Given a nonsingular complete curve over algebraically closed $\bar{k}$, which is interpreted as a field $\bar{k}(X)$ of transcendence degree 1 and its set of valuations trivial on $\bar{k}$, we may choose elements $x,y$ such that $x$ is trancendental, and $\bar{k}(X)=\bar{k}(x)(y)$. Letting $F$ be the homogenization of the minimal polynomial of $y$ over $\bar{k}(x)$, using these coordinates we get a function from the set of valuations to the points of the projective curve $X_F(\bar{k})$, given by mapping a valuation $v$ to the unique point $P$ of $X_F(\bar{k})$ such that $O_P \subset O_v$ and inclusion of maximal ideals.

The first question is to show that if $x,y$ are in $O_v$, then the valuation $v$ is mapped to is $(x(v):y(v):1)$, where $x(v),y(v)$ are the cosets of $x,y$ in $O_v / M_v\cong \bar{k}$.

For this, my thought was that if $\psi =G/H$ with $H(x(v):y(v):1)\neq 0$, then we get the coset of $H(x,y,1)$ in $O_v / M_v$ is nonzero, so we can conclude $G/H$ is defined at $v$, in the sense of having nonnegative valuation.

I feel okay with this, and the homogenisation/dehomogenisation is just from our isomorphism of fields $\bar{k}(X)\cong \bar{k}(X_F)$, but the next part I am feeling stuck, which is to show that if $y\notin O_v, x/y\in O_v$, then the point is $((x/y)(v):1:0)$.

Any assistance with how to think about this would also be very much appreciated.


Your Answer

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

Browse other questions tagged or ask your own question.