On blow-ups and cohomology of almost complex manifolds (Q714446)

An almost complex structure \(J\) on a compact manifold \(M\) is called \(\mathcal{C}^{\infty}\)-pure and full if the second de Rham cohomology group of \(M\), as a vector space, splits as \[ H^2(M,\mathbb{R}) = H^{1,1}_J(M)_{\mathbb{R}} \oplus H^{(2,0),(0,2)}_J(M)_{\mathbb{R}}, \] where the first summand consists of such cohomology classes of \(M\) which are represented by real valued exterior differential \(2\)-forms of \(J\)-bidegree \((1,1)\), and the second summand of those represented by sums of real valued exterior differential \(2\)-forms of \(J\)-bidegree \((2,0)\) and \((0,2)\), i.e., of classes represented by sums of almost holomorphic and almost anti-holomorphic real valued exterior differential 2-forms. An almost complex structure on a compact manifold \(M\) is called \(\mathcal{C}^{\infty}\)-pure if \[ H^{1,1}_J(M)_{\mathbb{R}} \cap H^{(2,0),(0,2)}_J(M)_{\mathbb{R}} = \{0\}. \] Note, that in the case of a Kähler manifold, the de Rham cohomology splits in the way described above. In the case of a general complex manifold, this decomposition need not hold. It is known that all compact complex manifolds of real dimension four are \(\mathcal{C}^{\infty}\)-pure and full, cf. [\textit{T. Draghici} et al., Int. Math. Res. Not. 2010, No. 1, 1--17 (2010; Zbl 1190.32021)]. In the article under review, the authors study the behaviour of \(\mathcal{C}^{\infty}\)-pureness and fullness. They derive a sufficient condition for a quotient \(M\) of a complex Lie group \(G\) by a co-compact discrete Lie subgroup \(\Gamma\) to be \(\mathcal{C}^{\infty}\)-pure. The condition is expressed as \(H^2(\mathfrak{g}) \simeq H^2(M,\mathbb{R}),\) where \(H^2(\mathfrak{g})\) denotes the Chevalley-Eilenberg cohomology of the Lie algebra \(\mathfrak{g}\) of \(G.\) The authors recall an example of a compact complex manifold the complex structure of which is \(\mathcal{C}^{\infty}\)-full and pure. They also give an example of a complex structure which is even not \(\mathcal{C}^{\infty}\)-pure. Both are obtained as co-compact quotients of a Heisenberg group by a lattice. They also prove that in the case of a complex manifold, the property of being \(\mathcal{C}^{\infty}\)-pure and full survives the complex blow-up along a complex compact submanifold. In the case of a symplectic manifold \((M, \omega)\), the authors prove that the property of being \(\mathcal{C}^{\infty}\)-pure and full of any almost complex structure \(J\) calibrated by the symplectic structure \(\omega\) is inherited under the symplectic blow-up along any compact symplectic submanifold if a certain map, arising during the blowing-up process, is pseudoholomorphic with respect to \(J\) and with respect to an almost complex structure \(\tilde{J}\) on the blow-up produced naturally during the blowing-up process. As proved in the text, one can change the almost-complex structure \(\tilde{J}\) to make the mentioned map pseudoholomorphic. The calibration of \(J\) by \(\omega\) is not mentioned in the formulation of Theorem 5.3.