Abstract
We investigate when the Chevalley-Eilenberg differential of a complex Lie algebroid on a manifold with boundary admits a Hodge decomposition. We introduce the concepts of Cauchy-Riemann structures, elliptic and non-elliptic boundary points and Levi-forms, which we use to define the notion of q-convexity. We show that the Chevalley-Eilenberg complex of an elliptic, q-convex Lie algebroid admits a Hodge decomposition in degree q. This generalizes the well-known Hodge decompositions for the exterior derivative on real manifolds and the delbar-operator on q-convex complex manifolds. We establish the results in a more general setting, where the differential does not necessarily square to zero and moreover varies in a family, including an analysis of the behaviour on the deformation parameter. As application we give a proof of a classical holomorphic tubular neighbourhood theorem (which implies the Newlander-Nirenberg theorem) based on the Moser trick, and we provide a finite-dimensionality result for certain holomorphic Poisson cohomology groups.
Similar content being viewed by others
Data Availability Statement
Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.
Notes
In fact, one can show that the Jacobi identity for \([\cdot ,\cdot ]\) itself already implies that \(\rho \) is bracket preserving.
By definition, the eigenvalues of a real (1, 1)-form \(\tau \) are the eigenvalues of the Hermitian matrix \(\tau _{ij}\) with respect to a decomposition \(\tau =i\sum _{i,j}\tau _{ij}dz^i\wedge d{\bar{z}}^j\) in local coordinates.
Specifically, we need that every real (1, 1)-form which is d-exact is also \(\partial {\bar{\partial }}\)-exact.
This is actually not immediately obvious, but can be seen e.g. using Remark 2.27.
See Remark 2.57 for why Čech-cohomology agrees with \({\bar{\partial }}\)-cohomology in this context. We abbreviate \(TU=T^{1,0}U\).
By definition, \({\mathcal {S}}({\mathbb {R}}^m_-)\) is the set of functions obtained by restricting elements of \({\mathcal {S}}\) to \({\mathbb {R}}^m_{-}\).
Note that \(\sigma (\Delta _t,dr)\) may only be invertible around \(\partial M\), but we can extend it in an arbitrary (but smooth) way to the interior of M.
Since \(\varphi _{I,\varepsilon }^\mu |_{\partial M}=0\) whenever \(l\in I\), the sum over the eigenvalues is well-defined.
References
Bailey, M.: Local classification of generalized complex structures. J. Differ. Geom. 95(1), 1–37 (2013)
Bailey, M., Cavalcanti, G.R., van der Leer Durán, J.L.: A neighbourhood theorem in generalized complex geometry. arXiv:1906.12069 (under review)
Conner, P.E.: The Neumann’s problem for differential forms on Riemannian manifolds. Mem. Am. Math. Soc. 20, 56 (1956)
Folland, G.B.: The tangential Cauchy-Riemann complex on spheres. Am. Math. Soc. 171, 83–133 (1972). https://doi.org/10.2307/1996376Trans
Folland, G.B., Kohn, J.J.: The Neumann problem for the Cauchy-Riemann complex: annals of mathematics studies, no. 75. Princeton University Press, Princeton (1972)
Friedrichs, K.: Spektraltheorie halbbeschränkter Operatoren und Anwendung auf die Spektralzerlegung von Differentialoperatoren. Math. Ann. 109(1), 465–487 (1934). https://doi.org/10.1007/BF01449150
Grauert, H.: On Levi’s problem and the imbedding of real-analytic manifolds. Ann. Math. 2(68), 460–472 (1958). https://doi.org/10.2307/1970257
Grauert, H.: Über Modifikationen und exzeptionelle analytische Mengen. Math. Ann. 146, 331–368 (1962)
Griffiths, P.A.: The extension problem in complex analysis. II. Embeddings with positive normal bundle. Am. J. Math. 88, 366–446 (1966). https://doi.org/10.2307/2373200
Gualtieri, M.: Generalized complex geometry. Ann. Math. 174(1), 75–123 (2011). https://doi.org/10.4007/annals.2011.174.1.3
Hitchin, N.: Generalized Calabi-Yau manifolds. J. Math. 54(3), 281–308 (2003). https://doi.org/10.1093/qjmath/54.3.281Q
Hörmander, L.: Linear partial differential operators, Die Grundlehren der mathematischen Wissenschaften, vol. 116. Springer, New York (1963)
Hörmander, L.: \(L^{2}\) estimates and existence theorems for the \(\bar{\partial }\) operator. Mathematics 113, 89–152 (1965). https://doi.org/10.1007/BF02391775Acta
Kohn, J.J.: Harmonic integrals on strongly pseudo-convex manifolds. II. Ann. Math. 79, 450–472 (1964). https://doi.org/10.2307/1970404
Levi, E.E.: Studii sui punti singolari essenziali delle funzioni analitiche di due o più variabili complesse. Annali di Matematica Pura ed Applicata (1898-1922) 17(1), 61–87 (1910). https://doi.org/10.1007/BF02419336
Lopatinsky, Y.B.: On a method of reducing boundary problems for a system of differential equations of elliptic type to regular integral equations, Ukrain. Math. Z. 5, 123–151 (1953)
Malgrange, B.: La cohomologie d’une variété analytique complexe à bord pseudo-convexe n’est pas nécessairement séparée, C. R. Acad. Sci. Paris Sér. A-B 280, A93–A95 (1975)
Morrey, C.B., Jr.: The analytic embedding of abstract real-analytic manifolds. Ann. Math. 68, 159–201 (1958). https://doi.org/10.2307/1970048
Moser, J.: A rapidly convergent iteration method and non-linear partial differential equations. I. Ann. Scuola Norm. Sup. Pisa 20, 265–315 (1966)
Nash, J.: The imbedding problem for Riemannian manifolds. Ann. Math. 63, 20–63 (1956). https://doi.org/10.2307/1969989
Newlander, A., Nirenberg, L.: Complex analytic coordinates in almost complex manifolds. Ann. Math. 65, 391–404 (1957)
Pradines, J.: Théorie de Lie pour les groupoïdes différentiables. Calcul différenetiel dans la catégorie des groupoïdes infinitésimaux, C. R. Acad. Sci. Paris Sér. A-B 264, A245–A248 (1967)
Weil, A.: Introduction ã l’étude des variétés kählériennes. (French) Publications de l’Institut de Mathématique de l’Université de Nancago, VI. Actualités Sci. Ind. no. 1267. Hermann, Paris (1958)
Acknowledgements
The author is thankful to Micheal Bailey, Gil Cavalcanti, Marco Gualtieri and Brent Pym for useful conversations.
Funding
This research is supported by NSERC Discovery Grant 355576.
Author information
Authors and Affiliations
Ethics declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Code availability
Not applicable.
Additional information
Communicated by F. C. Marques.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
A Appendix
A Appendix
In this appendix we collect some “Leibniz” rules involving Sobolev norms. We will use the notation of Sect. 3.3, and start with some numerical estimates.
Lemma A.1
Let \(k\in {\mathbb {R}}\) be a real number.
-
i)
For all \(\xi ,\eta \in {\mathbb {R}}^m\) we have \(\Big (\frac{1+|\xi |^2}{1+|\eta |^2}\Big )^k \le 2^{|k|} (1+|\xi -\eta |^2)^{|k|}\).
-
ii)
Define \(K_1(\xi ,\eta ):= (1+|\xi |^2)^{k/2}-(1+|\eta |^2)^{k/2}\), where \(\xi ,\eta \in {\mathbb {R}}^m\). Then
$$\begin{aligned} |K_1(\xi ,\eta )|&\le |k|\cdot |\xi -\eta | \cdot \big ((1+|\xi |^2)^{\frac{k-1}{2}}+(1+|\eta |^2)^{\frac{k-1}{2}}\big ). \end{aligned}$$ -
iii)
Define \(K_3(\xi ,\eta _1,\eta _2):= (1+|\xi |^2)^{\frac{k}{2}}+(1+|\eta _1|^2)^{\frac{k}{2}}-(1+|\eta _2|^2)^{\frac{k}{2}}-(1+|\xi +\eta _1-\eta _2|^2)^{\frac{k}{2}}\), where \(\xi ,\eta _1,\eta _2\in {\mathbb {R}}^m\). Setting \(C:=k(k-1)\), we have
$$\begin{aligned}&|K_3(\xi ,\eta _1,\eta _2)|\le C |\xi -\eta _2| \cdot |\eta _1-\eta _2|\cdot \int _0^1 \int _0^1 (1+\big |\xi +t(\eta _1-\eta _2)\nonumber \\&\quad +t'(\eta _2-\xi )\big |^2)^{\frac{k-2}{2}} dtdt'. \end{aligned}$$
Proof
i) See [5, Lemma (A.1.3)].
ii) Consider the smooth function \(f(x):=(1+x^2)^{k/2}\) on \({\mathbb {R}}_{\ge 0}\). For all \(x,y\in {\mathbb {R}}_{\ge 0}\) we have
Setting \(x=|\xi |\), \(y=|\eta |\) and using \(| |\xi |- |\eta | |\le |\xi -\eta |\), we obtain (ii).
iii): Define \(g(x):=(1+|x|^2)^{k/2}\) so that \(K_3(\xi ,\eta _1,\eta _2)=g(\xi )+g(\eta _1)-g(\eta _2)-g(\xi +\eta _1-\eta _2)\). A couple of applications of the fundamental theorem of calculus gives
from which the desired estimate readily follows. \(\square \)
Below we will write \(\alpha \lesssim \beta \) if there is a constant C such that \(\alpha \le C \beta \). The only constraint on C is that it is independent of the functions f, g and \(\varphi \) appearing in the inequalities below.
Proposition A.2
Set \(a:=1+\frac{m}{2}\), where m is the dimension of Euclidean space.
i) For \(s\ge 0\) and \(f,\varphi \in {\mathcal {S}}\) we have
For \(s\in {\mathbb {R}}\) and \(f,\varphi \in {\mathcal {S}}\) we have
ii) For \(s\in {\mathbb {R}}_{\ge 0}\), \(k\in {\mathbb {R}}_{\ge 1}\) and \(f,\varphi \in {\mathcal {S}}\) we have
For \(s,k\in {\mathbb {R}}\) and \(f,\varphi \in {\mathcal {S}}\) we have
iii) For \(s\in {\mathbb {R}}_{\ge 0}\), \(k\in {\mathbb {R}}_{\ge 1}\) and \(f,\varphi \in {\mathcal {S}}\) we have
For \(s,k\in {\mathbb {R}}\) and \(f,\varphi \in {\mathcal {S}}\) we have
iv) For \(s\in {\mathbb {R}}_{\ge 0}\), \(k\in {\mathbb {R}}_{\ge 2}\) and \(f,g,\varphi \in {\mathcal {S}}\) we have
For \(s,k\in {\mathbb {R}}\) and \(f,g,\varphi \in {\mathcal {S}}\) we have
Proof
i) Using \(\widehat{f\varphi }={\widehat{f}}\star {\widehat{\varphi }}\), where \(\star \) denotes convolution product, we obtain
If \(s\ge 0\) we can use the triangle inequality (raised to a non-negative power) to estimate
Using the Cauchy-Schwarz inequality we get
which implies that
Here in the last step we applied the Cauchy-Schwarz inequality again to estimate
where C is finite because \(a>m/2\). In a similar fashion one proves that
proving (A.1). For arbitrary \(s\in {\mathbb {R}}\) we have to replace the triangle inequality with the bound
which follows from Lemma A.1i). The rest of the steps are then similar to the ones above.
ii) The strategy is the same as in (i). First we observe that
where \(K_1\) was defined in Lemma A.1ii). When \(k-1\ge 0\) and \(s\ge 0\), the triangle inequality together with Lem.A.1ii) imply that
The same steps as in i) then yield (A.3). For arbitrary s and k we use Lemma A.1ii) to estimate
giving (A.4).
iii) We have
where \(K_2(\xi ,\eta ):= \big ( (1+|\xi |^2)^{k/2}-(1+|\eta |^2)^{k/2} \big )^2=K_1(\xi ,\eta )^2\). Using Lemma A.1ii) we obtain
valid for \(s\ge 0\) and \(k\ge 1\). The same steps as in (i) then give (A.5). For arbitrary s and k we use Lemma A.1i) to estimate
which gives (A.6).
iv): We have
where \(K_3\) was defined in Lemma A.1iv). Consequently, for \(s\ge 0\) and \(k\ge 2\) we have
Expanding this out gives nine terms, leading to (A.7) by using the same steps as in i).
For arbitrary s and k we use Lemma A.1 ii) to bound
where we used \(\xi -\eta _1+t(\eta _1-\eta _2)+t'(\eta _2-\xi )=(1-t')(\xi -\eta _2)+(1-t)(\eta _2-\eta _1)\). Proceeding as before yields (A.8). \(\square \)
There is also a boundary version of Proposition A.2. We continue with the notation of Sect. 3.3, and denote by \(\lceil s \rceil \) the smallest integer greater or equal than s. For \(K\subset {\mathbb {R}}^m_-\) we denote by \(C^\infty _K({\mathbb {R}}^m_-) \subset {\mathcal {S}}({\mathbb {R}}^m_-)\) the functions with support in K.
Proposition A.3
Let \(K\subset {\mathbb {R}}^m_-\) be a compact subset and let \(a=1+\frac{m}{2}\).
i) For \(s\ge 0\) and \(f\in C^\infty _K({\mathbb {R}}^m_-)\), \(\varphi \in {\mathcal {S}}({\mathbb {R}}^m_-)\) we have
For \(s\in {\mathbb {R}}\) and \(f\in C^\infty _K({\mathbb {R}}^m_-)\), \(\varphi \in {\mathcal {S}}({\mathbb {R}}^m_-)\) we have
ii) For \(s\in {\mathbb {R}}_{\ge 0}, k\in {\mathbb {R}}_{\ge 1}\) and \(f\in C^\infty _K({\mathbb {R}}^m_-) ,\varphi \in {\mathcal {S}}({\mathbb {R}}^m_-)\) we have
For \(s,k\in {\mathbb {R}}\) and \(f\in C^\infty _K({\mathbb {R}}^m_-),\varphi \in {\mathcal {S}}({\mathbb {R}}^m_-)\) we have
iii) For \(s\in {\mathbb {R}}_{\ge 0}\), \(k\in {\mathbb {R}}_{\ge 1}\) and \(f\in C^\infty _K({\mathbb {R}}^m_-), \varphi \in {\mathcal {S}}({\mathbb {R}}^m_-)\) we have
For \(s,k\in {\mathbb {R}}\) and \(f\in C^\infty _K({\mathbb {R}}^m_-), \varphi \in {\mathcal {S}}({\mathbb {R}}^m_-)\) we have
iv) For \(s\in {\mathbb {R}}_{\ge 0},k\in {\mathbb {R}}_{\ge 2}\) and \(f,g\in C^\infty _K({\mathbb {R}}^m_-), \varphi \in {\mathcal {S}}({\mathbb {R}}^m_-)\) we have
For \(s,k\in {\mathbb {R}}\) and \(f,g\in C^\infty _K({\mathbb {R}}^m_-), \varphi \in {\mathcal {S}}({\mathbb {R}}^m_-)\) we have
Proof
For each value of r we can consider the inequalities of Proposition A.2 for the functions \(f(\cdot ,r)\), \(g(\cdot ,r)\) and \(\varphi (\cdot ,r)\) on \({\mathbb {R}}^{m-1}\). Integrating these over r and using \(||f(\cdot ,r) ||_{s}\le C |f(\cdot ,r)|_{\lceil s \rceil }\le C |f|_{\lceil s \rceil }\) for \(s\in {\mathbb {R}}_{\ge 0}\) (where C depends on \(K\subset {\mathbb {R}}^m_-\)), we obtain (A.11)–(A.16). \(\square \)
Finally we mention a version of Rellich’s lemma for the norm \(||\text {D}(\cdot )||_{\partial ,-1/2}\), whose proof is outlined in the appendix of [5] (see the discussion preceding Proposition A.3.1).
Proposition A.4
If a sequence \(\varphi _j\in C^\infty _K({\mathbb {R}}^m_-)\) satisfies a uniform bound \(||\text {D}\varphi _j||_{\partial ,-1/2}\le C\), then a subsequence converges in \(L^2({\mathbb {R}}^m)\).
Rights and permissions
About this article
Cite this article
van der Leer Durán, J.L. Hodge decompositions for Lie algebroids on manifolds with boundary. Math. Ann. 382, 303–356 (2022). https://doi.org/10.1007/s00208-021-02293-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00208-021-02293-5