Problem 31
Question
Let \(D\) be a divisor with \(\operatorname{deg}(D)=2 g-2\) and \(l(D)=g\). Show that \(D\) is a canonical divisor. So these properties characterize canonical divisors.
Step-by-Step Solution
Verified Answer
Question: Show that a divisor D with degree \(2g - 2\) and a dimension of the Riemann-Roch space associated with the divisor equal to g is a canonical divisor.
Answer: Based on the given properties of D, and by applying the Riemann-Roch theorem, we have shown that divisor D has the degree of a canonical divisor (\(2g - 2\)) and the dimension of the Riemann-Roch space associated with the divisor equal to g. Therefore, D is a canonical divisor.
1Step 1: Apply the Riemann-Roch theorem to the given divisor D
We are given that the degree of D is \(2g - 2\), and \(l(D) = g\). We can now apply the Riemann-Roch theorem to the divisor D:
\(l(D) - l(K_C - D) = \operatorname{deg}(D) + 1 - g\)
Now, substitute the given information:
\(g - l(K_C - D)= (2g - 2) + 1 - g\)
2Step 2: Simplify the equation
Simplify the equation to isolate \(l(K_C - D)\):
\(l(K_C - D) = g - (2g - 2) + 1 - g = -g + 2 + 1\)
3Step 3: Evaluate the dimension of the Riemann-Roch space associated with K_C - D
We know that the dimension of a Riemann-Roch space is non-negative, so \(l(K_C - D) \geq 0\). From the previous step, we have found that \(l(K_C - D) = -g + 3\). Since \(l(K_C - D) \geq 0\), it must be the case that \(-g + 3 \geq 0\), which leads to:
\(g \leq 3\)
Now recall that \(l(D) = g\). Therefore, \(l(K_C - D) = g - 3\).
4Step 4: Apply Riemann-Roch theorem to K_C - D
Since we know \(l(K_C - D) = g - 3\), we can apply the Riemann-Roch theorem to the divisor \(K_C - D\):
\(l(K_C - D) - l(K_C - (K_C - D)) = \operatorname{deg}(K_C - D) + 1 - g\)
\(l(K_C - D) - l(D) = \operatorname{deg}(K_C - D) + 1 - g\)
Substitute the values we found in the previous steps:
\((g - 3) - g = \operatorname{deg}(K_C - D) + 1 - g\)
5Step 5: Simplify the equation and find the degree of K_C - D
Simplify the equation to find the degree of \(K_C - D\):
\(-3 = \operatorname{deg}(K_C - D) + 1 - g\)
\(\operatorname{deg}(K_C - D) = g -4\)
6Step 6: Show that D is a canonical divisor
Now recall that the degree of a canonical divisor is \(2g - 2\). By substituting our result for the degree of \(K_C - D\), we can show that D is a canonical divisor:
\(\operatorname{deg}(D) = \operatorname{deg}(K_C) - (\operatorname{deg}(K_C - D))\)
\(\operatorname{deg}(D) = (2g - 2) - (g - 4)\)
\(\operatorname{deg}(D) = 2g - 2\)
This shows that divisor D has the degree of a canonical divisor, and since \(l(D) = g\), D is a canonical divisor. This completes the proof, and the given properties characterize canonical divisors.
Key Concepts
Riemann-Roch TheoremCanonical DivisorDegree of Divisor
Riemann-Roch Theorem
The Riemann-Roch Theorem is a fundamental tool in algebraic geometry that relates the dimensions of spaces of divisors on a curve. It essentially provides a way to calculate the number of linearly independent meromorphic functions that can be associated with a given divisor on a curve. The theorem can be expressed as:\[l(D) - l(K_C - D) = \operatorname{deg}(D) + 1 - g\]where:- \(l(D)\) is the number of meromorphic functions associated with the divisor \(D\) which do not have poles except perhaps where prescribed by \(D\).- \(K_C\) is a canonical divisor associated with the curve \(C\).- \(g\) is the genus of the curve, which is a topological invariant indicating the number of "holes" in the curve. The theorem is particularly useful because it relates the linear equivalence class of a divisor to its degree, providing information about possible mappings from the curve to projective space. In this exercise, the Riemann-Roch Theorem was utilized to demonstrate that divisor \(D\) has properties consistent with a canonical divisor by confirming \(l(D) = g\). This shows that \(D\) consists entirely of those divisors for which the considered meromorphic functions exist.
Canonical Divisor
A canonical divisor is a special type of divisor on an algebraic curve. It is closely related to the differential forms on the curve. Essentially, the canonical divisor \(K_C\) can be thought of as the divisor of zeroes and poles of a certain kind of differential form on the curve.Some key points about canonical divisors include:
- The degree of a canonical divisor on a curve of genus \(g\) is \(2g - 2\).
- It represents a fundamental class of divisors that are invariant under isomorphisms, meaning the canonical divisor remains consistent even if the curve undergoes certain transformations.
- Divisors which are linearly equivalent to a canonical divisor share the same degree, indicating they can be expressed in terms of one another through the addition or subtraction of principal divisors.
Degree of Divisor
The degree of a divisor is a fundamental concept in algebraic geometry. It is essentially a measure of the "total weight" of the zeros and poles encoded by a divisor on an algebraic curve. If \(D\) is a divisor, it can be expressed as:\[D = \sum_i n_i P_i\]where each \( n_i \) is an integer and \( P_i \) are the points on the curve. The degree is then given by:\[\operatorname{deg}(D) = \sum_i n_i\]Key points include:
- A positive degree often indicates the net presence of zeros on the curve.
- A negative degree usually suggests an oversupply of poles.
- A divisor of degree zero represents a balance between zeros and poles.
Other exercises in this chapter
Problem 19
If \(f: C \rightarrow V\) is a morphism from a projective curve to a variety \(V\), then \(f(C)\) is a closed subvariety of \(V\). (Hint: Consider \(C^{\prime}=
View solution Problem 30
Suppose \(D\) and \(D^{\prime}\) are divisors, and \(D+D^{\prime}=W\) is a canonical divisor. Then \(l(D)-l\left(D^{\prime}\right)=\frac{1}{2}\left(\operatornam
View solution Problem 32
Let \(P_{1}, \ldots, P_{m} \in \mathbb{P}^{2}, r_{1}, \ldots, r_{m}\) nonnegative integers. Let \(V\left(d ; r_{1} P_{1}, \ldots, r_{m} P_{m}\right)\) be the pr
View solution Problem 33
Let \(D\) be a divisor, and let \(V\) be a subspace of \(L(D)\) (as a vector space). The set of effective divisors \(\\{\operatorname{div}(f)+D \mid f \in V, f
View solution