Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

Let $X$ be riemann surface (not supposed compact) and $\mathcal D$ be sheaf of divisors on $X$.Remind that this means for $U\subset X$ open then $\mathcal D(U)$ is group of divisors on $U$. How to prove that $H^1(X,\mathcal D)=0$ . This is exercise in §16 of Forster book Riemann Surfaces (but is not homework for me) and he says hint is to use discontinuous partitions of unit but I don't understand. Any other method for proving is for me also satisfying . I suppose $H^2(X,\mathcal D)$ is also zero, no?

Addedd: Problem is in analytic category, not algebraic ( where is trivial : $\mathcal D$ is flabby). So I don't understand Matt E's proof (but thank you very much for answer Matt) because I don't know if cohomology commutes with co-limits in non noetherian case.

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

The hint suggests modifying the proof that $H^1$ for $C^\infty$-sheaves vanishes. This also makes me guess that the problem is about Cech cohomology. The $C^\infty$ proof appears in many places, one reference is Proposition 4.1 in chapter IX of Miranda's book: Algebraic Curves and Riemann Surfaces.

There's probably a way of doing this more cleanly and in more generally but as a start: on $X$ fix a locally finite covering $\{U_i\}_{i \in I}$ where the index set is linearly ordered: $I \subset \mathbb{Z}$.

In this case, choose for each natural number $n$ a set of integers $a_1,..., a_n$ such that $\sum_{1}^n a_i = 1$; e.g. $a_1 = 1$ and all others $0$.

Then define functions $\phi_i \colon U_i \to \mathbb{Z}$ as follows:

1) For $p \in U_i$ let $U_{j_1}, ..., U_{j_n}$ be sets containing $p$ with $j_1 < ... < j_n$.

2) let $k$ be the index such that $j_k = i$

3) let $a_1,..., a_n$ be the previously determined integers summing to $1$

Define $\phi_i(p) = a_k$. These are discontinuous functions and by construction $\sum_i \phi_i \equiv 1$.

To show $H^1(X, \mathcal{D}) = H^1(\{U_i\}, \mathcal{D}) = 0$ we have to show that every $1$-cocycle $(f_{ij})$ is a co-boundary.

The proof now proceeds as in the $C^\infty$ case: on $U_i$ define $g_i = -\sum_j \phi_j f_{ij}$. Then

$g_i - g_j = -\sum_k \phi_kf_{ik} + \sum_k \phi_kf_{jk} = \sum_k \phi_k(f_{jk} - f_{ik})$

Now using that $f_{jk} - f_{ik} = f_{ji}$ we get

$g_i - g_j$ = $f_{ji}\sum_k \phi_k = f_{ji}$.

share|cite|improve this answer

The sheaf $\mathcal D$ is equal to $\bigoplus_{x \in X} i_{x*} \mathbb Z,$ where $\mathbb Z$ denotes the constant sheaf on the point $x$, and $i_x:\{x \} \to X$ is the embedding of the one point space $\{x\}$ into the Riemann surface $X$. (So $i_{x*}\mathbb Z$ is the skyscraper sheaf supported at $x$, whose stalk is $\mathbb Z$). The vanishing of $H^1$ now follows from the vanishing of higher cohomology for skyscraper sheaves.

share|cite|improve this answer

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.