Geometric invariant theory
Geometric Invariant Theory Learning Seminar
Fall 2020
Notes by Patrick Lei
Lectures by Various
Columbia University
Disclaimer
These notes were taken during the seminar using the vimtex package of the editor neovim. Any errors are mine and not the speakers’. In addition, my notes are picture-free (but will include commutative diagrams) and are a mix of my mathematical style and that of the lecturers. If you find any errors, please contact me at plei@math.columbia.edu.
Seminar Website: http://www.math.columbia.edu/~plei/f20-GIT.html
1 Anna (Sep 25): Three Approaches to GIT: Overview and Examples
Let be an algebraic variety and suppose acts on . We want to produce a quotient , which is an algebraic variety. However, this is not the set of orbits of .
We will assume that is a reductive group. Some examples of reductive groups are , etc, and an example of non-reductive groups is ,
1.1 Some Basic Examples
Example 1.1.1.
Consider the action of on . Then we expect , but of course we know that is not a geometric quotient of . However, it is the quotient .
This example can be generalized to toric varieties, which are varieties with a torus action such that there exists a dense orbit. To construct toric varieties, choose a polytope in a real vectir space of dimension . We know that is given by inequalities where and . We want the covectors to be “integral” so we will choose a lattice and ask that .
Consider the exact sequence , where is the number of the . Taking the lattice , we can take . Taking quotients, we obtain
If we complexify this picture, we obtain an exact sequence
Consider the action of on . We can restrict the action to , so we will define the toric variety corresponding to the polytope to be .
Our second example will be the moduli of points in . We want an algebraic variety which parameterizes subsets of points of up to . When , all sets of points are equivalent (by basic facts about projective space). The first interesting case is .
Each points in define a point in , where denotes the symmetric product. We will construct the moduli as .
1.2 Approaches to Taking Quotients
We will discuss various approaches to taking quotients.
1.2.1 Invariant Theory
Suppose that and that acts on . Then acts on , so we can consider the ring of invariants . Then we will define the quotient .
Example 1.2.1.
Suppose acts on with weight , Then acts on a monomial by
Here, the invariants are only the constants, so the quotient is simply a point.
Theorem 1.2.2.
Suppose is of finite type over a field. Then is also of finite type over a field. In addition, the map is surjective and sends -invariant closed subsets to closed sets. Finally, disjoint closed invariant subsets are sent to disjoint closed sets. In particular, if are closed -orbits, they are sent to distinct points.
Remark 1.2.3.
This says morally that the quotient parameterizes closed orbits.
Returning to the original example of acting on , the only closed orbit is the origin. If we remove the origin, the other orbits become closed, so we obtain projective space.
Now suppose that and that the action of on extends to an action on that preserves the grading. Then we will define . This quotient map is generally not a morphism and is only a rational map.
Define where have degree and has degree . Then and . Then invariants are given by . Thus the ring of invariants is generated by monomials of the form and so , which is clearly projective space.
To construct toric varieties, let and , where the are from the definition of the polytope. We want to compute the invariants with respect to .
Proposition 1.2.4.
if and only if for all , . Differentiating at the origin, we see that if and only if for all , .
A monomial is invariant if and only if
for all . In particular, this is equivalent to for some .
For the moduli of points, consider . When , there are no invariants, so is empty. For , note that the invariants are given by , where and thus is a point. When , there is also only one invariant.
When , we have more interesting behavior. Recall that we can define the cross-ration of four points in . Thus there exists a unique automorphism of sending , and is the cross-ratio. Clearly this depends on the order of the four points. We can fix this by defining the -invariant. There is another invariant called the -invariant, so .
In general, when , we can describe the invariants, but when , nobody knows.
1.2.2 (Semi)stability
We have the following problem:
Question 1.2.5.
Describe the maximal open such that is a morphism.
Definition 1.2.6.
is called semistable if there exists a -invariant such that .
Note this equivalent to the rational map being defined at . Usually, it is too difficult to -invariants, so we would like another way to describe semistable points.
Here is an analytic description. Let be a quasiprojective variety and consider the pullback of the Fubini-Study form (when is projective) or the constant symplectic form when is affine. Then is a symplectic manifold. Choose a maximal compact subgroup that preserves . Then any defines a vector field on , and we can rewrite this as
Therefore, we see that is closed. Often, this is even exact. This motivates the following definition:
Definition 1.2.7.
Let be a symplectic manifold. Then an action of on that preserves is called Hamiltonian if for all there exists some function such that . Moreover, there exists such that . Finally, require .
Consider the action of on . Then . We want to construct satisfying , where is tangent to the action of . It is easy to see that
and that
Then define . This is -invariant and is a sphere , so we define . We claim that is horizontal, and therefore descends to the .
Theorem 1.2.8 (Marsden-Weinstein).
- 1.
Assume that acts on freely. Then is a symplectic manifold.
- 2.
If is Kähler, then inherits the Kähler structure.
In our example, , and . This is the same as the quotient from our algebraic procedure.
We conclude with another definition of semistability.
Definition 1.2.9.
The point is called semistable if . Here, is an algebraic group (not necessarily compact).
In conclusion, our algebraic methods of describing invariants and the analytic method of moment maps give a quotient.
2 Caleb (Oct 2): GIT Preliminaries
Today, we will discuss preliminaries for GIT. These may be familiar from other contexts, but they may be different in the category of schemes. We will assume the reader knows what a scheme is.
We will begin with some notation.
- 1.
By a scheme , we mean a morphism of schemes .
- 2.
An =valued point is a morphism . In particular, when for an algebraically closed field , this is called a geometric point.
- 3.
A morphism os -schemes is a morphism of schemes that commutes with the maps to .
- 4.
The product of two morphisms is simply the fiber product.
Definition 2.0.1.
A group scheme is a scheme with multiplication , inverse , and identity that satisfies the group axioms. This is simply a group object in the category of schemes.
Definition 2.0.2.
An algebraic group over is a smooth separated group scheme over .
Definition 2.0.3.
A group acts on if there exists a morphism such that the diagram
| (2.1) |
commutes.
Given a group action , let be . Then the orbit is the image of and the stabilizer is the pullback of
Given a map , we can define to be .
Definition 2.0.4.
A categorical quotient of is a pair that is the pushout of .
Definition 2.0.5.
A geometric quotient of is a pair such that the diagram
| (2.2) |
commutes and
- (1)
is surjective and the image of is .
- (2)
is submersive, or is open if and only if is open.
- (3)
is the subsheaf of invariant functions of .
The second condition means that the preimage of a point in is a single orbit, and the last condition says that functions on are -invariant functions on .
2.1 Properties of Quotients
Proposition 2.1.1.
The geometric quotient is a categorical quotient.
Proof.
We need to prove that a geometric quotient is the pushout. First, cover by affines. If a map exists, we will prove it is unique. First, we will show that if , then . This follows from surjectivity and condition (1) of the geometric quotient. Thus , so is open. Then we have a diagram
| (2.3) |
Here, is the cover of pulled back from . Also, is injective. But then is unique.
To prove that exists, note that exists if is an invariant function for . Then the diagram
| (2.4) |
commutes, and then checking gluing is just the sheaf property. □
Example 2.1.2.
Let act on . Then the categorical quotient is , which means that there cannot be a geometric quotient. If we remove the origin from , then we obtain as the geometric quotient.
Definition 2.1.3.
A universal geometric (resp. categorical) quotient is a geometric (resp. categorical) quotient that is preserved by base change under . A quotient is called uniform if it is preserved by flat base change.
Remarks 2.1.4.
- 1.
The base scheme doesn’t matter.
- 2.
If is a categorical quotient of , then is reduced (resp. connected, irreducible, etc) if and only if is. However, the property of being Noetherian is not preserved (Hilbert’s 14th problem).
- 3.
Let be a universal categorical quotient. Then condition (3) of the geometric quotient is satisfied.
- 4.
Read the next remark yourself in Mumford’s book. This is used to give a criterion for geometric quotients.
- 5.
The property of being a universal categorical quotient can be checked on affines.
- 6.
Read the next remark yourself. It is not very important.
- 7.
Suppose we have a geometric quotient . When is it universal? The diagram always commutes and (1) holds under any extension. Then (2) does not always hold and (3) holds for flat extensions. To see this, we can rephrase (3) as exactness of
Then a flat extension is tensoring with a flat algebra.
- 8.
If is faithfully flat and quasicompact, then is universal if and only if is. This is proved in SGA 8.
Definition 2.1.5.
Suppose is a group action. This action is closed if for all geometric points , the orbit of in is closed.
Definition 2.1.6.
A group action is separated if the image of is closed.
Definition 2.1.7.
A group action is proper if is proper. Here, recall that proper means separated, universally closed, and of finite type.
Definition 2.1.8.
A group action is free if is a closed immersion. In particular, it is set-theoretically free.
Example 2.1.9.
Here, we will have a separated action that is set-theoretically free with a geometric quotient that is not proper. Let and define . Then acts on by matrix multiplication on .
Now define to be the set of points where and is the equare of a homogeneous quadratic form with discriminant . Then is a geometric quotient, but is not free or closed. This example shows that things do not necessarily work the way we expect set-theoretically.
Proposition 2.1.10.
Let be a geometric quotient. Then is closed. This means that the orbit of any geometric point is closed. In addition, is separated if and only if is separated.
Proof.
We have the diagram
| (2.5) |
It remains to check that . Then if lives over for a field , then any closed point is a maximal ideal.
Then we show that is closed. If we have maps , then the diagram
| (2.6) |
is a fiber product. Applying this to , we obtain the desired result because is universally submersive. The rest is omitted, and is Lemma 0.6 in Mumford. □
Remark 2.1.11.
Some of the definitions given may not be the current conventions.
Proposition 2.1.12.
is affine and is proper if and only if is affine.
Definition 2.1.13.
A principal fiber bundle with group is a geometric quotient such that is flat and finite type, and is an isomorphism.
Let act on with weights . Then is a geometric quotient but not a principal fiber bundle.
Remark 2.1.14.
The topics at the end of this chapter include Hilbert schemes and Picard schemes. To learn more about these, attend Caleb’s seminar on FGA Explained.
3 Alex (Oct 9): Actions of Reductive Algebraic Groups
In this talk, we will let be a field that is not necessarily algebraically closed. All of our schemes will live over . Sometimes will be embedded in some algebraically closed field , and we will write . We will denote algebraic groups by and invertible sheaves by .
3.1 Basic Definitions
Definition 3.1.1.
A representation of an algebraic group is a group homomorphism .
A representation of a group can be an action of on as a scheme, an action of on as a vector space, and a third action, called the dual action. Here, we let . Then is a coalgebra. Denote the multiplication by and the counit by . Then any representation of is a comodule for . Define the coaction by .
Definition 3.1.2.
Let be a dual action. Then is invariant if , and a vector is invariant if .
Lemma 3.1.3 (Cartier).
is a union of finite-dimensional invariant subspaces.
Proof.
Let be a basis for . Then write . Then define . At this point, we cut the proof short because. □
Definition 3.1.4.
An algebraic group is reductive if the radical is a torus. Here, the radical is the unique largest connected normal solvable subgroup.
Definition 3.1.5.
A group is linearly reductive if every representation is completely reducible.
Definition 3.1.6.
Let be a linearly reductive group and . Let . Then is a Reynolds operator if
- 1.
;
- 2.
;
- 3.
if and only if .
The important thing to know is that Reynolds operators always exist and that if is a -algebra, then .
Remark 3.1.7.
There is a classification of linear reductive algebraic groups due to Nagata. In characteristic , we have and is prime to . If , then linearly reductive is equivalent to reductive and , where is the commutator subgroup.
3.2 Affine Case
Fix . We will consider the action of a reductive group on an affine scheme . Denote the action by .
Theorem 3.2.1.
There exists a uniform categorical quotient where is universally submersive. If is finite type (resp. Noetherian), so is . In characteristic , the quotient is actually universal.
Recall that induces a dual action of on . Define to be the set of invariant elements. Then and is induced by the inclusion.
Remark 3.2.2.
From last time, it suffices to show that is the invariant subsheaf of and that is closed. Also, we need to show that .
- 1.
If we bet be an -algebra, then is the invariant ring in .
- 2.
If is a set of invariant ideals, then .
To prove that is Noetherian if is, we can choose an ideal . Then , so we have an ideal correspondence between and .
To prove that being of finite type is preserved, note that if is a graded -algebra, then is a graded subalgebra. Then is finite type, so it is finite type, so then is also Noetherian, so it is finite type. In general, there is a procedure to put a grading on any ring so that this works.
Theorem 3.2.3.
The categorical quotient is a geometric quotient if and only if is a closed action. In characteristic , this is a universal quotient.
3.3 Linearization
Definition 3.3.1.
Let be an invertible sheaf over . Then a -linearization is a map satisfying a nasty cocycle condition Alex didn’t bother to write down.
We can interpret a linearization in the following ways:
- 1.
For , we have . Also, we have a commutative diagram
- 2.
extends the -action to the -action to the line bundle defined by .
Example 3.3.2.
If and , then . A -linearization will be a character .
This defines three actions:
- 1.
An action on defined by .
- 2.
For , write for .
- 3.
A dual action
which is the composition
We observe that if are -linearized invertible sheavs, then and also carry a -linearized structure. Thus we can define a group inside the Picard group.
Remark 3.3.3.
If we have a morphism , then we can pull back . If acts trivially on , then this is a map from . If is a geometric quotient of , then pullback is an isomorphism of -linearized Picard groups.
Example 3.3.4.
Let acting on . Then define with
and
where are the coordinates on and are the homogeneous coordinates on .
It is known that has order . If , then it would have order dividing . However, if , then has a -linear structure. Then the total space of and clearly there is an action here.
Proposition 3.3.5.
If is connected and of finite type and is of finite type, there exists no nontrivial character of , and is geometrically reduced, then has at most one -linearization.
Corollary 3.3.6.
If is a normal variety, then there exists such that is -linearizable.
Proposition 3.3.7.
Let be basepoint-free. Then there exists an action of on and a -linearization of such that the map defined by satisfies . If is proper, then is induced from . If is very ample, is an immersion.
Proof.
Mumford does some setup and then uses some big fact with citation from three Japanese guys whose names I don’t know. Let be a basepoint free finite-dimensional subspace. Then we can extend to a a finite-dimensional invariant subspace.
Then we use the fact that there exists a natural equivalence between dual actions of on and actions of on with -linearizations of . □
3.4 The General Case
Definition 3.4.1.
Let be a geometric point. Then is a prestable point if there exists an invariant affine neighborhood such that acting on is a closed action.
Definition 3.4.2.
A geometric point is a semistable point () if there exists an invariant such that and is affine. The point is stable if it is semistable and the action of on is closed.
If the action of on is closed, the dimension of the stabilizers is constant over . Then for opens . A similar statement holds for . Finally, is “properly stable.’
Remark 3.4.3.
We can form a graded ring with invaraint sections such that is affine.
Proposition 3.4.4.
A uniform geometric quotient of exists and is an affine map. Also, is algebraic. Conversely, if has a geometric quotient with affine, then is contained in the prestable locus.
There is no proof of this because it is unimportant.
Theorem 3.4.5.
A universal categorical quotient of exists.
- 1.
The map is universally submersive.
- 2.
There exists an ample line byndle over such that .
- 3.
There exists open such that . Then is a uniform geometric quotient (and is universal in the charateristic case).
Proof relies on doing everything into the affine case and gluing the affines together.
Proposition 3.4.6.
Let be a geometric point. The following are equivalent:
- 1.
is a stable point.
- 2.
is regular for the action and the orbit of is closed.
- 3.
is regular and there exists an invariant section such that , is affine, and the orbit of is closed.
Moreover, if are geometric points, there exists an invariant such that is affine and .
There is a converse: If is a categorical quotient of and is affine and is quasiprojective, then there exists such that . If is a geometric quotient, then . This is not quite good enough because if has a categorical quotient, we only obtain an invertible sheaf on .
Here is a better converse. Let be connected, algebraic, and smooth. Let be connected, reductive, and algebraic. Then let be invariant open. The following are equivalent:
- 1.
There exists such that .
- 2.
There exists a geometric quotient of such that is affine and is quasiprojective.
- 3.
The action of on is proper and there exists a quasiprojective geometric quotient .
4 Patrick (Oct 16): Stability
Note: These are the speaker’s notes.
Throughout this talk, we will let be an algebraically closed field. All schemes will live over . Unless otherwise stated, all schemes are proper over . The goal of this talk is to give a numerical criterion for stability and to describe where stability breaks down, and then present examples.
4.1 Hilbert-Mumford Stability Criterion
Definition 4.1.1.
Let be an algebraic group. Then a -parameter subgroup of is a map .
Then for any point and -parameter subgroup of , we obtain a morphism (action of on via ). By the valuative criterion for properness, there is a unique extension to a morphism . Clearly is fixed under the -action on induced by . Now if , then induces a -linearization of . Restricting to , acts on the fiber of (as a line bundle) by some character , where is the integer associated to under the isomorphism .
Definition 4.1.2.
Let be proper over with an action of . Then if is a closed point, is a -parameter subgroup of , and , then we define
This satisfies the following properties:
- 1.
For , we have .
- 2.
Fix and . Then defines a group homomorphism .
- 3.
Let be a -equivariant morphism, , and . Then
- 4.
If as , then .
This is the main theorem of this talk:
Theorem 4.1.3 (Hilbert-Mumford Stability Criterion).
Let be proper over with an action of a reductive group . Let be ample. Then for any , the following hold:
- 1.
( is semistable) if and only if for all -parameter subgroups .
- 2.
The inequality is strict for all if and only if is stable.
To prove this result, we will choose a -equivariant immersion and reduce to the case of projective space. In Mumford, the proof is broken up into two propositions:
Proposition 4.1.4.
A point is semistable if and only if is not in the closure of the orbit for all points such that . Similarly, is properly stable if and only if is proper for all living over .
Proof of this fact is technical computation with the definitions.
The next step is to interpret the function when .
Proposition 4.1.5.
Let and a -parameter subgroup of . Suppose that as . Then let live over . Finally, suppose that acts on with weights . Then
In addition, has (no specialization, some specialization, specialization ) in when if and only if (, , ).
Proof.
The idea is to prove that . We will omit the technical computation. □
Remark 4.1.6.
There is an alternative method of proof. If we consider the orbit , then we can normalize its closure and reduce to the case when is dense in .
We will now state a result whose motivation is completely unknown to me. However, there are consequences that fit into the philosophy of determining properties of an action of on using -parameter subgroups.
Let and . Then if is a -parameter subgroup, then we can define a point via the morphism .
Theorem 4.1.7 (Iwahori).
Let be a semisimple algebraic group over of adjoint type. Every double coset of with respect to is represented by a point .
Here, a semisimple group is of adjoint type if its center is trivial. Somehow we have the following corollaries that are of interest to us.
Corollary 4.1.8.
Let be a reductive algebraic group and be an action of on a separated algebraic scheme over . Then is proper if and only if for every nontrivial -parameter subgroup , the induced action of on is proper.
Corollary 4.1.9.
Let be a reductive algebraic group acting on an algebraic scheme and . Then acts properly on .
4.2 Flag Complex
When we construct moduli of subschemes of projective space, we will explore relationships between stability and flags in projective space. Here, we will simply state various propositions. In modern terminology, this is the Tits building of .
Definition 4.2.1.
Let be a reductive group and be a -parameter subgroup. Then there is a unique subgroup (which is parabolic) such that if and only if has a specialization in whenever . In addition, lies in the radical of and the specialization of centralizes .
Proposition 4.2.2.
Let be a reductive group and be an action of on a proper separated scheme . Then for all ,
if .
We are now ready to define the Tits building for .
Definition 4.2.3.
Let be a reductive group. The rational flag complex is the set of non-trivial modulo the following equivalence relation:
if there exists and such that .
In the rest of the chapter, it is shown that is the set of rational points in a geometric realization of a simplicial complex (which is the spherical Tits building). On this geometric realization, we may construct various metric space structures. This Tits building is supposed to generalize the notion of a Riemannian symmetric space to groups over an arbitrary field, which I will now explain.
Definition 4.2.4.
Let be a Riemannian manifold such that for all , there exists an isometric involution that fixes and reverses all geodesics through . Then is called a symmetric space.
It turns out that the classification of (simply-connected) symmetric spaces is essentially the classification of Lie groups. First, there are three types: Euclidean (zero curvature), compact type (nonnegative curvature), and noncompact type (nonpositive curvature). Then every symmetric space is of the form , where is a Lie group and is some compact subgroup. In the noncompact type case, is always a maximal compact subgroup of .
Example 4.2.5.
Some examples of symmetric spaces are the upper half plane, hyperbolic space, spheres, and .
4.3 Examples of Stability
From now on, we will work over . If you care about other ground fields, go find a reference and read yourself.
4.3.1 Points on the Line
The first example we will consider is the space of points in . Recall that any set of points in . We will describe the space of (semi)stable points with respect to the action of . In Anna’s lecture [chapter 1], we discussed the approach of computing invariants for this action, but this problem is extremely hard.
Recall that a set of points in determines a degree polynomial in the variables . We will assume that because the story for a single point in is very boring (really this should be ). First, we recall some basic Lie theory. Every simple Lie algebra has a maximal solvable subalgebra and the Cartan subalgebra . Lifting to the situation for groups, this gives a Borel subgroup and a maximal torus . In particular, all Borels are conjugate and thus all maximal tori are also conjugate. For , the Lie algebra is the set of traceless matrices, is the set of upper triangular matrices, and is the set of diagonal matrices.
This means that WLOG, we may assume that for any , that the image of is the maximal torus
Thus all -parameter subgroups are determined by their weight . By Remark 6.12 in Hoskins, it suffices to check the Hilbert-Mumford criterion for primitive , or those with weight . This reduces the computation to checking one case.
Write our polynomial as . Then it is easy to see that
and therefore, we have
In particular, is determined completely by the multiplicity of the root . We have
- 1.
if and only if has multiplicity less than or equal to .
- 2.
if and only if has multiplicity strictly less than .
Generalizing to all roots of , we see that:
- 1.
is semistable if and only if all of its roots have multiplicity at most .
- 2.
is stable if and only if all of its roots have multiplicity strictly less than .
The actual quotients were described in Anna’s lecture [chapter 1].
4.3.2 Plane Cubics
We would like to describe the space of plane cubics up to projective equivalence. First we will describe the reducible (or nonreduced) plane cubics. Every conic is equivalent to the conic given by , and so any union of a conic and a line is given by either
- •
and the line meets the conic in two distinct points;
- •
, and the line is tangent to the conic.
These are the possibilities for the unions of three lines:
- •
a triple line;
- •
a double line and a distinct line;
- •
three concurrent lines;
- •
three distinct lines.
The singularities of all non-irreducible possibilities are: two double points, a single double point (with tangency ), a line of triple points, a line of double points with a triple point, a single triple point, two double points.
Next, all irreducible plane curves are given by a Weierstrass form
Thus we have three cases: smooth (elliptic curve), nodal, and cuspidal. This gives a complete classification, but it does not tell us which curves are (semi)stable.
Theorem 4.3.1 ([H], Lemma 7.25).
A plane cubic curve is semistable if and only if it has no triple point and no double point with a unique tangent. A plane cubic curve is stable if and only if it is smooth.
Proof.
We will represent the curve by a homogeneous polynomial . In addition, all -parameter subgroups of are conjugate to those of the form
where . Also, we may assume that . If we write
Then acts by
Thus we have
Because we assumed , the monomials that do not appear are if . Then if the term also vanishes, then is a triple point. Otherwise, it is a double point with a unique tangent (if we affinize with , then the lowest degree homogeneous term is , which means we have a double tangent).
Then if is a singular point, we can assume and then by the Jacobian criterion, we have . With weights , we can see that . On the other hand, if is not stable, then there exists weights such that . Then have strictly negative weights. Then if is nonzero, we have . However, because , we have , . This implies that
and because , we see that . This implies that all terms without vanish, so is reducible and thus the curve is singular. □
There are three strictly semistable orbits: nodal cubics (), union of conic and secant line, and three non-concurrent lines. It turns out that the first orbit contains both of the others in its orbit, and this corresponds to the point at infinity in the compactification of the quotient of the stable locus. It is known that the quotient of the stable locus by is isomorphic to , so the compactification is .
5 Nicolás (Oct 23): More Examples of Stability
Let be a reductive algebraic group, a smooth projective variety, and an invertible sheaf over with a -linearization. Throughout this talk we work over an algebraically closed field of characteristic zero.
Definition 5.0.1.
is semistable if there exists such that .
Then we have a categorical quotient of . The semistable locus is dependent on the line bundle. Also, we can write . Then . Here, we have
To find , it suffices to compute .
5.1 Matrices and Conjugation
Let . Consider the action of on by conjugation. The action on functions is given by . To write the quotient , we compute the ring of invariants. This is simply
where is the -th coefficient of the characteristic polynomial.
Proof.
Let and let . We can replace by its Jordan canonical form
Then if we multiply all by some , we take to see that . But then
Now to compute the stable locus, we note that if for all , then is nilpotent. Thus is unstable if and only if it is nilpotent.
Recall the Hilbert-Mumford criterion and the following proposition from Patrick’s lecture [chapter 4]:
Proposition 5.1.1.
Let , . Then if and only if .
Now write . The following proposition is from the notes of Jenia Tevelev at http://people.math.umass.edu/~tevelev/moduli797.pdf.
Proposition 5.1.2.
The following are equivalent:
- 1.
is unstable.
- 2.
There exists such that .
This means that up to changing in its orbit, we may assume . Then recall that
If , then we write . Then we will write for the convex hull.
Theorem 5.1.3.
([D], Theorem 9.2) is semistable if and only if .
We now return to the example of matrices and conjugation. Fix to be the standard maximal torus in . First, take the decomposition
Then the conjugation is
If , then there exists taking to its Jordan form. Then if , then
Then note that and . If for some , then .
5.2 Subschemes of Projective Space
Let act on . This gives an action on , which is the same thing as an action on degree hypersurfaces in . Letting be the standard maximal torus, we obtain the following grid for the values of :
Then semistable monomials are given by sweeping out a line through the origin and seeing which monomials lie on one side. For example considering the line , the allowable monomials are . If we have such a curve , then if we set , the point has multiplicity at least .
The next option is to include but not or . One of the choices gives us the monomials , and thus the curve is reducible.
Now we will consider points in projective space. We will construct . We will choose the line bundle . We know that is ample if and only if for all . We say that is democratic if .1
Choosing a -PS with weights , fix . Writing with , we see that
Using some basic manipulation, this becomes
Theorem 5.2.1.
(D, Theorem 11.1) A point if and only if for all ,
Proof.
Suppose and . The other direction is similar. We will neglect all of the explicit computations. □
6 Morena (Oct 30): GIT and Moduli Problems
6.1 Moduli Problems
One of the central problems in algebraic geometry is to classify various geometric objects: curves, schemes, sheaves. We want to find a space such that is in bijection with isomorphism classes of our objects. However, this is trivially solved by taking a disjoint union of points, so we need a stronger notion of moduli problem. This is given by a functor
that sends a scheme to the set of families of objects over .
Example 6.1.1.
Consider the Hilbert functor that sends a scheme to
Theorem 6.1.2.
The Hilbert functor is represented by a quasi-projected scheme.
Example 6.1.3.
Consider the functor that sends
This functor is represented by the scheme .
Recall that there is a functor . Then is the same thing as specifying sections of basepoint-free line bundle. For such that , we obtain sections
Because projective space does not see constants, we see that these maps glue and we obtain a global map.
Conversely, consider and take the sections . These do not simultaneously vanish, so we obtain an invertible sheaf with generating sections.
An obstruction to the solution of a moduli problem is the existence of nontrivial automorphisms. Suppose we want to classify vector bundles with rank over an algebraically closed field . Call this functor
If represents this functor, then a curve with nontrivial vector bundle is a map . Then given a cover of by affines, then any vector bundle is trivial when restricted to . But then the trivial bundle is a closed point , and so if we attempt to glue the maps , then we obtain the map and thus must have been trivial in the first place. Therefore, there is no fine moduli space.
The next moduli problem we will consider is the moduli problem of elliptic curves over . These are curves of genus with one marked point. To generalize this to families, we will take
Then a morphism is a cartesian diagram
that commutes with the sections . Here, the criterion of being Cartesian means that defining the functor on morphisms is well-defined, because pullbacks are unique.
6.2 Dealing with Nontrivial Automorphisms
Unfortunately, there are nontrivial automorphisms of elliptic curves. Over (or any algebraically closed of characterstic ), we can write elliptic curves with the equation . We then have an action of on , where we have
Then the orbit of is the set
Generically, this has six points, but sometimes there are fewer (for example . Then we have the -invariant
and when are in the same orbit, we have . If we solve for , we obtain a polynomial of degree , and thus generically there are six roots. Then we can see that all solutions belong to the orbit of . Therefore, we have
This discussion can be found in more detail in Hartshorne, Chapter 3, Section 4. We then see that if and only if . Therefore, we have a bijection . In fact, Silverman proves this identity for all fields.
However, we can show that has no fine moduli space. Consider the elliptic curves and over . We see that and are not isomorphic at the level of -points because contains no roots of . However, if we base change to , then we obtain an isomorphism of the two curves by
This gives us a contradiction because for any field extension and scheme , we have .
There are two ways to resolve this problem of nontrivial isomorphisms:
6.2.1 Stacks
We will in some sense, allow automorphisms. We can enlarge the category of schemes to algebraic spaces or stacks, in which we can restate the same problem. However, if we work with stacks, we will have to consider functors valued in the -category of groupoids. For a functor , we will consider the functor
where is the category with objects the geometric objects over and morphisms the -morphisms of such objects. The study of such things often becomes very topological. For elliptic curves, we can also rigidify our elliptic curves, we obtain a functor . This can then be represented by a scheme.
In this case, a rigidified elliptic curve is an elliptic curve with nowhere vanishing section . Then automorphisms of rigidified elliptic curves are always trivial. Then writing in Weierstrass form, any map over extends to a map . This is represented by a matrix
where . Then we can find a global nowhere vanishing section . If preserves , we see that
and thus . Thus we have killed all automorphisms. In fact, is represented by an affine scheme , where is the discriminant. Note here that are here because under an automorphism, we make the substitution , and become .
Now consider the action of on that scales the holomorphic -form by (weight ). Then we can define the morphism that simply forgets . Then we can define a quotient and this gives a map to . In this case, this is not a scheme, so we should really write the quotient stack .
6.2.2 Coarse Moduli
Some of us are plebs who can only think about schemes.1 We will loosen our requirement for representability and find a scheme which approximates the problem as well as possible. We say a scheme coarsely represents if
- 1.
is a bijection over algebraically closed fields.
- 2.
For all such that the first item holds, there exists a unique morphism that makes the diagram commute.
There are several problems that may arise: either may be non-separated or it may have infinitely many components. In addition, even if is separated with finitely many components, it may not be a scheme. Here are some more problems:
- 1.
If we take varieties with no extra data, we run into the problem of having many automorphisms. This can be fixed by adding the datum of a polarization, which rigidifies the problem. For proper, we can consider composed of line bundles such that for some . Then an inhomogeneous polarization is a coset formed by ample line bundles. For example, for of genus at least , the dualizing sheaf is ample, and then is very ample.
In Mumford, if we follow references [29] and [135], we see that requiring smoothness makes the the coarse moduli space is separated and has finitely many components. However, smoothness is an open condition, so our moduli space will not be compact and thus will not have a good intersection theory. For example, take the intersection of two lines in the plane. Two parallel lines do not intersect in , but they do intersect at infinity in . Thus, we will define the space , which is the moduli space of stable curves, where we allow nodes. In stack form, this becomes a proper Deligne-Mumford stack..
7 Morena (Nov 6): GIT and Coarse Moduli Problems (Part 2)
7.1 Moduli of Curves
Recall that the functor sends a scheme to the set of isomorphism classes of smooth curves of genus over . More precisely, these are flat and proper morphisms such that geometric fibers are curves of genus . Then recall that morphisms are given by cartesian diagrams. We will now rigidify this problem to find a coarse moduli space.
Theorem 7.1.1 (Olsson, Lemma 8.4.6).
Let be a curve of genus . Then consider .
- 1.
is locally free of rank .
- 2.
The map is surjective and is a closed embedding. Here, is simply the relative Proj of the sheaf over .
Definition 7.1.2.
Define the functor of rigidified smooth curves by
where the map of projective spaces is an -isomorphism. Here, morphisms are cartesian squares
such that
| (7.1) |
commutes and the right square is commutative.
Fortunately, there are no nontrivial automorphisms, so there is hope that this functor is represented by a scheme.
Theorem 7.1.3.
The functor is represented by a locally closed subscheme of , where
by the Riemann-Roch theorem. The embedding is given by
Now note that is a projective scheme. Then the identity map corresponds to the universal closed subscheme . However, is not smooth (but it is flat and proper), so we can take the non-smooth locus, so then we can take parameterizing smooth subschemes.
Now to impose geometrically connected fibers, we use Stein factorization (Stacks Project, 37.48.5). This says many things, but what we need is that if we have proper, then we can find such that is proper and has geometrically connected fibers and such that is integral and is an isomorphism over an open subscheme of . This implies using the Hilbert polynomial that and that the genus is . Thus we have an open subscheme parameterizing smooth curves.
Now the identity gives a universal smooth curve
To recover , we need this to factor through . Therefore, we need the pushforwards of and to pushforward to the same sheaf under . Here, the two sheaves give morphisms and the square
is cartesian. Then is separated, so is a closed embedding. Thus we have a closed subscheme . Now we ask for a surjective map
This is the same as the cokernel vanishing, so now we have an open subscheme. This is also Proposition 5.1 in Mumford.
Now we have shown that the rigidified moduli problem is representable. Then we consider the map that forgets , and so we have an action of on by postcomposition with . Now we define . Then we have a map compatible with . Here is a “principal bundle” and so is injective and locally surjective, which means that for all there exists and Zariski cover such that .
Then checking , we use the large commutative diagram (7.1) to see that are related by some element of . Then there exists a geometric quotient and this is a coarse modulu space for . Thus we have a bijection between maps and -invariant maps . This is equivalent to the orbits of to be bijective with , where .
Now recall from Theorem 3.4.5 that if is of finite and is a reductive group with , then there is a universal categorical quotient of and a uniform geometric quotient of (in characteristic , uniform is upgraded to universal). Now we want to express as the locus of stable points of something, and in particular as .
To do this, we will introduce the Chow scheme. The Chow variety parameterizes subvarieties of of dimension and degree . This is defined by
Then projecting down to , we obtain a hypersurface defined by the Cayley form . Then the coefficients of map to .
Now we have a map . Over , we use a push-pull with respect to
where is defined by first defining , then pulling considering the maps , and taking .
Now recall from Proposition 1.14 of Mumford that and similarly for the stable locus. Thus we can reduce to the case where is algebraically closed, and by Proposition 1.18 in Mumford, if is -equivariant, then is quasi-affine. Thus . Then by Theorem 6.2 in SGA 8, an injective morphism of finite type is quasi-affine. If we require that is finite, is proper, and is ample, the inclusion becomes an equality by Proposition 1.19 in Mumford.
Now recall that we have . Then is quasi-affine and . Then by Theorem 4.5 of Mumford, we use the Hilbert-Mumford stability criterion to see that for any smooth curve , the point is stable if
- 1.
is not contained in a hyperplane.
- 2.
is defined by a complete linear system.
- 3.
and . This is equivalent to .
Thus the image of is contained in the properly stable locus and thus we have is equal to the properly stable locus. Thus we obtain a geometric quotient.
8 Kuan-Wen (Nov 13): GIT and the moment map
8.1 Moment maps
Let be a symplectic manifold. Here, this means that is a smooth manifold and is a closed nondegenerate -form on . Now Let be a compact Lie group acting on symplectically, which means for all . Define the Lie algebra of to be .
Definition 8.1.1.
The moment map is a map which is
- 1.
Equivariant. Here, we use the coadjoint action on .
- 2.
The map satisfies the identity
for any and .
Now we will discuss existence and uniqueness of the moment map. For uniqueness, we know that
- 1.
is unique up to addition of a constant in .
- 2.
If is semisimple, then the only fixed point of the coadjoint action is , and thus is in fact unique.
- 3.
If we require for compact, then we obtain a unique moment map.
Now we will discuss existence. The following result is from reference [723] of Mumford’s book.
Theorem 8.1.2 (Marsden-Weinstein).
If is semisimple, then the moment map exists.
Next, it is known that if and is connected, then the moment map exists. If is a torus, this comes from a differential equation, but the obstruction comes from and so we are fine here. In general, we have an exact sequence
so by the previous discussion, we now have the existence of a moment map.
Example 8.1.3.
Consider the standard action of on . Then the mment map is given by
where is any lift of in . This is unique up the the addition of the trace map , which is the nontrivial fixed point of the coadjoint action. This follows from the fact that all characters of are powers of the determinant map. To see invariance, note that
Using the Fubini-Study metric on and the fact tha tthe action is transitive, we will compute at the point with coordinates . Recall that the Fubini-Study form is given by
Now we can directly compute
This gives the desired result.
Now suppose and acts on with an action induced from a linear action on . Then we can assume that acts unitarily, so then se can write the moment map as
This can be written explicitly as
where is the linear action.
We will now discuss quotients in the symplectic category.
Theorem 8.1.4 (Marsden-Weinstein).
Let act on and be the moment map. Let be a fixed point of the coadjoint action. If acts freely and properly on , then the quotient is a symplectic manifold of dimension . If is the inclusion and is the projection, then we have the identity . If is not fixed, we can replace by the stabilizer of .
Proof.
We have the equality
This comes from the differential equation in the definition of the moment map, and then we also have
To get the dimension equality, we have a short exact sequence for any given by
and thus . Now we replace with to obtain the desired result. On the other hand, we also have
By these computations, we know that is a regular value if and only if the action of on is locally free.
Now is an isotropic subspace of because we have
Next, for isotropic subspaces of , there is a symplectic form induced by on . This gives a symplectic form on , and so we have a symplectic form on the quotient.
Finally, to prove closedness of this form, this follows from the fact that
and the fact that is injective. □
8.2 Comparison of symplectic quotients and GIT quotients
Now let is reductive with a linear action such that acts on some projective variety . Now let be a maximal compact subgroup.
Theorem 8.2.1.
- 1.
A point is semistable if and only if .
- 2.
There is an inclusion (by Proposition 2.2 in Mumford). This induces a homeomorphism with the GIT quotient.
Proof.
First, for all , define . Then from Kempf-Ness (reference [175] in Mumford), we have
- (i)
All critical points of are minima.
- (ii)
If attains a minimum, it does so on exactly one double coset in .
- (iii)
attains a minimum if and only if the orbit of is closed in .
Also, because we can write the moment map explicitly, we have if and only if . Finally, we know that if is closed in , then is semistable and is closed in . To prove the second part, if , then there exists invariant such that . This means , so we can choose such that . Multiplying by roots of unity, we may assume that . Projecting down, we have .
Using all of these observations, we can prove the two parts of the theorem.
- 1.
If , then contains some closed orbit . This implies that . The converse simply follows from the observations and Proposition 2.2 in Mumford.
- 2.
We know is a continuous map from a compact space to a Hausdorff space. It suffices to show taht this is a bijection. Then recall that satisfies the condition that if and only if their orbit closures intersect, so by the first part of this theorem, we have surjectivity.
For injectivity, if but , then we know that by the previous discussion, but also are closed in , which gives a contradiction. □
We conclude with some remarks. It is easier to compute the cohomology of than it is to compute the cohomology of . To do this, if we consider the Morse function , where the norm comes from the Killing form, then we have a Morse stratification and then we define the functions
This allows us to compute the cohomology.
9 Alex (Nov 20): GIT quotients of symplectic manifolds
9.1 Kähler and Hyperkähler quotients
Let be a compact Kähler manifold and be a complex reductive group. Then is a maximal compact subgroup and suppose that the action of on preserves . Let be the moment map. Recall the Marsden-Weinstein theorem from last time. Define
This should be thought of as the stable locus in the symplectic quotient.
Theorem 9.1.1.
There exist a Kähler structure on
where the equivalence here is diffeomorphism and is open.
To sketch a proof of this, we first note that if , then if we call the quotient , we have
This gives us an almost complex structure on the quotient, and by reference [624] of Mumford, this is integrable and compatible with the symplectic structure. Here, the second isomorphism comes from the identity
Now we also need to define this mysterious . Define the function , where we fix an invariant inner product on . Then define
This should be thought of as the normal bundle of . In fact, we have and then we have .
We now discuss the hyperkähler case.
Definition 9.1.2.
A manifold is hyperkähler if are complex structures and . This gives us an entire of complex structures.1
The standard picture for a hyperkähler manifold is the space . Note that a hyperkähler structure gives three Kähler structures. Now if we have the corresponding moment maps , the hyperkähler quotient is
Example 9.1.3.
An interesting hyperkähler quotient is the space of anti-self-dual instantons, or principal bundles over . Alternatively, this is the space of holomorphic -bundles over with . There is a construction of Atiyah, Drinfeld, Hitchin, and Manin (ADHM).
This is a hyperkähler quotient of . We consider elements of as complex matrices complex matrices of size satisfying
- 1.
.
- 2.
.
- 3.
The matrix
is injective and the matrix is surjective.
Then define the action of on by
All of this is in a paper by Donaldson.
9.2 Desingularization
There are two reasons that can have singularities:
- 1.
is not a regular point.
- 2.
The action of on is not a free action.
Remark 9.2.1.
If is surjective for all , then .
Note that not all finite quotient singularities cannot be resolved by equivariant blowups.
Theorem 9.2.2.
If , then there exists which is a blowup of along -invariant nonsingular subvarieties with a suitable -linearization such that and such that is a birational morphism and only has finite quotient singularities.
To prove this, if we let be a reductive subgroup, then we define . Now if we define , we define to be a choice of representatives of conjugacy classes of reductive subgroups such that and . Then we outsource the fact that the set
is a disjoint union of -invariant subvarieties and thus we can set to be the blowup of along these subvarieties. Now this procedure decreases , and so by a finite process we end up with only finite stabilizers.
Remark 9.2.3.
This quotient is sometimes quite nice for algebraic geometry but not for symplectic geometry. Let act on by
Then we know that because the ring of invariant functions is generated by .
However, if we take the symplectic quotient, we have the moment map
and thus we have
Thus we have nontrivial stabilizers, where for example the point is fixed by and the point is fixed by . Therefore the symplectic quotient is an orbifold, and it can be resolved by using a Poisson bracket. To do this, the space has a stratification given by conjugacy classes of stabilizers such that each stratum is a symplectic manifold.
9.3 Yang-Mills functional
In some sense this is an infinite-dimensional version of what we were doing before. Here, is a compact Kähler manifold and is a holomorphic vector bundle with a Hermitian metric.
Definition 9.3.1.
A connection on is a map satisfying the Leibniz rule:
Note that by extending the Leibniz rule, we can define by
Note that if are connections, then and thus we have . Also note that . This is called the curvature form. Now if is a unitary connection, then everything lives in , where . Furthermore, the space of connections is an affine space . Now we define the Yang-Mills functional to be
Then the Yang-Mills equation is , where is the Hodge star. The gauge group is the group of unitary automorphisms of , and the corresponding Lie algebra is .
Now a complex structure on gives us an operator with kernel the holomorphic sections. Then there exists a unique unitary connection such that . Because , this implies that . We have no identified holomorphic structures on with the space of unitary connections with curvature. Finally, there exists a action such that holomorphic structures are biholomorphic if and only if they are in the same orbit. Finally, the Hermitian Yang-Mills connections are those that minimize the Y-M functional and those exist if and only if the bundle is slope-stable.
10 Patrick (Dec 04): Everything you need to know about Morena’s lectures
Note: These are the speaker’s notes. Throughout this lecture, we work over an algebraically closed field of characteristic .
10.1 Chow Varieties
Recall that the Chow variety parameterizes algebraic cycles of of dimension and degree . Denote the dual projective space by . Now points of are hyperplanes in , and so we define
Now projecting , obtain a hypersurface defined by a Cayley Form . Now the coefficients of give us a point .
Now we will discuss the stability of the Chow point of a nonsingular space curve. There are two key things we need to know:
- 1.
The Chow form is compatible with specialization.
- 2.
How to compute the Chow form of a cycle of linear subspaces.
For the second, we will do this explicitly. Let be coordinates on and be coordinates on . Now if is the linear subspace defined by , we can compute
Then if is the -PS of given by diagonal matrices with weights , we have
Now for a smooth curve in , we will find the cycle that is the specialization of by as . This will typically be a sum of lines and then we can compute easily. Assume that the weights of are . We will assume that is not contained in any hyperplane. Now recall that points on but not on specialize to , where we have and for . Thus almost all points of specialize to and therefore the specialization of is determined by .
Now consider the composition , which now defines a rational map with basepoints . Suppose this map is given at the point by (abusing notation and treating this list as an ideal). Then if is the normalization of the blowup of at the ideals . Now on , the divisors specialize to the divisor given by the zeroes of the function on . Thus is the image of the union of the proper transform of an the exceptional divisors . Because is taken to a point, we have , where is the replacement of the rational map .
Now we need to describe . To do this, we will represent by a power series. If is a local coordinate at , then suppose that is given by
Now at least one and thus the rational map is given by
where is the coordiante on .
Lemma 10.1.1.
Let be a smooth surface and . Let are local coordinates and
for integers . Let be the normalization of the blowup of at . If we plot the points on and construct the Newton polygon, then there is one exceptional divisor for each edge of the Newton polygon.
Now if we compute the Newton polygon of the , then the divisors correspond to slopes such that for some ,
| (10.1) |
This allows us to compute
where joins the two points and . WE are now able to compute
By combinatorial magic, can be rewritten as
where . Here is some more magic:
Lemma 10.1.2.
The number if the satisfy
- 1.
;
- 2.
for ;
- 3.
for ;
and if and and , , and not all .
This then tells us that
Theorem 10.1.3.
Let be a smooth curve in . Then is stable with respect to if
- 1.
is not contained in any hyperplane;
- 2.
is defined by a complete linear system;
- 3.
and .
The proof of this result relies on Riemann-Roch and the Weierstrass Gap Theorem (about linear systems associated to points).
Remark 10.1.4.
If and , then any smooth curve spanning is stable. However, there are elliptic curves in that are unstable.
10.2 Construction of
Recall that Chow points are divisors in that come from sections of the sheaf . Denote the projective space of such divisors by .
First we will construct a morphism for any Hilbert polynomial of the form . This morphism will send a subscheme to its Chow point and will be -equivariant. The construction will define a functor from the Hilbert functor to the functor represented by . Essentially, to a flat subscheme with Hilbert polynomial , we will assign a flat effective Cartier divisor which induces divisors of degree on the fibers over .
Define be the hypersurface given by the vanishing of the polynomial . Now consider the variety and the morphisms
We will define . We now have the following facts:
- 1.
Points of are triples , where .
- 2.
is a fiber bundle with fiber .
Now consider the diagram
Now given a closed subscheme flat over , we will compute a Cartier divisor. To do this, let be the fiber product of the following diagram:
Now we write , where the Div of a morphism was defined in Section 5.3 of Mumford. Here if is a projective morphism of Noetherian schemes such that
- 1.
is of finite Tor-dimension. This means that for all , the ring is a -module of finite Tor-dimension, and this dimension is bounded over all ;
- 2.
is empty if has depth and finite if has depth ;
then there is a canonical effective Cartier divisor whose support is contained in .
Therefore, we need to check that
- 1.
is of finite Tor-dimension;
- 2.
is a relative divisor over .
- 3.
When for an algebraically closed field , has empty fibers over points of depth and finite fibers over points of depth .
- 4.
When , the divisor has degree .
The first item follows from
Lemma 10.2.1 (Fogarty).
Let
be morphisms of finite type between Noetherian schemes. Suppose is a smooth morphism with fibers of dimension and that is flat. Then has Tor-dimension at most .
Proof of this follows from SGA 2, Theorem 4.15. Next, 2 follows from 3. The proof of 3 and 4 is apparently the “classical theory of Chow forms.” Because , we know that there exists some set of hyperplanes such that . Thus is empty when is the generic point. If is the generic point of a divisor, then the set of hyperplanes such that is the set of closed points on a subset of codimension at least , and this is also classical.
Finally, to show that is of degree , we will show that it is the Chow form of the cycle of -dimensional components of .
Definition 10.2.2.
Let be of finite type of dimension over . Let be the components of and the underlying set of . Write be the generic point of and . This is an Artinian local ring of length . Now define
If is an effective Cartier divisor, then this construction recovers the corresponding Weil divisor.
Now recalling the diagram
we define the Chow divisor of by . The following lemma will complete the construction of our morphism .
Lemma 10.2.3.
Let be smooth varieties and be a projective morphism. Let be a closed subscheme such that . Let . Then .
The rest of the construction of can be found in Morena’s second lecture.
11 Nicolás (Dec 11): The moduli space of stable curves
11.1 Introductions
Recall the functor . We constructed a coarse moduli space , which is a smooth quasiprojective variety contained in some . This allows us to compactify our moduli space in a natural way. We want our compactification to have a geometric meaning, which means some meaning in terms of moduli problems.
Example 11.1.1.
Let us consider the moduli problem of distinct points up to projective isomorphism. Then we can embed . Recall that are semistable as long as no three points are equal. Thus we can consider the semistable orbits in the GIT quotient as paramerizing tuples with repeated points.
11.2 Stable Curves
From here, we will follow the book Moduli of Curves by Harris and Morrison.
Definition 11.2.1.
A stable curve is a complete connected curve with only finitely many nodal singularities and finitely many automorphisms.
Recall that if is a component and , then we know . When , then has infinitely many automorphisms, but if we choose a marked point, then has finitely many automorphisms. When , we need to fix three points. In the smooth case, this kills all automorphisms.
Theorem 11.2.2 (Deligne-Mumford-Knudsen).
There exists a coarse moduli space of stable curves for that is a projective variety. Here, is the arithmetic genus of the curve.
We may alternative define as follows. If and there are nodes, we can define
The previous theorem is one of the most important results in the history of the algebraic geometry of curves.1 By the valuative criterion of properness, we can fill in families of curves with missing fibers.
Proposition 11.2.3 (HM, Prop 3.49).
Let be a smooth curve and fix . Let be a family of stable curves. Then there exists totally ramified at and a family which extends such that all fibers are stable curves. If two such coverings exist, they are dominated by a third, so the central fiber is unique.
Remark 11.2.4.
A reason this theory is well-known is because the theory of surfaces is well-understood. If we want to perform analogous constructions for the moduli of surfaces, we need to understand the minimal model program for threefolds.
11.3 Some Examples
In our setup, we may have some non-allowed singularities. Alternatively, we may have multiple fibers. In the first scenario, we can resolve the problem by blowing up. If we have multiple fibers, we need to remove the multiplicity by taking a base change. There is an explicit algorithm for this, described in [HM, p.124–126].
- 1.
Begin with , and let be the fiber at .
- 2.
Fix some prime such that for some .
- 3.
Let be locally given by and consider the fiber product . Take the normalization .
- 4.
The map ramifies , where with . The new fiber can be computed as follows:
- •
Take the preimage of .
- •
Multiply by all components contained in the branch locus.
- •
Divide all components by .
- •
Example 11.3.1.
Consider a family of curves with smooth fibers for but a triple point over . Define by the equation (locally)
where vanishes to degree at least at . Let be the singular fiber over . We now blow up at the singular point, and the new fiber is . To remove the multiplicity, we choose . Then is ramified over and the fiber of is a generically to cover of ramified at the intersection points with . By Riemann-Hurwitz, we have
and thus . Finally, we see that the fiber is .
Remark 11.3.2.
There are not many curves with automorphisms of order , and so we have . This uniquely determines .
Example 11.3.3.
Let be a smooth conic and a general smooth quartic. Consider the family . When , this is a generic quartic, but when , the fiber is a double conic. Assume that has points (so all intersections are transverse). Then the total space is not smooth because around each intersection, we have an intersection that is locally , which is an singularity.
Now we have the fiber over with marked points. We blow up each of the singularities and obtain the new fiber (here, each is a -curve). Now choose and follow the procedure, and we obtain a new family ramified at . Now the fiber above is a cover with ramification points. By Riemann-Hurwitz, we compute . The new fiber is . Then the are now -curves because , so we can contract them to resolve the instability of the by Castelnuovo’s criterion. Once we contract the , the new fiber is .
Thus the central fiber of the family is a genus hyperelliptic curve ramified over the intersection points .
Remark 11.3.4.
We know that a generic quartic curve has bitangents. Under the limit , the lines become the lines joining two points in .