Cohomology of Projective Space
The cohomology groups of twisted structure sheaves on projective space are among the most fundamental computations in algebraic geometry. These groups control the geometry of projective varieties and provide the foundation for Serre duality, Riemann-Roch, and many other essential theorems.
The Main Theorem
Let be a field and be projective -space over . For any integer , we have:
- for , and for
- for , and for
- for and all
In particular:
- for and for
- for and all
- for and for
The "gap" phenomenon is striking: except at the extremes ( and ), all cohomology groups vanish. This is a special feature of projective space not shared by general projective varieties.
Proof via Δech Cohomology
We compute using Δech cohomology with respect to the standard affine cover. Let for be the standard affine open cover of .
Step 1: Structure of the cover. Each with coordinate ring . The intersections are: where .
Step 2: Sections of on opens. For , sections over are:
For , this is the localization of degree homogeneous polynomials. For , we need Laurent polynomials.
Step 3: Δech complex for . The Δech complex is:
For , a global section is determined by homogeneous polynomials of degree that agree on overlaps. Since is connected, the must glue to give a global homogeneous polynomial of degree .
Thus has dimension .
Step 4: Higher cohomology for . For , the Δech groups involve sections over intersections of opens. These are Laurent polynomial rings, but for , the cochains are determined by polynomial data. The complex is exact at each stage by the following key fact: on the punctured affine space , the ring has trivial higher cohomology for coherent sheaves.
By induction on and using the acyclicity of affine schemes, we conclude for and .
Step 5: Using Serre duality. By Serre duality, .
For , write with . Then:
This is nonzero only when , i.e., , giving . In this case:
Step 6: The middle range. For , we have , so . By step 4, might be nonzero, but we need for this, i.e., . However, for , we have , and specifically:
- If , then , so .
- For , similar analysis shows vanishing.
Direct computation using the Δech complex confirms for and all .
Complete Tables of Cohomology
For projective line :
| | | | |-----|--------------------------------------|--------------------------------------| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Note the symmetry: for all .
For projective plane :
| | | | | |-----|--------------------------------------|--------------------------------------|--------------------------------------| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
The middle cohomology vanishes identically. The symmetry is: .
For projective 3-space :
| | | | | | |-----|-------|-------|-------|-------| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Both and vanish identically, illustrating the gap phenomenon.
Detailed Computations
Global sections of correspond to homogeneous polynomials of degree in variables . The dimension is:
Explicit examples:
- : . Basis: .
- : . Basis: .
- : . Basis: .
For , there are no nonzero global sections since a homogeneous polynomial of negative degree in positive degree variables is trivial.
For , a global section of would be given by a collection of Laurent polynomials on each affine chart that glue compatibly. On , sections are of the form where is homogeneous of degree .
For , such a section must have poles along the hyperplane in each chart. However, since is covered by the and the complement of is the hyperplane , a global section cannot have poles anywhere.
Therefore, for .
By Serre duality:
For , we have , so:
Concrete cases:
- : .
- : .
- : .
Consider and compute for various :
For any :
- If : By the main theorem, .
- If : By Serre duality, .
- For , i.e., , we would need .
- But for all by the main theorem.
Therefore, for all .
Similarly, for all . The only nonzero cohomology occurs at (for ) and (for ).
Δech Cohomology Computation in Detail
Consider the standard cover and of .
The Δech complex is:
We have:
- (Laurent monomials )
- (Laurent monomials )
- (all Laurent monomials)
The coboundary map .
A 1-cocycle is an element such that is not in the image of .
The image of consists of Laurent polynomials that can be written as a difference of a polynomial in and a polynomial in (after multiplying by or ).
Consider . In the chart, this is , which has a pole at . Similarly, it has a pole at in the chart. Therefore, cannot be written as .
Thus is generated by the class of .
For with cover , the Δech complex for is: where:
For , we compute :
- (Laurent polynomials)
- Elements of degree in that are not in form .
The element has poles everywhere and cannot be a coboundary. It generates .
Serre Duality Reflected in the Tables
The canonical sheaf of is . Serre duality states:
Verifying from the table:
The duality is perfect.
The canonical sheaf of is . Serre duality states:
Verifying:
The middle cohomology satisfies: Since both sides vanish for all , the duality is trivially satisfied.
Applications to Hilbert Polynomials
The Hilbert polynomial of is defined as:
For :
This is a polynomial in of degree with leading coefficient .
For : For : For :
For negative :
- If : (all terms vanish)
- If :
For and :
The Hilbert series of is:
For : For :
The coefficients match the dimensions from the tables.
Cohomology of the Euler Sequence
The Euler sequence on is the exact sequence: where is the tangent sheaf of .
Taking the long exact sequence in cohomology and using for :
For , the Euler sequence gives:
Since and :
Computing dimensions:
- Thus
These are the infinitesimal automorphisms of , corresponding to the -dimensional group .
Twisting the Euler sequence by gives:
For and :
Taking cohomology:
Since , , and , :
Thus .
For the next cohomology:
All terms are zero, so .
Vanishing Theorems and Consequences
A key consequence is that for any and :
This is a special case of Serre's vanishing theorem. It implies that for sufficiently positive twists, coherent sheaves on projective varieties have vanishing higher cohomology.
For example, on :
- for all
- More generally, if is coherent on , then for
Consider the ideal sheaf sequence of a point :
Twisting by and taking cohomology:
For :
- has dimension
- (evaluating at )
Therefore:
For :
- (no global functions vanishing at a point)
- From , we get
For :
- The sequence gives
For a curve of degree given by a homogeneous polynomial, the arithmetic genus is:
This can be computed using the exact sequence:
The arithmetic genus is . From the long exact sequence:
We have:
For :
- (by Serre duality with )
Wait, we need to be careful. For :
Thus:
Therefore:
Actually, there's a sign issue. Let me recalculate. For , , so:
For : For :
So for .
Thus:
And:
For a line (): . For a conic (): . For a cubic (): .
The complete understanding of is fundamental for:
- Computing cohomology of arbitrary coherent sheaves via resolutions
- Proving Serre's vanishing theorem
- Understanding ampleness and embeddings
- Computing Hilbert polynomials and dimensions of linear systems
- Applications to curve and surface theory