Torsion models for tensor-triangulated categories: the one-step case (Q2106578)

It is a common phenomenon, when trying to understand various algebraic structures, that it is easier to compute certain local information, for example localization at primes. However, the question then becomes how to reconstruct the full algebraic data from such local information. A motivating example comes from chromatic homotopy theory, where one considers spectra that are local with respect to a fixed prime, but might wish to then obtain a spectra from all this local information. Here, the authors develop a new approach to this question in the setting of tensor-triangulated categories, or, more generally, stable monoidal model categories, whose homotopy categories have this structure. Associated to such a category is its Balmer spectrum, which can be viewed as an enhancement of the Zariski spectrum of a commutative ring. In the special case of the derived category of a commutative Noetherian ring \(R\), there are many ways to go about reconstructing an \(R\)-module \(M\) from local data, including the Zariski standard model and the Zariski torsion model. The authors focus their attention here on the adelic torsion model and its relationship to the adelic standard model developed by Balchin and Greenlees. As the authors explain, these two models have various advantanges and disadvantages with respect to one another. The strength of the adelic torsion model that they describe here is that the objects produced tend to be smaller and more manageable, although it does not preserve information about the monoidal structure. A comparison is also made between the adelic torsion model and the Zariski torsion model, with the advantage there being in the nice properties of the spectrum produced from the adelic version. In general, one might have several different localizations to assemble, but here the authors focus on what they call the ``one-step'' context, where there is only one kind of local information to consider. For example, in the motivating example of the derived category, one could apply rationalization to \(p\)-local information. Their main result is that a stable monoidal model category is Quillen equivalent to a cellularization of a particular diagram of modules. A motivating example for the authors is that of \(G\)-equivariant spectra for a compact Lie group \(G\), where the localization is given with respect to a family of subgroups. A consequence of the main results of this paper is new and more direct proof of the Quillen equivalence between circle-equivariant spectra and a suitable category of differential objeccts in a certain abelian category first described by Greenlees.