Naively A 1 -Connected Components of Varieties
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
IISER Mohali
Abstract
A 1 -homotopy theory is a homotopy theory for schemes in which the affine line
A 1 plays the role of the unit interval. The main objects of study are simplicial
sheaves on the Nisnevich site of smooth schemes of finite type over a field. For
these objects, one constructs analogues of various devices from the classical ho-
motopy theory of topological spaces. One such device is the sheaf of A 1 -connected
components of a simplicial sheaves.
1
For a general simplicial sheaf X , the sheaf π 0 A (X ) of A 1 -connected components
of X is generally hard to compute. However, one can attempt to study it by means
of the sheaf of naively A 1 -connected components, denoted by S(X ). The sheaf
1
S(X ) may be viewed as a crude approximation to π 0 A (X ), but it is easier to define
and compute, at least when X is a sheaf of sets. The functor S is the main object
of study in this thesis.
When X is a sheaf of sets, the direct limit of the sheaves S n (X ), which we
denote by L(X ) can be proved to be A 1 -invariant. In fact, this is the universal
1
A 1 -homotopic quotient of X . When π 0 A (X ) is A 1 -invariant, it can be proved to
1
be isomorphic to L(X ). A recent example of Ayoub has shown that π 0 A (X ) is not
always A 1 -invariant. However, we show that there is a natural bijection between
1
field valued points of the sheaves L(X ) and π 0 A (X ) for any sheaf of sets X .
The sheaf L(X ) is obtained by iterating S on a the sheaf X infinitely many
times. Our second main result is to show that the infinitely many iterations are
indeed necessary. We achieve this by constructing a family of sheaves {X n } n ,
indexed by the positive integers, such that S i (X n ) ̸ = S i+1 (X n ) for any i < n.
The third main result of this thesis is regarding retract rational varieties over
an infinite field k. A result of Kahn and Sujatha shows that for a retract rational
1
variety X, the sheaf π 0 A (X) is the point sheaf. We strengthen this result by
showing that S(X) is the point sheaf.