Intersection theory
Intersection Theory Learning Seminar
Spring 2021
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.
Acknowledgements
I would like to thank Nicolás Vilches for pointing out that I was missing the accent in his name.
Seminar Website: http://www.math.columbia.edu/~plei/s21-INT.html
1 Caleb (Jan 22): Rational Equivalence
The idea of intersection theory is to define a well-defined intersection product on a scheme or variety that behaves like the intersection form in ordinary homology of a manifold.
Example 1.0.1.
Let be two plane curves. Then we can define the intersection multiplicity by
which is the dimension of the intersection scheme. However, this does not generalize, so we will need to use Serre’s formula, which involves the functor.
1.1 Chow Groups
We want to define zeroes and poles for a singular variety. If is codimension in for a nonsingular variety , then consider , where , where this local ring is a discrete valuation ring. Then we can write , which is a Weil divisor. For a curve, we can compute . In general, we can write , where .
Example 1.1.1.
If and , we see that along , we have and thus .
Proposition 1.1.2.
The map is a homomorphism.
Proposition 1.1.3.
For any , there are finitely many such that .
Examples 1.1.4.
- 1.
Let be irreducible and be plane curves. Then .
- 2.
Let be normalizations. Then .
Now we will define rational equivalence. Let be a -dimensional subvariety of and . Then we may obtain cycles of the form , and such cycles form the group .
Definition 1.1.5.
The Chow groups of are defined by .
Example 1.1.6 (Not Mayer-Vietoris).
Let be closed subschemes of . Then the sequence
is exact.
1.2 Pushforward of Cycles
Let be a morphism. We want to define a pushforward . This would imply that . This is true when is proper. Then is a map of cycles because proper implies universally closed. We obtain a field , which is finite if are of equal dimension. Then we define
Definition 1.2.1.
The pushforward of a cycle is .
Theorem 1.2.2.
Let is proper. If , then .
Proposition 1.2.3.
Let be proper and surjective and let . Then
- 1.
if ;
- 2.
if .
Proof.
Consider the map . Because the order map is a homomorphism, let be an irreducible polynomial, where has degree . Then and then we have
Now suppose is finite. Given of codimension , write . Construct a domain such that is finite and such that are obtained by . Here, we need to show that
Now if , we want , and then the LHS becomes . In general, we need to apply results from EGA III to show that exists.
Now in the different-dimensional case, we replace with and reduce to the case of a curve over . □
Corollary 1.2.4 (Bezout’s Theorem).
Let and suppose are plane curves of degree with no common component. Then .
Proof.
Assume is irreducible and replace with . Then and therefore
so we may as well assume are lines and then the result is obvious. □
Definition 1.2.5.
We can define the degree to be .
1.3 Alternative Definition of Rational Equivalence
Let be a scheme with irreducible components . Then let . We define the fundamental class .
Proposition 1.3.1.
A cycle is rationally equivalent to if and only if there exist -dimensional subvarieties of such that are dominant and .
Sketch of Proof.
Suppose for some . Now defines , so we will define to be the closure of the graph. This gives a dominant rational map and thus . The other direction is easy. □
Remark 1.3.2.
This discussion and more machinery allows us to show that if is flat of relative dimension , then we can define pullbacks of cycles. Here, if is a -cycle on , then in , so we have
2 Avi (Jan 29): Intersecting with divisors and the first Chern class
Note: These are the speaker’s notes. Minor edits to the TeX source were made.
2.1 Cartier and Weil divisors
Let be a variety of dimension over a field . We want to introduce two notions of divisors, one familiar from the last chapter.
Definition 2.1.1.
A Weil divisor of is an -cycle on , i.e. a finite formal linear combination of codimension subvarieties of . Thus the Weil divisors form a group .
Definition 2.1.2.
A Cartier divisor consists of the following data:
- •
an open cover of ;
- •
for each a nonzero rational function on , defined up to multiplication by a unit , i.e. a function without zeros or poles, such that for any we have a unit on .
Like the Weil divisors, the Cartier divisors form an abelian group: (we can assume that the open covers are the same, since if not they refine to ). We call this abelian group .
Given a Cartier divisor and a codimension subvariety of , we define
for such that is nonempty; since each is defined up to a unit, this order is well-defined. We define the associated Weil divsior
This defines a homomorphism
For any rational function on , we get a principal Cartier divisor by choosing any cover and defining . It is immediate that the image of this divisor under the map to is the Weil principal divisor. Say that two Cartier divisors and are linearly equivalent if for some ; then we define to be the group of Cartier divisors modulo linear equivalence, and the above then shows that the map descends to a map . This map is in general neither injective nor surjective.
Notice that the definition of a Cartier divisor yields that of a line bundle on : given a divisor , define a line bundle to be trivialized on each with transition functions . Two Cartier divisors and are linearly equivalent if and only if , and so we get the alternate description of as the abelian group of line bundles on with group operation given by the tensor product. Conversely, given a line bundle , this determines a Cartier divisor up to some additional data: a nonzero rational section of . Therefore we can also think of Cartier divisors as the data of a line bundle together with a nonzero rational section.
We define the support or of a Cartier divisor to be the union of codimension subvarieties of such that is not a unit for nontrivially intersecting , i.e. is nonzero.
We say that a Cartier divisor if all of the are regular, i.e. have no poles.
2.2 Pseudo-divisors
In general, Cartier divisors are not well-behaved under pullbacks (although line bundles are). In particular, given the data of a line bundle and a nonzero rational section and a morphism , there is no guarantee that the pullback is nonzero. Therefore we enlarge the notion to make it behave better: let be a line bundle on , be a closed subset, and be a nowhere vanishing section of restricted to , or equivalently a trivialization of . A pseudo-divisor on consists of the data of such a triple , up to the following equivalence: two triples and define the same pseudo-divisor if and there exists an isomorphism such that restricted to we have . Note that this is well-behaved under pullback.
Example 2.2.1.
Let be a Cartier divisor, with support . Then each away from gives a local section of the associated line bundle , and so these glue to a section of on ; this makes a pseudo-divisor.
We say that a Cartier divisor represents a pseudo-divisor when and there exists an isomorphism such that restricted to we have , with notation as above.
Lemma 2.2.2.
If is a variety, then every pseudo-divisor on is represented by a Cartier divisor . If , then is unique; if , then is unique up to linear equivalence.
Proof.
If , then is a section on , and so a pseudo-divisor is just a line bundle; and we saw in the previous section that the group of Cartier divisors up to linear equivalence is isomorphic to the group of line bundles, so corresponds to a unique linear equivalence class of Cartier divisors.
If , let . As above, choose a Cartier divisor with . The section consists of a collection of functions on such that on ; thus on each intersection, i.e. there exists some rational function such that on each . Then is the Cartier divisor and by definition on each ; therefore using the definition above . Since is linearly equivalent to , it corresponds to the same line bundle, and since is regular on each the support of is contained in ; therefore represents .
For uniqueness, suppose that two Cartier divisors and both represent . Then similarly there must exist some rational function such that on each . But since , if , i.e. is nonempty, then and must agree on every , and so restricted to must be ; since is rational it follows that and . □
For any pseudo-divisor , as for Weil divisors we will write , , and .
If and are two pseudo-divisors, we can define their sum
This agrees with the sum on Cartier divisors, except that the supports may be larger in this case. Similarly defining
makes the set of pseudo-divisors into an abelian group.
Given a pseudo-divisor on a variety of dimension , we can define the Weil class divisor by taking to be the Cartier divisor which represents and setting , the associated Weil divisor from the previous section. The above lemma shows that this yields a well-defined element of ; this gives a homomorphism from the group of pseudo-divisors to .
2.3 Intersecting with divisors
Let be a variety of dimension , be a pseudo-divisor on , and be a subvariety of dimension . Let be the inclusion of into ; then the pullback is a pseudo-divisor on with support . We define the class in given by the Weil class divisor of :
For any closed subscheme containing , we can also view this as an element of ; we will also denote this by .
Let be a -cycle on , with support the union of the subvarieties such that is nonzero. For a pseudo-divisor on , we define the intersection class in by
As above, we can also view this as an element of for any containing .
We will apply this in two main cases. First: f , then the data of is just that of a line bundle as above; in this case the action of on a -cycle is called that of the first Chern class, written .
Second: if is the inclusion of into , then is called the Gysin pullback .
Theorem 2.3.1.
Let be a scheme, be a pseudo-divisor on , and be a -cycle on .
- (a)
Let be a -cycle on . Then
in .
- (b)
Let be a pseudo-divisor on . Then
in .
- (c)
Let be a proper morphism, be a -cycle on , and be the restriction of to . Then
in .
- (d)
Let be a flat morphism of relative dimension and be the restriction of to . Then
in .
- (e)
If the line bundle is trivial, then
in .
Proof.
Part (a) is immediate from the definition. Using part (a), then, we can assume by linearity that for some -dimensional subvariety . Restricting to , (b) is just the statement that taking the Weil class divisor is compatible with sums.
For part (c), we can likewise assume that for some -dimensional subvariety ; then is the restriction of the Cartier divisor representing to , and so we can assume that . Similarly on the right-hand side and so concerns only the restriction of to , and so we can assume that . In this case on the support of and so the statement is
since and . If is a map of degree and for some function on some open subset of , then from last time we know that locally
where is the determinant map from functions on subsets of to functions on their images, since since has degree . But locally we can always assume that is principal, and so as desired.
For (d), we can again assume that , so the statement similarly becomes
By linearity, we can assume for some subvariety of , at which point the statement is , which is true whenever is flat.
Finally for (e) we can again assume , so that the statement is in whenever is trivial, where is the dimension of . Letting be the Cartier divisor representing , we know from section 1 that is trivial precisely when is linearly equivalent to the trivial Cartier divisor for which every local function is a unit; and we know that the associated Weil divisor map descends to a map , i.e. whenever is trivial. □
2.4 Commutativity
Suppose that we have two Cartier divisors on an -dimensional variety . Then they both determine associated Weil divisors (and thus in ), and so it is natural to consider the intersections
Theorem 2.4.1.
In , we have
Corollary 2.4.2.
Let be a pseudo-divisor on a scheme , and be a -cycle on rationally equivalent to . Then
in .
Proof.
Given a closed subscheme and a -cycle on , we can construct its intersection for any pseudo-divisor on . This gives a map
The above corollary shows that in fact this map descends to a map
this is called intersecting with .
Corollary 2.4.3.
For two pseudo-divisors on a scheme and a -cycle on , we have
in .
Proof.
We can assume without loss of generality that for some subvariety of dimension . Then we can restrict and to , so that and similarly ; and then applying Theorem 2.4.1 immediately gives the result. □
For pseudo-divisors on and a -cycle on , we can then define inductively
in . Theorem 2.4.1 implies that the order of the is unimportant, and parts (a) and (b) of Theorem 2.3.1 implies that the action is linear in each and in . More generally if is a homogeneous polynomial of degree and is a closed subscheme of containing , then we can define in .
Definition 2.4.4.
We say that an algebraic variety is complete if for any variety the projection is a closed map.
For example, any projective variety is complete.
If and is complete, then we can define the intersection number
Similarly if is a homogeneous polynomial of degree in variables then we can define
For a subvariety purely of dimension , we will sometimes write simply instead of ; similarly we will sometimes write instead of .
Example 2.4.5.
Let be the projective completion of the affine surface defined by . Consider the Cartier divisor on defined everywhere by the equation , corresponding to the subvariety cut out by . Define the lines by and respectively, and let be the origin . Along the subvariety , from the defining equation we also have (in affine space), and so ; we have
where (in the affine variety) . Thus which has length , with maximal proper subsequence of modules given by . Therefore . We can compute
where is the inclusion of into , since restricted to the line the equation specifies only the point with multiplicity . Therefore there cannot exist any Cartier divisor with , since if there were we would have
in either or , by Theorem 2.4.1 and the above calculation. This proves our above claim that the maps and are not in general surjective.
2.5 The first Chern class
Let be a scheme, a subvariety of dimension , and a line bundle on . The restriction of to is a line bundle on and so is isomorphic to for some Cartier divisor on , determined up to linear equivalence. This in turn defines a well-defined element of ; we write . More generally, if is a -cycle on then define for each as above, and write
If for some pseudo-divisor , then if is the inclusion then the Cartier divisor on representing satisfies by construction; by definition, this means that and so
in .
Theorem 2.5.1.
Let be a scheme, be a line bundle on , and be a -cycle on .
- (a)
If is rationally equivalent to , then . Therefore there is an induced homomorphism .
- (b)
If is a second line bundle on , then
in .
- (c)
If is a proper morphism and is a -cycle on , then
in .
- (d)
If is a flat morphism of relative dimension , then
in .
- (e)
If is a second line bundle on , then
and
in .
Proof.
A line bundle on defines a pseudo-divisor with support , and so the analogous properties from Theorem 2.3.1 and its corollaries immediately imply these. □
2.6 The Gysin map
Fix an effective Cartier divisor on a scheme , with the inclusion given by . Then we define the “Gysin homomorphism”
for -cycles on .
Proposition 2.6.1.
With notation as above:
- (a)
If is rationally equivalent to , then , and so there is an induced homomorphism .
- (b)
We have
- (c)
If is a -cycle on , then
- (d)
If is purely -dimensional, then
in .
- (e)
If is a line bundle on , then
in .
All of these follow immediately from the definitions and the results above.
3 Alex (Feb 05): Chern classes and Segre classes of vector bundles
3.1 Chern and Segre Classes
Let be a line bundle over a scheme . If is a subvariety, then there exists a Cartier divisor such that , which gives us a cycle .
Definition 3.1.1.
We will define the first Chern class of by . Therefore we obtain a map
Proposition 3.1.2.
Here are some properties of Chern classes:
- 1.
The Chern class gives a well-defined map .
- 2.
We have .
- 3.
Let be proper and . For a line bundle over , we have
- 4.
If is flat of relative dimension , , and , we have
- 5.
Let be line bundles. Then
In the smooth case, Segre classes agree with Chern classes, but they have a nice generalization to the singular case. Let be a rank vector bundle over a scheme and let be the associated projective bundle. Then we have a line bundle , and so we define
If is a line bundle, we see that .
Proposition 3.1.3.
Let be a line bundle. Then
Now we will define Chern classes for an arbitrary vector bundle of rank . Set
and set
This tells us that . This gives us a map . Here are some new properties of Chern classes:
- •
For all , we have .
- •
If is an exact sequence of vector bundles, then .
Remark 3.1.4.
The projection formula, flat pullback, and give us uniqueness for all of the Chern classes.
3.2 Splitting Principle
Let be a vector bundle over . Then there exists a flat such that
- 1.
The pushforward is injective.
- 2.
There exists a filtration such that for some line bundle . In addition, we have
This gives us a splitting principle:
To prove a universal formula for Chern clases of some vector bundles with certain relations, it suffices to show that the formula is true for filtrations with line bundle quotients and the relations are preserved under flat pullback.
Here are some properties of Chern classes:
- 1.
If is the dual of , we have and , where are the Chern roots of .
- 2.
Let be vector bundles with Chern roots . Then are Chern roots for .
- 3.
Let be a rank vector bundle. Then has Chern roots .
Example 3.2.1.
Recall the exact sequence
From the fact that and , we can compute the Chern character of .
Let be a vector bundle on a nonsingular variety and suppose is the projection. Then we have an exact sequence
This allows us to calculate
3.3 Rational equivalence of bundles
This generalizes the Gysin map from last week. Let be a vector bundle of rank and let be the projection. Then we obtain a commutative diagram
Theorem 3.3.1.
- 1.
The map is an isomorphism for all .
- 2.
A cycle is uniquely expressible in the form
where .
Definition 3.3.2.
Let be . This is the Gysin map.
Proposition 3.3.3.
Let and that restricts to . Then where is the universal rank quotient bundle of . Here, we have an exact sequence
4 Patrick (Feb 12): Cones: because not every coherent sheaf is locally free
Note: These are the speaker’s notes.
4.1 Cones and Segre Classes
Our goal is to define a Segre class of a subvariety and study its properties.
4.1.1 Cones
Definition 4.1.1.
Let be a sheaf of graded -algebras such that is surjective, is coherent, and is generated by . Then any scheme of the form is called a cone.
If is a cone, then is the projective completion with projection . Let be the canonical line bundle on .
Definition 4.1.2.
The Segre class of is defined as
Proposition 4.1.3.
- 1.
If is a vector bundle on , then , where is the total Chern class.
- 2.
Let by the irreducible components of with geometric multiplicity . Then
Example 4.1.4.
Let be coherent sheaves and let be locally free. Then we may define . We may define . Now if
is exact, then .
4.1.2 Segre Class of a Subvariety
Let be a closed subscheme of defined by the ideal sheaf and let
be the normal cone. Note that if is regularly embedded in , then is a vector bundle.
Definition 4.1.5.
The Segre class of in is defined by
Lemma 4.1.6.
Let be a scheme of pure dimension and let be the irreducible components of with multiplicity . If is a closed subscheme of and , then
Proposition 4.1.7.
Let be a morphism of pure-dimensional schemes, a closed subscheme, and be the induced morphism.
- 1.
If is proper, is irreducible, and maps each irreducible component of onto , then
- 2.
If is flat, then .
Remark 4.1.8.
If is birational, then . This says that Segre classes are unchanged by pushforward along birational modifications.
Corollary 4.1.9.
Let be a variety and be a proper closed subsecheme. Then let and be the exceptional divisor with projection . Then
Example 4.1.10.
Let be effective Cartier divisors on a surface . Then let , and let . Suppose that meet transversally at a single smooth point . Then if and is the blowup with exceptional divisor , we see that , so we have
If both have multiplicity at and no common tangents at , then
In general, the answer is more complicated.
4.1.3 Multiplicity
Let be an (irreducible) subvariety. Then the coefficient of in the class is called the algebraic multiplicity of on and is denoted .
Suppose has positive codimension , and are the projections to , and with exceptional divisor . Then we have
For example, if is a point, then we have
Example 4.1.11.
Let be a smooth curve of genus and be the -th symmetric power of . Then let , be the Jacobian, and be given by . We know that the fibers of are the linear systems ; if , then is a projective bundle; and if , then is birational onto its image . Now if and , we have
where . When is large, this follows from the second bullet, but if is small, then we may embed
and then consider the normal bundle to this embedding restricted to . Combined with Proposition 4.1.7, this gives us the Riemann-Kempf formula, which says that the multiplicity of at is given by .
Remark 4.1.12.
The previous example can be generalized to the Fano varieties of lines on a cubic threefold . In particular if is the Fano variety of lines on , then there is a morphism of degree from to the theta divisor, and we can calculate (following Clemens-Griffiths) that
and then the theta divisor has a singular point of multiplicity .
4.1.4 Linear Systems
Let be a line bundle on a variety (of dimension ) and let be a partial linear system of dimension . Then let be the base locus of . Then if , we obtain a morphism resolving the rational map . By definition, we have . Define to be the degree of .
Proposition 4.1.13.
We have the identity
Example 4.1.14.
Let be the rational normal curve. Then let be the linear system of quadrics containing . If , we see that
If , then .
4.2 Deformation to the Normal Cone
Let be a closed subscheme and be the normal cone. We will construct a scheme and a closed embedding such that
comutes and such that
- 1.
Away from , we have and the embedding is the trivial embedding .
- 2.
Over , is a sum of two Cartier divisors, where . The embedding of is given by . We also have , which is embedded as the hyperplane at in and as the exceptional divisor in .
We will now construct this deformation. Let . Clearly we have . But now we can embed . The first property is obvious by the blowup construction, so now we need to show the second property.
We may assume is affine and is defined by the ideal . Identify . Then if we write , then we see that . But now this is covered by affines
Now for , we see that is defined by the equation , while is defined by , and now we see that
as desired.
Now this allows us to define a specialization morphism
Proposition 4.2.1.
Specialization preserves rational equivalence. Therefore we have a specialization morphism
Remark 4.2.2.
Supposing that are smooth, then the embedding of is nicer than in several ways:
- 1.
There is a retraction .
- 2.
There is a vector bundle on or rank and a section such that . Therefore is represented by the top Chern class of .
5 Caleb (Feb 19): Chern classes and intersection products
5.1 Chern Classes
We will recall the notion of Chern classes in topology. Recall that characteristic classes are natural transformations from . Because complex vector bundles of rank are reprsented by , the Yoneda lemma tells us that characteristic classes are given by . Now Chern classes satisfy the following properties:
- 1.
, where denotes the trivial bundle.
- 2.
If has rank , then is nonzero only in degrees .
- 3.
.
- 4.
generates .
Remark 5.1.1.
Chern classes can be computed using the exponential sequence
where
under the isomorphism .
We now return to algebraic geometry. Let be a line bundle and be a rational section. Then .
Example 5.1.2.
If is a smooth variety over , then we can write , and . For example, if , then we can choose , and .
Here are some properties of Chern classes of vector bundles. Let be a vector bundle on a smooth quasiprojective variety . Then we define
- 1.
If is a line bundle, then .
- 2.
If are global sections with degeneracy locus having codimension , then
- 3.
If is an exact sequence, then .
- 4.
If is a morphism, then .
This axiomatic construction of Chern classes was done first by Grothendieck in 1958.
Now we discuss degeneracy loci. We know that if and only if , whic happens if and only if has a nonvanishing section. Now if has rank and we have general global sections , the degeneracy locus is where
Now recall that we defined the map and the map . Finally, we defined . We will now return to splitting. If is a vector bundle of rank , then there exists such that is surjective and is injective on and has a filtration by line bundles. The proof is by induction and is thus omitted. This allows us to write
where are the Chern roots.
Example 5.1.3.
If has roots , then has roots .
Example 5.1.4.
We will now compute the lines on a cubic surface. Note that there is a -dimensional locus of cubic hypersurfaces in . On the other hand, there is a -dimensional space of cubic forms on a line . Therefore, lies on a cubic surface if and only if .
As varies, the cubic forms give a vector bundle . Now each gives us a section, so the locus where is contained in is precisely the zero-locus of . Now we have a vector bundle on the Grassmannian of lines in . Now we may assume that
(the number of lines on the cubic surface) is finite, so we want to calculate it. We know has roots . Then , so
Therefore we have . But then is the number of lines contained in two planes. On the other hand, is the number of lines intersecting two lines in a given plane, so we see that , as desired.
5.2 Intersection Products
Let be a regular closed embedding of codimension with normal bundle . Let be purely -dimensional with a map . Our goal is to construct . If we want this to look like cohomology, use formal Poincare duality. We will use deformation to the normal cone for our construction. Consider the pullback
Now define be a rank vector bundle on and be the projection. Then if is the ideal sheaf of , this generates the ideal sheaf of . This means that
is surjective. This gives a closed embedding of the normal cone . By a result in the appendix, is purely -dimensional. Now consider te cycle on and let be the zero-section of . Now we will define
where is the Gysin map.
Here are some properties of the intersection product:
- 1.
The intersection product is compactible with proper pushforwards.
- 2.
Intersection products are compatible with flat pullbacks.
- 3.
The intersection product is commutative.
- 4.
Functoriality.
6 Nicolás (Mar 12): families of algebraic cycles
Note: these notes were adapted from the source for the beamer provided by the speaker. Very light edits were made to the source.
6.1 Introduction
“A typical problem in enumerative geometry is to find the number of geometric figures in a given family which satisfy certain conditions”. One of the classical examples is that given five points in general position in , there exists a unique smooth conic passing through them.
The idea is that “conics are parametrized by , and passing through a point is a degree equation in ”. But this might be dangerous:
- •
parametrizes conics, not smooth conics.
- •
We need transversal intersections.
For instance, the argument does not work for smooth conics tangent to five lines. Each tangency is a degree equation on , but they are not transversal (the conics of the form are “tangent” to all lines).
The correct number is , which may be seen by taking the dual conic (the set of tangent lines, as a subset of ).
6.1.1 Conservation of number
The classical principle is called conservation of number: if the problem has a finite numerical answer, this number is constant (or jumps to infinity).
Sadly, this does not work. Given four lines and a point in general position, there exists smooth conics tangent to the lines and passing through the point (as one can show by taking the dual problem). But if the point lies in the diagonals of the quadrilateral given by the lines, then the number of smooth solutions decreases to or .
Today we will discuss strong foundations for this principle, and some applications in enumerative geometry.
6.2 Families of cycle classes
During this section, will denote an irreducible variety of dimension . We take a regular closed point, and we denote
for the point and the inclusion.
We will use script letters (e.g. ) for schemes over , and the corresponding latin letters (e.g. ) for the corresponding fibers over (as schemes over ). If is a morphism, we denote the map on the fibers.
6.2.1 Specialization
Let . We define by
where is the refined Gysin homomorphism induced by
For instance, if and , then .
Here are some basic properties.
Proposition 6.2.1.
- 1.
If is proper, , then
- 2.
If is flat of relative dimension ,
- 3.
If is a regular embedding of codimension , such that is also a regular embedding of codimension , a morphism, , then
- 4.
If is a vector bundle over , , then
The proof follows directly from similar statements for the refined Gysin homomorphism (see § 6.2–6.4).
We would now like to relate different fibers. Given a family and , it is natural to compare for different values of . It is not obvious that such relation exists, even if is the trivial family.
Example 6.2.2.
Let be a projective curve of genus , and the diagonal. If , then . But for , we have that and are not rationally equivalent.
This can be solved if we assume that , and if for every , they can be connected by a chain of rational curves in (see Example 10.1.7).
Here is a useful corollary.
Corollary 6.2.3.
Assume is non-singular, rational over the ground field, smooth over with relative dimension . If , then
This gives us a strategy to show that in a non-singular variety . We construct a family with for some , and such that can be lifted to . Then, it suffices to show that , which we can try to prove generically.
6.2.2 Sample Application
Let be a non-singular curve, its symmetric product (which points are effective divisors of degree over ). If is an effective divisor on of degree , define
One can show that if and have disjoint support, then and intersect transversally, and so
This is true even if and intersect, by using Corollary 6.2.3 and by “moving” .
6.3 Conservation of number
We have seen that for , it is not clear that are related, even if is the trivial family. We have the following substitute.
Proposition 6.3.1 (Conservation of number).
Let be a proper morphism, as before. Let be an -cycle on . Then all have the same degree (which is obtained by ).
The idea of the proof is write , for some . Then, by Proposition 6.2.1 we get
This proposition can be improved to compute the degree of intersections with Chern classes or some divisors (see § 10.2 for precise statements). We will need the following result.
Corollary 6.3.2.
Let be a scheme, effective Cartier divisors which are flat over . Let be a -cycle on . Assume that
is proper over . Then
is independent of .
6.4 An enumerative problem
Our main application of these techniques will be to solve the following problem.
Given an -dimensional family of plane curves, and curves in general position in the plane, how many curves in the family are tangent to the given curves?
The answer will require to compute the characteristics of the family, which are the number of curves in the family passing through general points and tangent to general lines.
For instance, if we consider the family of smooth conics, then
- 1.
We will study the incidence correspondence
This can be seen as a -bundle over . In fact, if is the kernel of
then .
This allows us to compute (see Example 8.3.4), with a basis
where , and the pullbacks of .
Now, if is a line and a point, consider
One can show that
- 2.
Let be a curve without multiple components. Define as the closure of
We claim that
where is the degree and the class of (the number of tangents from a general point to ). The idea is to compute
which has generically points.
The equivalence can be computed explicitely. Take a general point, a general line, and let the intersections of with the tangents from .
Figure 6.2: Projection from to the line . The projection from to gives a family with . (There is a explicit computation in § 10.4.)
- 3.
Let be a flat family of plane curves, , non-singular. Assume has no multiple compontents for general , and let an open set with reduced for .
Let given by such that is a simple point of , and is tangent in . Note that .
Take reduced curves, and consider
We can move , so that the interseccion between and is transversal (by taking a general element in ). This way, has (reduced) points.
Now, compactify , and . If is a closed subsed of dimension less than , which contains all , then the number does not change after we remove .
- 4.
We now degenerate each to a multiple line (as we did for ). This gives a diagram
The space is complete, so is proper over . This way, we may take an open neighborhood of and , so that is proper over and disjoint from .
The right hand side is just
where each curve has degree and class .
The left hand side is the number of points , provided that we take a convenient (which avoids technical difficulties such as bitangents).
6.4.1 A famous example
The most famous example is the Steiner’s conic problem, which tries to determine the number of conics tangent to five smooth conics in general position.
The natural family here is the family of smooth conics (as a subset of ), which has characteristics
(in characteristic zero!)
This way, the number of conics tangent to five non-singular curves of degree in general position is
which for gives the famous number .
7 Patrick (Mar 19): Doing Italian-style algebraic geometry rigorously
Note: these are the speaker’s notes.
7.1 Intersection multiplicities
Consider a Cartesian square
where is a regular embedding of codimension and has pure dimension . Let have components with multiplicity . Let be the support of . We call the the distinguished varieties of the intersection.
Lemma 7.1.1.
- (a)
Every irreducible component of is distinguished.
- (b)
For any distinguished variety , we have .
Definition 7.1.2.
An irreducible component of is a proper component of intersection of by if . The intersection multiplicity of in , denoted is the coefficient of in the class .
If is the pullback of to , then is the coefficient of in . Now let and be the ideal of . Then has finite length when is an irreducible component of .
Proposition 7.1.3.
Assume is a proper component of .
- (a)
If is the length of , then .
- (b)
If is generated by a regular sequence of length , then .
If is Cohen-Macaulay, then local equations for in give a regular sequence generating and equality in (b) holds.
Now suppose is a proper component of the intersection of by on . Let and be the ideal in generated by the ideal (sheaf) of in , and let be the maximal ideal of .
Proposition 7.1.4.
Suppose is a variety. Then if and only if is regular and .
7.2 Nonsingular varieties
Let be a smooth variety of dimension . Then is regularly embedded with codimension . The global intersection product is the map
where is the Gysin homomorphism.
More generally, let be a scheme and a morphism to a smooth variety. Then is regularly embedded in , so we can define a cap product
If is smooth, then we write .
Remark 7.2.1.
We may also replace the Gysin homomorphisms with the refined Gysin homomorphisms .
Definition 7.2.2.
Let be a morphism with smooth of dimension . Let be morphisms of schemes. Then form the square
Now define the refined intersection product by
for . When , this is the global product.
Proposition 7.2.3.
The refined products satisfy the following formal properties:
- (a)
(Associativity) If with smooth, then
- (b)
(Commutativity) If with smooth, then
- (c)
(Projection) Let with smooth. Let be proper with and . Then
- (d)
(Compatibility) Let with smooth and be a regular embedding. Then
Corollary 7.2.4.
Let be smooth and be a regular embedding. If is a cycle on , then .
Corollary 7.2.5.
Let with smooth. Then .
Corollary 7.2.6.
Let with smooth and a cycle on . Then .
Definition 7.2.7.
Let be a morphism with purely -dimensional and a smooth -dimensional variety . For any morphism , define a refined Gysin homomorphism
Proposition 7.2.8.
- (a)
If is flat, then , where is the base change.
- (b)
If is a local complete intersection morphism, then agrees with the morphism constructed in Section 6.6 of Fulton.
Now let be a smooth variety of dimension . Let be closed subschemes of of pure dimension . Now a component is a proper component if . If is proper, then the coefficient of in is called the intersection multiplicity . If every component of is proper, then the intersection class is
Proposition 7.2.9.
Assume is a proper component of . Then
- (a)
.
- (b)
If the local ring is Cohen-Macaulay, then .
- (c)
If are varieties, then if and only if the maximal ideal of is the sum of the prime ideals of and . In fact, are regular.
Now let be a smooth variety of dimension . Set . Now this indexing, the intersection product is now . In addition, if is a morphism, the cap product is now . If is also smooth, the pullback now reads .
Proposition 7.2.10.
- (a)
Suppose is a smooth variety. Then the intersection product makes into a commutative graded ring with unit . Then is a contravariant functor from smooth varieties to rings.
- (b)
If is a morphism from a scheme to a smooth variety , then is a -module with action
- (c)
If is a proper morphism of smooth varieties, then
for all classes on and on .
7.3 Bézout’s Theorem
We will now use this theory to discuss something classical. It is easy to see that and is generated by for a linear subspace . If is a -cycle on , we define the degree to be the integer satisfying . Equivalently, we may define
Theorem 7.3.1 (Bézout).
Let for . If , then
Proof.
We have an isomorphism , where . Thus , and the desired result follows. □
Now if subschemes representing meet properly, then
where are the components of . Then Bézout’s theorem gives us the identity
If are hypersurfaces intersecting properly (so the intersection is a finite number of points), then consider the local ring . Then complete intersections are Cohen-Macaulay, so
Now , and thus we obtain
This recovers the very classical Bézout’s theorem.
Example 7.3.2.
Let be the hyperplane class on and be the hyperplane class on . Then
- 1.
.
- 2.
If are hypersurfaces in with bidegree , then
- 3.
If is the diagonal in , then . This formula follows from intersecting with , where are linear subspaces of complementary dimension.
- 4.
Let be the Segre embedding. It is the hyperplane class on , then . Also, the degree of is .
- 5.
If is the Veronese embedding and are the hyperplane classes on the source and target, then . If is a -dimensional subvarity of of degree , then .
8 Morena (Mar 26): Grothendieck-Riemann-Roch
We will begin by reviewing some notions from algebraic topology.
- 1.
Let be a vector bundle over with Chern roots . Then we may define the Chern character
where is the -th Newton polynomial. Recall that if
is exact, then and .
- 2.
We may define the Todd class
Similarly to Chern classes, if
is exact, then .
- 3.
Now define the ring
This is well-behaved under pullback, but pushforward is not well-defined.
In order to fix the problems with pushforwards of vector bundles, we will attempt to define the Grothendieck group of coherent sheaves. Unfortunately, pullbacks of coherent sheaves are only right exacts. Also, pushforwards do exist for proper morphisms, but this is only left exact. Therefore we need to consider the higher derived functors, and now we define
This is a finite sum because we assume all of our schemes are Noetherian.
For a smooth variety, there is always a map . In fact, there is an inverse, because we can resolve any coherent sheaf by vector bundles. In this case, we will write . If is resolved by
we write and .
8.1 Statement of the theorem
Recall the Chern character . In topology, we need the Todd class to make the diagram
commute. Now we are able to state the Grothendieck-Riemann-Roch theorem:
Theorem 8.1.1 (Grothendieck-Riemann-Roch).
Let be proper where are smooth quasiprojective varieties over . Then the diagram
commutes. In other words, .
First, we will see why this implies the classical (Hirzebruch)-Riemann-Roch theorem. First, let be a curve, , and be a line bundle. Then the right hand side is
On the other side, we have
This recovers the classical Riemann-Roch theorem.
Now let be a smooth surface. Then our formula becomes
This recovers the Hirzebruch-Riemann-Roch theorem.
8.2 Proof for
First, we have the following theorem:
Theorem 8.2.1.
is generated by for .
To prove this, we need to find a resolution of a coherent sheaf by line bundles, so we have
Because is ample, is generated by global sections, so we have the exact sequence
and we can tensor this by . Now if we set , we can take the Koszul complex
Now we have the exact sequence
In particular, this means that we can generate by .
Now it remains to prove Grothendieck-Riemann-Roch just for . Here, because is a point, we have . In our case, we know that for , and , so
by basic combinatorics.1
On the other hand, we want to compute . Recall the Euler exact sequence
Therefore we have . Also, . Looking for the -coefficient in , we can multiply by and compute the residue to obtain .
Remark 8.2.2.
Grothendieck wrote the following about this result:
Um dieser Aussage über einen approximativen Sunn zu geben, musste ich nahezu zwei Stunden lang die Geduld der Zuhörer missbrauchen. Schwartz auf weiss (in Springer’s Lecture Notes) nimmt’s wohl an die 400, 500 Seiten. Ein packendes Beispiel dafür, wie unser Wissens - und Entdeckungsdrang sich immer mehr in einem lebensentrücken logischen Delirium auslabt, während das Leben selbst auf tausendfache Art zum Teufel geht – und mit endgültiger Vernichtung bedroht ist. Höchste Zeit, unsern Kurs zu ändern!
The English translation of this,2 keeping sentence structure intact as much as possible, is
To give this statement about an approximate meaning I’d have to abuse the patience of the listeners for nearly two hours. Black on white (in Springer’s Lecture Notes) it should take about 400, 500 pages. A gripping example of how our thirst for knowledge and discovery indulges more and more in a life-divorced logical delirium, while life itself is going to hell in a thousand different ways – and is threatened with absolute extinction. High time to change our course!
9 Patrick (Apr 02): Fine moduli memes for -categorical teens
Note: these are the speaker’s notes. Throughout this lecture, we work over .
9.1 The moduli space
The space parameterizes curves that look like this:
More precisely, these are reduced connected curves that are a tree of s such that each has at least three marked points or nodes. In addition, at most two components meet at each node and we want . If all of these conditions are satisfied, we call our curve stable. More precisely, we want to represent the functor
Theorem 9.1.1 (Knudsen).
There exists a smooth complete variety
In fact, Knudsen also shows that
Theorem 9.1.2 (Knudsen).
Contraction and stabilization are functorial! Moreover, they commute with base change.
Here are some pictorial depictions of our operations:
9.2 Keel construction of M ¯ 0 , n
First, we will describe the boundary divisors of
It is easy to see that
Now consider the map
where
Set
Lemma 9.2.1 (Keel).
The divisors
Now we set
Theorem 9.2.2 (Keel).
The map
9.3 Intersection theory of M ¯ 0 , n
There are several major results about the intersection theory of
Theorem 9.3.1.
We have an isomorphism
Theorem 9.3.2.
For any scheme
Theorem 9.3.3.
For all
which is induced by the maps
where
Theorem 9.3.4.
The Chow groups
In particular, we have the Picard rank
Theorem 9.3.5.
The Chow ring
- 1.
;D T = D ( T C ) - 2.
For any distinct
, we have the equalityi , j , k , ℓ ∈ [ n ] ∑ i . j ∈ T k , ℓ ∉ T D T = ∑ i , k ∈ T j , ℓ ∉ T D T = ∑ i , ℓ ∈ T j , k ∉ T D T . - 3.
For
,T 1 , T 2 ⊂ [ n ] unless one ofD T 1 D T 2 = 0 holds.T 1 ⊂ T 2 , T 2 ⊂ T 1 , T 1 ⊂ T 2 C , T 2 ⊂ T 1 C
Remark 9.3.6.
All of the relations encode geometric content:
- 1.
As divisors, we already know that
.D T = D ( T C ) - 2.
If we consider the map
, then the three sums are the pullbacks of the three boundary divisorsπ i , j , k , ℓ : M ¯ 0 , n → M ¯ 0 , 4 .D i , j , D i , k , D i , ℓ ⊂ M ¯ 0 , 4 = ℙ 1 - 3.
The final relation encodes the fact that
unless one of the four inclusions holds. Pictorially, this is encoded in the diagram below:D T 1 ∩ D T 2 = ∅ Figure 9.5: Degeneration to a common stable curve
9.4 Intersection theory of regular blowups
Let
Theorem 9.4.1.
Suppose
where
Theorem 9.4.2.
A scheme
Theorem 9.4.3.
The map
is an isomorphism.
10 Patrick (Apr 09): Do we even need derived categories?
Note: these are the speaker’s notes.
10.1 Serre formula
Recall the following from my March 19 lecture (this is Proposition 7.2.9 in my notes):
Proposition.
Let
where
However, most rings are not Cohen-Macaulay, so we would like a formula to compute the intersection multiplicities in all cases. Serre gives a formula in terms of higher Tor functors (because
First, if
This follows from the construction of the derived tensor product
Lemma 10.1.1.
Let
Proof of this fact is pure homological algebra, and again can be found in the Stacks project.
Lemma 10.1.2.
Let
Proof.
The support condition is clear by looking at the stalks, so we need to consider when the stalks are nonzero. Here, we note that because
Now we can compute the intersection multiplicities as (Stacks gives this as the definition of intersection multiplicity)
This formula is due to Serre, and Stacks writes the total intersection as
Remark 10.1.3.
Stacks writes the intersection multiplicity as
Lemma 10.1.4.
Assume that
Proof.
Write
By Theorem 4.4.2 from commutative algebra, we only have
Example 10.1.5.
Suppose
10.2 Some algebra
Let
Recall that
Definition 10.2.1.
For
Lemma 10.2.2.
For all
Lemma 10.2.3.
Let
for any
Proof.
Write
The desired claim follows from Pascal’s identity. □
Theorem 10.2.4.
Let
There is a very long proof of this statement in Stacks using spectral sequences.
10.3 Computing intersection multiplicities without derived categories
We give some cases where intersection multiplicities can be computed without using derived categories.
Lemma 10.3.1.
Suppose
Proof.
Write
Lemma 10.3.2.
Let
In what follows, we will assume
Lemma 10.3.3.
In this case, we have
Proof.
By the previous lemma,
Lemma 10.3.4.
Assume
Proof.
Note that
Lemma 10.3.5.
Asssume
Proof.
Because
Now take the leading coefficient. □
11 Caleb (Apr 16): Motivic cohomology
Note: these are the speaker’s notes. No changes were made except to bring notational and stylistic conventions in line with the rest of the notes and to adapt pictures that were not available to me. Also, the quote of Grothendieck was changed to the original French.
These are my notes on motivic cohomology. Essentially everything here is based off of Voevodsky’s lectures, now turned into a book by Mazza and Weibel.
11.1 Bloch’s higher Chow groups
11.1.1 Some topological motivation
Recall the following exact sequence.
Proposition 11.1.1 (Fulton, Prop. 1.8).
Let
is exact for all
This is great, but if we want homology to continue to the left. We certainly cannot put these sequences together. In fact, the indexing is a bit misleading in this way – instead, we should put the
Definition 11.1.2 (Rat ( X ) ).
Let
Then rationally equivalent cycles are those which differ by something in
11.1.2 Definition
Motivated by algebraic topology, we define the algebraic simplex
Let
Definition 11.1.3.
The higher Chow groups
11.1.3 Properties
- 1.
Homotopy invariance:
The projection
induces an isomorphismX × 𝔸 1 → X CH i ( X , m ) ≅ CH i ( X × 𝔸 1 , m ) . - 2.
Long exact sequence:
There is a distinguished triangle
z p ( Y , ∗ ) → z p ( X , ∗ ) → z p ( U , ∗ ) → z p ( Y , ∗ ) [ 1 ] . - 3.
Isomorphism with rational K-theory:
( K i ( X ) ⊗ ℚ ) ( q ) ≅ CH q ( X , i ) ⊗ ℚ .
We will see that
11.2 The category of correspondences
11.2.1 Correspondences
Let
Definition 11.2.1.
An elementary correspondence between a smooth connected scheme
If
The group
Then given a closed subscheme
11.2.2 The category of correspondences
We compose correspondences
It is not difficult to check that
Furthermore,
11.2.3 Examples
- 1.
is generated by the𝖢𝗈𝗋 k ( Spec k , X ) -cycles of0 .X - 2.
is generated by the irreducible components of𝖢𝗈𝗋 k ( X , Spec k ) .X - 3.
Take
and twoW ∈ 𝖢𝗈𝗋 k ( 𝔸 1 , X ) -pointsk . Then the zero-cycless , t : Spec k → 𝔸 1 andW ∘ Γ s are rationally equivalent.W ∘ Γ t
11.3 Presheaves with transfers
11.3.1 Definition
Definition 11.3.1.
A presheaf with transfers is a contravariant additive functor
Additivity gives a map
Thus there are extra “transfer maps"
Theorem 11.3.2.
11.3.2 Examples
Example 11.3.3.
The constant presheaf
For
Example 11.3.4.
Example 11.3.5.
Example 11.3.6.
Representable functors:
11.3.3 Representable functors of Cor k ( X )
Take
By Yoneda,
Note that
The structure map
Out of laziness we screenshot the following definitions from Voevodsky’s lectures.
Definition 11.3.7.
If
Lemma 11.3.8.
The presheaf
Moreover, the following sequence of presheaves with transfers is split-exact:
Consider the pointed scheme
Before continuing, we recall our construction
Recall that a simplicial object of a category
is a simplicial presheaf with transfers. Similarly,
11.3.4 Homotopy invariant presheaves
Definition 11.3.9.
A presheaf
Note that this is equivalent to
Furthermore, if
are homotopy invariant for all
Definition 11.3.10.
Two finite correspondences from
This is an equivalence relation on
11.4 Motivic cohomology
11.4.1 The motivic complex
Definition 11.4.1.
For
We can change coefficients to
These are actually complexes of sheaves with respect to the Zariski topology. In fact, they are also sheaves in the étale topology.
For example when
which is quasi-isomorphic to just
11.4.2 Motivic cohomology groups
Definition 11.4.2.
The motivic cohomology groups
If
11.4.3 Weight 1
There is a quasi-isomorphism
Thus we have the following table.
11.5 Relation to other fields
11.5.1 Algebraic K-theory
Atiyah-Hirzebruch:
In the algebraic setting, it is much more difficult. Indeed, both algebraic K-theory and motivic cohomology are significantly harder to define than their topological counterparts. In 2002, Suslin and Friedlander built upon previous work of Bloch and Lichtenbaum to show the following spectral sequence.
11.5.2 Motives
C’est pour parvenir à exprimer cette intuition de “parenté” entre théories cohomologiques différentes, que j’ai dégagé la notion du “motif” associé à une variété algébrique. Par ce terme, j’entends suggérer qu’il s’agit du “motif commun” (or de la “raison commune”) sous-jacent à cette multitude d’invariants cohomologiques différents associés à là variété, à l’aide de la multitude des toutes les théories cohomologiques possibles a priori.Alexander Grothendieck
Note: there was an English translation originally in the slideshow, but I don’t actually think the translation is very good. I provided my own translation at some point, but that was also bad so it was removed.
Grothendieck constructed Chow motives by replacing morphisms of schemes with correspondences (defined under rational equivalence, different from the correspondences discussed earlier), augmenting the category to look like an abelian category, and taking the opposite category. By construction, this works in that cohomology theories factor through it. However, to truly achieve what is desired from them, one must assume the standard conjectures on algebraic cycles (or some variants), which have been open for over 50 years!
Voevodsky used motivic cohomology to construct a triangulated category
11.5.3 Arithmetic geometry
There’s the Bloch-Kato conjecture and the Bloch-Kato conjectures, which are different!
The Bloch-Kato conjecture is now a theorem: the norm residue isomorphism theorem, proven by Voevodsky. Through proving it, Voevodsky developed motivic cohomology, motivic homotopy theory, motivic Steenrod algebra…
The norm residue isomorphism theorem (or Bloch-Kato conjecture) states that for a field
and an integer k that is inverible in ℓ , the norm residue map k
∂ n : K n M ( k ) / ℓ → H ét n ( k , μ ℓ ⊗ n ) from Milnor K-theory mod-
to étale cohomology is an isomorphism. The case ℓ is the Milnor conjecture, and the case ℓ = 2 is the Murkurjev-Suslin theorem.Wikipedia n = 2
The Bloch-Kato conjectures are on special values of L-functions.
Conjecture 11.5.1 (Soulé).
If
Applied to elliptic curves, this implies (one of the two parts of) the Birch-Swinnerton Dyer conjecture!