MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am trying to find an explicit way to view global holomorphic sections of $\Omega^{1} \otimes \mathcal{O} (2)$ over $\mathbb{CP}^{2}$. I guess what I mean by "explicit" would be a formulation over an affine open $U_i \subset \mathbb{CP}^{2}$. According to what I found in Okoneck, Schneider and Spindler, there is a 3-dimensional space of such sections, but I want this for a computation in differential geometry.

share|cite|improve this question
up vote 4 down vote accepted

If $x,y,z$ are coordinates on $P^2$ then the 3 sections of $\Omega(2)$ are given by $xdy - ydx$, $ydz-zdy$, and $zdx-xdz$.

share|cite|improve this answer
That is elegant, and the sort of answer I wanted. Can you give me a reference as to why this is true? – David Johnson Nov 29 '10 at 5:43
It follows from Yuhao's answer plus the Koszul complex exactness. – Torsten Ekedahl Nov 29 '10 at 6:08
I think it is necessary to clarify the notations here: as $x,y,z$ are only sections of $\mathcal{O}(1)$, "d" of them doesn't make sense in the usual way, in fact, as sections of $\Omega(2)$ the above sections are $x^2d(\dfrac{y}{x})$, etc. which can be written formally as $xdy-ydx$. – Yuhao Huang Nov 29 '10 at 6:22
If $P^2 = (V - \{0\})/C^*$ then you can think about $x$, $y$, and $z$ as about the coordinates on $V$. Then $dx$, $dy$ and $dz$ are forms on $V$. – Sasha Nov 29 '10 at 6:27
Between Sasha's answers and Torsten and Yuhao's explanations, I believe this makes sense to me now. To Yuhao, as a section of $\mathcal{O}(1)$ is a linear polynomial in the homogeneous coordinates, I think that the formulation of Sasha is correct, and dx has meaning in terms of the pullback via a local section of $\mathbb{C}^3\\{(0,0,0)}\to \mathbb{P}^2$ over a coordinate neighborhood. Thanks for the answers. – David Johnson Nov 30 '10 at 4:36

Use the Euler sequence:

$0 \to \Omega^1_{\mathbb{P}^n_A/A} \to \mathcal{O}_{\mathbb{P}^n_A}(-1)^{\oplus n+1} \to \mathcal{O}_{\mathbb{P}^n_A} \to 0.$

Everything can be seen explicitly from here. Tensoring the sequence with $\mathcal{O}(2)$ gives an exact sequence, which is still exact if you take global sections because every monomial of degree 2 is a multiple of a monomial of degree 1.:) So the dim you want = 9-6 =3.

share|cite|improve this answer
The dimension I can see, but what I am looking for is explicit formulas, like Sasha's --- which is probably exactly what I need. – David Johnson Nov 29 '10 at 5:45
Well, the 3 sections Sasha write down is precisely a basis for the kernel of the map obtained by taking global sections. The wikipedia page of Euler sequence contains a description of the maps of the Euler sequence, which is precisely what you missed, I guess. – Yuhao Huang Nov 29 '10 at 6:01
You get the kernel of the map of global sections $\mathcal O_{\mathbb P^n_A}(1)^{\oplus n+1}\to \mathcal O_{\mathbb P^n_A}(2)$ by using the exactness of the Koszul complex of the $x_0,\dots,x_n$. – Torsten Ekedahl Nov 29 '10 at 6:07

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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