1}{}_{4}=0 into the last two of the second structure equations (2.10). ∎

Remark 2.11.

Note that the curvature conditions

P11=P12=P22=P33=P34=P44=Ψ0=Ψ1=Ψ3=Ψ4=Ψ2+112R=0,{\mbox{\sf P}}_{11}={\mbox{\sf P}}_{12}={\mbox{\sf P}}_{22}={\mbox{\sf P}}_{33}={\mbox{\sf P}}_{34}={\mbox{\sf P}}_{44}=\Psi_{0}^{\prime}=\Psi_{1}^{\prime}=\Psi_{3}^{\prime}=\Psi_{4}^{\prime}=\Psi_{2}^{\prime}+\tfrac{1}{12}R=0,

implied by the para-Kähler condition Γ4=1Γ1=40\Gamma^{4}{}_{1}=\Gamma^{1}{}_{4}=0, when inserted to the second structure equations (2.10)-(2.11), give that the entire curvature 12Raθcbcdθd\tfrac{1}{2}R^{a}{}_{bcd}\theta^{c}\wedge\theta^{d} of the para-Kähler structure is a 𝔤𝔩2()\mathfrak{gl}_{2}(\mathbb{R})-valued 2-form i.e. 12RaθcbcdθdΛ(1,1).\tfrac{1}{2}R^{a}{}_{bcd}\theta^{c}\wedge\theta^{d}\in\Lambda^{(1,1)}.

Recall that a 4-dimensional pseudo-Riemannian manifold (M,g)(M,g) where gg has split signature is called Einstein if and only if its traceless Ricci curvature vanishes, i.e. Ricci=0\overset{\circ}{Ricci}=0 in (2.9). Therefore, one obtains the following.

Corollary 2.12.

The curvature of a 4-dimensional para-Kähler-Einstein structure decomposes to

Riemann20=(Scal1Weyl31)Weyl15.Riemann^{20}=\Big{(}Scal^{1}\cong Weyl^{1}_{3}\Big{)}\oplus Weyl^{5}_{1}.

When written in an adapted coframe it reads

(2.28) Riemann=([c||c][c|c|c]00003Ψ2000000Ψ2Ψ22Ψ1Ψ0Ψ32Ψ2Ψ2Ψ1Ψ42Ψ3Ψ2Ψ2).\displaystyle Riemann=\begin{pmatrix}[c||c]\begin{matrix}[c|c|c]0&0&0\\ \cline{1-3}\cr 0&-3\Psi_{2}^{\prime}&0\\ \cline{1-3}\cr 0&0&0\end{matrix}&0\\ \cline{1-2}\cr\cline{1-2}\cr 0&\begin{matrix}\Psi_{2}-\Psi_{2}^{\prime}&2\Psi_{1}&\Psi_{0}\\ -\Psi_{3}&-2\Psi_{2}-\Psi_{2}^{\prime}&-\Psi_{1}\\ \Psi_{4}&2\Psi_{3}&\Psi_{2}-\Psi_{2}^{\prime}\end{matrix}\end{pmatrix}.
Remark 2.13.

It follows from § 2.3.2 that for para-Kähler-Einstein manifolds the two constant curvature components are related by R=12Ψ2.R=-12\Psi_{2}^{\prime}. From now on, we restrict ourselves to para-Kähler-Einstein 4-manifolds with non-vanishing Weyl+Weyl^{+} i.e. we always assume

Ψ2=const0.\Psi_{2}^{\prime}=\mathrm{const}\neq 0.

Moreover, following the discussion in § 2.2.4 on the Petrov type of the anti-self-dual Weyl curvature, Weyl=Weyl15,Weyl^{-}=Weyl^{5}_{1}, one obtains that the Petrov type of the quartic representation of Weyl+,Weyl^{+}, as the self-dual Weyl curvature of the metric g,g, is DD if Ψ20,\Psi^{\prime}_{2}\neq 0, and OO if Ψ2=0.\Psi_{2}^{\prime}=0.

2.3.3. Para-Kähler structure in a coordinate system

One of the features of Kähler metrics is that they can be locally expressed in terms of a function, called the Kähler potential. An analogous feature for the para-Kähler structures in 4 dimensions is described in the following two propositions.

Proposition 2.14.

Let 𝒰\mathcal{U} be an open set of 4\mathbb{R}^{4}, and let (a,b,x,y)(a,b,x,y) be Cartesian coordinates in 𝒰\mathcal{U}. Consider a real-valued sufficiently differentiable function V=V(a,b,x,y)V=V(a,b,x,y) on 𝒰\mathcal{U} such that

det(VaxVayVbxVby)0in𝒰.\det\begin{pmatrix}V_{ax}&V_{ay}\\ V_{bx}&V_{by}\end{pmatrix}\neq 0\quad\mathrm{in}\quad{\mathcal{U}}.

Define

g=\displaystyle g= 2da(Vaxdx+Vaydy)+2db(Vbxdx+Vbydy),\displaystyle~{}2{\rm d}a~{}(V_{ax}{\rm d}x+V_{ay}{\rm d}y)+2{\rm d}b~{}(V_{bx}{\rm d}x+V_{by}{\rm d}y),
K=\displaystyle K= ada+bdbxdxydy,\displaystyle~{}\partial_{a}\otimes{\rm d}a+\partial_{b}\otimes{\rm d}b-\partial_{x}\otimes{\rm d}x-\partial_{y}\otimes{\rm d}y,
ρ=\displaystyle\rho= da(Vaxdx+Vaydy)+db(Vbxdx+Vbydy).\displaystyle~{}{\rm d}a\wedge(V_{ax}{\rm d}x+V_{ay}{\rm d}y)+{\rm d}b\wedge(V_{bx}{\rm d}x+V_{by}{\rm d}y).

Then the pair (g,K)(g,K) defines a para-Kähler structure on 𝒰\mathcal{U} with ρ(,)=g(K(),)\rho(\cdot,\cdot)=g(K(\cdot),\cdot).

The para-Kähler structure (𝒰,g,K)({\mathcal{U}},g,K) is Einstein i.e. Ric(g)=Λg,Ric(g)=\Lambda g, if and only if the potential function VV satisfies

(2.29) det(VaxVayVbxVby)=c1c2eΛV\det\begin{pmatrix}V_{ax}&V_{ay}\\ V_{bx}&V_{by}\end{pmatrix}=c_{1}c_{2}~{}\mathrm{e}^{-\Lambda V}

for a real number Λ\Lambda and real-valued functions c1=c1(a,b)c_{1}=c_{1}(a,b), c2=c2(x,y).c_{2}=c_{2}(x,y).

Proof.

In the adapted coframe

α1\displaystyle\alpha^{1} =da,\displaystyle={\rm d}a,\qquad α2\displaystyle\alpha^{2} =db\displaystyle={\rm d}b
α¯1\displaystyle\bar{\alpha}{}^{1} =Vaxdx+Vaydy,\displaystyle=V_{ax}{\rm d}x+V_{ay}{\rm d}y,\qquad α¯2\displaystyle\bar{\alpha}{}^{2} =Vbxdx+Vbydy,\displaystyle=V_{bx}{\rm d}x+V_{by}{\rm d}y,

the 1-forms Γba\Gamma^{a}_{~{}b}, constituting the 𝔤𝔩2()\mathfrak{gl}_{2}(\mathbb{R}) part of the Levi-Civita connection, read

Γ1=1\displaystyle\Gamma^{1}{}_{1}= VaayVbxVaaxVbyVayVbxVaxVbyα1+VabyVbxVabxVbyVayVbxVaxVbyα2\displaystyle\frac{V_{aay}V_{bx}-V_{aax}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}\alpha^{1}+\frac{V_{aby}V_{bx}-V_{abx}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}\alpha^{2}
Γ1=2\displaystyle\Gamma^{1}{}_{2}= VabyVbxVabxVbyVayVbxVaxVbyα1+VbbyVbxVbbxVbyVayVbxVaxVbyα2\displaystyle\frac{V_{aby}V_{bx}-V_{abx}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}\alpha^{1}+\frac{V_{bby}V_{bx}-V_{bbx}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}\alpha^{2}
Γ2=1\displaystyle\Gamma^{2}{}_{1}= VaayVaxVaaxVayVayVbx+VaxVbyα1+VabyVaxVabxVayVayVbx+VaxVbyα2\displaystyle\frac{V_{aay}V_{ax}-V_{aax}V_{ay}}{-V_{ay}V_{bx}+V_{ax}V_{by}}\alpha^{1}+\frac{V_{aby}V_{ax}-V_{abx}V_{ay}}{-V_{ay}V_{bx}+V_{ax}V_{by}}\alpha^{2}
Γ2=2\displaystyle\Gamma^{2}{}_{2}= VabyVaxVabxVayVayVbx+VaxVbyα1+VbbyVaxVbbxVayVayVbx+VaxVbyα2.\displaystyle\frac{V_{aby}V_{ax}-V_{abx}V_{ay}}{-V_{ay}V_{bx}+V_{ax}V_{by}}\alpha^{1}+\frac{V_{bby}V_{ax}-V_{bbx}V_{ay}}{-V_{ay}V_{bx}+V_{ax}V_{by}}\alpha^{2}.

It is straightforward to check that dρ=0{\rm d}\rho=0, and Γ1=4Γ4=10\Gamma^{1}{}_{4}=\Gamma^{4}{}_{1}=0, as it should be for the Levi-Civita connection in an adapted coframe of a para-Kähler structure.

For the calculation of the Ricci tensor it is more convenient to work in the coordinate frame (da,db,dx,dy)({\rm d}a,{\rm d}b,{\rm d}x,{\rm d}y) rather than in the adapted frame (α1,α2,α¯,1α¯)2(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}). Thus, we need to display the Levi-Civita connection 1-forms in the coordinate frame as well. Let us use the following notation for the coordinates

xA=(a,b),xA˙=(x,y),A=1,2,A˙=1,2.x^{A}=(a,b),\quad x^{\dot{A}}=(x,y),\quad A=1,2,\quad\dot{A}=1,2.

It follows that

ΓA=A˙ΓA˙=A0,\Gamma^{A}{}_{\dot{A}}=\Gamma^{\dot{A}}{}_{A}=0,

and the connection 1-forms ΓAB\Gamma^{A}{}_{B} and ΓA˙B˙\Gamma^{\dot{A}}{}_{\dot{B}} are given by (A.1) in the Appendix.

Using the expressions for the Levi-Civita connection in (A.1) the curvature can be calculated easily and the Ricci tensor satisfies

RAB=RA˙B˙=0,R_{AB}=R_{\dot{A}\dot{B}}=0,

and

RAA˙=RA˙A=2xAxA˙log(VaxVbyVayVbx).R_{A\dot{A}}=R_{\dot{A}A}=-\frac{\partial^{2}}{\partial_{x^{A}}\partial_{x^{\dot{A}}}}\log\big{(}V_{ax}V_{by}-V_{ay}V_{bx}\big{)}.

Since in the (A,A˙)(A,\dot{A}) notation the metric gg reads as

g=A,A˙=1,22VxAxA˙(dxAdxA˙+dxA˙dxA),g=\sum_{A,\dot{A}=1,2}\frac{\partial^{2}V}{\partial_{x^{A}}\partial_{x^{\dot{A}}}}({\rm d}x^{A}\otimes{\rm d}x^{\dot{A}}+{\rm d}x^{\dot{A}}\otimes{\rm d}x^{A}),

the Einstein equations are

2xAxA˙log(VaxVbyVayVbx)=Λ2VxAxA˙,-\frac{\partial^{2}}{\partial_{x^{A}}\partial_{x^{\dot{A}}}}\log\big{(}V_{ax}V_{by}-V_{ay}V_{bx}\big{)}=\Lambda~{}\frac{\partial^{2}V}{\partial_{x^{A}}\partial_{x^{\dot{A}}}},

which after integration give

VaxVbyVayVbx=c1c2eΛV.V_{ax}V_{by}-V_{ay}V_{bx}=c_{1}c_{2}~{}\mathrm{e}^{-\Lambda V}.

There is a converse to this proposition:

Proposition 2.15.

Every para-Kähler structure (M,g,K)(M,g,K) in dimension four is locally expressible in terms of a para-Kähler potential function VV as in Proposition 2.14.

Proof.

The integrability of the distributions =Ker{α1,α2}\mathscr{H}=\mathrm{Ker}\{\alpha^{1},\alpha^{2}\} and ¯=Ker{α¯,1α¯}2\bar{\mathscr{H}}=\mathrm{Ker}\{\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}\} in a para-Kähler structure implies that in some neighbourhood 𝒰M{\mathcal{U}}\subset M there exists a coordinate system (a,b,x,y)(a,b,x,y) such that

[ll]α1=A1da+B1db,α2=A2da+B2dbα¯=1P1dx+Q1dy,α¯=2P2dx+Q2dy,\begin{matrix}[ll]\alpha^{1}=A_{1}{\rm d}a+B_{1}{\rm d}b,\quad&\alpha^{2}=A_{2}{\rm d}a+B_{2}{\rm d}b\\ \bar{\alpha}{}^{1}=P_{1}{\rm d}x+Q_{1}{\rm d}y,\quad&\bar{\alpha}{}^{2}=P_{2}{\rm d}x+Q_{2}{\rm d}y,\end{matrix}

for some functions Ai,Bi,Pi,QiA_{i},B_{i},P_{i},Q_{i} defined in 𝒰\mathcal{U}. Since the coframe (α1,α2,α¯,1α¯)2(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}) is defined up to the 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}) action (2.17), we can use this transformation to bring the coframe into the form

[ll]α1=da,α2=dbα¯=1Pdx+Qdy,α¯=2Rdx+Sdy,\begin{matrix}[ll]\alpha^{1}={\rm d}a,\quad&\alpha^{2}={\rm d}b\\ \bar{\alpha}{}^{1}=P{\rm d}x+Q{\rm d}y,\quad&\bar{\alpha}{}^{2}=R{\rm d}x+S{\rm d}y,\end{matrix}

with new functions P,Q,R,SP,Q,R,S on 𝒰\mathcal{U} such that PSQR0PS-QR\neq 0. In this new adapted frame we have dα1=0{\rm d}\alpha^{1}=0, dα2=0{\rm d}\alpha^{2}=0. Inserting this into (2.8) with Γ41=Γ14=0,\Gamma^{1}_{4}=\Gamma^{4}_{1}=0, we get

Γ11α1+Γ12α2=0&Γ21α1+Γ22α2=0,\Gamma^{1}{}_{1}\wedge\alpha^{1}+\Gamma^{1}{}_{2}\wedge\alpha^{2}=0\quad\&\quad\Gamma^{2}{}_{1}\wedge\alpha^{1}+\Gamma^{2}{}_{2}\wedge\alpha^{2}=0,

which implies

Γ11\displaystyle\Gamma^{1}{}_{1} =a1α1+a2α2,Γ1=2a3α1+a4α2\displaystyle=a_{1}\alpha^{1}+a_{2}\alpha^{2},\qquad\Gamma^{1}{}_{2}=a_{3}\alpha^{1}+a_{4}\alpha^{2}
Γ21\displaystyle\Gamma^{2}{}_{1} =a5α1+a6α2,Γ2=2a7α1+a8α2,\displaystyle=a_{5}\alpha^{1}+a_{6}\alpha^{2},\qquad\Gamma^{2}{}_{2}=a_{7}\alpha^{1}+a_{8}\alpha^{2},

for some unknown functions a1,a2,,a8a_{1},a_{2},\dots,a_{8} on 𝒰\mathcal{U}. Inserting this back into the last two of the structure equations (2.8) gives

0=\displaystyle 0= (QxPy)dxdy+\displaystyle(Q_{x}-P_{y}){\rm d}x\wedge{\rm d}y+
(Pba2Pa6R)dbdx+(Qba2Qa6S)dbdy+\displaystyle\big{(}P_{b}-a_{2}P-a_{6}R\big{)}{\rm d}b\wedge{\rm d}x+\big{(}Q_{b}-a_{2}Q-a_{6}S\big{)}{\rm d}b\wedge{\rm d}y+
(Paa1Pa5R)dadx+(Qaa1Qa5S)dady\displaystyle\big{(}P_{a}-a_{1}P-a_{5}R\big{)}{\rm d}a\wedge{\rm d}x+\big{(}Q_{a}-a_{1}Q-a_{5}S\big{)}{\rm d}a\wedge{\rm d}y
0=\displaystyle 0= (SxRy)dxdy+\displaystyle(S_{x}-R_{y}){\rm d}x\wedge{\rm d}y+
(Rba8Ra4P)dbdx+(Sba8Sa4Q)dbdy+\displaystyle\big{(}R_{b}-a_{8}R-a_{4}P\big{)}{\rm d}b\wedge{\rm d}x+\big{(}S_{b}-a_{8}S-a_{4}Q\big{)}{\rm d}b\wedge{\rm d}y+
(Raa7Ra3P)dadx+(Saa7Sa3Q)dady.\displaystyle\big{(}R_{a}-a_{7}R-a_{3}P\big{)}{\rm d}a\wedge{\rm d}x+\big{(}S_{a}-a_{7}S-a_{3}Q\big{)}{\rm d}a\wedge{\rm d}y.

This in particular means that

(QxPy)=0and(SxRy)=0.(Q_{x}-P_{y})=0\quad\mathrm{and}\quad(S_{x}-R_{y})=0.

As a result, locally, there exist functions UU and WW on 𝒰\mathcal{U} such that

Q=Uy,P=Ux,S=Wy,R=Wx.Q=U_{y},\quad P=U_{x},\quad S=W_{y},\quad R=W_{x}.

Thus, one obtains

[ll]α1=da,α2=dbα¯=1Uxdx+Uydy,α¯=2Wxdx+Wydy.\begin{matrix}[ll]\alpha^{1}={\rm d}a,\quad&\alpha^{2}={\rm d}b\\ \bar{\alpha}{}^{1}=U_{x}{\rm d}x+U_{y}{\rm d}y,\quad&\bar{\alpha}{}^{2}=W_{x}{\rm d}x+W_{y}{\rm d}y.\end{matrix}

Since the 2-form ρ\rho is given by

ρ=α1α¯+1α2α¯,2\rho=\alpha^{1}\wedge\bar{\alpha}{}^{1}+\alpha^{2}\wedge\bar{\alpha}{}^{2},

the para-Kähler condition dρ=0{\rm d}\rho=0 implies

0=dρ=(WaUb)ydadbdy+(WaUb)xdadbdx.0={\rm d}\rho=(W_{a}-U_{b})_{y}{\rm d}a\wedge{\rm d}b\wedge{\rm d}y+(W_{a}-U_{b})_{x}{\rm d}a\wedge{\rm d}b\wedge{\rm d}x.

This means that WaUb=f(a,b)W_{a}-U_{b}=f(a,b) for some function ff of variables a,ba,b only. But since in the coframe (α1,α2,α¯,1α¯)2(\alpha^{1},\alpha^{2},\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}) functions WW and UU appear only in terms of their xx and yy derivatives, they can be chosen so that f(a,b)0.f(a,b)\equiv 0. Hence there exists a differentiable function V=V(a,b,x,y)V=V(a,b,x,y) on 𝒰\mathcal{U} such that

W=VbandU=Va.W=V_{b}\quad\mathrm{and}\quad U=V_{a}.

Thus, the adapted coframe can be expressed as

[ll]α1=da,α2=db,α¯=1Vaxdx+Vaydy,α¯=2Vbxdx+Vbydy.\begin{matrix}[ll]\alpha^{1}={\rm d}a,\quad&\alpha^{2}={\rm d}b,\\ \bar{\alpha}{}^{1}=V_{ax}{\rm d}x+V_{ay}{\rm d}y,\quad&\bar{\alpha}{}^{2}=V_{bx}{\rm d}x+V_{by}{\rm d}y.\end{matrix}

Expressing gg, KK and ρ\rho in terms of the adapted coframe as in (2.14)-(2.15) gives the Proposition. ∎

2.3.4. Homogeneous models

Two particular solutions of the Einstein condition (2.29) are given by the potentials

  • i)

    V1=1Ψ2log(b+axy)V_{1}=-\frac{1}{\Psi_{2}^{\prime}}\log(b+ax-y),

  • ii)

    V2=23Ψ2log((132Ψ2ax)(132Ψ2by))V_{2}=-\frac{2}{3\Psi_{2}^{\prime}}\log\big{(}(1-\tfrac{3}{2}\Psi_{2}^{\prime}ax)(1-\tfrac{3}{2}\Psi_{2}^{\prime}by)\big{)},

where Ψ2=const0\Psi_{2}^{\prime}=\mathrm{const}\neq 0.

Both potentials are solutions of (2.29) with Λ=3Ψ2\Lambda=-3\Psi_{2}^{\prime} and c1c2=1Ψ22c_{1}c_{2}=\tfrac{1}{\Psi_{2}^{\prime}{}^{2}} for V1V_{1}, and c1c2=1c_{1}c_{2}=1 for V2V_{2}. Let the para-Kähler structure (gi,Ki,ρi)(g_{i},K_{i},\rho_{i}) correspond to ViV_{i} where i=1,2.i=1,2. Then, in an open set of 4\mathbb{R}^{4} parametrized by (a,b,x,y)(a,b,x,y) one finds

K1=K2=ada+bdbxdxydy.K_{1}=K_{2}=\partial_{a}\otimes{\rm d}a+\partial_{b}\otimes{\rm d}b-\partial_{x}\otimes{\rm d}x-\partial_{y}\otimes{\rm d}y.

Straightforward computation gives

g1=\displaystyle g_{1}= 2da((yb)dxxdy)+2db(adxdy)Ψ2(b+axy)2,ρ1=\displaystyle\textstyle{\frac{2{\rm d}a\big{(}(y-b){\rm d}x-x{\rm d}y\big{)}+2{\rm d}b\big{(}a{\rm d}x-{\rm d}y\big{)}}{\Psi_{2}^{\prime}(b+ax-y)^{2}}},\qquad\rho_{1}= da((yb)dxxdy)+db(adxdy)Ψ2(b+axy)2\displaystyle\textstyle{\frac{{\rm d}a\wedge\big{(}(y-b){\rm d}x-x{\rm d}y\big{)}+{\rm d}b\wedge\big{(}a{\rm d}x-{\rm d}y\big{)}}{\Psi_{2}^{\prime}(b+ax-y)^{2}}}

and

g2=\displaystyle g_{2}= 2dadx(132Ψ2ax)2+2dbdy(132Ψ2bx)2,ρ2=\displaystyle\textstyle{\frac{2{\rm d}a{\rm d}x}{(1-\tfrac{3}{2}\Psi_{2}^{\prime}ax)^{2}}+\frac{2{\rm d}b{\rm d}y}{(1-\tfrac{3}{2}\Psi_{2}^{\prime}bx)^{2}}},\qquad\rho_{2}= dadx(132Ψ2ax)2+dbdy(132Ψ2by)2.\displaystyle\textstyle{\frac{{\rm d}a\wedge{\rm d}x}{(1-\tfrac{3}{2}\Psi_{2}^{\prime}ax)^{2}}+\frac{{\rm d}b\wedge{\rm d}y}{(1-\tfrac{3}{2}\Psi_{2}^{\prime}by)^{2}}}.

The potential V1V_{1} corresponds to the homogeneous para-Kähler-Einstein structure referred to as the dancing metric in [BHLN18] which is the unique homogeneous model that is self-dual, i.e. Weyl=0,Weyl^{-}=0, and not anti-self-dual for which Ψ2=1.\Psi^{\prime}_{2}=1. The potential V2V_{2} corresponds to the only other homogeneous para-Kähler-Einstein structure. It has the property that the Petrov type of WeylWeyl^{-} is D. A derivation of these potential functions is outlined in § 3.2.2 and § 3.2.6. Finding explicit examples of pKE structures in terms of potential functions satisfying the PDE (2.29) is not an easy task. In the next section we use an alternative technique to give more explicit examples of pKE structures.

3. Para-Kähler-Einstein (pKE) metrics in dimension 4

This section is the heart of the article, in which we describe pKE structures as Cartan geometries, give an in-depth study when the Petrov type is real and special and provide explicit examples. To be more specific, in § 3.1 pKE structures are interpreted as Cartan geometries of type (𝐒𝐋3(),𝐆𝐋2())(\mathbf{SL}_{3}(\mathbb{R}),\mathbf{GL}_{2}(\mathbb{R})). If the Einstein constant is 3,-3, then they satisfy the Yang-Mills equations for the associated 𝔰𝔩3()\mathfrak{sl}_{3}(\mathbb{R})-valued Cartan connection. In § 3.2 we focus on pKE structures for which WeylWeyl^{-} has special real Petrov type and give examples of each type. In particular, we find all homogeneous models, give a local normal form for all real Petrov type DD pKE metrics and present examples of real Petrov type II that satisfy Yang-Mills equations. Moreover, we use Cartan-Kähler machinery to find the local generality of all Petrov types assuming analyticity.

3.1. Cartan geometries of type (𝐒𝐋3(),𝐆𝐋2())(\mathbf{SL}_{3}(\mathbb{R}),\mathbf{GL}_{2}(\mathbb{R}))

In order to view pKE structures as a Cartan geometry, let us specialize the EDS (2.8), (2.10)-(2.11) to the case of para-Kähler-Einstein metrics. We have

(3.1) dα1\displaystyle{\rm d}\alpha^{1} =Γ11α1Γ21α2,\displaystyle=-\Gamma^{1}_{~{}1}\wedge\alpha^{1}-\Gamma^{1}_{~{}2}\wedge\alpha^{2},
dα2\displaystyle{\rm d}\alpha^{2} =Γ12α1Γ22α2,\displaystyle=-\Gamma^{2}_{~{}1}\wedge\alpha^{1}-\Gamma^{2}_{~{}2}\wedge\alpha^{2},
dα¯1\displaystyle{\rm d}\bar{\alpha}{}^{1} =Γ11α¯+1Γ12α¯,2\displaystyle=\Gamma^{1}_{~{}1}\wedge\bar{\alpha}{}^{1}+\Gamma^{2}_{~{}1}\wedge\bar{\alpha}{}^{2},
dα¯2\displaystyle{\rm d}\bar{\alpha}{}^{2} =Γ21α¯+1Γ22α¯,2\displaystyle=\Gamma^{1}_{~{}2}\wedge\bar{\alpha}{}^{1}+\Gamma^{2}_{~{}2}\wedge\bar{\alpha}{}^{2},
dΓ11\displaystyle\mathrm{d}\Gamma^{1}_{~{}1} =Γ21Γ12+(2Ψ2Ψ2)α1α¯+1Ψ1α1α¯2Ψ3α2α¯+1(Ψ2Ψ2)α2α¯2\displaystyle=-\Gamma^{1}_{~{}2}{\,{\wedge}\;}\Gamma^{2}_{~{}1}+(-2\Psi^{\prime}_{2}-\Psi_{2})\alpha^{1}{\,{\wedge}\;}\bar{\alpha}{}^{1}+\Psi_{1}\alpha^{1}{\,{\wedge}\;}\bar{\alpha}{}^{2}-\Psi_{3}\alpha^{2}{\,{\wedge}\;}\bar{\alpha}{}^{1}+(\Psi_{2}-\Psi^{\prime}_{2})\alpha^{2}{\,{\wedge}\;}\bar{\alpha}{}^{2}
dΓ21\displaystyle\mathrm{d}\Gamma^{1}_{~{}2} =Γ11Γ21Γ21Γ22Ψ3α1α¯+1(Ψ2Ψ2)α1α¯2Ψ4α2α¯+1Ψ3α2α¯2\displaystyle=-\Gamma^{1}_{~{}1}{\,{\wedge}\;}\Gamma^{1}_{~{}2}-\Gamma^{1}_{~{}2}{\,{\wedge}\;}\Gamma^{2}_{~{}2}-\Psi_{3}\alpha^{1}{\,{\wedge}\;}\bar{\alpha}{}^{1}+(\Psi_{2}-\Psi^{\prime}_{2})\alpha^{1}{\,{\wedge}\;}\bar{\alpha}{}^{2}-\Psi_{4}\alpha^{2}{\,{\wedge}\;}\bar{\alpha}{}^{1}+\Psi_{3}\alpha^{2}{\,{\wedge}\;}\bar{\alpha}{}^{2}
dΓ12\displaystyle\mathrm{d}\Gamma^{2}_{~{}1} =Γ11Γ12+Γ12Γ22+Ψ1α1α¯1Ψ0α1α¯+2(Ψ2Ψ2)α2α¯1Ψ1α2α¯2\displaystyle=\Gamma^{1}_{~{}1}{\,{\wedge}\;}\Gamma^{2}_{~{}1}+\Gamma^{2}_{~{}1}{\,{\wedge}\;}\Gamma^{2}_{~{}2}+\Psi_{1}\alpha^{1}{\,{\wedge}\;}\bar{\alpha}{}^{1}-\Psi_{0}\alpha^{1}{\,{\wedge}\;}\bar{\alpha}{}^{2}+(\Psi_{2}-\Psi^{\prime}_{2})\alpha^{2}{\,{\wedge}\;}\bar{\alpha}{}^{1}-\Psi_{1}\alpha^{2}{\,{\wedge}\;}\bar{\alpha}{}^{2}
dΓ22\displaystyle\mathrm{d}\Gamma^{2}_{~{}2} =Γ21Γ12+(Ψ2Ψ2)α1α¯1Ψ1α1α¯+2Ψ3α2α¯+1(2Ψ2Ψ2)α2α¯2\displaystyle=\Gamma^{1}_{~{}2}{\,{\wedge}\;}\Gamma^{2}_{1}+(\Psi_{2}-\Psi^{\prime}_{2})\alpha^{1}{\,{\wedge}\;}\bar{\alpha}{}^{1}-\Psi_{1}\alpha^{1}{\,{\wedge}\;}\bar{\alpha}{}^{2}+\Psi_{3}\alpha^{2}{\,{\wedge}\;}\bar{\alpha}{}^{1}+(-2\Psi^{\prime}_{2}-\Psi_{2})\alpha^{2}{\,{\wedge}\;}\bar{\alpha}{}^{2}

where we have used Γ41=Γ14=0\Gamma^{1}_{~{}4}=\Gamma^{4}_{~{}1}=0 from Proposition 2.7

As discussed in Remark 2.9 and 2.1, the EDS (3.1) can be regarded as the structure equations for the coframe on the principal 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R})-bundle 8M\mathcal{F}^{8}\to M which is the 8-dimensional bundle of adapted null frames for para-Kähler-Einstein structures. In fact, one can show that para-Kähler-Einstein structures correspond to Cartan geometries of type (𝐒𝐋3(),𝐆𝐋2())(\mathbf{SL}_{3}(\mathbb{R}),\mathbf{GL}_{2}(\mathbb{R})). First let us define a Cartan geometry [Sha97, ČS09].

Definition 3.1.

A Cartan geometry (𝒢,S,ψ),(\mathcal{G},S,\psi), of type (G,H)(G,H) is a principal HH-bundle 𝒢S,\mathcal{G}\to S, equipped with a 𝔤\mathfrak{g}-valued 1-form 𝒜,\mathcal{A}, which is a Cartan connection, i.e.,

  1. (1)

    𝒜u:Tu𝒢𝔤\mathcal{A}_{u}:T_{u}\mathcal{G}\rightarrow\mathfrak{g} is linear isomorphism for all u𝒢.u\in\mathcal{G}.

  2. (2)

    𝒜\mathcal{A} is HH-equivariant, i.e., Rh𝒜=Ad(h1)𝒜,R_{h}^{*}\mathcal{A}=\mathrm{Ad}(h^{-1})\circ\mathcal{A}, where RhR_{h} denotes the right action by hH.h\in H.

  3. (3)

    𝒜(Xv)=v,\mathcal{A}(X_{v})=v, for every fundamental vector field XvX_{v} of τ:𝒢S,v𝔥.\tau:\mathcal{G}\rightarrow S,v\in\mathfrak{h}.

The curvature of the Cartan connection 𝒜\mathcal{A} is given by K𝒜=d𝒜+𝒜𝒜Ω2(𝒢,𝔤)K_{\mathcal{A}}=\mathrm{d}\mathcal{A}+\mathcal{A}{\,{\wedge}\;}\mathcal{A}\in\Omega^{2}(\mathcal{G},\mathfrak{g}) which is horizontal and defines the curvature function κ𝒜:𝒢2(𝔤/𝔥)𝔤.\kappa_{\mathcal{A}}:\mathcal{G}\to\bigwedge^{2}(\mathfrak{g}/\mathfrak{h})^{*}\otimes\mathfrak{g}.

Let us now consider the 𝔰𝔩3()\mathfrak{sl}_{3}(\mathbb{R})-valued 1-form

(3.2) 𝒜:=([c|c]Γ13Tr(Γ)id2×2αα¯13Tr(Γ)),\mathcal{A}:=\begin{pmatrix}[c|c]\Gamma-\tfrac{1}{3}\mathrm{Tr}(\Gamma)\operatorname{id}_{2\times 2}&\alpha\\ \cline{1-2}\cr\bar{\alpha}{}&-\tfrac{1}{3}\mathrm{Tr}(\Gamma)\end{pmatrix},

where

(3.3) Γ=(Γ11Γ12Γ21Γ22),α=(α1α2),α¯=(α¯,1α¯)2,\Gamma=\begin{pmatrix}\Gamma^{1}{}_{1}&\Gamma^{1}{}_{2}\\ \Gamma^{2}{}_{1}&\Gamma^{2}{}_{2}\end{pmatrix},\qquad\alpha=\begin{pmatrix}\alpha^{1}\\ \alpha^{2}\end{pmatrix},\qquad\bar{\alpha}=(\bar{\alpha}{}^{1},\bar{\alpha}{}^{2}),

Using 𝒜\mathcal{A}, the structure equations (3.1) can be expressed as

(3.4) K𝒜=d𝒜+𝒜𝒜,K_{\mathcal{A}}={\rm d}\mathcal{A}+\mathcal{A}\wedge\mathcal{A},

where, using the 2-forms σ+i\sigma_{+}^{i}’s and σi\sigma_{-}^{i}’s in (2.2), one has

(3.5) K𝒜=\displaystyle K_{\mathcal{A}}= ([c|c]Ψ11+Ψ2Ψ2Ψ0Ψ1000)σ1+([c|c]Ψ3Ψ41Ψ2+Ψ2Ψ3000)σ2+\displaystyle\begin{pmatrix}[c|c]\begin{matrix}\Psi_{1}&1+\Psi_{2}-\Psi_{2}^{\prime}\\ -\Psi_{0}&-\Psi_{1}\end{matrix}&0\\ \cline{1-2}\cr 0&0\end{pmatrix}\sigma^{1}_{-}+\begin{pmatrix}[c|c]\begin{matrix}\Psi_{3}&\Psi_{4}\\ -1-\Psi_{2}+\Psi_{2}^{\prime}&-\Psi_{3}\end{matrix}&0\\ \cline{1-2}\cr 0&0\end{pmatrix}\sigma^{2}_{-}+
([c|c]12(12Ψ2Ψ2)Ψ3Ψ112(Ψ2+2Ψ21)000)σ3+\displaystyle\begin{pmatrix}[c|c]\begin{matrix}\tfrac{1}{2}(1-2\Psi_{2}-\Psi_{2}^{\prime})&-\Psi_{3}\\ \Psi_{1}&\tfrac{1}{2}(\Psi_{2}^{\prime}+2\Psi_{2}-1)\end{matrix}&0\\ \cline{1-2}\cr 0&0\end{pmatrix}\sigma^{3}_{-}+
12(1Ψ2)([c|c]id2×2002)σ+3.\displaystyle\tfrac{1}{2}(1-\Psi_{2}^{\prime})\begin{pmatrix}[c|c]\operatorname{id}_{2\times 2}&0\\ \cline{1-2}\cr 0&-2\end{pmatrix}\sigma^{3}_{+}.

One can interpret the 1-form 𝒜\mathcal{A} as an 𝔰𝔩3()\mathfrak{sl}_{3}(\mathbb{R})-valued Cartan connection on the principal 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R})-bundle 8M\mathcal{F}^{8}\to M of null frames adapted to a pKE structure. As a result, one obtains the following theorem.

Theorem 3.2.

Every pKE 4-manifold defines a Cartan geometry of type (𝐒𝐋3(),𝐆𝐋2())(\mathbf{SL}_{3}(\mathbb{R}),\mathbf{GL}_{2}(\mathbb{R})) for which the structure bundle

𝐆𝐋2()8M\mathbf{GL}_{2}(\mathbb{R})\to\mathcal{F}^{8}\to M

is the bundle of adapted null frames. The curvature K𝒜K_{\mathcal{A}} vanishes, i.e. the Cartan geometry is flat, if and only if the Einstein constant is -3 and the anti-self-dual Weyl tensor vanishes, i.e.

Ψ2=1andWeyl=0.\Psi^{\prime}_{2}=1\qquad\mathrm{and}\qquad Weyl^{-}=0.

The flat model, i.e. K𝒜=0,K_{\mathcal{A}}=0, is locally equivalent to the para-Kähler-Einstein structure induced by the dancing metric discussed in § 2.3.4.

Recall that the exterior derivative of (3.4) gives the Bianchi identity

(3.6) DK𝒜:=dK𝒜+𝒜K𝒜K𝒜𝒜=0.DK_{\mathcal{A}}:={\rm d}K_{\mathcal{A}}+\mathcal{A}\wedge K_{\mathcal{A}}-K_{\mathcal{A}}\wedge\mathcal{A}=0.

As a result of the theorem above one obtains the following.

Proposition 3.3.

A 4-dimensional para-Kähler-Einstein structure satisfies the Cartan connection Yang-Mills equations, DK𝒜=0D*K_{\mathcal{A}}=0, if and only if Ψ2=1.\Psi_{2}^{\prime}=1.

Proof.

By (3.5) and the definition of self-dual and anti-self-dual null planes in (2.2), it follows that

Ψ2=1K𝒜=K𝒜.\Psi^{\prime}_{2}=1\Longleftrightarrow*K_{\mathcal{A}}=-K_{\mathcal{A}}.

Since the curvature K𝒜K_{\mathcal{A}} of a Cartan connection is always horizontal (see Definition 3.1 and (3.5)), one can apply the Hodge star to K𝒜K_{\mathcal{A}} defined on 8{\mathcal{F}}^{8}. Applying the Bianchi identity (3.6), one obtains DK𝒜=DK𝒜=0.D*K_{\mathcal{A}}=-DK_{\mathcal{A}}=0. Alternatively, by taking a section s:M8,s\colon M\to\mathcal{F}^{8}, computing the curvature and applying the Hodge star one can verify the claim.

Conversely, it is a matter of straightforward computation to show that DK𝒜=0D*K_{\mathcal{A}}=0 combined with the Bianchi identities (3.6), or equivalently equations (A.2), and the EDS (3.1) imply Ψ2=1.\Psi^{\prime}_{2}=1.

Remark 3.4.

Note that pKE structures can also be associated to Cartan geometries of type (4𝐆𝐋2(),𝐆𝐋2())(\mathbb{R}^{4}\rtimes\mathbf{GL}_{2}(\mathbb{R}),\mathbf{GL}_{2}(\mathbb{R})) whose flat model satisfies Ψ2=0\Psi^{\prime}_{2}=0 and Weyl=0Weyl^{-}=0. This point of view is however not desirable for the purpose of this article, since we always assume Ψ20.\Psi_{2}^{\prime}\neq 0.

3.2. Cartan reduction: homogeneous models, examples and local generality

In this section we carry out the Cartan reduction procedure for pKE metrics whose anti-self-dual Weyl curvature has non-generic real Petrov type. Our reduction will not be exhaustive and will omit Petrov type GG. We will describe the reduction procedure for Petrov type IIII and DD in detail, give a complete local normal form for type DD pKE metrics, and use the same method to find examples of Petrov types IIIIII and NN. The reduction allows us to find all homogeneous models and use the Cartan-Kähler theory to find the local generality of all Petrov types assuming analyticity.

3.2.1. Reduction for special real Petrov types

Recall from (2.24) the quartic

(3.7) W(λ)=Ψ4λ4+4Ψ3λ3+6Ψ2λ2+4Ψ1λ+Ψ0W(\lambda)=\Psi_{4}\lambda^{4}+4\Psi_{3}\lambda^{3}+6\Psi_{2}\lambda^{2}+4\Psi_{1}\lambda+\Psi_{0}

where λ\lambda is the affine parameter for the anti-self-dual null planes in (2.3) for a given choice of adapted coframe. Taking a different choice of coframe, θ1~,,θ~4,\tilde{\theta_{1}},\dots,\tilde{\theta}_{4}, by the action of the structure group, as in (2.17), one obtains a quartic whose coefficients,Ψ~0,,Ψ~4,\tilde{\Psi}_{0},\dots,\tilde{\Psi}_{4}, can be expressed in terms of Ψi\Psi_{i}’s and the elements a11,a12,a21,a22a_{11},a_{12},a_{21},a_{22} of the matrix A𝐆𝐋2().A\in\mathbf{GL}_{2}(\mathbb{R}). For instance, one obtains

(3.8) Ψ~0=1det(A)(a214Ψ44a213a22Ψ3+6a212a222Ψ24a21a223Ψ1+a224Ψ0).\tilde{\Psi}_{0}=\textstyle{\frac{1}{\mathrm{det}(A)}(a^{4}_{21}\Psi_{4}-4a_{21}^{3}a_{22}\Psi_{3}+6a_{21}^{2}a_{22}^{2}\Psi_{2}-4a_{21}a_{22}^{3}\Psi_{1}+a_{22}^{4}\Psi_{0})}.
Remark 3.5.

Note that the infinitesimal form of the transformation law for Ψ0\Psi_{0} given in (3.8) is represented by the Bianchi identity for Ψ0\Psi_{0} in (A.2b). For a discussion on obtaining the group action from its infinitesimal see [Gar89].

It is clear from (3.8) that if W(λ0)=0W(\lambda_{0})=0 then with respect to the coframe obtained from the action of (2.16) where a11=a22=1,a12=0,a_{11}=a_{22}=1,a_{12}=0, and a21=λ0a_{21}=-\lambda_{0}, the root λ0\lambda_{0} would be translated to zero, i.e. Ψ~0=0\tilde{\Psi}_{0}=0 in this choice of coframe.

If WeylWeyl^{-} has a repeated root then by our discussion above, there is a coframe adaptation with respect to which the double root is translated to zero. If the root has multiplicity k4k\leq 4, in the newly adapted coframe we have

(3.9) Ψ0==Ψk1=0,andΨk0.\Psi_{0}=\dots=\Psi_{k-1}=0,\quad\mathrm{and}\quad\Psi_{k}\neq 0.

Using the group action on Ψi\Psi_{i}’s or, equivalently the Bianchi identities (A.2), it follows that the bundle of adapted coframes that preserves the condition (3.9) gives rise to a 7-dimensional principal bundle 7M.\mathcal{F}^{7}\to M. More precisely, the new adapted coframes were determined by a choice of a21a_{21} in (2.16) as a result of which the structure group is reduced to 𝐇(1)𝐒𝐎2,2\mathbf{H}_{(1)}\subset\mathbf{SO}_{2,2} defined as

(3.10) 𝐇(1)={T(U)=(A00AT)A=(a11a120a22)𝐆𝐋2()}.\mathbf{H}_{(1)}=\left\{T(U)=\begin{pmatrix}A&0\\ 0&-A^{T}\end{pmatrix}\ \vline\ \ A=\begin{pmatrix}a_{11}&a_{12}\\ 0&a_{22}\end{pmatrix}\in\mathbf{GL}_{2}(\mathbb{R})\right\}.

Using the gauge transformations (2.18b) arising from the structure group 𝐇(1)\mathbf{H}_{(1)} for adapted coframes with respect to which (3.9) holds one obtains that the transformation of the connection 1-form Γ12\Gamma^{2}_{~{}1} does not involve the inhomogeneous terms d(T(U))T(U)1\mathrm{d}(T(U))T(U)^{-1} and therefore the following proposition holds.

Proposition 3.6.

Given a pKE metric, if the anti-self-dual Weyl curvature has special real Petrov type, i.e. it has a repeated root whose multiplicity is at least 2, then the bundle of adapted coframes which preserves the condition (3.9) is given by a principal 𝐇(1)\mathbf{H}_{(1)}-bundle 7M.\mathcal{F}^{7}\to M. Adapted coframes (α,α¯)(\alpha,\bar{\alpha}) arising as sections of 7\mathcal{F}^{7} satisfy (3.1) where

(3.11) Γ12=J1α2+J2α¯,1\Gamma^{2}_{~{}1}=J_{1}\alpha^{2}+J_{2}\bar{\alpha}{}^{1},

for some functions J1J_{1} and J2J_{2} on MM.

Proof.

The proof of the proposition simply follows from the Bianchi identity for Ψk1\Psi_{k-1} in (A.2) by inserting (3.9), which results in (3.11). ∎

Remark 3.7.

As a result of the proof above, one can express the quantities J1J_{1} and J2J_{2} in (3.11) in terms of Ψij\Psi_{ij}’s in (A.2). For instance, for k=2k=2 in (3.9) one obtains

J1=13Ψ2Ψ21,J2=13Ψ2Ψ24.J_{1}=\textstyle{\frac{-1}{3\Psi_{2}}\Psi_{21}},\qquad J_{2}=\textstyle{\frac{1}{3\Psi_{2}}\Psi_{24}}.

where Ψ24\Psi_{24} and Ψ21\Psi_{21} are referred to as the coframe derivatives of Ψ2.\Psi_{2}. However, we will not take this point of view in order to avoid fractional expressions.

A second coframe adaptation will result in the following proposition.

Proposition 3.8.

Every 4-dimensional pKE structure whose anti-self-dual Weyl tensor is of special real Petrov type defines a Cartan geometry of type (𝐒𝐎2,2,𝐓2)(\mathbf{SO}_{2,2},\mathbf{T}^{2}) where 𝐓2𝕊1×𝕊1\mathbf{T}^{2}\cong\mathbb{S}^{1}\times\mathbb{S}^{1} is the maximal torus in 𝐒𝐎2,2.\mathbf{SO}_{2,2}. The structure equations are given by (3.1) where

(3.12) Γ12=\displaystyle\Gamma^{2}_{~{}1}= J1α2+J2α¯,1\displaystyle J_{1}\alpha^{2}+J_{2}\bar{\alpha}{}^{1},
Γ21=\displaystyle\Gamma^{1}_{~{}2}= J3α1+J6α2+J5α¯+1J4α¯2\displaystyle-J_{3}\alpha^{1}+J_{6}\alpha^{2}+J_{5}\bar{\alpha}{}^{1}+J_{4}\bar{\alpha}{}^{2}

for some functions (J1,J2,J3,J4,J5,J6)(J_{1},J_{2},J_{3},J_{4},J_{5},J_{6}) on M.M. The 𝔰𝔬2,2\mathfrak{so}_{2,2}-valued Cartan connection can be represented as

(3.13) =([c|c]12Γ1132|Ψ2|α132|Ψ2|α¯112Γ110012Γ2232|Ψ2|α232|Ψ2|α¯212Γ22).\mathcal{B}=\begin{pmatrix}[c|c]\begin{matrix}\textstyle{\frac{1}{2}\Gamma^{1}_{~{}1}}&\sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}\alpha^{1}\\ \sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}\bar{\alpha}{}^{1}&-\textstyle{\frac{1}{2}\Gamma^{1}_{~{}1}}\end{matrix}&0\\ \cline{1-2}\cr 0&\begin{matrix}\textstyle{\frac{1}{2}\Gamma^{2}_{~{}2}}&\sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}\alpha^{2}\\ \sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}\bar{\alpha}{}^{2}&-\textstyle{\frac{1}{2}\Gamma^{2}_{~{}2}}\end{matrix}\end{pmatrix}.

The flatness condition for the resulting Cartan geometry, K=d+=0,K_{\mathcal{B}}={\rm d}\mathcal{B}+\mathcal{B}\wedge\mathcal{B}=0, is equivalent to J1=J2=J3=J4=J5=J6=Ψ4=0J_{1}=J_{2}=J_{3}=J_{4}=J_{5}=J_{6}=\Psi_{4}=0, and Ψ2=Ψ2\Psi_{2}=\Psi_{2}^{\prime} and implies that WeylWeyl^{-} has Petrov type DD.

Proof.

Assuming that the Petrov type is IIII or DD, the adaptation (3.9) reads

(3.14) Ψ0=Ψ1=0,andΨ20.\Psi_{0}=\Psi_{1}=0,\quad\mathrm{and}\quad\Psi_{2}\neq 0.

If the conditions (3.14) are to be preserved, by Proposition 3.6, one has (3.11) and, using the Bianchi identities (A.2), one obtains the differential relations

(3.15) dJ1=\displaystyle{\rm d}J_{1}= J1Γ11J12α1+J12α2+J13α¯+1J1J2α¯,2\displaystyle J_{1}\Gamma^{1}_{~{}1}-J_{1}^{2}\alpha^{1}+J_{12}\alpha^{2}+J_{13}\bar{\alpha}{}^{1}+J_{1}J_{2}\bar{\alpha}{}^{2},
dJ2=\displaystyle{\rm d}J_{2}= J2Γ22J1J2α1+(J13+Ψ2Ψ2)α2+J23α¯+1J22α¯,2\displaystyle-J_{2}\Gamma^{2}_{~{}2}-J_{1}J_{2}\alpha^{1}+(J_{13}+\Psi_{2}-\Psi_{2}^{\prime})\alpha^{2}+J_{23}\bar{\alpha}{}^{1}+J_{2}^{2}\bar{\alpha}{}^{2},
dΨ2=\displaystyle{\rm d}\Psi_{2}^{\prime}= 0,\displaystyle 0,
dΨ2=\displaystyle{\rm d}\Psi_{2}= 3J1Ψ2α1+(2J1Ψ3+Ψ31)α2+(2J2Ψ3Ψ34)α¯+13J2Ψ2α¯,2\displaystyle-3J_{1}\Psi_{2}\alpha^{1}+(2J_{1}\Psi_{3}+\Psi_{31})\alpha^{2}+(2J_{2}\Psi_{3}-\Psi_{34})\bar{\alpha}{}^{1}+3J_{2}\Psi_{2}\bar{\alpha}{}^{2},
dΨ3=\displaystyle{\rm d}\Psi_{3}= 3Ψ2Γ21Ψ3(Γ11Γ22)Ψ31α1(J1Ψ4+Ψ41)α2+(J2Ψ4Ψ44)α¯1Ψ34α¯,2\displaystyle 3\Psi_{2}\Gamma^{1}_{~{}2}-\Psi_{3}(\Gamma^{1}_{~{}1}-\Gamma^{2}_{~{}2})-\Psi_{31}\alpha^{1}-(J_{1}\Psi_{4}+\Psi_{41})\alpha^{2}+(J_{2}\Psi_{4}-\Psi_{44})\bar{\alpha}{}^{1}-\Psi_{34}\bar{\alpha}{}^{2},
dΨ4=\displaystyle{\rm d}\Psi_{4}= 4Ψ3Γ212Ψ4(Γ11Γ22)+Ψ41α1+Ψ42α2+Ψ43α¯+1Ψ44α¯2\displaystyle 4\Psi_{3}\Gamma^{1}_{~{}2}-2\Psi_{4}(\Gamma^{1}_{~{}1}-\Gamma^{2}_{~{}2})+\Psi_{41}\alpha^{1}+\Psi_{42}\alpha^{2}+\Psi_{43}\bar{\alpha}{}^{1}+\Psi_{44}\bar{\alpha}{}^{2}

for some functions J12,J13,J23J_{12},J_{13},J_{23} on M.M. These relations are obtained by inserting (3.11) in the structure equations (3.1) and requiring that the exterior derivative of the right hand side of the equations are zero.

Let the quantities Ji,Ψj,Ψ2J_{i},\Psi_{j},\Psi^{\prime}_{2} and J~i,Ψ~j,Ψ~2\tilde{J}_{i},\tilde{\Psi}_{j},\tilde{\Psi}^{\prime}_{2} be the quantities appearing in the structure equations for two choices of such adapted coframe related via θ~a=T(U)aθbb\tilde{\theta}^{a}=T(U)^{a}{}_{b}\theta^{b}. Using the structure group 𝐇(1)\mathbf{H}_{(1)} as in (3.10) and its induced gauge transformation (2.18), one obtains

(3.16) J~1\displaystyle\tilde{J}_{1} =a111J1,\displaystyle=a_{11}^{-1}J_{1},\qquad J~2=a22J2,\displaystyle\tilde{J}_{2}=a_{22}J_{2},
Ψ~2\displaystyle\tilde{\Psi}_{2}^{\prime} =Ψ2,\displaystyle=\Psi_{2}^{\prime},\qquad Ψ~2=Ψ2,\displaystyle\tilde{\Psi}_{2}=\Psi_{2},
Ψ~3\displaystyle\tilde{\Psi}_{3} =1a22(3a12Ψ2+a11Ψ3),\displaystyle=\textstyle{\frac{1}{a_{22}}(-3a_{12}\Psi_{2}+a_{11}\Psi_{3})},\qquad Ψ~4=1a222(6a122Ψ24a11a12Ψ3+a112Ψ4).\displaystyle\tilde{\Psi}_{4}=\textstyle{\frac{1}{a_{22}^{2}}(6a_{12}^{2}\Psi_{2}-4a_{11}a_{12}\Psi_{3}+a_{11}^{2}\Psi_{4})}.

As was mentioned in Remark 3.5, the infinitesimal version of the transformations above are given by the Bianchi identities (3.15).

Consequently, using the transformation law for Ψ3\Psi_{3} given in (3.16), we can further restrict to the bundle of adapted coframes with respect to which

(3.17) Ψ0=Ψ1=Ψ3=0,Ψ20.\Psi_{0}=\Psi_{1}=\Psi_{3}=0,\qquad\Psi_{2}\neq 0.

This can be seen explicitly from (3.16) by setting

a12=Ψ33Ψ2a11.a_{12}=\frac{\Psi_{3}}{3\Psi_{2}}a_{11}.

As a result, when the Petrov type of WeylWeyl^{-} is IIII or D,D, the bundle of adapted coframes preserving (3.17) gives rise to a principal 𝐇(2)\mathbf{H}_{(2)}-bundle 6M\mathcal{F}^{6}\to M where

(3.18) 𝐇(2):={T(U)=(A00AT)A=(a1100a22)𝐆𝐋2()}.\mathbf{H}_{(2)}:=\left\{T(U)=\begin{pmatrix}A&0\\ 0&-A^{T}\end{pmatrix}\ \vline\ \ A=\begin{pmatrix}a_{11}&0\\ 0&a_{22}\end{pmatrix}\in\mathbf{GL}_{2}(\mathbb{R})\right\}.

Since for such coframes the gauge transformations (2.18b) do not affect Γ21\Gamma^{1}_{~{}2} and Γ12\Gamma^{2}_{~{}1} by the inhomogeneous term d(T(U))T(U)1\mathrm{d}(T(U))T(U)^{-1}, one obtains the expressions (3.12) for some functions (J1,J2,J3,J4,J5,J6)(J_{1},J_{2},J_{3},J_{4},J_{5},J_{6}) satisfying the Bianchi identities (A.3).

It follows that the set of coframes adapted to the condition (3.17) gives a principal 𝐓2\mathbf{T}^{2}-bundle which is equipped by the Cartan connection .\mathcal{B}. The flatness condition follows from a straightforward computation.

For real Petrov types IIIIII and NN one needs to find the appropriate reduction of the structure bundle for pKE metrics and proceeds analogously to find the Cartan connection (3.13). If the Petrov type of WeylWeyl^{-} is III,III, the desired principal 𝐓2\mathbf{T}^{2}-bundle is given by the set of adapted null coframes with respect to which

Ψ0=Ψ1=Ψ2=Ψ4=0,Ψ30.\Psi_{0}=\Psi_{1}=\Psi_{2}=\Psi_{4}=0,\qquad\Psi_{3}\neq 0.

If the Petrov type is N,N, by Proposition 3.6 one can consider the bundle of adapted coframes with respect to which

Ψ3=Ψ2=Ψ1=Ψ0=0,Ψ40.\Psi_{3}=\Psi_{2}=\Psi_{1}=\Psi_{0}=0,\qquad\Psi_{4}\neq 0.

For such adapted coframes the 1-forms Γ12\Gamma^{2}_{~{}1} is reduced as in (3.11). Subsequently, one can use identities (A.2) to obtain

dΨ43\displaystyle\mathrm{d}\Psi_{43} =(2Γ223Γ11)Ψ435J2Ψ4Γ21+(3Ψ2Ψ4Ψ442)α1\displaystyle=(2\Gamma^{2}_{~{}2}-3\Gamma^{1}_{~{}1})\Psi_{43}-5J_{2}\Psi_{4}\Gamma^{1}_{2}+(-3\Psi_{2}^{\prime}\Psi_{4}-\Psi_{442})\alpha^{1}
+Ψ432α2+Ψ433α¯+1Ψ434α¯,2\displaystyle+\Psi_{432}\alpha^{2}+\Psi_{433}\bar{\alpha}{}^{1}+\Psi_{434}\bar{\alpha}{}^{2},

where Ψ43\Psi_{43} is the coframe derivative of Ψ4\Psi_{4} with respect to α¯1\bar{\alpha}{}^{1}. Therefore the principal 𝐓2\mathbf{T}^{2}-bundle 6M\mathcal{F}^{6}\to M is given by adapted null coframes with respect to which one additionally has Ψ43=0.\Psi_{43}=0. After these reductions Γ21\Gamma^{1}_{2} and Γ12\Gamma^{2}_{~{}1} can be expressed as (3.12) for some functions J1,,J6J_{1},\dots,J_{6} on M.M. We will not express the Bianchi identities for JiJ_{i}’s when the Petrov type is IIIIII or N.N. It is a matter of straightforward computation to show that K=0K_{\mathcal{B}}=0 for real Petrov types IIIIII and NN implies Ψ2=0\Psi^{\prime}_{2}=0 which is not being considered in this article.

3.2.2. Petrov type DD

PKE metrics of real Petrov type DD are particularly interesting because we find an explicit local normal form for them as shown below. Suppose that WeylWeyl^{-} has Petrov type DD i.e. the quartic (3.7) has two real roots with double multiplicity. Although Proposition 3.8 describes such pKE metrics as a Cartan geometry of type (𝐒𝐎2,2,𝐓2)(\mathbf{SO}_{2,2},\mathbf{T}^{2}), here we carry out the reduction procedure further and describe all such pKE metrics in some normal coordinate system. First, we have the following.

Proposition 3.9.

Given a pKE metric whose anti-self-dual Weyl curvature has real Petrov type DD everywhere, the Cartan connection \mathcal{B} on 6M\mathcal{F}^{6}\to M satisfies the structure equations (3.1) where

(3.19) Γ21=J3θ1+J4θ4,Γ12=J1θ2+J2θ3.\Gamma^{1}_{~{}2}=-J_{3}\theta^{1}+J_{4}\theta^{4},\qquad\Gamma^{2}_{~{}1}=J_{1}\theta^{2}+J_{2}\theta^{3}.

The EDS obtained from the reduced structure equations together with the differential relations among the functions J1,,J4,Ψ2,Ψ2J_{1},\dots,J_{4},\Psi_{2},\Psi^{\prime}_{2} given by

(3.20) dJ1\displaystyle{\rm d}J_{1} =J12θ1+J1J2θ4+Γ11J1+2J1J3θ2J41θ3\displaystyle=-J_{1}^{2}\theta^{1}+J_{1}J_{2}\theta^{4}+\Gamma^{1}_{~{}1}J_{1}+2J_{1}J_{3}\theta^{2}-J_{41}\theta^{3}
dJ2\displaystyle{\rm d}J_{2} =Γ22J2J1J2θ1+(J41+Ψ2Ψ2)θ2+2J2J4θ3+J22θ4\displaystyle=-\Gamma^{2}_{~{}2}J_{2}-J_{1}J_{2}\theta^{1}+(-J_{41}+\Psi_{2}-\Psi^{\prime}_{2})\theta^{2}+2J_{2}J_{4}\theta^{3}+J_{2}^{2}\theta^{4}
dJ3\displaystyle{\rm d}J_{3} =Γ22J32J1J3θ1+J32θ2+J3J4θ3+(J41+Ψ2Ψ2)θ4\displaystyle=\Gamma^{2}_{~{}2}J_{3}-2J_{1}J_{3}\theta^{1}+J_{3}^{2}\theta^{2}+J_{3}J_{4}\theta^{3}+(-J_{41}+\Psi_{2}-\Psi^{\prime}_{2})\theta^{4}
dJ4\displaystyle{\rm d}J_{4} =J3J4θ2+J42θ3Γ11J4+J41θ1+2J2J4θ4\displaystyle=J_{3}J_{4}\theta^{2}+J_{4}^{2}\theta^{3}-\Gamma^{1}_{~{}1}J_{4}+J_{41}\theta^{1}+2J_{2}J_{4}\theta^{4}
dJ41\displaystyle{\rm d}J_{41} =(2J1J412J1Ψ2J1Ψ2+2J1J2J3)θ1+(2J3J412J1J3J4)θ2\displaystyle=(-2J_{1}J_{41}-2J_{1}\Psi^{\prime}_{2}-J_{1}\Psi_{2}+2J_{1}J_{2}J_{3})\theta^{1}+(2J_{3}J_{41}-2J_{1}J_{3}J_{4})\theta^{2}
+(2J2J3J4+2J4J41+2J4Ψ2+J4Ψ2)θ3+(2J1J2J4+2J2J41)θ4\displaystyle\ \ +(-2J_{2}J_{3}J_{4}+2J_{4}J_{41}+2J_{4}\Psi^{\prime}_{2}+J_{4}\Psi_{2})\theta^{3}+(-2J_{1}J_{2}J_{4}+2J_{2}J_{41})\theta^{4}
dΨ2\displaystyle{\rm d}\Psi_{2} =3J1Ψ2θ1+3J2Ψ2θ4+3J3Ψ2θ2+3J4Ψ2θ3\displaystyle=-3J_{1}\Psi_{2}\theta^{1}+3J_{2}\Psi_{2}\theta^{4}+3J_{3}\Psi_{2}\theta^{2}+3J_{4}\Psi_{2}\theta^{3}
dΨ2\displaystyle{\rm d}\Psi_{2}^{\prime} =0\displaystyle=0

for some function J41J_{41} is closed under the exterior derivative operator d\mathrm{d}. As a result, the local moduli space of type DD pKE metrics is 5-dimensional.

Proof.

The differential identities (3.20) are obtained via straightforward computations discussed previously. The fact that the space of such pKE metrics is 5-dimensional follows from the Frobenius theorem applied to the resulting closed EDS. More precisely, define the 13-dimensional bundle 136\mathcal{E}^{13}\to\mathcal{F}^{6} whose fibers are parametrized by J=(J1,J2,J3,J4,J41,Ψ2,Ψ2).J=(J_{1},J_{2},J_{3},J_{4},J_{41},\Psi_{2},\Psi^{\prime}_{2}). Since the Pfaffian system (3.20) is integrable, its leaf space is 7-dimensional parametrized by J.J. As the infinitesimal group actions in (3.20) suggests, one obtains that the action of the structure group 𝐇(2)\mathbf{H}_{(2)} transforms the quantities J2J_{2} and J4J_{4} by

J~2=a22J2,J~4=a11J4.\tilde{J}_{2}=a_{22}J_{2},\qquad\tilde{J}_{4}=a_{11}J_{4}.

To find the local generality one considers generic pKE metrics of type D.D. As a result, it can be assumed that J2,J40,J_{2},J_{4}\neq 0, after restricting to sufficiently small neighborhoods. Hence, setting a11=1J4a_{11}=\frac{1}{J_{4}} and a22=1J2,a_{22}=\frac{1}{J_{2}}, one can normalize J2=J4=1J_{2}=J_{4}=1 and obtain a canonical coframe at each point of such pKE metrics. Consequently, the remaining 5 parameters of JJ can be used to distinguish every generic type DD pKE metric up to isomorphism. Therefore, the local generality of type DD pKE metrics depends on 5 constants. ∎

It turns out that the pKE metrics of type D locally belong to one of four branches which can be characterized according to the vanishing of the quantities J2J_{2} and J4.J_{4}. Each branch can be locally expressed using normal coordinates given below.

Theorem 3.10.

In sufficiently small open sets, every pKE metric

g=2θ1θ3+2θ2θ4,g=2\theta^{1}\theta^{3}+2\theta^{2}\theta^{4},

whose anti-self-dual Weyl curvature has real Petrov type DD, belongs to one of the following four branches. The first branch is characterized by the property that J2J_{2} and J4J_{4} are non-vanishing and is comprised of 5-parameter family of pKE metrics for which a choice of normalized coframe can be expressed as

(3.21) θ1\displaystyle\theta^{1} =(y4)2(1y3)dy1y4dy2,\displaystyle=(y^{4})^{2}\left(1-y^{3}\right)\mathrm{d}y^{1}-y^{4}\mathrm{d}y^{2},
θ2\displaystyle\theta^{2} =(y4)2dy1θ1,\displaystyle=(y^{4})^{2}\mathrm{d}y^{1}-\theta^{1},
θ3\displaystyle\theta^{3} =(Ψ2(y3)3+k1(y3)2y4k2y3(y4)2k3+12k4(y4)3)θ1dy3y3y4dy4\displaystyle=\textstyle{\left(-\Psi_{2}^{\prime}(y^{3})^{3}+\frac{k_{1}(y^{3})^{2}}{y^{4}}-\frac{k_{2}y^{3}}{(y^{4})^{2}}-\frac{k_{3}+\frac{1}{2}k_{4}}{(y^{4})^{3}}\right)\theta^{1}-\mathrm{d}y^{3}-\frac{y^{3}}{y^{4}}\mathrm{d}y^{4}}
θ4\displaystyle\theta^{4} =(Ψ2(y31)3k1(y31)2y4+k2(y31)(y4)2+k3(y4)3)θ2+dy3+y31y4dy4\displaystyle=\textstyle{\left(\Psi_{2}^{\prime}(y^{3}-1)^{3}-\frac{k_{1}(y^{3}-1)^{2}}{y^{4}}+\frac{k_{2}(y^{3}-1)}{(y^{4})^{2}}+\frac{k_{3}}{(y^{4})^{3}}\right)\theta^{2}+\mathrm{d}y^{3}+\frac{y^{3}-1}{y^{4}}\mathrm{d}y^{4}}

for some constants k1,,k4k_{1},\dots,k_{4} and Ψ2.\Psi^{\prime}_{2}. The second branch is characterized by the condition that J4=0J_{4}=0 and J2J_{2} non-vanishing and is comprised of a 3-parameter family of pKE metrics for which a choice of coframe can be expressed as

(3.22) θ1=k2(y1)2y4dy3+y1y4dy4+dy1,θ2=y1dy3+1y4dy2,θ3=dy3θ4=k1(y4)3+2k2y4+2Ψ22y4dy2k1(y4)3+2k2y4+2Ψ22y1dy3+1y4dy4\begin{gathered}\theta^{1}=\textstyle{k_{2}(y^{1})^{2}y^{4}\mathrm{d}y^{3}+\frac{y^{1}}{y^{4}}\mathrm{d}y^{4}+\mathrm{d}y^{1}},\qquad\theta^{2}=\textstyle{y^{1}\mathrm{d}y^{3}+\frac{1}{y^{4}}\mathrm{d}y^{2}},\qquad\theta^{3}=\mathrm{d}y^{3}\\ \theta^{4}=\textstyle{-\frac{k_{1}(y^{4})^{3}+2k_{2}y^{4}+2\Psi^{\prime}_{2}}{2y^{4}}\mathrm{d}y^{2}-\frac{k_{1}(y^{4})^{3}+2k_{2}y^{4}+2\Psi^{\prime}_{2}}{2}y^{1}\mathrm{d}y^{3}+\frac{1}{y^{4}}\mathrm{d}y^{4}}\end{gathered}

where k1,k2k_{1},k_{2} and Ψ2\Psi_{2}^{\prime} are constants. Similarly, the third branch is characterized by J2=0J_{2}=0 and J4J_{4} non-vanishing, which is comprised of 3-parameter family of pKE metrics and can be expressed as in (3.22) after switching θ1θ2\theta^{1}\leftrightarrow\theta^{2} and θ3θ4.\theta^{3}\leftrightarrow\theta^{4}.

Lastly, the fourth branch, characterized by J2=J4=0,J_{2}=J_{4}=0, is comprised of the only homogeneous pKE metrics of type D.D. They form a 1-parameter family parametrized by Ψ2,\Psi^{\prime}_{2}, for which a choice of coframe is given by

(3.23) θ1\displaystyle\theta^{1} =dy1132Ψ2y1y3,\displaystyle=\textstyle{\frac{{\rm d}y^{1}}{1-\tfrac{3}{2}\Psi_{2}^{\prime}y^{1}y^{3}}},\qquad θ2=dy2132Ψ2y2y4,θ3\displaystyle\theta^{2}=\textstyle{\frac{{\rm d}y^{2}}{1-\tfrac{3}{2}\Psi_{2}^{\prime}y^{2}y^{4}}},\qquad\theta^{3} =dy3132Ψ2y1y3,\displaystyle=\textstyle{\frac{{\rm d}y^{3}}{1-\tfrac{3}{2}\Psi_{2}^{\prime}y^{1}y^{3}}}, θ4=dy4132Ψ2y2y4,\displaystyle\theta^{4}=\textstyle{\frac{{\rm d}y^{4}}{1-\tfrac{3}{2}\Psi_{2}^{\prime}y^{2}y^{4}}},
Proof.

Let us first work in a neighborhood UU in which J2,J4J_{2},J_{4} are nowhere vanishing. Using the action of a11a_{11} and a22,a_{22}, whose infinitesimal version is given in (3.20), there is a unique coframe with respect to which J2=J4=1J_{2}=J_{4}=1 and therefore

(3.24) Γ11=J41θ1+J3θ2+θ3+2θ4,Γ22=J1θ1+(J41+Ψ2Ψ2)θ2+2θ3+θ4.\Gamma^{1}_{~{}1}=J_{41}\theta^{1}+J_{3}\theta^{2}+\theta^{3}+2\theta^{4},\qquad\Gamma^{2}_{~{}2}=-J_{1}\theta^{1}+(-J_{41}+\Psi_{2}-\Psi^{\prime}_{2})\theta^{2}+2\theta^{3}+\theta^{4}.

To find our normal coordinate system we make use of the orbits of the Killing vector fields of these pKE metrics. Let v=viθiv=v^{i}\frac{\partial}{\partial\theta^{i}} be a Killing vector field i.e. vθi=0\mathscr{L}_{v}\theta^{i}=0 where \mathscr{L} denotes the Lie derivative. It is straightforward to use the structure equations (3.1) and reductions (3.19) and (3.24) when J2=J4=1J_{2}=J_{4}=1 to obtain dvi\mathrm{d}v^{i} for i=1,,4i=1,\dots,4. Subsequently, the identities d2vi=0\mathrm{d}^{2}v^{i}=0 imply that the isometry group for such pKE metrics is two dimensional since

v=v1(θ1+J1θ3)+v2(θ2J3θ4),v=\textstyle{v^{1}(\frac{\partial}{\partial\theta^{1}}+J_{1}\frac{\partial}{\partial\theta^{3}})+v^{2}(\frac{\partial}{\partial\theta^{2}}-J_{3}\frac{\partial}{\partial\theta^{4}}),}

where

(3.25) dv1=v1(J1θ1θ3)(2v1+v2)(J3θ2+θ4),dv2=v1(θ1+J1θ3)+v2(θ2J3θ4).\mathrm{d}v^{1}=v^{1}(J_{1}\theta^{1}-\theta^{3})-(2v^{1}+v^{2})(J_{3}\theta^{2}+\theta^{4}),\quad\mathrm{d}v^{2}=v^{1}(\theta^{1}+J_{1}\theta^{3})+v^{2}(\theta^{2}-J_{3}\theta^{4}).

Because of the relation d(v1+v2)2(v1+v2)=J1θ1J3θ2θ3θ4,\textstyle{\frac{\mathrm{d}(v^{1}+v^{2})}{2(v^{1}+v^{2})}=J_{1}\theta^{1}-J_{3}\theta^{2}-\theta^{3}-\theta^{4}}, we restrict ourselves to an open set where d(v1+v2)\mathrm{d}(v^{1}+v^{2}) and v1+v2v^{1}+v^{2} are non-zero. As a result, one can assume v1+v2>0v^{1}+v^{2}>0 and define

y3=v2v1+v2,y4=v1+v2.\textstyle{y^{3}=\frac{v^{2}}{{v^{1}+v^{2}}}},\qquad y^{4}=\sqrt{v^{1}+v^{2}}.

From (3.25) it follows that

(3.26) θ3=J1θ1dy3y3y4dy4,θ4=J3θ2+dy3+y31y4dy4\theta^{3}=\textstyle{J_{1}\theta^{1}-\mathrm{d}y^{3}-\frac{y^{3}}{y^{4}}\mathrm{d}y^{4}},\qquad\theta^{4}=\textstyle{-J_{3}\theta^{2}+\mathrm{d}y^{3}+\frac{y^{3}-1}{y^{4}}\mathrm{d}y^{4}}

Using the reduced 1-forms (3.24) and normalized values J2=J4=1J_{2}=J_{4}=1 in the Bianchi identities (3.20) together with the expressions (3.26), it follows that J1,J3,J41J_{1},J_{3},J_{41} and Ψ2\Psi_{2} are functions of y3y^{3} and y4.y^{4}. For instance, one obtains dΨ2=3Ψ2y4dy4\mathrm{d}\Psi_{2}=-\frac{3\Psi_{2}}{y^{4}}\mathrm{d}y^{4} which implies Ψ2=k4(y4)3\Psi_{2}=\frac{k_{4}}{(y^{4})^{3}} for a constant k4.k_{4}. Similarly, elementary calculations can be carried out to show

(3.27) J1\displaystyle J_{1} =Ψ2(y3)3+k1y4(y3)2k2(y4)2y32k3+k42(y4)3,\displaystyle=-\Psi_{2}^{\prime}(y^{3})^{3}+\textstyle{\frac{k_{1}}{y^{4}}(y^{3})^{2}-\frac{k_{2}}{(y^{4})^{2}}y^{3}-\frac{2k_{3}+k_{4}}{2(y^{4})^{3}}},
J3\displaystyle J_{3} =Ψ2(y3)3+(3Ψ2+k1y4)(y3)2(3Ψ2+2k1y4+k2(y4)2)y3+Ψ2+k1y4+k2(y4)2k3(y4)3,\displaystyle=\textstyle{-\Psi^{\prime}_{2}(y^{3})^{3}+(3\Psi^{\prime}_{2}+\frac{k_{1}}{y^{4}})(y^{3})^{2}-(3\Psi^{\prime}_{2}+\frac{2k_{1}}{y^{4}}+\frac{k_{2}}{(y^{4})^{2}})y^{3}+\Psi^{\prime}_{2}+\frac{k_{1}}{y^{4}}+\frac{k_{2}}{(y^{4})^{2}}-\frac{k_{3}}{(y^{4})^{3}}},
J41\displaystyle J_{41} =2Ψ2(y3)3(3Ψ2+2k1y4)(y3)2+(2k1y4+2k2(y4)2)y3k2(y4)2+2k3+k4(y4)3,\displaystyle=\textstyle{2\Psi^{\prime}_{2}(y^{3})^{3}-(3\Psi^{\prime}_{2}+\frac{2k_{1}}{y^{4}})(y^{3})^{2}+(\frac{2k_{1}}{y^{4}}+\frac{2k_{2}}{(y^{4})^{2}})y^{3}-\frac{k_{2}}{(y^{4})^{2}}+\frac{2k_{3}+k_{4}}{(y^{4})^{3}}},

for constants k1,k2,k3k_{1},k_{2},k_{3} and k4.k_{4}. To express θ1\theta^{1} and θ2\theta^{2} in a local normal form, one makes use of the structure equations (3.1) and the reductions (3.19) and (3.24) when J2=J4=1J_{2}=J_{4}=1 to obtain

d(θ1+θ2)\displaystyle\mathrm{d}(\theta^{1}+\theta^{2}) =2y4dy4(θ1+θ2).\displaystyle=\textstyle{\frac{2}{y^{4}}\mathrm{d}y^{4}{\,{\wedge}\;}(\theta^{1}+\theta^{2})}.

By Darboux’ theorem, locally one obtains θ1=(y4)2dy1θ2\theta^{1}=(y^{4})^{2}\mathrm{d}y^{1}-\theta^{2} for a local coordinate y1.y^{1}. Lastly, the relation

dθ2=1y4dy4θ2+y4(y3dy4+y4dy4)dy1\mathrm{d}\theta^{2}=\textstyle{\frac{1}{y^{4}}\mathrm{d}y^{4}{\,{\wedge}\;}\theta^{2}+y^{4}\left(y^{3}\mathrm{d}y^{4}+y^{4}\mathrm{d}y^{4}\right){\,{\wedge}\;}\mathrm{d}y^{1}}

implies that θ2=(y4)2y3dy1+y4dy2\theta^{2}=(y^{4})^{2}y^{3}\mathrm{d}y^{1}+y^{4}\mathrm{d}y^{2} for a local coordinate y2.y^{2}. This proves the local normal form presented in (3.21). Furthermore, it is straightforward to verify that y1\frac{\partial}{\partial y^{1}} and y2\frac{\partial}{\partial y^{2}} are the Killing vector fields for these pKE metrics.

For the second branch we restrict ourselves to open sets UMU\subseteq M in which J4=0J_{4}=0 and J2J_{2} is nowhere vanishing. Over UU one obtains J41=J1=0.J_{41}=J_{1}=0. Using the relations (3.20), consider the set of coframes with respect to which J2=1J_{2}=1 and

(3.28) Γ22=(Ψ2Ψ2)θ2+θ4.\Gamma^{2}_{~{}2}=(\Psi_{2}-\Psi^{\prime}_{2})\theta^{2}+\theta^{4}.

As described before, it is straightforward to express the Killing vector fields of such pKE metrics which are of the form

(3.29) v=v1θ1+v2θ2+v3θ1+J3v2θ4+v5Γ11v=\textstyle{v^{1}\frac{\partial}{\partial\theta^{1}}+v^{2}\frac{\partial}{\partial\theta^{2}}+v^{3}\frac{\partial}{\partial\theta^{1}}+J_{3}v^{2}\frac{\partial}{\partial\theta^{4}}+v^{5}\frac{\partial}{\partial\Gamma^{1}_{~{}1}}}

and, as a result, have to satisfy the differential relations

(3.30) dv1\displaystyle\mathrm{d}v^{1} =v1Γ11+(J3v2+v5)θ1v1J3θ2,\displaystyle=-v^{1}\Gamma^{1}_{~{}1}+(J_{3}v^{2}+v^{5})\theta^{1}-v^{1}J_{3}\theta^{2},
dv2\displaystyle\mathrm{d}v^{2} =J3v2θ2v1θ3v2θ4+v3θ1,\displaystyle=-J_{3}v^{2}\theta^{2}-v^{1}\theta^{3}-v^{2}\theta^{4}+v^{3}\theta^{1},
dv3\displaystyle\mathrm{d}v^{3} =v3Γ11(J3v2+v5)θ3v3θ4,\displaystyle=v^{3}\Gamma^{1}_{~{}1}-(J_{3}v^{2}+v^{5})\theta^{3}-v^{3}\theta^{4},
dv5\displaystyle\mathrm{d}v^{5} =(J32Ψ2Ψ2)(v3θ1v1θ3)+v2(Ψ2Ψ2)(J3θ2+θ4).\displaystyle=(J_{3}-2\Psi^{\prime}_{2}-\Psi_{2})(v^{3}\theta^{1}-v^{1}\theta^{3})+v^{2}(\Psi^{\prime}_{2}-\Psi_{2})(J_{3}\theta^{2}+\theta^{4}).

Using the structure equations (3.1) and the reduced 1-forms (3.19) and (3.28) when J4=0J_{4}=0 and J2=1,J_{2}=1, together with the Bianchi identities (3.20), it follows that

(3.31) d(J3θ2+θ4)=0θ4=1y4dy4J3θ2\mathrm{d}(J_{3}\theta^{2}+\theta^{4})=0\Rightarrow\theta^{4}=\textstyle{\frac{1}{y^{4}}\mathrm{d}y^{4}}-J_{3}\theta^{2}

for a local coordinate y4.y^{4}. It follows from (3.29) that the orbits of the isometry group for such pKE metrics are level sets of y4.y^{4}. This implies that such pKE metrics have cohomogeneity one. Using the Bianchi identities (3.20) and the reduction (3.28), one obtains that J3J_{3} and Ψ2\Psi_{2} are given by

Ψ2=k1(y4)3,J3=12k1(y4)3+k2y4+Ψ2.\Psi_{2}=k_{1}(y^{4})^{3},\qquad J_{3}=\textstyle{\frac{1}{2}k_{1}(y^{4})^{3}+k_{2}y^{4}+\Psi^{\prime}_{2}.}

It remains to express θ1,θ2,θ3\theta^{1},\theta^{2},\theta^{3} in a local normal form which will be done using (3.30). Restricting to open sets where v30,v^{3}\neq 0, the relations (3.30) imply that

(3.32) θ1\displaystyle\theta^{1} =v1v3θ3+v2v3y4dy4+1v3dv2\displaystyle=\textstyle{\frac{v^{1}}{v^{3}}\theta^{3}+\frac{v^{2}}{v^{3}y^{4}}\mathrm{d}y^{4}+\frac{1}{v^{3}}\mathrm{d}v^{2}}
Γ11\displaystyle\Gamma^{1}_{1} =(Ψ2+(y4)32k1+k2y4)θ2+v2(12k1(y4)3+k2y+Ψ2)+v5v3θ3+dy4y4+dv3v3\displaystyle=\textstyle{-(\Psi^{\prime}_{2}+\frac{(y^{4})^{3}}{2k_{1}}+k_{2}y^{4})\theta^{2}+\frac{v^{2}(\frac{1}{2}k_{1}(y^{4})^{3}+k_{2}y^{+}\Psi^{\prime}_{2})+v^{5}}{v^{3}}\theta^{3}+\frac{\mathrm{d}y^{4}}{y^{4}}+\frac{\mathrm{d}v^{3}}{v^{3}}}

Using (3.31), (3.32) in (3.30), it is a matter of elementary calculation to show

v1=k2y4(v2)2v3,v5=k1(y4)32k2y4+2Ψ22v2.v^{1}=\textstyle{\frac{k_{2}y^{4}(v^{2})^{2}}{v^{3}}},\qquad v^{5}=\textstyle{-\frac{k_{1}(y^{4})^{3}-2k_{2}y^{4}+2\Psi^{\prime}_{2}}{2}v^{2}}.

Lastly, the reduced structure equations imply that

dθ2=θ3(dv2v3v2dy4v3y4)+θ2dy4y4,dθ3=θ3dv3v3.\mathrm{d}\theta^{2}=-\textstyle{\theta^{3}{\,{\wedge}\;}(\frac{dv^{2}}{v^{3}}-\frac{v^{2}\mathrm{d}y^{4}}{v^{3}y^{4}})+\theta^{2}{\,{\wedge}\;}\frac{\mathrm{d}y^{4}}{y^{4}}},\qquad\mathrm{d}\theta^{3}=-\theta^{3}{\,{\wedge}\;}\textstyle{\frac{\mathrm{d}v^{3}}{v^{3}}}.

Using Darboux’ theorem, one can find local coordinate system with respect to which

θ2=dy2y4+v2dy3,θ3=v3dy3.\theta^{2}=\textstyle{\frac{\mathrm{d}y^{2}}{y^{4}}}+v^{2}\mathrm{d}y^{3},\qquad\theta^{3}=v^{3}\mathrm{d}y^{3}.

As a result, we have a local coordinate system (y1,,y5),(y^{1},\dots,y^{5}), where y1=v2y^{1}=v^{2} and y5=v3y^{5}=v^{3} with respect to which we have expressed θi\theta^{i}’s and Γ11\Gamma^{1}_{~{}1}. It is clear that v3v^{3} acts by scaling on θ1\theta^{1} and θ3\theta^{3} and therefore, corresponds to the element a11a_{11} in the reduced structure group (3.18). The coframe (3.22) is obtained by setting v3=1.v^{3}=1. Finally, note that in terms of the local coordinates (y1,,y4)(y^{1},\dots,y^{4}), the trajectories of the Killing vector fields are given by the level sets of y2y^{2} and y3y^{3} and (y1)2(y3)2.(y^{1})^{2}-(y^{3})^{2}. The third branch characterized by J2=0J_{2}=0 and J4J_{4} nowhere vanishing can be treated similarly.

Finally, when J2=J4=0,J_{2}=J_{4}=0, straightforward computation shows that all JiJ_{i}’s vanish and Ψ2=Ψ2.\Psi_{2}=\Psi^{\prime}_{2}. As a result, such metrics are homogeneous. The structure equations are given by

(3.33a) dθ1=Γ11θ1,dθ3=Γ11θ3,dΓ11=3Ψ2θ1θ3,\displaystyle\mathrm{d}\theta^{1}=-\Gamma^{1}_{~{}1}\wedge\theta^{1},\qquad\mathrm{d}\theta^{3}=\Gamma^{1}_{~{}1}\wedge\theta^{3},\qquad\mathrm{d}\Gamma^{1}_{~{}1}=-3\Psi_{2}^{\prime}\theta^{1}\wedge\theta^{3},
(3.33b) dθ2=Γ22θ2,dθ4=Γ22θ4,dΓ22=3Ψ2θ2θ4.\displaystyle\mathrm{d}\theta^{2}=-\Gamma^{2}_{~{}2}\wedge\theta^{2},\qquad\mathrm{d}\theta^{4}=\Gamma^{2}_{~{}2}\wedge\theta^{4},\qquad\mathrm{d}\Gamma^{2}_{~{}2}=-3\Psi_{2}^{\prime}\theta^{2}\wedge\theta^{4}.

Hence, it is sufficient to find a normal form for (3.33a). Using Darboux’s theorem, (3.33a) imply that there are coordinates y1,y3,y5y^{1},y^{3},y^{5} with respect to which

θ1=ey5+F1dy1,θ3=ey5F1+F2dy3,Γ11=dy5(F1F2)y1dy1F1y3dy3\theta^{1}=e^{-y^{5}+F_{1}}\mathrm{d}y^{1},\qquad\theta^{3}=e^{y^{5}-F_{1}+F_{2}}\mathrm{d}y^{3},\qquad\Gamma^{1}_{~{}1}=\mathrm{d}y^{5}-\textstyle{\frac{\partial(F_{1}-F_{2})}{\partial y^{1}}\mathrm{d}y^{1}-\frac{\partial F_{1}}{\partial y^{3}}\mathrm{d}y^{3}}

for arbitrary functions F1=F1(y1,y3)F_{1}=F_{1}(y^{1},y^{3}) and F2=F2(y1,y3)F_{2}=F_{2}(y^{1},y^{3}) which satisfy

(3.34) 2y1y3F23Ψ2eF2=0.\textstyle{\frac{\partial^{2}}{\partial y^{1}\partial y^{3}}F_{2}-3\Psi^{\prime}_{2}e^{F_{2}}=0}.

The equation above is known as Liouville’s equation [Hen86] whose solutions can be expressed as

(3.35) F2(y1,y3)=ln(2pq3Ψ2(pq)2)F_{2}(y^{1},y^{3})=\ln\left(\frac{2p^{\prime}q^{\prime}}{-3\Psi^{\prime}_{2}(p-q)^{2}}\right)

for two arbitrary functions p=p(y1)p=p(y^{1}) and q=q(y3).q=q(y^{3}). Setting

p(y1)=6Ψ2y1,q(y3)=4y3,F1(y1,y3)=ln(1132Ψ2y1y3),y5=0\textstyle{p(y^{1})=6\Psi_{2}^{\prime}y^{1},\qquad q(y^{3})=\frac{4}{y^{3}},\qquad F_{1}(y^{1},y^{3})=\textstyle{\ln\left(\frac{1}{1-\frac{3}{2}\Psi^{\prime}_{2}y^{1}y^{3}}\right)},\qquad y^{5}=0}

one obtains the expression (3.23) for θ1\theta^{1} and θ3.\theta^{3}. The expressions of θ2\theta^{2} and θ4\theta^{4} are obtained similarly.

It is straightforward to use the coframe (3.23) in order to arrive at the potential function V2V_{2} in § 2.3.4. Moreover, one can characterize the branchings in Theorem 3.10 in terms of the vanishing of J1J_{1} and J3.J_{3}.

Remark 3.11.

Theorem 3.10 is yet another instance of explicit local normal form for certain classes of (pseudo-)Riemannian metrics whose Weyl curvature has algebraic type D,D, which includes the Plebański-Demiański metrics [Kin69, Deb71, PD76] in the Lorentzian signature (see [GP06, KK09] for a survey of all the results), and ambitoric metrics in Riemannian signature [ACG16].

3.2.3. Petrov type IIII

Now we proceed to pKE metrics whose anti-self-dual Weyl curvature has Petrov type IIII.

Theorem 3.12.

Given a pKE metric of Petrov type IIII, the 𝐒𝐎2,2\mathbf{SO}_{2,2}-valued Cartan connection \mathcal{B} on the principal bundle 𝕊1×𝕊16M\mathbb{S}^{1}\times\mathbb{S}^{1}\to\mathcal{F}^{6}\to M, as defined in Proposition 3.8, satisfies the Yang-Mills equations DK=0D*K_{\mathcal{B}}=0, where K=d+K_{\mathcal{B}}={\rm d}\mathcal{B}+\mathcal{B}\wedge\mathcal{B}, if and only if

(3.36) Ψ2=Ψ2,andJ1=J2=J3=J4=0.\Psi_{2}=\Psi_{2}^{\prime},\quad\mathrm{and}\quad J_{1}=J_{2}=J_{3}=J_{4}=0.

in (3.12). Examples of such pKE structure are given by g=2θ1θ3+2θ2θ4g=2\theta^{1}\theta^{3}+2\theta^{2}\theta^{4} where

(3.37) θ1=\displaystyle\theta^{1}= dx32Ψ2(x2+xf1(a)+f2(b))da\displaystyle{\rm d}x-\tfrac{3}{2}\Psi_{2}^{\prime}\big{(}x^{2}+xf_{1}(a)+f_{2}(b)\big{)}{\rm d}a
θ2=\displaystyle\theta^{2}= db\displaystyle{\rm d}b
θ3=\displaystyle\theta^{3}= da\displaystyle{\rm d}a
θ4=\displaystyle\theta^{4}= dy32Ψ2(y2+yf3(b)+f4(a))db\displaystyle{\rm d}y-\tfrac{3}{2}\Psi_{2}^{\prime}\big{(}y^{2}+yf_{3}(b)+f_{4}(a)\big{)}{\rm d}b

for some arbitrary functions f1,,f4f_{1},\dots,f_{4} where f2,f4f^{\prime}_{2},f^{\prime}_{4} are nowhere vanishing.

Proof.

Using the expression of KK_{\mathcal{B}} in (A.4), one immediately obtains (3.36) (see Remark 3.13). One can integrate the structure equations for pKE metrics of Petrov type IIII that are Yang-Mills assuming some simplifying conditions. The integration procedure can be carried out similarly to what was explained in Theorem 3.9 and will not be explained here. It turns out that pKE metrics arising from the coframe (3.37) satisfy

Ψ2\displaystyle\Psi_{2} =Ψ2=const,Ψ0=Ψ1=Ψ3=0,\displaystyle=\Psi_{2}^{\prime}=\mathrm{const},\qquad\Psi_{0}=\Psi_{1}=\Psi_{3}=0,
Ψ4\displaystyle\Psi_{4} =34(3(f3f2+f1f4+2xf4+2yf2)Ψ22f4′′2f2′′)Ψ2\displaystyle=-\tfrac{3}{4}\Big{(}3\left(f_{3}f^{\prime}_{2}+f_{1}f_{4}^{\prime}+2xf_{4}^{\prime}+2yf_{2}^{\prime}\right)\Psi^{\prime}_{2}-2f_{4}^{\prime\prime}-2f_{2}^{\prime\prime}\Big{)}\Psi_{2}^{\prime}
J5\displaystyle J_{5} =32Ψ2f2,J6=32Ψ2f4\displaystyle=\textstyle{-\frac{3}{2}\Psi_{2}^{\prime}f^{\prime}_{2}},\qquad J_{6}=\textstyle{\frac{3}{2}\Psi_{2}^{\prime}f^{\prime}_{4}}

Note that if f2=f4=0f^{\prime}_{2}=f^{\prime}_{4}=0 one obtains homogeneous pKE metrics of type DD.

The 𝐒𝐎2,2\mathbf{SO}_{2,2}-valued Cartan connection \mathcal{B}, as defined in (3.13), has curvature

K=([c|c]03232|Ψ2|f23/20000003232|Ψ2|f43/20)σ2,K_{\mathcal{B}}=\begin{pmatrix}[c|c]\begin{matrix}0&\tfrac{3}{2}\sqrt{\tfrac{3}{2}}|\Psi_{2}^{\prime}|{}^{3/2}~{}f^{\prime}_{2}\\ 0&0\end{matrix}&0\\ \cline{1-2}\cr 0&\begin{matrix}0&0\\ -\tfrac{3}{2}\sqrt{\tfrac{3}{2}}|\Psi_{2}^{\prime}|{}^{3/2}~{}f^{\prime}_{4}&0\end{matrix}\end{pmatrix}\sigma^{2}_{-},

which is anti-self-dual, and therefore satisfies the Yang-Mills equations DK=0D*K_{\mathcal{B}}=0. ∎

Remark 3.13.

Note that if the conditions (3.36) for J1,,J6J_{1},\dots,J_{6} in  (3.12), which arise from the Yang-Mills equation, are replaced by J5=J6=0,J_{5}=J_{6}=0, then the WeylWeyl^{-} of the pKE metric has type DD as discussed in Theorem 3.9. Furthermore, as shown in Proposition 3.8, one can always associate a Cartan geometry of type (𝐒𝐎2,2,𝐓2)(\mathbf{SO}_{2,2},\mathbf{T}^{2}), with a canonical Cartan connection, to pKE metrics of any Petrov type by appropriately reducing the structure group. However, except for type IIII, it can be shown that the set of Yang-Mills solutions among other types is empty.

3.2.4. Petrov type IIIIII

Assume that the quartic W(λ)W(\lambda) in (3.7) has a repeated root of multiplicity three. As we did in § 3.2.2, by coframe adaptation one can translate the multiple root to zero, which is equivalent to finding an adapted coframe with respect to which

Ψ0=Ψ1=Ψ2=0,andΨ30.\Psi_{0}=\Psi_{1}=\Psi_{2}=0,\quad\mathrm{and}\quad\Psi_{3}\neq 0.

In this case Proposition 3.6 still remains valid and the following differential relations hold

(3.38) dΨ3\displaystyle{\rm d}\Psi_{3} =Γ11Ψ3+Γ22Ψ32J1Ψ3θ1+(J1Ψ4+Ψ41)θ2+(J2Ψ4Ψ44)θ3+2J2Ψ3θ4\displaystyle=-\Gamma^{1}_{~{}1}\Psi_{3}+\Gamma^{2}_{~{}2}\Psi_{3}-2J_{1}\Psi_{3}\theta^{1}+(J_{1}\Psi_{4}+\Psi_{41})\theta^{2}+(J_{2}\Psi_{4}-\Psi_{44})\theta^{3}+2J_{2}\Psi_{3}\theta^{4}
dΨ4\displaystyle{\rm d}\Psi_{4} =2Γ11Ψ4+4Γ21Ψ3+2Γ22Ψ4+Ψ41θ1+Ψ42θ2+Ψ43θ3+Ψ44θ4\displaystyle=-2\Gamma^{1}_{~{}1}\Psi_{4}+4\Gamma^{1}_{~{}2}\Psi_{3}+2\Gamma^{2}_{~{}2}\Psi_{4}+\Psi_{41}\theta^{1}+\Psi_{42}\theta^{2}+\Psi_{43}\theta^{3}+\Psi_{44}\theta^{4}
dJ2\displaystyle{\rm d}J_{2} =J1J2θ1+J22θ4Γ22J2+J22θ2+J23θ3\displaystyle=-J_{1}J_{2}\theta^{1}+J_{2}^{2}\theta^{4}-\Gamma^{2}_{~{}2}J_{2}+J_{22}\theta^{2}+J_{23}\theta^{3}

for some functions J22,J23.J_{22},J_{23}.

Using the action of a11a_{11} and a22a_{22}, one can find the set of adapted coframes with respect to which the quantities Ψ3\Psi_{3} and J2J_{2} are normalized to constants. The set of such adapted coframes give rise to a line bundle 5M\mathcal{F}^{5}\to M with the group parameter a12a_{12} in (3.10) as the fiber coordinate.

Proposition 3.14.

Given a pKE metric whose anti-self-dual Weyl curvature has Petrov type III and for which J20J_{2}\neq 0, the bundle of adapted coframes preserving

Ψ0=Ψ1=Ψ2=0,Ψ3=const0,J2=const0\Psi_{0}=\Psi_{1}=\Psi_{2}=0,\qquad\Psi_{3}=\mathrm{const}\neq 0,\qquad J_{2}=\mathrm{const}\neq 0

is a line bundle 5M,\mathcal{F}^{5}\to M, whose sections satisfy the structure equations (3.1) wherein

Γ12=J1θ2+J2θ3,Γ11=3J1θ1+J3θ2+J4θ3+3J2θ4,Γ22=J1θ1+J5θ3+J6θ2+J2θ4.\Gamma^{2}_{~{}1}=J_{1}\theta^{2}+J_{2}\theta^{3},\qquad\Gamma^{1}_{~{}1}=-3J_{1}\theta^{1}+J_{3}\theta^{2}+J_{4}\theta^{3}+3J_{2}\theta^{4},\qquad\Gamma^{2}_{~{}2}=-J_{1}\theta^{1}+J_{5}\theta^{3}+J_{6}\theta^{2}+J_{2}\theta^{4}.

Examples of such pKE metrics for which J1=0J_{1}=0, J6=Ψ2,J_{6}=-\Psi^{\prime}_{2}, J2=Ψ3=1J_{2}=\Psi_{3}=1 are given by g=2θ1θ3+2θ2θ4g=2\theta^{1}\theta^{3}+2\theta^{2}\theta^{4} where

θ1\displaystyle\theta^{1} =14(3e12y4Ψ2y1+y2e12y4Ψ2+6Ψ22y1+2y2ey42e12y4Wy3Ψ2Wy3+ey4Uy1+Uy1,y1)e2y4dy1\displaystyle=-\scriptstyle{\frac{1}{4}\left(3\mathrm{e}^{\frac{1}{2}y^{4}}\Psi^{\prime}_{2}y^{1}+y^{2}\mathrm{e}^{\frac{1}{2}y^{4}}\Psi^{\prime}_{2}+6\Psi^{\prime 2}_{2}y^{1}+2y^{2}\mathrm{e}^{y^{4}}-2\mathrm{e}^{\frac{1}{2}y^{4}}W_{y^{3}}-\Psi^{\prime}_{2}W_{y^{3}}+\mathrm{e}^{y^{4}}-U_{y^{1}}+U_{y^{1},y^{1}}\right)\mathrm{e}^{-2y^{4}}\mathrm{d}y^{1}}
+ey4dy2+12e32y4(y2e12y42Wy3,y4)dy4\displaystyle\phantom{=}+\textstyle{\,\mathrm{e}^{y^{4}}\mathrm{d}y^{2}+\frac{1}{2}\mathrm{e}^{-\frac{3}{2}y^{4}}\left(y^{2}\mathrm{e}^{\frac{1}{2}y^{4}}-2W_{y^{3},y^{4}}\right)\mathrm{d}y^{4}}
14(6Ψ22y1y2e12y49y2ey4Ψ2y1+ey4Ψ2(y2)2+9Ψ2y1Wy3e12y42Wy3Ψ2y2e12y46Ψ22y1Wy3\displaystyle\phantom{=}\scriptstyle{-\frac{1}{4}\left(6\Psi^{\prime 2}_{2}y^{1}y^{2}\mathrm{e}^{\frac{1}{2}y^{4}}-9y^{2}\mathrm{e}^{y^{4}}\Psi^{\prime}_{2}y^{1}+\mathrm{e}^{y^{4}}\Psi^{\prime}_{2}(y^{2})^{2}+9\Psi^{\prime}_{2}y^{1}W_{y^{3}}\mathrm{e}^{\frac{1}{2}y^{4}}-2W_{y^{3}}\Psi^{\prime}_{2}y^{2}\mathrm{e}^{\frac{1}{2}y^{4}}-6\Psi^{\prime 2}_{2}y^{1}W_{y^{3}}\right.}
+4y2ey4Wy32Wy32e12y4Uy1y2e12y4+Uy1,y1y2e12y4+Wy32Ψ22(y2)2e32y4ey4Wy3\displaystyle\phantom{=}\scriptstyle{\left.+4y^{2}\mathrm{e}^{y^{4}}W_{y^{3}}-2W_{y^{3}}^{2}\mathrm{e}^{\frac{1}{2}y^{4}}-U_{y^{1}}y^{2}\mathrm{e}^{\frac{1}{2}y^{4}}+U_{y^{1},y^{1}}y^{2}\mathrm{e}^{\frac{1}{2}y^{4}}+W_{y^{3}}^{2}\Psi^{\prime}_{2}-2(y^{2})^{2}\mathrm{e}^{\frac{3}{2}y^{4}}-\mathrm{e}^{y^{4}}W_{y^{3}}\right.}
4Ue12y4+4Wy3,y3e12y4+Uy1Wy3Wy3Uy1,y1+y2e32y4)e2y4dy3\displaystyle\phantom{=}\scriptstyle{\left.-4U\mathrm{e}^{\frac{1}{2}y^{4}}+4W_{y^{3},y^{3}}\mathrm{e}^{\frac{1}{2}y^{4}}+U_{y^{1}}W_{y^{3}}-W_{y^{3}}U_{y^{1},y^{1}}+y^{2}\mathrm{e}^{\frac{3}{2}y^{4}}\right)\mathrm{e}^{-2y^{4}}\mathrm{d}y^{3}}
θ2\displaystyle\theta^{2} =e12y4dy1+(y2Wy3)e12y4dy3\displaystyle=\textstyle{\mathrm{e}^{-\frac{1}{2}y^{4}}\mathrm{d}y^{1}+\left(y^{2}-W_{y^{3}}\right)\mathrm{e}^{-\frac{1}{2}y^{4}}\mathrm{d}y^{3}}
θ3\displaystyle\theta^{3} =ey4dy3\displaystyle=\mathrm{e}^{y^{4}}\mathrm{d}y^{3}
θ4\displaystyle\theta^{4} =(Ψ2e12y41)dy1+12dy4\displaystyle=\textstyle{\left(-\Psi^{\prime}_{2}\mathrm{e}^{-\frac{1}{2}y^{4}}-1\right)\mathrm{d}y^{1}+\frac{1}{2}\mathrm{d}y^{4}}
+(32e12y4Ψ22y1+34e12y4Wy3Ψ214e12y4Uy1+14e12y4Uy112y2e12y4\displaystyle\textstyle{\phantom{=}+\left(\frac{3}{2}\mathrm{e}^{-\frac{1}{2}y^{4}}\Psi^{\prime 2}_{2}y^{1}+\frac{3}{4}\mathrm{e}^{-\frac{1}{2}y^{4}}W_{y^{3}}\Psi^{\prime}_{2}-\frac{1}{4}\mathrm{e}^{-\frac{1}{2}y^{4}}U_{y^{1}}+\frac{1}{4}\mathrm{e}^{-\frac{1}{2}y^{4}}U_{y^{1}}-\frac{1}{2}y^{2}\mathrm{e}^{\frac{1}{2}y^{4}}\right.}
+34Ψ2y134Ψ2y234e12y4+12Wy3)dy3\displaystyle\phantom{=}\textstyle{\left.+\frac{3}{4}\Psi^{\prime}_{2}y^{1}-\frac{3}{4}\Psi^{\prime}_{2}y^{2}-\frac{3}{4}\mathrm{e}^{\frac{1}{2}y^{4}}+\frac{1}{2}W_{y^{3}}\right)\mathrm{d}y^{3}}

where U=U(y1,y3),W=W(y3,y4)U=U(y^{1},y^{3}),W=W(y^{3},y^{4}) are arbitrary functions.

Proof.

We skip the proof due to its similarity to that of Theorems 3.9 and 3.12. ∎

Remark 3.15.

We point out that by the action of the structure group, infinitesimally given in (3.38), one can reduce the structure group to identity by translating Ψ4\Psi_{4} to zero and obtain a unique choice of coframe at each point. However, to obtain examples above one does not need to carry out full reduction. Some solutions satisfying these conditions where obtained earlier by A. Chudecki in [Chu17].

3.2.5. Petrov type NN

Assume that the quartic W(λ)W(\lambda) in (3.7) has a repeated root of multiplicity four. As we did in § 3.2.2, by coframe adaptation one can translate the multiple root to zero, which is equivalent to finding an adapted coframe with respect to which

Ψ0=Ψ1=Ψ2=Ψ3=0,andΨ40.\Psi_{0}=\Psi_{1}=\Psi_{2}=\Psi_{3}=0,\quad\mathrm{and}\quad\Psi_{4}\neq 0.

In this case Proposition 3.6 still remains valid and the following differential relations hold

(3.39) dΨ4\displaystyle{\rm d}\Psi_{4} =2Γ11Ψ4+2Γ22Ψ4J1Ψ4θ1+Ψ42θ2+Ψ43θ3+J2Ψ4θ4\displaystyle=-2\Gamma^{1}_{~{}1}\Psi_{4}+2\Gamma^{2}_{~{}2}\Psi_{4}-J_{1}\Psi_{4}\theta^{1}+\Psi_{42}\theta^{2}+\Psi_{43}\theta^{3}+J_{2}\Psi_{4}\theta^{4}
dJ2\displaystyle{\rm d}J_{2} =J1J2θ1+J22θ2+J23θ3+J22θ4Γ22J2\displaystyle=-J_{1}J_{2}\theta^{1}+J_{22}\theta^{2}+J_{23}\theta^{3}+J_{2}^{2}\theta^{4}-\Gamma^{2}_{~{}2}J_{2}

As a result, by normalizing Ψ4\Psi_{4} and J2J_{2} to non-zero constants we can reduce the parameters a11a_{11} and a22a_{22} in the structure group (3.10) which reduces the bundle of adapted coframes to a line bundle 5M\mathcal{F}^{5}\to M with the element a12a_{12} in (3.10) as the fiber coordinate. It follows that

(3.40) Γ11=32J1θ1+J3θ2+J4θ3+32J2θ4,Γ22=J1θ1+J5θ2+J6θ3+J2θ4.\displaystyle\Gamma^{1}_{~{}1}=-\textstyle{\frac{3}{2}J_{1}\theta^{1}+J_{3}\theta^{2}+J_{4}\theta^{3}+\frac{3}{2}J_{2}\theta^{4}},\quad\Gamma^{2}_{~{}2}=-J_{1}\theta^{1}+J_{5}\theta^{2}+J_{6}\theta^{3}+J_{2}\theta^{4}.

for some functions J3,,J6.J_{3},\dots,J_{6}. It is straightforward to obtain

dJ452J2Γ21mod{θ1,θ2,θ3,θ4}.\mathrm{d}J_{4}\equiv-\textstyle{\frac{5}{2}J_{2}\Gamma^{1}_{2}}\qquad\mathrm{mod}\qquad\{\theta^{1},\theta^{2},\theta^{3},\theta^{4}\}.

Because the quantity J2J_{2} is normalized to a non-zero constant, the above differential relation can be interpreted as the infinitesimal action of the 1-dimensional structure group on J4J_{4} (see Remark 3.5). Hence, by choosing a12a_{12} appropriately, one can translate J4J_{4} to zero which would reduce the structure group to identity. In other words there is a unique coframe at each point with respect to which one has the relations equations (3.11), (3.40) and

(3.41) Γ21=15(2J3+3J5+7Ψ2J2)θ1+J7θ2+J8θ3+35J6θ4.\Gamma^{1}_{~{}2}=\textstyle{-\frac{1}{5}\left(2J_{3}+3J_{5}+7\frac{\Psi^{\prime}_{2}}{J_{2}}\right)\theta^{1}+J_{7}\theta^{2}+J_{8}\theta^{3}+\frac{3}{5}J_{6}\theta^{4}}.

for some functions J7J_{7} and J8J_{8} on M.M. As a result we obtain the following.

Theorem 3.16.

Given a pKE metric whose anti-self-dual Weyl curvature has Petrov type NN and for which J20J_{2}\neq 0, there is a unique adapted coframe that preserves

Ψ0=Ψ1=Ψ2=Ψ3=0,Ψ4=const0,J2=const0,J4=0\Psi_{0}=\Psi_{1}=\Psi_{2}=\Psi_{3}=0,\qquad\Psi_{4}=\mathrm{const}\neq 0,\qquad J_{2}=\mathrm{const}\neq 0,\qquad J_{4}=0

with respect to which the relations (3.11), (3.40) and (3.41) hold. A class of examples for which J1=J6=J7=0J_{1}=J_{6}=J_{7}=0, Ψ2=8J3=4J5,\Psi^{\prime}_{2}=-8J_{3}=4J_{5}, J2=4,J_{2}=-4, and Ψ3=1\Psi_{3}=1 is given by g=2θ1θ3+2θ2θ4g=2\theta^{1}\theta^{3}+2\theta^{2}\theta^{4} where

θ1\displaystyle\theta^{1} =2e3y4dy1+(16(y3)2+F1(y2)+y3F2(y2))dy2\displaystyle=2\mathrm{e}^{-3y^{4}}\mathrm{d}y^{1}+\left(-16(y^{3})^{2}+F_{1}(y^{2})+y^{3}F_{2}(y^{2})\right)\mathrm{d}y^{2}
θ2\displaystyle\theta^{2} =8e2y4(dy3y1dy2)\displaystyle=8\mathrm{e}^{-2y^{4}}\left(\mathrm{d}y^{3}-y^{1}\mathrm{d}y^{2}\right)
θ3\displaystyle\theta^{3} =ey4dy2\displaystyle=\mathrm{e}^{y^{4}}\mathrm{d}y^{2}
θ4\displaystyle\theta^{4} =12dy412Ψ2e2y4(dy3y1dy2)\displaystyle=\textstyle{-\frac{1}{2}\mathrm{d}y^{4}-\frac{1}{2}\Psi^{\prime}_{2}\mathrm{e}^{-2y^{4}}\left(\mathrm{d}y^{3}-y^{1}\mathrm{d}y^{2}\right)}

where F1(y2)F_{1}(y^{2}) and F2(y2)F_{2}(y^{2}) are arbitrary functions.

Proof.

We skip the proof due to its similarity to that of Theorems 3.9 and 3.12. ∎

3.2.6. Petrov type OO

The Petrov type O corresponds to pKE metrics for which Ψ0==Ψ4=0.\Psi_{0}=\dots=\Psi_{4}=0. Since the only non-zero quantity in the structure equations (3.1) is the constant Ψ2,\Psi^{\prime}_{2}, it follows that such metrics are homogeneous and therefore no reduction of the structure bundle is possible. Nevertheless, one can follow the procedure explained before and integrate the structure equations from which the following choice of coframe is obtained

(3.42) θ1=dy1Ψ2(y2+y1y3y4),θ2=dy2Ψ2(y2+y1y3y4),θ3=(y4y2)dy3y3dy4Ψ2(y2+y1y3y4),θ4=y1dy3dy4Ψ2(y2+y1y3y4)\theta^{1}=\textstyle{\frac{\mathrm{d}y^{1}}{\Psi^{\prime}_{2}(y^{2}+y^{1}y^{3}-y^{4})}},\ \ \theta^{2}=\textstyle{\frac{\mathrm{d}y^{2}}{\Psi^{\prime}_{2}(y^{2}+y^{1}y^{3}-y^{4})}},\ \ \theta^{3}=\textstyle{\frac{(y^{4}-y^{2})\mathrm{d}y^{3}-y^{3}\mathrm{d}y^{4}}{\Psi^{\prime}_{2}(y^{2}+y^{1}y^{3}-y^{4})}},\ \ \theta^{4}=\textstyle{\frac{y^{1}\mathrm{d}y^{3}-\mathrm{d}y^{4}}{\Psi^{\prime}_{2}(y^{2}+y^{1}y^{3}-y^{4})}}

Using the coframe above one can recover the potential function V1V_{1} given in § 2.3.4 for the so-called dancing metric.

3.2.7. Homogeneous models and local generality of various Petrov types

The structure equations of pKE metrics in dimension four and reduced structure equations obtained for non-generic Petrov types enable one to use the Cartan-Kähler theory and obtain the local generality of analytic pKE metrics of each Petrov type. We will not give the details of how Cartan-Kähler theory is implemented and refer the reader to [Bry14] for details.

Assuming analyticity for pKE metrics, the following table gives the local generality of various Petrov types.

Petrov type Local generality
GG 2 functions of 3 variables
IIII 4 functions of 2 variables
IIIIII 3 functions of 2 variables
NN 2 functions of 2 variables
IIII and Yang-Mills 2 functions of 2 variables
DD 5 constants
OO 1 constant
Table 1. Local generality of pKE metrics

Furthermore, the reduced structure equations for each Petrov types allows one to look for homogeneous models. Finding homogeneous models involves a straightforward inspection of structure equations considering all possible normalizations which can be carried out algorithmically. We will not present all the necessary computation here. It turns out that the only homogeneous models of pKE metrics satisfying Ψ20\Psi_{2}^{\prime}\neq 0 are the 1-parameter families of pKE metrics of type DD and OO which correspond to the coframes (3.23) and (3.42). In particular, there is no homogeneous pKE metric of type G,G, II,II, IIIIII and NN for which Ψ20.\Psi^{\prime}_{2}\neq 0.

4. (2,3,5)-distributions arising from pKE metrics

This section contains the highlight of the article. In § 4.1 we give a brief review of the geometry of (2,3,5)-distributions. In § 4.2 we show that the naturally induced rank 2 twistor distribution on the space of self-dual null planes of any pKE metric is (2,3,5) in an open subset if Ψ20.\Psi^{\prime}_{2}\neq 0. Furthermore, the root type of the Cartan quartic of this twistor distribution agrees with the root type of the quartic representation of Weyl.Weyl^{-}. This remarkable and surprising coincidence is contrasted with the case of twistor distribution naturally arising on the space of anti-self-dual null planes of pKE metrics satisfying Weyl0,Weyl^{-}\neq 0, which is considered in § 4.3. In the latter case, the coefficients of the Cartan quartic depend on the fourth jet of the coefficients of WeylWeyl^{-} and there is no further simplification from the larger context of twistor distributions arising from indefinite conformal structures in dimension four satisfying Weyl0.Weyl^{-}\neq 0. In other words, a priori, no relation between the type of the Cartan quartic and the Petrov type of WeylWeyl^{-} or Weyl+Weyl^{+} can be made (see Remark 4.15). Moreover, our construction in § 4.2 gives rise to 5-dimensional para-Sasaki-Einstein structures and conformal structures with 𝐒𝐋3()\mathbf{SL}_{3}(\mathbb{R}) holonomy, as studied in [SW17]. Consequently, our explicit examples of pKE metrics of special real Petrov type in § 3.2, provide examples of 5-dimensional para-Sasaki-Einstein metrics.

4.1. A primer on (2,3,5)(2,3,5) distributions

In this section we recall the basic definitions and theorems about the local geometry of a generic 2-plane field on a 5-dimensional manifold, Q,Q, which involves a Cartan connection and a naturally induced conformal structure of signature (3,2).

Given a 5-dimensional manifold, Q,Q, with a rank 2 distribution 𝒟TQ\mathscr{D}\subset\mathrm{T}Q let 𝒟\partial\mathscr{D} denote its first derived system defined as the distribution whose sections are given by Γ(𝒟)+[Γ(𝒟),Γ(𝒟)],\Gamma(\mathscr{D})+[\Gamma(\mathscr{D}),\Gamma(\mathscr{D})], where Γ(𝒟)\Gamma(\mathscr{D}) denotes the sheaf of sections of the distribution 𝒟.\mathscr{D}. Moreover, define 2𝒟=(𝒟).\partial^{2}\mathscr{D}=\partial(\partial\mathscr{D}).

Definition 4.1.

A rank 2 distribution in dimension 5, 𝒟TQ,\mathscr{D}\subset\mathrm{T}Q, is called a (2,3,5)-distribution if

rank(𝒟)=3,andrank(2𝒟)=5.\mathrm{rank}(\partial\mathscr{D})=3,\qquad\mathrm{and}\qquad\mathrm{rank}(\partial^{2}\mathscr{D})=5.

Locally, a generic rank 2 distribution is a (2,3,5)-distribution. Given a (2,3,5)-distribution, locally, one can find a frame {𝐯1,,𝐯5}\{\mathbf{v}_{1},\cdots,\mathbf{v}_{5}\} for MM such that

𝒟=span{𝐯4,𝐯5},𝒟=span{𝐯3,𝐯4,𝐯5},2𝒟=span{𝐯1,,𝐯5}\mathscr{D}=\mathrm{span}\{\mathbf{v}_{4},\mathbf{v}_{5}\},\qquad\partial\mathscr{D}=\mathrm{span}\{\mathbf{v}_{3},\mathbf{v}_{4},\mathbf{v}_{5}\},\qquad\partial^{2}\mathscr{D}=\mathrm{span}\{\mathbf{v}_{1},\cdots,\mathbf{v}_{5}\}

where

𝐯3=[𝐯4,𝐯5],𝐯2=[𝐯3,𝐯4],𝐯1=[𝐯3,𝐯5].\mathbf{v}_{3}=-[\mathbf{v}_{4},\mathbf{v}_{5}],\qquad\mathbf{v}_{2}=-[\mathbf{v}_{3},\mathbf{v}_{4}],\qquad\mathbf{v}_{1}=-[\mathbf{v}_{3},\mathbf{v}_{5}].

As a result, the corresponding coframe {η1,,η5}\{\eta^{1},\cdots,\eta^{5}\} satisfies

(4.1) dη1\displaystyle\mathrm{d}\eta^{1} η3η4mod{η1,η2},\displaystyle\equiv\eta^{3}{\,{\wedge}\;}\eta^{4}\qquad\mathrm{mod}\qquad\{\eta^{1},\eta^{2}\},
dη2\displaystyle\mathrm{d}\eta^{2} η3η5mod{η1,η2},\displaystyle\equiv\eta^{3}{\,{\wedge}\;}\eta^{5}\qquad\mathrm{mod}\qquad\{\eta^{1},\eta^{2}\},
dη3\displaystyle\mathrm{d}\eta^{3} η4η5mod{η1,η2,η3}.\displaystyle\equiv\eta^{4}{\,{\wedge}\;}\eta^{5}\qquad\mathrm{mod}\qquad\{\eta^{1},\eta^{2},\eta^{3}\}.

Cartan in his famous ’five-variables’ paper [Car10] solved the equivalence problem for (2,3,5)-distributions and explicitly introduced the distribution 𝒟o\mathscr{D}_{o} whose algebra of infinitesimal symmetries is given by the split real form of the exceptional Lie algebra 𝔤2.\mathfrak{g}^{*}_{2}. Recall that the noncompact exceptional simple Lie group of dimension 14, 𝐆2𝐒𝐎4,3\mathbf{G}^{*}_{2}\subset\mathbf{SO}_{4,3} acts transitively on the projective quadric 𝖰3,26\mathsf{Q}_{3,2}\subset\mathbb{P}^{6} defined by the (3,2)-signature diagonal matrix. Let 𝐏1\mathbf{P}_{1} be the parabolic subgroup of 𝐆2\mathbf{G}^{*}_{2} that preserves a null line in 𝖰3,2\mathsf{Q}_{3,2}. Using his method of equivalence, Cartan associated an {e}\{e\}-structure on a 14-dimensional 𝐏1\mathbf{P}_{1}-principal bundle π:𝒢Q\pi:\mathcal{G}\to Q to any (2,3,5)-distribution.

Using the appropriate transformation Cartan’s original construction results in the following theorem.

Theorem 4.2 ([Car10, Nur05]).

Any (2,3,5)-distribution, 𝒟TQ,\mathscr{D}\subset\mathrm{T}Q, defines a Cartan geometry (𝒢,Q,ω𝐆2)(\mathcal{G},Q,\omega_{\mathbf{G}^{*}_{2}}) of type (𝐆2,𝐏1).(\mathbf{G}_{2}^{*},\mathbf{P}_{1}). Expressing the 𝔤2\mathfrak{g}_{2}^{*}-valued Cartan connection as

(4.2) ω𝐆2=(ζ1ζ4ζ8ζ913ζ713ζ513ζ60η1ζ1ζ213η413η3013ζ6η2ζ3ζ413η5013η313ζ523η323ζ523ζ6013η513η413ζ7η4ζ7023ζ6ζ4ζ2ζ9η50ζ723ζ5ζ3ζ1ζ80η5η423η3η2η1ζ1+ζ4)\omega_{\mathbf{G}^{*}_{2}}=\textstyle{\begin{pmatrix}-\zeta_{1}-\zeta_{4}&-\zeta_{8}&-\zeta_{9}&-\frac{1}{\sqrt{3}}\zeta_{7}&\frac{1}{3}\zeta_{5}&\frac{1}{3}\zeta_{6}&0\\ \eta^{1}&\zeta_{1}&\zeta_{2}&\frac{1}{\sqrt{3}}\eta^{4}&-\frac{1}{3}\eta^{3}&0&\frac{1}{3}\zeta_{6}\\ \eta^{2}&\zeta_{3}&\zeta_{4}&\frac{1}{\sqrt{3}}\eta^{5}&0&-\frac{1}{3}\eta^{3}&-\frac{1}{3}\zeta_{5}\\ \frac{2}{\sqrt{3}}\eta^{3}&\frac{2}{\sqrt{3}}\zeta_{5}&\frac{2}{\sqrt{3}}\zeta_{6}&0&\frac{1}{\sqrt{3}}\eta^{5}&-\frac{1}{\sqrt{3}}\eta^{4}&-\frac{1}{\sqrt{3}}\zeta_{7}\\ \eta^{4}&\zeta_{7}&0&\frac{2}{\sqrt{3}}\zeta_{6}&-\zeta_{4}&\zeta_{2}&\zeta_{9}\\ \eta^{5}&0&\zeta_{7}&-\frac{2}{\sqrt{3}}\zeta_{5}&\zeta_{3}&-\zeta_{1}&-\zeta_{8}\\ 0&\eta^{5}&-\eta^{4}&\frac{2}{\sqrt{3}}\eta^{3}&-\eta^{2}&\eta^{1}&\zeta_{1}+\zeta_{4}\end{pmatrix}}

the distribution 𝒟\mathscr{D} is the projection of Ker{η1,η2,η3}\mathrm{Ker}\{\eta^{1},\eta^{2},\eta^{3}\} from 𝒢\mathcal{G} to Q.Q.

Remark 4.3.

Cartan realized that the curvature Kω𝐆2=dω𝐆2+ω𝐆2ω𝐆2K_{\omega_{\mathbf{G}^{*}_{2}}}=\mathrm{d}\omega_{\mathbf{G}_{2}^{*}}+\omega_{\mathbf{G}_{2}^{*}}{\,{\wedge}\;}\omega_{\mathbf{G}_{2}^{*}} can be interpreted as a ternary quartic form 𝐖Sym4(𝒟)\mathbf{W}\in\mathrm{Sym}^{4}(\partial\mathscr{D})^{*} and expressed as

(4.3) 𝐖\displaystyle\mathbf{W} =i=04(i4)ai(η4)4i(η5)i+i=03(i3)bi(η4)3i(η5)iη3\displaystyle=\sum_{i=0}^{4}\begin{pmatrix}i\\ 4\end{pmatrix}a_{i}(\eta^{4})^{4-i}(\eta^{5})^{i}+\sum_{i=0}^{3}\begin{pmatrix}i\\ 3\end{pmatrix}b_{i}(\eta^{4})^{3-i}(\eta^{5})^{i}\eta^{3}
+i=02(i2)ci(η4)2i(η5)i(η3)2+i=01di(η4)1i(η5)i(η3)3+e(η3)4\displaystyle+\sum_{i=0}^{2}\begin{pmatrix}i\\ 2\end{pmatrix}c_{i}(\eta^{4})^{2-i}(\eta^{5})^{i}(\eta^{3})^{2}+\sum_{i=0}^{1}d_{i}(\eta^{4})^{1-i}(\eta^{5})^{i}(\eta^{3})^{3}+e(\eta^{3})^{4}

where the coefficients a0,,ea_{0},\cdots,e are components of the curvature Kω𝐆2K_{\omega_{\mathbf{G}^{*}_{2}}} (see [Nur05]). Moreover, the fundamental curvature tensor is a binary quartic form 𝐂Sym4(𝒟),\mathbf{C}\in\mathrm{Sym}^{4}(\mathscr{D}^{*}), referred to as the Cartan quartic, given by the first 5 terms in (4.3). If 𝐂\mathbf{C} is identically zero it follows that Kω𝐆2=0K_{\omega_{\mathbf{G}^{*}_{2}}}=0 i.e. the (2,3,5)-distribution is flat. It is convenient to express the Cartan quartic in 1-variable zz as follows

(4.4) C(z)\displaystyle C(z) :=𝐂(η4+zη5,η4+zη5,η4+zη5,η4+zη5)\displaystyle:=\textstyle{\mathbf{C}(\frac{\partial}{\partial\eta^{4}}+z\frac{\partial}{\partial\eta^{5}},\frac{\partial}{\partial\eta^{4}}+z\frac{\partial}{\partial\eta^{5}},\frac{\partial}{\partial\eta^{4}}+z\frac{\partial}{\partial\eta^{5}},\frac{\partial}{\partial\eta^{4}}+z\frac{\partial}{\partial\eta^{5}})}
=a0+4a1z+6a2z2+4a3z3+a4z4.\displaystyle=a_{0}+4a_{1}z+6a_{2}z^{2}+4a_{3}z^{3}+a_{4}z^{4}.

Using Cartan’s result and the embedding 𝐆2𝐒𝐎4,3,\mathbf{G}^{*}_{2}\hookrightarrow\mathbf{SO}_{4,3}, the following non-trivial link between (2,3,5)-distributions and conformal structures of signature (3,2)(3,2) can be obtained.

Theorem 4.4 ([Nur05]).

Any (2,3,5)-distribution 𝒟TQ\mathscr{D}\subset\mathrm{T}Q defines a conformal structure [h~][\tilde{h}] of signature (3,2) on Q,Q, which can be expressed as h~=sh\tilde{h}=s^{*}h for any section s:Q𝒢,s\colon Q\to\mathcal{G}, where

(4.5) h=η1η5η2η4+23η3η3Sym2(TQ).h=\eta^{1}\eta^{5}-\eta^{2}\eta^{4}+{\textstyle\frac{2}{3}}\,\eta^{3}\eta^{3}\in\mathrm{Sym}^{2}(\mathrm{T}^{*}Q).

The conformal holonomy of this conformal structure takes value in 𝐆2\mathbf{G}_{2}^{*} and its Weyl curvature can be expressed in terms of Kω𝐆2.K_{\omega_{\mathbf{G}^{*}_{2}}}.

Remark 4.5.

Using Theorem 4.4, we give another interpretation of the Cartan quartic (4.4) which will be important for analyzing non-integrable twistor distributions. At each point qQ,q\in Q, consider the 𝕊1\mathbb{S}^{1}-family of planes

(4.6) 𝒵q:=Ker{η2zη1,η5zη4,η3}TqQ,\mathcal{Z}_{q}:=\operatorname{Ker}\{\eta^{2}-z\eta^{1},\eta^{5}-z\eta^{4},\eta^{3}\}\subset\mathrm{T}_{q}Q,

where z{}.z\in\mathbb{R}\cup\{\infty\}. Such planes are null with respect to the conformal structure [h],[h], defined in (4.5), and intersect the distribution 𝒟q\mathscr{D}_{q} along the lines η4+zη5.\langle\frac{\partial}{\partial\eta^{4}}+z\frac{\partial}{\partial\eta^{5}}\rangle. The bundle

(4.7) γ:𝒵Q,\gamma\colon\mathcal{Z}\to Q,

where γ1(q)=𝒵q,\gamma^{-1}(q)=\mathcal{Z}_{q}, is the circle bundle of such null planes. Denote the components of the Weyl curvature for the conformal structure [h][h] by WjkliW^{i}_{~{}jkl} where 1i,j,k,l5.1\leq i,j,k,l\leq 5. Following our discussion in § 2.2.4, let Wijkl=himWjklmW_{ijkl}=h_{im}W^{m}_{~{}jkl} and define the multilinear map

𝐖~:=Wijkl(ηiηj)(ηkηl)Sym2(Λ2TQ)C(Q).\widetilde{\mathbf{W}}:=W_{ijkl}(\eta^{i}{\,{\wedge}\;}\eta^{j})\circ(\eta^{k}{\,{\wedge}\;}\eta^{l})\in\mathrm{Sym}^{2}(\Lambda^{2}\mathrm{T}Q)\to C^{\infty}(Q).

Restricting to 𝒵,\mathcal{Z}, one obtains the quartic polynomial

C(z)\displaystyle C(z) =𝐖~(η4+zη5,η1+zη2,η4+zη5,η1+zη2)\displaystyle=\widetilde{\mathbf{W}}(\textstyle{\frac{\partial}{\partial\eta^{4}}+z\frac{\partial}{\partial\eta^{5}},\frac{\partial}{\partial\eta^{1}}+z\frac{\partial}{\partial\eta^{2}},\frac{\partial}{\partial\eta^{4}}+z\frac{\partial}{\partial\eta^{5}},\frac{\partial}{\partial\eta^{1}}+z\frac{\partial}{\partial\eta^{2}}})
=a4z4+4a3z3+6a2z2+4a1z+a0,\displaystyle=a_{4}z^{4}+4a_{3}z^{3}+6a_{2}z^{2}+4a_{1}z+a_{0},

where

a4=W5225,a3=W4225,a2=W4125,a1=W4124,a0=W4114.a_{4}=W_{5225},\qquad a_{3}=W_{4225},\qquad a_{2}=W_{4125},\qquad a_{1}=W_{4124},\qquad a_{0}=W_{4114}.

Let us point out that the circle bundle 𝒵\mathcal{Z} is not preserved by the action of the full structure group for the geometry of (2,3,5)-distributions. In order to remedy this issue and define 𝒵\mathcal{Z} one can make a choice of splitting for 𝒟\partial\mathscr{D} given by 𝒟=𝒟.\partial\mathscr{D}=\mathscr{D}\oplus\langle\ell\rangle. Such splitting will reduce the structure group and allows one to define 𝒵\mathcal{Z} invariantly. We will see in the next section that twistor distributions arising from pKE metrics are naturally equipped with such splitting therefore enable one to define 𝒵.\mathcal{Z}.

Remark 4.6.

Using Theorem 4.4, the rank 2 distribution 𝒟=Ker{η1,η2,η3}\mathscr{D}=\operatorname{Ker}\{\eta^{1},\eta^{2},\eta^{3}\} is null with respect to the conformal structure [h~][\tilde{h}]. In fact, it has been shown [HS11] that 𝒟\mathscr{D} induces a parallel spin tractor. Conversely, it has been shown that conformal structures of signature (3,2) which are equipped with a parallel spin-tractor arise from the construction of Theorem 4.4. The existence of such parallel objects implies that the conformal holonomy of the conformal structure is a subgroup of 𝐆2\mathbf{G}_{2}^{*} (see [HS09, HS11] for more details.) This is an instance of an extension of the holonomy principle in pseudo-Riemannian geometry, as explained in Remark 2.9, to the context of Cartan geometries (see [ČS09].)

As was mentioned in Remark 2.1, in order to find the Cartan connection (4.2), one can either work with a lifted coframe defined on the bundle 𝒢\mathcal{G} or start with a choice of coframe on the manifold QQ and impose the structure equations to find the Cartan connection in terms of the coframe, which, if needed, can consequently be equivariantly lifted to 𝒢.\mathcal{G}. In this article we will follow the latter approach, as we did for the pKE structures.

4.2. Null self-dual planes and a remarkable coincidence

In this section we show the main result of this article by finding the Cartan connection of the twistor distribution on the space of self-dual null planes and showing that the root type of its Cartan quartic is the same as the root type of Weyl.Weyl^{-}. Furthermore, as a by-product of our construction, one obtains para-Sasaki-Einstein metrics in dimension five and 5-dimensional conformal structures with 𝐒𝐋3()\mathbf{SL}_{3}(\mathbb{R})-holonomy.

4.2.1. Twistor distribution on 𝒩+\mathcal{N}_{+}

It was observed in [AN14] that for a 4-dimensional conformal structure of split signature the circle bundles of self-dual and anti-self-dual null planes, 𝒩+\mathcal{N}_{+} and 𝒩,\mathcal{N}_{-}, are each equipped with a naturally defined rank 2 distribution which is referred to as the twistor distribution. The twistor distribution on 𝒩+\mathcal{N}_{+} and 𝒩\mathcal{N}_{-} is (2,3,5) in an open set UMU\subset M if the self-dual and anti-self-dual Weyl curvature of the conformal structure is nowhere vanishing in UU.

Given a pKE structure, in order to define the twistor distribution on 𝒩+\mathcal{N}_{+} we make use of the parametrization (2.3), where μ{}\mu\in\mathbb{R}\cup\{\infty\}. As a result, on 𝒩+\mathcal{N}_{+} one obtains the 0-adapted coframe

η01=θ1+μθ4,η02=θ2μθ3,η03=dμ,η04=θ4,η05=θ3,\eta^{1}_{0}=\theta^{1}+\mu\theta^{4},\quad\eta^{2}_{0}=\theta^{2}-\mu\theta^{3},\quad\eta_{0}^{3}={\rm d}\mu,\quad\eta^{4}_{0}=\theta^{4},\quad\eta^{5}_{0}=\theta^{3},

where the subscript 0 refers to the adaptation with respect to which the 2-distribution will be defined.

A coframe (η1,,η5)(\eta^{1},\dots,\eta^{5}) defines a (2,3,5) distribution 𝒟=Ker{η1,η2,η3}\mathscr{D}=\operatorname{Ker}\{\eta^{1},\eta^{2},\eta^{3}\} if

(4.8a) dη1\displaystyle{\rm d}\eta^{1} η3η4,mod{η1,η2},\displaystyle\equiv\eta^{3}\wedge\eta^{4},\quad\mathrm{mod}\quad\{\eta^{1},\eta^{2}\},
(4.8b) dη2\displaystyle{\rm d}\eta^{2} η3η5,mod{η1,η2},\displaystyle\equiv\eta^{3}\wedge\eta^{5},\quad\mathrm{mod}\quad\{\eta^{1},\eta^{2}\},
(4.8c) dη3\displaystyle{\rm d}\eta^{3} η4η5,mod{η1,η2,η3}.\displaystyle\equiv\eta^{4}\wedge\eta^{5},\quad\mathrm{mod}\quad\{\eta^{1},\eta^{2},\eta^{3}\}.

To define the twistor distribution on 𝒩\mathcal{N} we further adapt the coframe {η01,,η05}\{\eta^{1}_{0},\dots,\eta^{5}_{0}\} so that it satisfies (4.8). Using the structure equations (3.1), one obtains

dη01\displaystyle{\rm d}\eta^{1}_{0} (η03+μΓ22+μΓ11)η04,anddη02\displaystyle\equiv(\eta^{3}_{0}+\mu\Gamma^{2}_{~{}2}+\mu\Gamma^{1}_{~{}1})\wedge\eta^{4}_{0},\quad\mathrm{and}\quad{\rm d}\eta^{2}_{0} (η03+μΓ11+μΓ22)η05,\displaystyle\equiv-(\eta^{3}_{0}+\mu\Gamma^{1}_{~{}1}+\mu\Gamma^{2}_{~{}2})\wedge\eta^{5}_{0},

modulo {η01,η02}.\{\eta^{1}_{0},\eta^{2}_{0}\}. In order to obtain the relations (4.8a),(4.8b) define the 1-adapted coframe as

(4.9) η11=θ1+μθ4,η12=θ2μθ3,η13=dμ+μΓ11+μΓ22,η14=θ4,η15=θ3\eta^{1}_{1}=\theta^{1}+\mu\theta^{4},\quad\eta^{2}_{1}=\theta^{2}-\mu\theta^{3},\quad\eta_{1}^{3}={\rm d}\mu+\mu\Gamma^{1}_{~{}1}+\mu\Gamma^{2}_{~{}2},\quad\eta^{4}_{1}=\theta^{4},\quad\eta^{5}_{1}=-\theta^{3}

Using (3.1), the 1-adapted coframe satisfies

(4.10) dη11\displaystyle{\rm d}\eta^{1}_{1} η13η14,\displaystyle\equiv\eta^{3}_{1}\wedge\eta^{4}_{1},\quad mod{η11,η12},\displaystyle\mathrm{mod}\quad\{\eta^{1}_{1},\eta^{2}_{1}\},
dη12\displaystyle{\rm d}\eta^{2}_{1} η13η15,\displaystyle\equiv\eta^{3}_{1}\wedge\eta^{5}_{1},\quad mod{η11,η12},\displaystyle\mathrm{mod}\quad\{\eta^{1}_{1},\eta^{2}_{1}\},
dη13\displaystyle{\rm d}\eta^{3}_{1} 6μ2Ψ2η14η15,\displaystyle\equiv-6\mu^{2}\,\Psi_{2}^{\prime}\,\eta^{4}_{1}\wedge\eta^{5}_{1},\quad mod{η11,η12,η13}.\displaystyle\mathrm{mod}\quad\{\eta^{1}_{1},\eta^{2}_{1},\eta^{3}_{1}\}.

Using the fact that Ψ20\Psi^{\prime}_{2}\neq 0 and dΨ2=0,{\rm d}\Psi^{\prime}_{2}=0, the 2-adapted coframe defined by

(4.11) η21=16μ2Ψ2(θ1+μθ4),η22=16μ2Ψ2(θ2μθ3),η23=16μΨ2(dμμ+Γ11+Γ22),η24=θ4,η25=θ3\begin{gathered}\textstyle{\eta^{1}_{2}=\frac{-1}{6\mu^{2}\,\Psi_{2}^{\prime}}\left(\theta^{1}+\mu\theta^{4}\right),\qquad\eta^{2}_{2}=\frac{-1}{6\mu^{2}\,\Psi_{2}^{\prime}}\left(\theta^{2}-\mu\theta^{3}\right)},\\ \textstyle{\eta_{2}^{3}=\frac{-1}{6\mu\,\Psi_{2}^{\prime}}\left(\frac{{\rm d}\mu}{\mu}+\Gamma^{1}_{~{}1}+\Gamma^{2}_{~{}2}\right),\qquad\eta^{4}_{2}=\theta^{4},\qquad\eta^{5}_{2}=-\theta^{3}}\end{gathered}

satisfies (4.8), therefore, defines a (2,3,5)-distribution,

(4.12) 𝒟:=Ker{η21,η22,η23},\mathscr{D}:=\mathrm{Ker}\{\eta^{1}_{2},\eta^{2}_{2},\eta^{3}_{2}\},

on 𝒩+\mathcal{N}_{+} for μ\mu\in\mathbb{R}^{*}. It is straightforward to show that 𝒟\mathscr{D} is invariant under the induced action of the structure group 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R}). We have the following theorem.

Theorem 4.7.

The 𝕊1\mathbb{S}^{1}-bundle of self-dual null planes of a pKE metric, 𝒩+,\mathcal{N}_{+}, is naturally equipped with a rank 2 distribution 𝒟.\mathscr{D}. If the scalar curvature of the pKE metric is non-zero, the twistor distribution 𝒟\mathscr{D} is (2,3,5) on the open subset 𝒩+=𝒩+\{,¯}{\mathcal{N}^{*}_{+}}=\mathcal{N}_{+}\backslash\{\mathscr{H},\bar{\mathscr{H}}\} where {,¯}𝒩+\{\mathscr{H},\bar{\mathscr{H}}\}\subset\mathcal{N}_{+} are the ±1\pm 1-eigenspaces of the para-complex structure K,K, as defined in (2.12). In terms of the affine parameter μ\mu in (2.3), the open subset 𝒩+{\mathcal{N}^{*}_{+}} corresponds to μ,\mu\in\mathbb{R}^{*}, and the conformal structure of signature (3,2)(3,2) associated to the twistor distribution, as in Theorem 4.4, is given by [h][h] where

(4.13) h=16μ2Ψ2(θ1θ3+θ2θ4)+154μ2(Ψ2)2(dμμ+Γ11+Γ22)2.h=\textstyle{\frac{1}{6\mu^{2}\Psi^{\prime}_{2}}(\theta^{1}\theta^{3}+\theta^{2}\theta^{4})+\frac{1}{54\mu^{2}(\Psi^{\prime}_{2})^{2}}(\frac{\mathrm{d}\mu}{\mu}+\Gamma^{1}_{~{}1}+\Gamma^{2}_{~{}2})^{2}}.
Proof.

Using the coframe (4.9), the relations (4.10) imply that the 2-plane field Ker{η11,η12,η13}\operatorname{Ker}\{\eta^{1}_{1},\eta^{2}_{1},\eta^{3}_{1}\} is integrable on the hypersurfaces corresponding to μ=0\mu=0 and μ=,\mu=\infty, which by (2.3) are given by Ker{α1,α2}\mathrm{Ker}\{\alpha^{1},\alpha^{2}\} and Ker{α¯1,α¯2}\mathrm{Ker}\{\bar{\alpha}^{1},\bar{\alpha}^{2}\} i.e. 2-plane fields \mathscr{H} and ¯\bar{\mathscr{H}}. For μ>0\mu>0 and μ<0\mu<0 the twistor distribution is (2,3,5) by (4.12). Consequently, using the adapted coframe (4.11), the metric (4.5) gives (4.13). ∎

Remark 4.8.

The simple expression (4.13) for the metric hh is the key to what follows in Theorem 4.9. For general twistor distributions the expression for hh involves the second jet of the self-dual Weyl curvature of g,g, whose components are denoted by Ψi\Psi^{\prime}_{i}’s. However, in pKE metrics the only non-zero component of Weyl+Weyl^{+} is the constant Ψ2.\Psi^{\prime}_{2}. This point is explained further in § 4.3, in particular Remark 4.15.

Using the theorem above, one obtains the following theorem which is the main result of this section.

Theorem 4.9.

Given a pKE metric with non-zero scalar curvature, the Cartan quartic C(z)C(z) for the twistor distribution 𝒟T𝒩+\mathscr{D}\subset\mathrm{T}\mathcal{N}_{+} on 𝒩+=𝒩+\{,¯}{\mathcal{N}^{*}_{+}}=\mathcal{N}_{+}\backslash\{\mathscr{H},\bar{\mathscr{H}}\} is a non-zero multiple of the quartic representation of the anti-self-dual Weyl curvature W(z)W(z). More explicitly, one obtains

C(z)=6μ2Ψ2(Ψ0+4Ψ1z+6Ψ2z2+4Ψ3z3+Ψ4z4)=6μ2Ψ2W(z).C(z)=-6\mu^{2}\Psi_{2}^{\prime}\left(\Psi_{0}+4\Psi_{1}z+6\Psi_{2}z^{2}+4\Psi_{3}z^{3}+\Psi_{4}z^{4}\right)=-6\mu^{2}\Psi^{\prime}_{2}W(z).

In particular, the root types of the Cartan quartic and the anti-self-dual Weyl curvature coincide.

Remark 4.10.

We point out that there is an underlying bundle map behind Theorem 4.9 which allows one to express C(z)C(z) as a non-zero multiple of W(z)W(z). As discussed in § 2.2.4, W(z)W(z) is defined on 𝒩\mathcal{N}_{-} and, by Remark 4.5, C(z)C(z) is defined on 𝒵.\mathcal{Z}. However, using the twistorial nature of 𝒟\mathscr{D} one can naturally identify 𝒩\mathcal{N}_{-} and 𝒵\mathcal{Z} in the following way. First note that given a pKE structure, the derived system 𝒟:=Ker{η21,η22}\partial\mathscr{D}:=\operatorname{Ker}\{\eta^{1}_{2},\eta^{2}_{2}\} is equipped with a splitting 𝒟η23\mathscr{D}\oplus\langle\frac{\partial}{\partial\eta^{3}_{2}}\rangle which is invariant under the induced action of 𝐆𝐋2().\mathbf{GL}_{2}(\mathbb{R}). Therefore, by Remark 4.5, the circle bundle 𝒵,\mathcal{Z}, as defined in (4.7), is well-defined on 𝒩+{\mathcal{N}^{*}_{+}}. Consequently, via the bundle map ν+:𝒩+M,\nu_{+}\colon{\mathcal{N}^{*}_{+}}\to M, it is elementary to check that dν+(𝒵q)=𝒩p,\mathrm{d}\nu_{+}(\mathcal{Z}_{q})=\mathcal{N}_{p-}, where p=ν+(q),p=\nu_{+}(q), for all q𝒵,q\in\mathcal{Z}, using definitions (2.3) and (4.6). Therefore, the property of W(z)W(z) and C(z)C(z) being proportional everywhere is well-defined.

Proof of Theorem 4.9.

Having the twistor distribution 𝒟\mathscr{D} defined by (4.12) on 𝒩+{\mathcal{N}^{*}_{+}}, it is straightforward to find the explicit Cartan connection (4.2) for 𝒟\mathscr{D} starting with the 2-adapted coframe (4.11). Consequently, one obtains that the Cartan connection (4.2) for 𝒟\mathscr{D} is given by

(4.14) η1\displaystyle\eta^{1} =16μ2Ψ2(θ1+μθ4),\displaystyle=\textstyle{\frac{-1}{6\mu^{2}\,\Psi_{2}^{\prime}}\left(\theta^{1}+\mu\theta^{4}\right)},\qquad η2=16μ2Ψ2(θ2μθ3),\displaystyle\eta^{2}=\textstyle{\frac{-1}{6\mu^{2}\,\Psi_{2}^{\prime}}\left(\theta^{2}-\mu\theta^{3}\right)},
η3\displaystyle\eta^{3} =16μ2Ψ2(dμ+μΓ11+μΓ22),\displaystyle=\textstyle{\frac{-1}{6\mu^{2}\,\Psi_{2}^{\prime}}\left({\rm d}\mu+\mu\Gamma^{1}_{~{}1}+\mu\Gamma^{2}_{~{}2}\right)},\qquad η4=θ4,\displaystyle\eta^{4}=\theta^{4},\qquad η5=θ3,\displaystyle\eta^{5}=-\theta^{3},
ζ1\displaystyle\zeta_{1} =13(2Γ11Γ22+2dμμ),\displaystyle=\textstyle{\frac{1}{3}(2\Gamma^{1}_{~{}1}-\Gamma^{2}_{~{}2}+2\frac{{\rm d}\mu}{\mu})},\quad ζ2=Γ21,\displaystyle\zeta_{2}=\Gamma^{1}_{~{}2},\qquad ζ3=Γ12,\displaystyle\zeta_{3}=\Gamma^{2}_{~{}1},
ζ4\displaystyle\zeta_{4} =13(2Γ22Γ11+2dμμ),\displaystyle=\textstyle{\frac{1}{3}(2\Gamma^{2}_{~{}2}-\Gamma^{1}_{~{}1}+2\frac{{\rm d}\mu}{\mu}),} ζ5=3μΨ2θ3,\displaystyle\zeta_{5}=-3\mu\Psi^{\prime}_{2}\theta^{3},\qquad ζ6=3μΨ2θ4,\displaystyle\zeta_{6}=-3\mu\Psi^{\prime}_{2}\theta^{4},
ζ7\displaystyle\zeta_{7} =0,\displaystyle=0,\qquad ζ8=6μ2(Ψ2)2θ3,\displaystyle\zeta_{8}=6\mu^{2}(\Psi^{\prime}_{2})^{2}\theta^{3},\qquad ζ9=6μ2(Ψ2)2θ4.\displaystyle\zeta_{9}=6\mu^{2}(\Psi^{\prime}_{2})^{2}\theta^{4}.

Using the Cartan connection, one computes the curvature

K𝐆2=dω𝐆2+ω𝐆2ω𝐆2.K_{\mathbf{G}^{*}_{2}}=\mathrm{d}\omega_{\mathbf{G}^{*}_{2}}+\omega_{\mathbf{G}^{*}_{2}}{\,{\wedge}\;}\omega_{\mathbf{G}^{*}_{2}}.

Therefore, the coefficients a0,,a4a_{0},\dots,a_{4} of the Cartan quartic (4.4) are found to be

(4.15) ai=6μ2Ψ2Ψi,i=0,,4.a_{i}=-6\mu^{2}\Psi^{\prime}_{2}\Psi_{i},\quad i=0,\dots,4.

By the formulas for C(z)C(z) and W(z),W(z), as given in (4.15) and (3.7), it follows that

C(z)=6μ2Ψ2W(z).C(z)=-6\mu^{2}\Psi^{\prime}_{2}W(z).

Remark 4.11.

As was mentioned in the introduction, it was not known whether (2,3,5) distributions that arise as a twistor distribution of split signature metrics can have any fixed (Petrov) root type. Theorem 4.9 shows that any root type can be achieved via pKE metrics with non-zero scalar curvature for which WeylWeyl^{-} has the same root type. Moreover, our examples in § 3.2 provide explicit metrics whose twistor distributions have Cartan quartics of real root type II,III,N,DII,III,N,D and O.O.

4.2.2. An invariant description

To have an invariant understanding of 𝒩+=𝒩+\{,¯},{\mathcal{N}^{*}_{+}}=\mathcal{N}_{+}\backslash\{\mathscr{H},\bar{\mathscr{H}}\}, we define another space equipped with a rank 2 distribution which will be shown to be isomorphic to the twistor distribution on 𝒩+.{\mathcal{N}^{*}_{+}}. Consider the principal 𝐆𝐋2()\mathbf{GL}_{2}(\mathbb{R})-bundle 8\mathcal{F}^{8} of adapted null coframes for pKE metrics, equipped with the Cartan connection 𝒜,\mathcal{A}, as in (3.2). Using the structure equations (3.1), one can define the 5-dimensional leaf space, 𝒬,\mathcal{Q}, of the Pfaffian system

I={θ1,θ2,θ3,θ4,Γ11+Γ22}.I=\{\theta^{1},\theta^{2},\theta^{3},\theta^{4},\Gamma^{1}_{~{}1}+\Gamma^{2}_{~{}2}\}.

Note that by the structure equations (3.1) the Pfaffian system II is integrable and its 5-dimensional leaf space 𝒬\mathcal{Q} is the quotient of 8\mathcal{F}^{8} by the orbits of 𝐒𝐋2()𝐆𝐋2(),\mathbf{SL}_{2}(\mathbb{R})\subset\mathbf{GL}_{2}(\mathbb{R}), which would give an \mathbb{R}^{*}-bundle over MM. As a result, 𝒬\mathcal{Q} is a cone. Similar to our previous discussion, if Ψ20,\Psi^{\prime}_{2}\neq 0, after necessary adaptation, one obtains the following coframe on 𝒬\mathcal{Q}

(4.16) η1=16Ψ2(θ1+θ4),η2=16Ψ2(θ2θ3),η3=16Ψ2(Γ11+Γ22),η4=θ4,η5=θ3,\begin{gathered}\textstyle{\eta^{1}=\frac{-1}{6\,\Psi_{2}^{\prime}}\left(\theta^{1}+\theta^{4}\right),\quad\eta^{2}=\frac{-1}{6\,\Psi_{2}^{\prime}}\left(\theta^{2}-\theta^{3}\right)},\\ \textstyle{\eta^{3}=\frac{-1}{6\,\Psi_{2}^{\prime}}\left(\Gamma^{1}_{~{}1}+\Gamma^{2}_{~{}2}\right),\quad\eta^{4}=\theta^{4},\quad\eta^{5}=-\theta^{3}},\end{gathered}

which defines a rank 2 distribution 𝒟~T𝒬,\tilde{\mathscr{D}}\subset\mathrm{T}\mathcal{Q}, by 𝒟~=Ker{η1,η2,η3}.\tilde{\mathscr{D}}=\operatorname{Ker}\{\eta^{1},\eta^{2},\eta^{3}\}. The distribution 𝒟~\tilde{\mathscr{D}} is invariant under the induced action of the structure group 𝐆𝐋2().\mathbf{GL}_{2}(\mathbb{R}).

The Cartan connection for the 𝒟~\tilde{\mathscr{D}} is obtainable from (4.14) by setting μ=1.\mu=1. In the language of parabolic geometries [ČS09], this is the explicit form of the extension functor from a pKE structure to the corresponding (2,3,5)-distribution.

To relate our discussion to 𝒩+=𝒩+\{,¯},{\mathcal{N}^{*}_{+}}=\mathcal{N}_{+}\backslash\{\mathscr{H},\bar{\mathscr{H}}\}, note that the structure group of a pKE metric can be decomposed as

𝐆𝐋2()=𝐒𝐋2()×.\mathbf{GL}_{2}(\mathbb{R})=\mathbf{SL}_{2}(\mathbb{R})\times\mathbb{R}^{*}.

The \mathbb{R}^{*}-action transforms the pKE coframe in the following way

θ1sθ1,θ2sθ2,θ31sθ3,θ11sθ4,\theta^{1}\rightarrow s\theta^{1},\quad\theta^{2}\rightarrow s\theta^{2},\quad\theta^{3}\rightarrow\textstyle{\frac{1}{s}\theta^{3}},\quad\textstyle{\theta^{1}\rightarrow\frac{1}{s}\theta^{4}},

by setting a11=a22=sa_{11}=a_{22}=s and a12=a21=0a_{12}=a_{21}=0 in the structure group (2.16). This action results in the following change of coframe (4.11) on 𝒩+{\mathcal{N}^{*}_{+}}

(4.17) η21=16μ2Ψ2(sθ1+μsθ4),η22=16μ2Ψ2(sθ2μsθ3),η23=16μΨ2(dμμdss+Γ11+Γ22),η24=1sθ4,η25=1sθ3.\begin{gathered}\textstyle{\eta^{1}_{2}=\frac{-1}{6\mu^{2}\,\Psi_{2}^{\prime}}\left(s\theta^{1}+\frac{\mu}{s}\theta^{4}\right),\qquad\eta^{2}_{2}=\frac{-1}{6\mu^{2}\,\Psi_{2}^{\prime}}\left(s\theta^{2}-\frac{\mu}{s}\theta^{3}\right)},\\ \textstyle{\eta_{2}^{3}=\frac{-1}{6\mu\,\Psi_{2}^{\prime}}\left(\frac{{\rm d}\mu}{\mu}-\frac{\mathrm{d}s}{s}+\Gamma^{1}_{~{}1}+\Gamma^{2}_{~{}2}\right),\qquad\eta^{4}_{2}=\frac{1}{s}\theta^{4},\qquad\eta^{5}_{2}=-\frac{1}{s}\theta^{3}}.\end{gathered}

Using the expressions (4.17) one finds the bundle isomorphism τ:𝒩+𝒬\tau\colon{\mathcal{N}^{*}_{+}}\to\mathcal{Q} in terms of the fiber coordinates ss and μ,\mu, given by τ(μ)=1s2\tau(\mu)=\frac{1}{s^{2}} for s>0s>0 and τ(μ)=1s2\tau(\mu)=-\frac{1}{s^{2}} for s<0,s<0, via which 𝒟~=τ𝒟.\tilde{\mathscr{D}}=\tau_{*}\mathscr{D}. This gives the equivalence of the (2,3,5) distributions induced on 𝒩+{\mathcal{N}^{*}_{+}} and 𝒬\mathcal{Q}

4.2.3. para-Sasaki-Einstein structures and 𝐒𝐋3()\mathbf{SL}_{3}(\mathbb{R}) holonomy

To state the last result of this section we give the following definition of a para-Sasaki-Einstein structure.

Definition 4.12.

A para-Sasaki-Einstein structure on 𝒬\mathcal{Q} is (ϕ,ξ,β,h),(\phi,\xi,\beta,h), where ϕ:T𝒬T𝒬\phi:\mathrm{T}\mathcal{Q}\to\mathrm{T}\mathcal{Q} is an endomorphism, ξ\xi is a vector field, β\beta is a 1-form, and hh is a split signature metric. The quadruple (ϕ,ξ,β,h)(\phi,\xi,\beta,h) satisfies

(4.18) ϕ2=Idβξ,β(ξ)=1,ϕ(ξ)=0,βϕ=0,h(ξ,.)=β.\phi^{2}=\mathrm{Id}-\beta\otimes\xi,\quad\beta(\xi)=1,\quad\phi(\xi)=0,\quad\beta\circ\phi=0,\quad h(\xi,.)=\beta.

Additionally, the ±1\pm 1-eigenspaces of ϕ\phi define rank 2 integrable sub-distributions of Ker(β),\mathrm{Ker}(\beta), the metric hh is Einstein, and the compatibility conditions

(4.19) h(ϕX,ϕY)\displaystyle h(\phi X,\phi Y) =h(X,Y)+β(X)β(Y),\displaystyle=-h(X,Y)+\beta(X)\beta(Y),
dβ(X,Y)\displaystyle\mathrm{d}\beta(X,Y) =h(ϕX,Y)\displaystyle=h(\phi X,Y)\qquad X,YT𝒬\displaystyle\forall X,Y\in\mathrm{T}\mathcal{Q}

hold.

We have the following proposition.

Proposition 4.13.

The cone 𝒬\mathcal{Q} is equipped with a para-Sasaki-Einstein structure arising from the para-Kähler-Einstein structure on M.M.

Proof.

Using the coframe (4.16), the claimed para-Sasakian structure on 𝒬\mathcal{Q} is given by

ϕ\displaystyle\phi =η1η1+η2η2η4η4η5η5\displaystyle=\textstyle{\eta^{1}\otimes\frac{\partial}{\partial\eta^{1}}+\eta^{2}\otimes\frac{\partial}{\partial\eta^{2}}-\eta^{4}\otimes\frac{\partial}{\partial\eta^{4}}-\eta^{5}\otimes\frac{\partial}{\partial\eta^{5}}}
ξ\displaystyle\xi =32η3\displaystyle=\textstyle{\sqrt{\frac{3}{2}}\frac{\partial}{\partial\eta^{3}}}
β\displaystyle\beta =23η3\displaystyle=\textstyle{\sqrt{\frac{2}{3}}\eta^{3}}
h\displaystyle h =η1η5η2η4+23(η3)2\displaystyle=\eta^{1}\eta^{5}-\eta^{2}\eta^{4}+\textstyle{\frac{2}{3}}(\eta^{3})^{2}

The Levi-Civita connection of hh with respect to the coframe (4.16) is

(4.20) (Γ224Ψ2η3Γ212Ψ2η1+23η413η30Γ12Γ22+2Ψ2η32Ψ2η2+23η5013η332Ψ2η532Ψ2η4032Ψ2η2+12η532Ψ2η112η4002Ψ2η42Ψ2η3Γ22Γ21002Ψ2η5Γ124Ψ2η3+Γ22)\begin{pmatrix}-\Gamma^{2}_{~{}2}-4\Psi^{\prime}_{2}\eta^{3}&\Gamma^{1}_{~{}2}&2\Psi^{\prime}_{2}\eta^{1}+\frac{2}{3}\eta^{4}&-\frac{1}{3}\eta^{3}&0\\ \Gamma^{2}_{~{}1}&\Gamma^{2}_{~{}2}+2\Psi^{\prime}_{2}\eta^{3}&2\Psi^{\prime}_{2}\eta^{2}+\frac{2}{3}\eta^{5}&0&-\frac{1}{3}\eta^{3}\\ \frac{3}{2}\Psi^{\prime}_{2}\eta^{5}&-\frac{3}{2}\Psi^{\prime}_{2}\eta^{4}&0&\frac{3}{2}\Psi^{\prime}_{2}\eta^{2}+\frac{1}{2}\eta^{5}&-\frac{3}{2}\Psi^{\prime}_{2}\eta^{1}-\frac{1}{2}\eta^{4}\\ 0&0&-2\Psi_{2}^{\prime}\eta^{4}&-2\Psi^{\prime}_{2}\eta^{3}-\Gamma^{2}_{~{}2}&\Gamma^{1}_{~{}2}\\ 0&0&-2\Psi^{\prime}_{2}\eta^{5}&\Gamma^{2}_{~{}1}&4\Psi^{\prime}_{2}\eta^{3}+\Gamma^{2}_{~{}2}\end{pmatrix}

which is 𝔰𝔬3,2\mathfrak{so}_{3,2}-valued with respect to the matrix representation of the metric hh

(00001000100043000100010000).\begin{pmatrix}0&0&0&0&1\\ 0&0&0&-1&0\\ 0&0&\frac{4}{3}&0&0\\ 0&-1&0&0&0\\ 1&0&0&0&0\end{pmatrix}.

It follows that the metric hh on 𝒬\mathcal{Q} is Einstein whose Einstein constant is 24(Ψ2)2.-24(\Psi^{\prime}_{2})^{2}.

Remark 4.14.

The proposition above is an analogue of the well-known construction of Sasakian structures from Kähler metrics [GKN00]. Moreover, by the same analogy, following [GKN00], it can be checked that using the potential function of a para-Kähler-Einstein metric and the coordinate ss on the cone, as introduced in § 4.2.2, one obtains a potential function for the resulting para-Sasaki-Einstein metric on 𝒬.\mathcal{Q}.

Finally one can directly verify the well-known relation between the existence of an Einstein representative in the conformal structure [h][h] of the metric (4.13) defined by the (2,3,5) distribution, and the holonomy reduction of the Cartan conformal connection. This relation goes back to works of many people, including [Sas43] (see Section 5.2 in [ČS09] for an overview.) One can check that the Cartan connection (4.2) given by (4.14) takes value in 𝔰𝔩3()𝔤2𝔰𝔬4,3\mathfrak{sl}_{3}(\mathbb{R})\subset\mathfrak{g}_{2}\subset\mathfrak{so}_{4,3} for any non-zero value of Ψ2.\Psi^{\prime}_{2}. This implies that the conformal holonomy of the conformal structure [h][h] is reduced and the conformal class [h][h] must contain at least one Einstein metric. We refer the reader to [SW17] for a more detailed study of conformally Einstein structures arising from (2,3,5) distributions and their corresponding conformal holonomy reductions.

4.3. Null anti-self-dual planes

The purpose of this short section is to justify the remarkable nature of Theorem 4.9 by examining the twistor distribution induced on the 𝕊1\mathbb{S}^{1}-bundle of anti-self-dual null planes of a pKE metric, 𝒩.\mathcal{N}_{-}. Following the description in § 4.2.2, one can identify 𝒩\mathcal{N}_{-} as the leaf space of the Pfaffian system

Iasd={θ1,θ2,θ3,θ4,Γ21}.I_{asd}=\{\theta^{1},\theta^{2},\theta^{3},\theta^{4},\Gamma^{1}_{~{}2}\}.

Consequently, if Weyl0,Weyl^{-}\neq 0, one can follow the discussion in § 4.2.1 to find an adapted coframe. An adapted coframe for the twistor distribution satisfying (4.8) on the open set of 𝒩\mathcal{N}_{-} where Ψ40\Psi_{4}\neq 0 is given by

η1\displaystyle\eta^{1} =1Ψ4(α1+α¯2)η2=1Ψ4α¯2,η3=1Ψ4Γ21,\displaystyle=\textstyle{\frac{1}{\Psi_{4}}(\alpha^{1}+\bar{\alpha}^{2})\quad\eta^{2}=\frac{1}{\Psi_{4}}\bar{\alpha}^{2},\quad\eta^{3}=\frac{1}{\Psi_{4}}\Gamma^{1}_{2}},
η4\displaystyle\eta^{4} =α¯1α230Ψ3Ψ42+3Ψ4Ψ422Ψ4Ψ4335Ψ422+4Ψ43230Ψ42α1\displaystyle=\textstyle{\bar{\alpha}^{1}-\alpha^{2}-\frac{30\Psi_{3}\Psi_{4}^{2}+3\Psi_{4}\Psi_{422}-\Psi_{4}\Psi_{433}-5\Psi_{42}^{2}+4\Psi_{43}^{2}}{30\Psi_{4}^{2}}\alpha^{1}}
15Ψ3Ψ423Ψ4Ψ4323Ψ4Ψ4335Ψ42Ψ43+5Ψ43215Ψ42α¯2+Ψ42+Ψ433Ψ42Γ21\displaystyle\textstyle{-\frac{15\Psi_{3}\Psi_{4}^{2}-3\Psi_{4}\Psi_{432}-3\Psi_{4}\Psi_{433}-5\Psi_{42}\Psi_{43}+5\Psi_{43}^{2}}{15\Psi_{4}^{2}}\bar{\alpha}^{2}+\frac{\Psi_{42}+\Psi_{43}}{3\Psi_{4}^{2}}\Gamma^{1}_{2}}
η5\displaystyle\eta^{5} =α¯13Ψ4Ψ4225Ψ42230Ψ43α1Ψ423Ψ42Γ21\displaystyle=\textstyle{\bar{\alpha}^{1}-\frac{3\Psi_{4}\Psi_{422}-5\Psi_{42}^{2}}{30\Psi_{4}^{3}}\alpha^{1}-\frac{\Psi_{42}}{3\Psi_{4}^{2}}\Gamma^{1}_{2}}
30Ψ3Ψ426Ψ4Ψ4323Ψ4Ψ433+10Ψ43Ψ42+5Ψ43230Ψ43α¯2.\displaystyle\textstyle{-\frac{30\Psi_{3}\Psi_{4}^{2}-6\Psi_{4}\Psi_{432}-3\Psi_{4}\Psi_{433}+10\Psi_{43}\Psi_{42}+5\Psi_{43}^{2}}{30\Psi_{4}^{3}}\bar{\alpha}^{2}}.

Note that if Weyl0,Weyl^{-}\neq 0, one can always find a coframe with respect to which Ψ40.\Psi_{4}\neq 0.

Consequently, the expressions for the coefficients of the Cartan quartic, a0,,a4a_{0},\dots,a_{4}, are found to be very complicated and depend on the 4th jet of Ψi\Psi_{i}’s. For instance, in the coframe introduced above, one obtains

(4.21) a0\displaystyle a_{0} =10Ψ43Ψ4222270Ψ42Ψ42Ψ422249Ψ42Ψ4222+280Ψ4Ψ422Ψ422175Ψ424100Ψ44.\displaystyle=\textstyle{-\frac{10\Psi_{4}^{3}\Psi_{42222}-70\Psi_{4}^{2}\Psi_{42}\Psi_{4222}-49\Psi_{4}^{2}\Psi_{422}^{2}+280\Psi_{4}\Psi_{42}^{2}\Psi_{422}-175\Psi_{42}^{4}}{100\Psi_{4}^{4}}}.

The equation arising from the vanishing of a0a_{0} is referred to as Noth’s equation and its general solution can be presented by a certain family of rational sextics [AN18, DS11].

We could not identify any relation between the root types of Weyl±Weyl^{\pm} and the root type of the Cartan quartic of the twistor distribution on 𝒩,\mathcal{N}_{-}, with the exception of the homogeneous pKE metrics of Petrov type DrD^{r} where they coincide.

Remark 4.15.

Following our approach in this section, one can find the Cartan quartic for the twistor distribution on 𝒩\mathcal{N}_{-} and 𝒩+\mathcal{N}_{+} for any conformal structure of split signature provided that Weyl0Weyl^{-}\neq 0 and Weyl+0,Weyl^{+}\neq 0, respectively. The computations are extremely tedious. Restricting to 𝒩\mathcal{N}_{-}, if Weyl0,Weyl^{-}\neq 0, one obtains that the expression for the metric h,h, as defined in (4.5), involves the second jets of Weyl.Weyl^{-}. Furthermore, in an appropriate coframe, aia_{i}’s can be expressed in terms of the fourth jets of WeylWeyl^{-} and zeroth jets of Weyl+Weyl^{+} at each point. For instance, consider conformal structures [g],[g], where g=2θ1θ3+2θ2θ4,g=2\theta^{1}\theta^{3}+2\theta^{2}\theta^{4}, with Weyl0Weyl^{-}\neq 0 and denote the components of WeylWeyl^{-} and Weyl+Weyl^{+} by Ψi\Psi_{i}’s and Ψi\Psi^{\prime}_{i}’s respectively, as we did in § 2.1.2. Then on the open subset of 𝒩\mathcal{N}_{-} where Ψ40,\Psi_{4}\neq 0, there is an adapted coframe for the twistor distribution with respect to which

a0=Ψ010Ψ43Ψ4222270Ψ42Ψ42Ψ422249Ψ42Ψ4222+280Ψ4Ψ422Ψ422175Ψ424100Ψ44.a_{0}=\textstyle{-\Psi^{\prime}_{0}-\frac{10\Psi_{4}^{3}\Psi_{42222}-70\Psi_{4}^{2}\Psi_{42}\Psi_{4222}-49\Psi_{4}^{2}\Psi_{422}^{2}+280\Psi_{4}\Psi_{42}^{2}\Psi_{422}-175\Psi_{42}^{4}}{100\Psi_{4}^{4}}}.

Comparing the expression above to 4.21, it is clear that the case of twistor distribution induced on 𝒩\mathcal{N}_{-} for pKE metrics is nearly as complicated as in the case of general conformal structures. As a result, one cannot expect any relation between the root types of Weyl±Weyl^{\pm} for a conformal structure and the Cartan quartic of the twistor distribution on 𝒩±.\mathcal{N}_{\pm}.

Appendix

The connection 1-form referred to in § 2.3.3 are given by

(A.1) Γa=a\displaystyle\Gamma^{a}{}_{a}= VaayVbxVaaxVbyVayVbxVaxVbyda+VabyVbxVabxVbyVayVbxVaxVbydb\displaystyle\frac{V_{aay}V_{bx}-V_{aax}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}{\rm d}a+\frac{V_{aby}V_{bx}-V_{abx}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}{\rm d}b
Γa=b\displaystyle\Gamma^{a}{}_{b}= VabyVbxVabxVbyVayVbxVaxVbyda+VbbyVbxVbbxVbyVayVbxVaxVbydb\displaystyle\frac{V_{aby}V_{bx}-V_{abx}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}{\rm d}a+\frac{V_{bby}V_{bx}-V_{bbx}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}{\rm d}b
Γb=a\displaystyle\Gamma^{b}{}_{a}= VaayVaxVaaxVayVayVbx+VaxVbyda+VabyVaxVabxVayVayVbx+VaxVbydb\displaystyle\frac{V_{aay}V_{ax}-V_{aax}V_{ay}}{-V_{ay}V_{bx}+V_{ax}V_{by}}{\rm d}a+\frac{V_{aby}V_{ax}-V_{abx}V_{ay}}{-V_{ay}V_{bx}+V_{ax}V_{by}}{\rm d}b
Γb=b\displaystyle\Gamma^{b}{}_{b}= VabyVaxVabxVayVayVbx+VaxVbyda+VbbyVaxVbbxVayVayVbx+VaxVbydb\displaystyle\frac{V_{aby}V_{ax}-V_{abx}V_{ay}}{-V_{ay}V_{bx}+V_{ax}V_{by}}{\rm d}a+\frac{V_{bby}V_{ax}-V_{bbx}V_{ay}}{-V_{ay}V_{bx}+V_{ax}V_{by}}{\rm d}b
Γx=x\displaystyle\Gamma^{x}{}_{x}= VbxxVayVaxxVbyVayVbxVaxVbydx+VbxyVayVaxyVbyVayVbxVaxVbydy\displaystyle\frac{V_{bxx}V_{ay}-V_{axx}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}{\rm d}x+\frac{V_{bxy}V_{ay}-V_{axy}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}{\rm d}y
Γx=y\displaystyle\Gamma^{x}{}_{y}= VbxyVayVaxyVbyVayVbxVaxVbydx+VbyyVayVaxyVbyVayVbxVaxVbydy\displaystyle\frac{V_{bxy}V_{ay}-V_{axy}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}{\rm d}x+\frac{V_{byy}V_{ay}-V_{axy}V_{by}}{V_{ay}V_{bx}-V_{ax}V_{by}}{\rm d}y
Γy=x\displaystyle\Gamma^{y}{}_{x}= VbxxVaxVaxxVbxVayVbx+VaxVbydx+VbxyVaxVaxyVbxVayVbx+VaxVbydy\displaystyle\frac{V_{bxx}V_{ax}-V_{axx}V_{bx}}{-V_{ay}V_{bx}+V_{ax}V_{by}}{\rm d}x+\frac{V_{bxy}V_{ax}-V_{axy}V_{bx}}{-V_{ay}V_{bx}+V_{ax}V_{by}}{\rm d}y
Γy=y\displaystyle\Gamma^{y}{}_{y}= VbxyVaxVaxyVbxVayVbx+VaxVbydx+VbyyVaxVayyVbxVayVbx+VaxVbydy.\displaystyle\frac{V_{bxy}V_{ax}-V_{axy}V_{bx}}{-V_{ay}V_{bx}+V_{ax}V_{by}}{\rm d}x+\frac{V_{byy}V_{ax}-V_{ayy}V_{bx}}{-V_{ay}V_{bx}+V_{ax}V_{by}}{\rm d}y.

If (M,g,K)(M,g,K) is a para-Kähler-Einstein structure written in a null adapted frame satisfying (3.1) then the derivatives of the non-vanishing curvature coefficients are given by

(A.2a) dΨ2=0,\displaystyle{\rm d}\Psi_{2}^{\prime}=0,
(A.2b) dΨ0=2Ψ0(Γ11Γ2)2+4Ψ1Γ2+1Ψ01α1+Ψ11α2Ψ14α¯+1Ψ04α¯,2\displaystyle{\rm d}\Psi_{0}=2\Psi_{0}(\Gamma^{1}{}_{1}-\Gamma^{2}{}_{2})+4\Psi_{1}\Gamma^{2}{}_{1}+\Psi_{01}\alpha^{1}+\Psi_{11}\alpha^{2}-\Psi_{14}\bar{\alpha}{}^{1}+\Psi_{04}\bar{\alpha}{}^{2},
(A.2c) dΨ1=Ψ1(Γ11Γ2)2+Ψ0Γ1+23Ψ2Γ2+1Ψ11α1+Ψ21α2Ψ24α¯+1Ψ14α¯,2\displaystyle{\rm d}\Psi_{1}=\Psi_{1}(\Gamma^{1}{}_{1}-\Gamma^{2}{}_{2})+\Psi_{0}\Gamma^{1}{}_{2}+3\Psi_{2}\Gamma^{2}{}_{1}+\Psi_{11}\alpha^{1}+\Psi_{21}\alpha^{2}-\Psi_{24}\bar{\alpha}{}^{1}+\Psi_{14}\bar{\alpha}{}^{2},
(A.2d) dΨ2=2Ψ1Γ1+22Ψ3Γ2+1Ψ21α1+Ψ31α2Ψ34α¯+1Ψ24α¯,2\displaystyle{\rm d}\Psi_{2}=2\Psi_{1}\Gamma^{1}{}_{2}+2\Psi_{3}\Gamma^{2}{}_{1}+\Psi_{21}\alpha^{1}+\Psi_{31}\alpha^{2}-\Psi_{34}\bar{\alpha}{}^{1}+\Psi_{24}\bar{\alpha}{}^{2},
(A.2e) dΨ3=Ψ3(Γ11Γ2)2+3Ψ2Γ1+2Ψ4Γ2+1Ψ31α1+Ψ41α2Ψ44α¯+1Ψ34α¯,2\displaystyle{\rm d}\Psi_{3}=-\Psi_{3}(\Gamma^{1}{}_{1}-\Gamma^{2}{}_{2})+3\Psi_{2}\Gamma^{1}{}_{2}+\Psi_{4}\Gamma^{2}{}_{1}+\Psi_{31}\alpha^{1}+\Psi_{41}\alpha^{2}-\Psi_{44}\bar{\alpha}{}^{1}+\Psi_{34}\bar{\alpha}{}^{2},
(A.2f) dΨ4=2Ψ4(Γ11Γ2)2+4Ψ3Γ2+1Ψ41α1+Ψ42α2+Ψ43α¯+1Ψ44α¯,2\displaystyle{\rm d}\Psi_{4}=-2\Psi_{4}(\Gamma^{1}{}_{1}-\Gamma^{2}{}_{2})+4\Psi_{3}\Gamma^{2}{}_{1}+\Psi_{41}\alpha^{1}+\Psi_{42}\alpha^{2}+\Psi_{43}\bar{\alpha}{}^{1}+\Psi_{44}\bar{\alpha}{}^{2},

for some functions Ψia\Psi_{ia} on MM which represent the coframe derivatives of Ψi\Psi_{i}’s.

The differential relations among the quantities appearing in proof of the Proposition 3.8 for Petrov type IIII is as follows.

(A.3) dJ1=\displaystyle{\rm d}J_{1}= J1Γ11J12α1+J12α2+J13α¯+1J1J2α¯,2\displaystyle J_{1}\Gamma^{1}_{~{}1}-J_{1}^{2}\alpha^{1}+J_{12}\alpha^{2}+J_{13}\bar{\alpha}{}^{1}+J_{1}J_{2}\bar{\alpha}{}^{2},
dJ2=\displaystyle{\rm d}J_{2}= J2Γ22J1J2α1+(J13+Ψ2Ψ2)α2+J23α¯+1J22α¯,2\displaystyle-J_{2}\Gamma^{2}_{~{}2}-J_{1}J_{2}\alpha^{1}+(J_{13}+\Psi_{2}-\Psi_{2}^{\prime})\alpha^{2}+J_{23}\bar{\alpha}{}^{1}+J_{2}^{2}\bar{\alpha}{}^{2},
dJ3=\displaystyle{\rm d}J_{3}= J3Γ22+J31α1+J32α2+J33α¯+1J34α¯,2\displaystyle J_{3}\Gamma^{2}_{~{}2}+J_{31}\alpha^{1}+J_{32}\alpha^{2}+J_{33}\bar{\alpha}{}^{1}+J_{34}\bar{\alpha}{}^{2},
dJ4=\displaystyle{\rm d}J_{4}= J4Γ11(J34Ψ2+Ψ2)α1+J42α2+(J42J2J5+J54)α¯+1J44α¯,2\displaystyle-J_{4}\Gamma^{1}_{~{}1}-(J_{34}-\Psi_{2}+\Psi_{2}^{\prime})\alpha^{1}+J_{42}\alpha^{2}+(J_{4}^{2}-J_{2}J_{5}+J_{54})\bar{\alpha}{}^{1}+J_{44}\bar{\alpha}{}^{2},
dJ5=\displaystyle{\rm d}J_{5}= 2J5Γ11+J5Γ22(J33J3J4+J2J6)α1+J52α2+J53α¯+1J54α¯,2\displaystyle-2J_{5}\Gamma^{1}_{~{}1}+J_{5}\Gamma^{2}_{~{}2}-(J_{33}-J_{3}J_{4}+J_{2}J_{6})\alpha^{1}+J_{52}\alpha^{2}+J_{53}\bar{\alpha}{}^{1}+J_{54}\bar{\alpha}{}^{2},
dJ6=\displaystyle{\rm d}J_{6}= J6Γ11+2J6Γ22(J32J32+J1J6)α1+J62α2+\displaystyle-J_{6}\Gamma^{1}_{~{}1}+2J_{6}\Gamma^{2}_{~{}2}-(J_{32}-J_{3}^{2}+J_{1}J_{6})\alpha^{1}+J_{62}\alpha^{2}+
(J52J3J5+J4J6+Ψ4)α¯+1(J42+J1J5J3J4)α¯,2\displaystyle(J_{52}-J_{3}J_{5}+J_{4}J_{6}+\Psi_{4})\bar{\alpha}{}^{1}+(J_{42}+J_{1}J_{5}-J_{3}J_{4})\bar{\alpha}{}^{2},
dΨ2=\displaystyle{\rm d}\Psi_{2}^{\prime}= 0,\displaystyle 0,
dΨ2=\displaystyle{\rm d}\Psi_{2}= 3J1Ψ2α1+3J3Ψ2α2+3J4Ψ2α¯+13J2Ψ2α¯,2\displaystyle-3J_{1}\Psi_{2}\alpha^{1}+3J_{3}\Psi_{2}\alpha^{2}+3J_{4}\Psi_{2}\bar{\alpha}{}^{1}+3J_{2}\Psi_{2}\bar{\alpha}{}^{2},
dΨ4=\displaystyle{\rm d}\Psi_{4}= 2Ψ4Γ11+2Ψ4Γ22(3J6Ψ2+J1Ψ4)α1+Ψ42α2+Ψ43α¯+1(3J5ψ2+J2Ψ4)α¯.2\displaystyle-2\Psi_{4}\Gamma^{1}_{~{}1}+2\Psi_{4}\Gamma^{2}_{~{}2}-(3J_{6}\Psi_{2}+J_{1}\Psi_{4})\alpha^{1}+\Psi_{42}\alpha^{2}+\Psi_{43}\bar{\alpha}{}^{1}+(3J_{5}\psi_{2}+J_{2}\Psi_{4})\bar{\alpha}{}^{2}.

The curvature 2-form for the Cartan connection \mathcal{B} in (3.13) is given by

(A.4) KB\displaystyle K_{B} =\displaystyle=
([c|c]12(J2J6J1J5)32|Ψ2|J5012(J1J5J2J6)0012(J1J5J2J6)032|Ψ2|J612(J2J6J1J5))σ2+\displaystyle\begin{pmatrix}[c|c]\begin{matrix}\tfrac{1}{2}(J_{2}J_{6}-J_{1}J_{5})&-\sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}J_{5}\\ 0&\tfrac{1}{2}(J_{1}J_{5}-J_{2}J_{6})\end{matrix}&0\\ \cline{1-2}\cr 0&\begin{matrix}\tfrac{1}{2}(J_{1}J_{5}-J_{2}J_{6})&0\\ -\sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}J_{6}&\tfrac{1}{2}(J_{2}J_{6}-J_{1}J_{5})\end{matrix}\end{pmatrix}\sigma^{2}_{-}+
([c|c]14(J2J3J1J42Ψ2+2Ψ2)38|Ψ2|J438|Ψ2|J114(J1J4J2J3+2Ψ22Ψ2)0014(J1J4J2J3+2Ψ22Ψ2)38|Ψ2|J238|Ψ2|J314(J2J3J1J42Ψ2+2Ψ2))σ3+\displaystyle\begin{pmatrix}[c|c]\begin{matrix}\tfrac{1}{4}(J_{2}J_{3}-J_{1}J_{4}-2\Psi_{2}+2\Psi_{2}^{\prime})&-\sqrt{\tfrac{3}{8}|\Psi_{2}^{\prime}|}~{}J_{4}\\ -\sqrt{\tfrac{3}{8}|\Psi_{2}^{\prime}|}~{}J_{1}&\tfrac{1}{4}(J_{1}J_{4}-J_{2}J_{3}+2\Psi_{2}-2\Psi_{2}^{\prime})\end{matrix}&0\\ \cline{1-2}\cr 0&\begin{matrix}\tfrac{1}{4}(J_{1}J_{4}-J_{2}J_{3}+2\Psi_{2}-2\Psi_{2}^{\prime})&\sqrt{\tfrac{3}{8}|\Psi_{2}^{\prime}|}~{}J_{2}\\ -\sqrt{\tfrac{3}{8}|\Psi_{2}^{\prime}|}~{}J_{3}&\tfrac{1}{4}(J_{2}J_{3}-J_{1}J_{4}-2\Psi_{2}+2\Psi_{2}^{\prime})\end{matrix}\end{pmatrix}\sigma^{3}_{-}+
([c|c]12J1J332|Ψ2|J3012J1J30012J1J332|Ψ2|J1012J1J2)σ+1+([c|c]12J2J4032|Ψ2|J212J2J40012J2J4032|Ψ2|J412J2J4)σ+2+\displaystyle\begin{pmatrix}[c|c]\begin{matrix}\tfrac{1}{2}J_{1}J_{3}&\sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}J_{3}\\ 0&-\tfrac{1}{2}J_{1}J_{3}\end{matrix}&0\\ \cline{1-2}\cr 0&\begin{matrix}-\tfrac{1}{2}J_{1}J_{3}&\sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}J_{1}\\ 0&\tfrac{1}{2}J_{1}J_{2}\end{matrix}\end{pmatrix}\sigma^{1}_{+}+\begin{pmatrix}[c|c]\begin{matrix}\tfrac{1}{2}J_{2}J_{4}&0\\ \sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}J_{2}&-\tfrac{1}{2}J_{2}J_{4}\end{matrix}&0\\ \cline{1-2}\cr 0&\begin{matrix}-\tfrac{1}{2}J_{2}J_{4}&0\\ -\sqrt{\tfrac{3}{2}|\Psi_{2}^{\prime}|}~{}J_{4}&\tfrac{1}{2}J_{2}J_{4}\end{matrix}\end{pmatrix}\sigma^{2}_{+}+
([c|c]14(J2J3+J1J4)38|Ψ2|J438|Ψ2|J114(J2J3+J1J4)0014(J2J3+J1J4)38|Ψ2|J238|Ψ2|J314(J2J3+J1J4))σ+3.\displaystyle\begin{pmatrix}[c|c]\begin{matrix}\tfrac{1}{4}(J_{2}J_{3}+J_{1}J_{4})&\sqrt{\tfrac{3}{8}|\Psi_{2}^{\prime}|}~{}J_{4}\\ \sqrt{\tfrac{3}{8}|\Psi_{2}^{\prime}|}~{}J_{1}&-\tfrac{1}{4}(J_{2}J_{3}+J_{1}J_{4})\end{matrix}&0\\ \cline{1-2}\cr 0&\begin{matrix}-\tfrac{1}{4}(J_{2}J_{3}+J_{1}J_{4})&\sqrt{\tfrac{3}{8}|\Psi_{2}^{\prime}|}~{}J_{2}\\ -\sqrt{\tfrac{3}{8}|\Psi_{2}^{\prime}|}~{}J_{3}&\tfrac{1}{4}(J_{2}J_{3}+J_{1}J_{4})\end{matrix}\end{pmatrix}\sigma^{3}_{+}.

References

  • [ACG16] V. Apostolov, D. M. J. Calderbank, and P. Gauduchon. Ambitoric geometry I: Einstein metrics and extremal ambikähler structures. J. Reine Angew. Math., 721:109–147, 2016.
  • [AG96] M. A. Akivis and V. V. Goldberg. Conformal differential geometry and its generalizations. Pure and Applied Mathematics (New York). John Wiley & Sons Inc., New York, 1996. A Wiley-Interscience Publication.
  • [AN14] D. An and P. Nurowski. Twistor space for rolling bodies. Comm. Math. Phys., 326(2):393–414, 2014.
  • [AN18] D. An and P. Nurowski. Symmetric (2,3,5)(2,3,5) distributions, an interesting ODE of 7th order and Plebański metric. J. Geom. Phys., 126:93–100, 2018.
  • [BBI97] L. Bérard Bergery and A. Ikemakhen. Sur l’holonomie des variétés pseudo-riemanniennes de signature (n,n)(n,n). Bull. Soc. Math. France, 125(1):93–114, 1997.
  • [BH93] R. L. Bryant and L. Hsu. Rigidity of integral curves of rank 22 distributions. Invent. Math., 114(2):435–461, 1993.
  • [BHLN18] G. Bor, L. Hernández Lamoneda, and P. Nurowski. The dancing metric, G2G_{2}-symmetry and projective rolling. Trans. Amer. Math. Soc., 370(6):4433–4481, 2018.
  • [Bry14] R. L. Bryant. Notes on exterior differential systems. arXiv preprint, 2014.
  • [Car10] E. Cartan. Les systèmes de Pfaff, à cinq variables et les équations aux dérivées partielles du second ordre. Ann. Sci. École Norm. Sup. (3), 27:109–192, 1910.
  • [CFG96] V. Cruceanu, P. Fortuny, and P. M. Gadea. A survey on paracomplex geometry. Rocky Mountain J. Math., 26(1):83–115, 1996.
  • [Chu17] A. Chudecki. On some examples of para-Hermite and para-Kähler Einstein spaces with Λ0\Lambda\neq 0. J. Geom. Phys., 112:175–196, 2017.
  • [ČS09] A. Čap and J. Slovák. Parabolic geometries. I, volume 154 of Mathematical Surveys and Monographs. American Mathematical Society, Providence, RI, 2009. Background and general theory.
  • [Deb71] R. Debever. On type DD expanding solutions of Einstein-Maxwell equations. Bull. Soc. Math. Belg., 23:360–376, 1971.
  • [DS11] M. Dunajski and V. Sokolov. On the 7th order ODE with submaximal symmetry. J. Geom. Phys., 61(8):1258–1262, 2011.
  • [Gar89] R. B. Gardner. The method of equivalence and its applications, volume 58 of CBMS-NSF Regional Conference Series in Applied Mathematics. Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA, 1989.
  • [GKN00] M. Godliński, W. Kopczyński, and P. Nurowski. Locally Sasakian manifolds. Classical Quantum Gravity, 17(18):L105–L115, 2000.
  • [GP06] J. B Griffiths and J. Podolskỳ. A new look at the Plebański–Demiański family of solutions. International Journal of Modern Physics D, 15(03):335–369, 2006.
  • [Hen86] P. Henrici. Applied and computational complex analysis. Vol. 3. Pure and Applied Mathematics (New York). John Wiley & Sons, Inc., New York, 1986. Discrete Fourier analysis—Cauchy integrals—construction of conformal maps—univalent functions, A Wiley-Interscience Publication.
  • [HS09] M. Hammerl and K. Sagerschnig. Conformal structures associated to generic rank 2 distributions on 5-manifolds—characterization and Killing-field decomposition. SIGMA Symmetry Integrability Geom. Methods Appl., 5:Paper 081, 29, 2009.
  • [HS11] M. Hammerl and K. Sagerschnig. The twistor spinors of generic 2- and 3-distributions. Ann. Global Anal. Geom., 39(4):403–425, 2011.
  • [Kin69] W. Kinnersley. Type D vacuum metrics. Journal of Mathematical Physics, 10(7):1195–1203, 1969.
  • [KK09] N. Kamran and A. Krasiński. Editorial Note to: Brandon Carter, Black hole equilibrium states part I. Analytic and geometric properties of the Kerr solutions. General Relativity and Gravitation, 41(12):2867, 2009.
  • [Lib52] P. Libermann. Sur les structures presque paracomplexes. C. R. Acad. Sci. Paris, 234:2517–2519, 1952.
  • [Mer84] S A Merkulov. A conformally invariant theory of gravitation and electromagnetism. Classical and Quantum Gravity, 1(4):349–354, jul 1984.
  • [NS03] P. Nurowski and G. A J Sparling. Three-dimensional cauchy–riemann structures and second-order ordinary differential equations. Classical and Quantum Gravity, 20(23):4995–5016, oct 2003.
  • [Nur05] P. Nurowski. Differential equations and conformal structures. J. Geom. Phys., 55(1):19–49, 2005.
  • [Olv95] P. J. Olver. Equivalence, invariants, and symmetry. Cambridge University Press, Cambridge, 1995.
  • [PD76] J. F. Plebański and M. Demiański. Rotating, charged, and uniformly accelerating mass in general relativity. Ann. Physics, 98(1):98–127, 1976.
  • [Sas43] S. Sasaki. On the spaces with normal conformal connexions whose groups of holonomy fix a point or a hypersphere. I,II,III. Jpn. J. Math., 18:615–622, 1943.
  • [Sha97] R. W. Sharpe. Differential geometry, volume 166 of Graduate Texts in Mathematics. Springer-Verlag, New York, 1997. Cartan’s generalization of Klein’s Erlangen program, With a foreword by S. S. Chern.
  • [SW17] K. Sagerschnig and T. Willse. The geometry of almost Einstein (2,3,5)(2,3,5) distributions. SIGMA Symmetry Integrability Geom. Methods Appl., 13:Paper No. 004, 56, 2017.