Abstract:We define the log motivic nearby cycles functor. We show that this sends the motive of a proper smooth scheme over the fraction field of a DVR to the motive of the boundary of a log smooth model assuming absolute purity, which is unconditional in the equal characteristic case. In characteristic $0$, we show that the $\infty$-categories of motives over the standard log point and rigid analytic motives are equivalent, and we relate log motivic nearby cycles functor with Ayoub's motivic nearby cycles functor.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to define and study log motivic nearby cycles over discrete valuation rings (DVRs) with different characteristics. Specifically, the author Doosung Park defines the log motivic nearby cycle functor in the paper and shows how this functor maps the motive of a smooth proper scheme over the fraction field of a DVR to the motive of a log - smooth model of its boundary, assuming absolute purity holds. This result holds unconditionally in the equal - characteristic case. Moreover, in the characteristic 0 case, the author also proves that the infinity - category over the standard log - point is equivalent to the infinity - category of rigid - analytic motives, and explores the relationship between the log motivic nearby cycle functor and Ayoub's motivic nearby cycle functor.
### Main contributions
1. **Define the log motivic nearby cycle functor**:
- The author defines a new functor \(\Psi^{\text{log}}\), called the log motivic nearby cycle functor, which maps from \(SH(K)\) to \(SH(k^\dagger)\), where \(K\) is the fraction field of a DVR and \(k^\dagger\) is the standard log - point.
- This functor maps the motive of a smooth proper scheme over \(K\) to the motive of a log - smooth model of its boundary, assuming absolute purity holds.
2. **Equivalence in the equal - characteristic case**:
- In the equal - characteristic 0 case, the author proves that the infinity - category over the standard log - point is equivalent to the infinity - category of rigid - analytic motives, that is, \(SH(k^\dagger)\simeq\text{Rig}SH(k((x)))\) and \(DA(k^\dagger,\Lambda)\simeq\text{Rig}DA(k((x)),\Lambda)\).
3. **Relationship with Ayoub's motivic nearby cycle functor**:
- The author shows that in the equal - characteristic 0 case, the log motivic nearby cycle functor \(\Psi\) is naturally isomorphic to Ayoub's motivic nearby cycle functor \(\Psi_{\text{Ayo}}\), that is, \(\Psi\simeq\Psi_{\text{Ayo}}\).
### Technical details
- **Existence of log - smooth models**:
- The author points out that in the equal - characteristic 0 case, log - smooth models always exist, while in the mixed - characteristic or positive - characteristic case, log - smooth models may not exist unless certain additional conditions (such as absolute purity) are met.
- **Assumption of absolute purity**:
- The assumption of absolute purity is necessary in the mixed - characteristic case because it ensures that the log motivic nearby cycle functor maps the identity to the identity.
- **Properties of the log motivic nearby cycle functor**:
- The author proves that the log motivic nearby cycle functor is a tensor functor and is a monomorphism and an epimorphism under certain conditions.
### Applications
- **Extraction of motivic information**:
- Through the log motivic nearby cycle functor, motivic information can be extracted, especially in the equal - characteristic 0 case, elements in the Grothendieck group of motives can be calculated.
- **Comparison of different motivic theories**:
- The author's work provides a method to compare motivic theories from different sources, such as motives in log - geometry and motives in rigid - analytic geometry.
In conclusion, this paper provides a new tool for understanding the motivic behavior over DVRs with different characteristics by introducing the log motivic nearby cycle functor and establishes connections with existing motivic theories in the equal - characteristic 0 case.