paint-brush
Expansions for Hilbert Schemes: the Expanded Constructionby@eigenvector

Expansions for Hilbert Schemes: the Expanded Construction

Too Long; Didn't Read

This paper improves methods for degenerating "Hilbert schemes" (geometric objects) on surfaces, exploring stability and connections to other constructions.
featured image - Expansions for Hilbert Schemes: the Expanded Construction
Eigenvector Initialization Publication HackerNoon profile picture

Author:

(1) CALLA TSCHANZ.

3. The expanded construction


Output of expanded construction. The expanded degeneration X[n] ! C[n] which we construct in this section has the following properties:


3.1 The blow-ups


In this expanded degeneration construction, we will be blowing up schemes along Weil divisors. A consequence of the way these blow-ups are defined is that the blow-up morphisms contract only components of codimension at least 2.






the morphisms corresponding to each individual blow-up. We therefore have the equality



We now fix the following terminology.







Proposition 3.1.5. The following blow-up diagram commutes




Proof. This is immediate from the local description of the blow-ups above.



We now extend the definition of ∆1-components to the schemes X[n] and fix some additional terminology.




Before we continue we fix some terminology which will help us describe the expanded components.


Definition 3.1.11. We refer to an irreducible component of a ∆-component as a bubble. The notions of two bubbles being equal and a bubble being expanded out in a certain fibre are as in Definitions 3.1.4 and 3.1.9.




Now, we note that there is a natural inclusion



which, in turn, induces a natural inclusion




on the basis directions, and acts by



on the ∆-components.


Proof. This follows immediately from [GHH19].



we described in the previous section are equivariant under the group action.


Lemma 3.1.13. We have the isomorphism



Proof. This is immediate from the above description of the group action.


Remark 3.1.14. We abuse notation slightly by referring to the group acting on X[n] by G, instead of G[n]. It should always be clear from the context what group G is meant.

3.2 Embedding into product of projective bundles


Lemma 3.2.1. There is an embedding



From this, we deduce that there are embeddings



Hence we have embeddings




Linearisations. The following lemma gives a method to construct all the linearised line bundles we will need to vary the GIT stability condition.



This paper is available on arxiv under CC 4.0 license.