Given a Kähler manifold $X$, the blow up is a Kähler manifold aswell (see Hodge theory and CAG by Voision Prop. 3.24). The idea is of course to use the pull-back the Kähler form $\pi^*\omega_X$. It says that this form is not positive, but only semi-positive and I fail to see why exactly.

Positivity of $\omega_x$ means that locally we can write $\omega_x= \sum_i \alpha_i dz_i\wedge d\bar z_i$ with $\alpha_i$ real and non-negative. Now $\pi^*\omega_X= \sum_i (\alpha_i\circ\pi) d(z_i\circ\pi)\wedge d(\bar z_i\circ\pi)$. Obviously the coefficients $(\alpha_i\circ\pi)$ still remain positive, so somehow the differential forms $ d(z_i\circ\pi)\wedge d(\bar z_i\circ\pi)$ must vanish on some special tangent vectors in a neighborhood around the blow up?

I am also not sure, if my notion of positivity is correct or appropriate here.

  • $\begingroup$ Did you understand the proof in the end? I am currently trying to figure it out, but I have my problems. see math.stackexchange.com/questions/3276030/…. I am under the impression that what is proved in the consecutive lemma 3.25 is noth enough. Could you help me? $\endgroup$
    – Michael
    Jun 30, 2019 at 11:48

2 Answers 2


Let $\widetilde X$ be the blowup of $X$, and let $\pi\colon \widetilde X\to X$ denote the blow-down map. If the local coordinates $(z_i)$ are chosen so that $z=0$ is the point being blown up, then $D = \pi^{-1}(0)\subseteq \widetilde X$ is a complex hypersurface called the exceptional divisor. Because $z_i\circ\pi\equiv 0$, it follows that $d(z_i\circ \pi)$ annihilates every vector tangent to $D$, and the same goes for $d(\bar z_i\circ\pi)$.

  • $\begingroup$ This makes sense, but what do you mean by "every vector tangent to D". Just any $v\in T_x \widetilde X$ for $x\in D$? $\endgroup$
    – Notone
    Nov 26, 2017 at 19:24
  • $\begingroup$ @Notone: No, that's not what it means. For each $x\in D$, there is a natural way to identify $T_xD$ with the subspace $di_x(T_xD) \subseteq T_x\widetilde X$, where $i\colon D\to\widetilde X$ is the inclusion map. A vector in $T_x\widetilde X$ is said to be tangent to $D$ if it is in the image of this map. $\endgroup$
    – Jack Lee
    Nov 26, 2017 at 19:35
  • $\begingroup$ In your last sentence do you mean "it follows that $dz_i\circ\pi$ annihilates every vector tangent to D"? But other than that it's clear now, thanks alot $\endgroup$
    – Notone
    Nov 26, 2017 at 20:25
  • $\begingroup$ @Notone: Sorry, what I means was $d(z_i\circ\pi)$. Fixed now. $\endgroup$
    – Jack Lee
    Nov 26, 2017 at 20:28

If $M$ and $N$ be complex manifolds of the same dimension and $π:M→N$ is a holomorphic mapping, then for a volume form (as measure )$Ψ$ on $N$ the pull-back $π^∗Ψ$ is positive outside aramification divisor of $M$ and may not be a positive on the whole of $M$. There is a classical paper of P. Griffiths http://publications.ias.edu/sites/default/files/nevanlinna.pdf

From Principles of Algebraic Geometry by Phillip Griffiths and Joseph Harris, we have:

Let $Y\subset X$. If $Y$ is compact then blow up is Kahler but to construct the metric on $Bl_YX $substantially you use $π^∗ω+εc_1(\mathcal O(−E))$ where $E$ is the exceptional divisor and $π:Bl_YX\to X$ is the canonical surjection

Definition of positivity of pull back of kahler form as current is different with what you wrote , see this paper. See also the definition of pull back of current https://arxiv.org/pdf/math/0606248.pdf

See Lemma 34. of this paper


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.