Timeline for Curve with no embedding in a toric surface
Current License: CC BY-SA 4.0
9 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Sep 14, 2020 at 13:17 | vote | accept | Ben C | ||
Sep 13, 2020 at 15:07 | answer | added | David E Speyer | timeline score: 13 | |
Sep 13, 2020 at 13:31 | history | edited | David E Speyer |
edited tags
|
|
Sep 13, 2020 at 2:11 | comment | added | David E Speyer | To be clear, by the width of a lattice polytope $P$, I mean the following: the minimum, over nonzero lattice vectors $v$, of $\max_{p \in P} \langle p,v \rangle - \min_{q \in P} \langle q,v \rangle$. | |
Sep 12, 2020 at 23:50 | comment | added | David E Speyer | Some observations: The divisor $C$ on $S$ corresponds to a lattice polygon, call it $P$. The number of interior lattice points of this polygon is $g$. If the polygon has width $k$ in any lattice direction, then the curve has gonality $\leq k$. Thus, if we are to get a generic curve of genus $g$, the width must be $\geq g/2+1$ in every direction. Also, if $N$ is the number of lattice points and we want to get a generic curve, we must have $N-3 \geq 3g-3$, since the moduli space of curves is $3g-3$ dimensional, and we lose $3$ dimensions rescaling coordinates. | |
Sep 11, 2020 at 16:46 | comment | added | Ben C | @Jef thanks for the comment. I am familiar with Castryck's work and I think you are correct nondegenerate is a stronger condition it corresponds to an embedding that intersects the toric divisor transversally. There is a related notion of weakly nondegenerate which I do believe corresponds to what I am looking for. However, it appears less is know about when a curve admits a weakly nondegenerate equation. | |
Sep 11, 2020 at 10:56 | comment | added | Jef | You might already know this, but there is the notion of a nondegenerate curve which is related but (I think) stronger than your condition, for a definition see Castryck and Voight's article. In that paper they prove that every curve over an algebraically closed field is nondegenerate if its genus is $\leq 4$, so your example will necessarily have genus at least $5$. | |
Sep 10, 2020 at 20:22 | history | edited | Ben C | CC BY-SA 4.0 |
added 238 characters in body
|
Sep 10, 2020 at 20:10 | history | asked | Ben C | CC BY-SA 4.0 |