Log in

Hausdorff Morita equivalence of singular foliations

  • Published:
Annals of Global Analysis and Geometry Aims and scope Submit manuscript

Abstract

We introduce a notion of equivalence for singular foliations—understood as suitable families of vector fields—that preserves their transverse geometry. Associated with every singular foliation, there is a holonomy groupoid, by the work of Androulidakis–Skandalis. We show that our notion of equivalence is compatible with this assignment, and as a consequence, we obtain several invariants. Further, we show that it unifies some of the notions of transverse equivalence for regular foliations that appeared in the 1980s.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
EUR 32.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or Ebook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (France)

Instant access to the full article PDF.

Fig. 1

Similar content being viewed by others

Notes

  1. Globally this is not the case [3, Lemma 1.3].

  2. The actions can be both right actions, both left actions, or one right and one left action.

  3. Notice that the singular foliation induced by a right action on a manifold agrees with the one induced by the corresponding left action.

  4. This follows from the fact that, under the diffeomorphism \(G/G_{p_2}\cong M_2\), the left action of G on \(M_2\) corresponds to the action on \(G/G_{p_2}\) induced by left multiplication on G.

  5. For instance, \(\mathbf {t}\left( \frac{v}{N},x\right) = \hbox {exp}_x\left( \frac{1}{N}X_1\right) =\gamma \left( \frac{1}{N}\right) \).

  6. Notice that a Hausdorfness assumption is needed also to match the Morita equivalence of Lie groupoids and Lie algebroids (see Proposition 2.27.

References

  1. Androulidakis, I., Skandalis, G.: The holonomy groupoid of a singular foliation. J. Reine Angew. Math. 626, 1–37 (2009)

    Article  MathSciNet  MATH  Google Scholar 

  2. Androulidakis, I., Zambon, M.: Integration of Singular Subalgebroids. In preparation

  3. Androulidakis, I., Zambon, M.: Smoothness of holonomy covers for singular foliations and essential isotropy. Math. Z. 275(3–4), 921–951 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  4. Androulidakis, I., Zambon, M.: Holonomy transformations for singular foliations. Adv. Math. 256, 348–397 (2014)

    Article  MathSciNet  MATH  Google Scholar 

  5. Androulidakis, I., Zambon, M.: Stefan–Sussmann singular foliations, singular subalgebroids and their associated sheaves. Int. J. Geom. Methods Mod. Phys. 13(Supp. 1), 1641001 (2016)

    Article  MathSciNet  MATH  Google Scholar 

  6. Androulidakis, I., Zambon, M.: Almost regular Poisson manifolds and their holonomy groupoids. Sel. Math. (N.S.) 23(3), 2291–2330 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  7. Crainic, M., Fernandes, R.L.: Integrability of Poisson brackets. J. Differ. Geom. 66(1), 71–137 (2004)

    Article  MathSciNet  MATH  Google Scholar 

  8. Crainic, M., Mestre, J.N.: Orbispaces as differentiable stratified spaces. Lett. Math. Phys. 108(3), 805–859 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  9. Debord, C.: Holonomy groupoids of singular foliations. J. Diff. Geom. 58(3), 467–500 (2001)

    Article  MathSciNet  MATH  Google Scholar 

  10. Debord, C.: Longitudinal smoothness of the holonomy groupoid. C. R. Math. Acad. Sci. Paris 351(15–16), 613–616 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  11. del Hoyo, M.L.: Lie groupoids and their orbispaces. Port. Math. 70(2), 161–209 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  12. Garmendia, A.: Ph.D. thesis, KU Leuven (in progress)

  13. Ginzburg, V.L.: Grothendieck groups of Poisson vector bundles. J. Symplectic Geom. 1(1), 121–169 (2001)

    Article  MathSciNet  MATH  Google Scholar 

  14. Haefliger, A.: Groupoïdes d’holonomie et classifiants. Astérisque 116, 70–97 (1984). [transversal structure of foliations (Toulouse, 1982)]

    MATH  Google Scholar 

  15. Landsman, N.P.: The Muhly–Renault–Williams theorem for Lie groupoids and its classical counterpart. Lett. Math. Phys. 54(1), 43–59 (2000)

    Article  MathSciNet  MATH  Google Scholar 

  16. Laurent-Gengoux, C., Stiénon, M., Xu, P.: Non-abelian differentiable gerbes. Adv. Math. 220(5), 1357–1427 (2009)

    Article  MathSciNet  MATH  Google Scholar 

  17. Mackenzie, K.C.H.: General Theory of Lie Groupoids and Lie Algebroids, London Mathematical Society Lecture Note Series, vol. 213. Cambridge University Press, Cambridge (2005)

    Book  Google Scholar 

  18. Moerdijk, I., Mrčun, J.: Lie groupoids, sheaves and cohomology. In: Gutt, S., Rawnsley, J., Sternheimer, D. (eds.) Poisson Geometry, Deformation Quantisation and Group Representations, London Mathematical Society Lecture Note Series, vol. 323, pp. 145–272. Cambridge University Press, Cambridge (2005)

    MATH  Google Scholar 

  19. Molino, P.: Orbit-like foliations. In: Mizutani, T., Masuda, K., Matsumoto, S., Inaba, T., Tsuboi, T., Mitsumatsu, Y. (eds.) Geometric Study of Foliations (Tokyo, 1993), pp. 97–119. World Scientific Publishing, River Edge (1994)

    Google Scholar 

  20. Pronk, D.A.: Etendues and stacks as bicategories of fractions. Compos. Math. 102(3), 243–303 (1996)

    MathSciNet  MATH  Google Scholar 

  21. Tu, J.-L.: Non-Hausdorff groupoids, proper actions and \(K\)-theory. Doc. Math. 9, 565–597 (2004)

    MathSciNet  MATH  Google Scholar 

  22. Wang, R.: On Integrable Systems Rigidity for PDEs with Symmetry. Ph.D. thesis, Utrecht University, p. 12 (2017), Arxiv:1712.00808

  23. Weinstein, A.: The local structure of Poisson manifolds. J. Differ. Geom. 18(3), 523–557 (1983)

    Article  MathSciNet  MATH  Google Scholar 

  24. Xu, P.: Morita equivalence of Poisson manifolds. Commun. Math. Phys. 142(3), 493–509 (1991)

    Article  MathSciNet  MATH  Google Scholar 

Download references

Acknowledgements

We are grateful to Dorette Pronk for explanations about topological groupoids, to George Skandalis for his constructive remarks on this work, and to Kirsten Wang for pointing out reference [19]. Further, we thank Iakovos Androulidakis and Ori Yudilevich for useful discussions. The authors acknowledge partial support by Pesquisador Visitante Especial Grant 88881.030367/2013-01 (CAPES/Brazil) and by IAP Dygest, the long-term structural funding—Methusalem grant of the Flemish Government, the FWO under EOS Project G0H4518N, the FWO research Project G083118N (Belgium).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Marco Zambon.

Appendix

Appendix

1.1 Morita equivalence for open topological groupoids and Lie groupoids

Definition 2.25, on Morita equivalence of Lie groupoids, is equivalent to several other characterizations, as was proved in [18], (see also [16]). An analogue statement holds also for open topological groupoids, upon replacing submersions with continuous open maps. In this Appendix, we recall these facts and prove some implications that are used in the main body of the paper, the main one being Corollary A.7.

We start recalling the notion of weak equivalence, as given in [20, §1.3], and of bitorsor.

Definition A.1

Let \(G\rightrightarrows M\) and \(\varGamma \rightrightarrows P\) be two Lie groupoids (respectively, topological groupoids). A morphism \(\widehat{\pi }:\varGamma \rightarrow G\) is a weak equivalence if:

  1. (i)

    \(\varGamma \rightarrow \pi ^{-1} G; \;\;\gamma \mapsto (\mathbf {t}(\gamma ),\widehat{\pi }(\gamma ),\mathbf {s}(\gamma ))\) is an isomorphism,

  2. (ii)

    \(\mathbf {t}\circ \textit{Pr}_1:G {}_{\mathbf {s}}\!\times _{\pi } P\rightarrow M\) is a surjective submersion (resp. a surjective continuous and open map).

Here \(\pi :P\rightarrow M\) denotes the base map covered by \( \widehat{\pi }\).

Remark A.2

  1. (i)

    Looking at a groupoid as a small category, a weak equivalence is the same thing as a fully faithful and essentially surjective functor.

  2. (ii)

    When a map \(\pi :P\rightarrow M\) is completely transverse (transverse to the orbits and meeting every orbit) to a Lie groupoid \(G\rightrightarrows M\), then the projection \(\pi ^{-1} G \rightarrow G\) is a weak equivalence.

  3. (iii)

    If G is an open topological groupoid and \({\pi }\) is a continuous, open and surjective map, then condition (ii) in Definition A.1 is automatically satisfied, as can be shown using Lemma 3.35.

Definition A.3

Let \(G\rightrightarrows M\) be a Lie groupoid (respectively, a topological open groupoid) and \(\pi :P\rightarrow M\) a surjective submersion (resp. a surjective continuous and open map). A G-action over a not necessarily Hausdorff manifoldP is a smooth (resp. continuous) map \(\star :G {}_\mathbf {s}\!\times _\pi P \rightarrow P\) such that for all \(g,h\in G\) and \(p\in P\):

$$\begin{aligned} \pi (g\star p)=\mathbf {t}(g),\quad g\star (h \star p)=(gh)\star p,\quad e_{\pi (p)}\star p=p. \end{aligned}$$

Such a manifold P with a G-action is called a G-module, and \(\pi \) is called its moment map. If the G-action is free and proper, then P / G is a manifold and we say that P is a G-principal bundle.

A (GH)-bimodule for the Lie groupoids \(G\rightrightarrows M\) and \(H\rightrightarrows N\) is a (not necessarily Hausdorff) manifold P with two actions commuting with each other. A (GH)-bimodule P that is principal with respect to both actions and such that \(G{\backslash } P\cong N\) and \(P/H\cong M\) is called a (GH)-bitorsor.

The following statement can be found in [18, §2.5]

Lemma A.4

Consider a Lie groupoid \(\varGamma \rightrightarrows K\), a \(\varGamma \)-principal bundle S, a \(\varGamma \)-module Q and a map \(f:Q\rightarrow S\) preserving the \(\varGamma \) actions. Then, \(Q/\varGamma \) is a manifold.

Proposition A.5

Let \(G\rightrightarrows M\) and \(H\rightrightarrows N\) be Lie groupoids. The following statements are equivalent:

  1. (i)

    There exists a Lie groupoid \(\varGamma \) and two weak equivalences \(\varGamma \rightarrow G\) and \(\varGamma \rightarrow H\).

  2. (ii)

    There exists a (GH)-bitorsor P.

  3. (iii)

    G and H are Morita equivalent (Definition 2.25).

The proof of this statement can be found in [18] and [16, prop. 2.4]; nevertheless, we review its proof here.

Proof

\(\mathrm{(i)}\Rightarrow \mathrm{(ii)}\): Consider a Lie groupoid \(\varGamma \rightrightarrows K\) with weak equivalences \(\widehat{\pi }_M:\varGamma \rightarrow G\) and \(\widehat{\pi }_N:\varGamma \rightarrow H\). Therefore, \(\varGamma =\pi _M^{-1} G\cong \pi _N^{-1} H\). We get that \(Q_G:=G {}_{\mathbf {s}}\!\times _{\pi _M} K\) is a \((G,\varGamma )\)-bitorsor. Using a similar argument, we get that \(Q_H\) a \((\varGamma ,H)\)-bitorsor. The (not necessarily Hausdorff) manifold \(Q:=(Q_G\times _K Q_H)\) has a diagonal \(\varGamma \)-action with the canonical map to K as moment map. Applying Lemma A.4 to the map \(Q\rightarrow Q_G\), we see that

$$\begin{aligned} P:=(Q_G\times _K Q_H)/\varGamma \end{aligned}$$

is a (not necessarily Hausdorff) manifold. One can check that it is a (GH)-bitorsor.

\(\mathrm{(ii)}\Rightarrow \mathrm{(iii)}\): Consider a (GH)-bitorsor P, with moment maps \(\pi _M:P\rightarrow M\),\(\pi _N:P\rightarrow N\). Then,

$$\begin{aligned} \varGamma = G {}_\mathbf {s}\!\times _{\pi _M} P {}_{\pi _N}\!\times _\mathbf {t}H \end{aligned}$$

has a natural structure of Lie groupoid over P with \(\mathbf {t}(g,p,h)=gph\), \(\mathbf {s}(g,p,h)=p\) and multiplication given canonically by G and H. Then, the maps \(\varGamma \rightarrow \pi _M^{-1}G; \;(g,p,h)\mapsto (p,g^{-1},gph)\) and \(\varGamma \rightarrow \pi _N^{-1} H;\;(g,p,h)\mapsto (p,h,gph)\) are isomorphisms of Lie groupoids.

This shows that \(\pi _M^{-1}G\cong \pi _N^{-1} H\) as Lie groupoids over the not necessarily Hausdorff manifold P. Now take a Hausdorff cover \(\{U_i\}_{i\in I}\) of P and let \(\tilde{P}:=\sqcup _i U_i\). There is a canonical submersion \(\pi :\tilde{P}\rightarrow P\). It is easy tho see that \((\tilde{P},\pi _M\circ \pi , \pi _N\circ \pi )\) is a Morita equivalence.

\(\mathrm{(iii)}\Rightarrow \mathrm{(i)}\): Given a Morita equivalence \((P,\pi _M,\pi _N)\) between G and H, call \(\varGamma :=\pi _M^{-1}G\cong \pi _N^{-1}H\). The natural projections \(\varGamma \rightarrow G\) and \(\varGamma \rightarrow H\) are weak equivalences. \(\square \)

Remark A.6

Proposition A.5 also holds for open topological groupoids, as can be proven using Lemma 3.35. For arbitrary topological groupoids, this is not the case.

Corollary A.7

Let \(k\ge 0\). If \(G\rightrightarrows M\) and \(H\rightrightarrows N\) are source k-connected Morita equivalent Hausdorff Lie groupoids, then there exists a Hausdorff (GH)-bitorsor P. Moreover, this bitorsor is a Morita equivalence with k-connected fibres (in the sense of Definition 2.25).

Proof

Following the implications \(\mathrm{(iii)}\Rightarrow \mathrm{(i)}\Rightarrow \mathrm{(ii)}\) in the proof of Proposition A.5, one sees that the bitorsor P constructed there is Hausdorff. Then, use the implication \(\mathrm{(ii)}\Rightarrow \mathrm{(iii)}\) to prove that P is a Morita equivalence.

Note that being P a bitorsor, the fibres of \(\pi _M:P\rightarrow P/H{\cong M}\) are diffeomorphic to the source fibres of H, which are k-connected by assumption. A similar argument holds for the fibres of \(\pi _N:P\rightarrow P/G\cong N\). \(\square \)

Remark A.8

The Morita equivalence P in Corollary A.7 is a (global) bisubmersion for the underlying foliations, as we now show. Using the implication “\(\mathrm{(ii)}\Rightarrow \mathrm{(iii)}\)” in Proposition A.5, we get an isomorphism of Lie groupoids

$$\begin{aligned} \pi _M^{-1}G\cong \pi _N^{-1} H\cong G\ltimes P\rtimes H. \end{aligned}$$

Denote by \(\mathcal {F}_M\) and \(\mathcal {F}_N\) the foliations underlying G and H. Using Lemmas 1.14(i) and 1.15, we get that the foliations underlying \(\pi _M^{-1}G\) and \(\pi _N^{-1} H\) are \(\pi _M^{-1}\mathcal {F}_M\) and \(\pi _N^{-1}\mathcal {F}_N\), respectively. Since P is a (GH)-bitorsor, the foliation underlying the Lie groupoid \(G\ltimes P\rtimes H\) is \(\varGamma _\mathrm{c}(\hbox {ker}(d\pi _M))+\varGamma _\mathrm{c}(\hbox {ker}(d\pi _N))\). Hence,

$$\begin{aligned} \pi _M^{-1}\mathcal {F}_M=\pi _N^{-1}\mathcal {F}_N=\varGamma _\mathrm{c}(\hbox {ker}(d\pi _M))+\varGamma _\mathrm{c}(\hbox {ker}(d\pi _N)). \end{aligned}$$

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Garmendia, A., Zambon, M. Hausdorff Morita equivalence of singular foliations. Ann Glob Anal Geom 55, 99–132 (2019). https://doi.org/10.1007/s10455-018-9620-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10455-018-9620-6

Keywords

Navigation