[Paper Review] Orientation data for coherent sheaves on the local projective plane
This paper establishes a canonical orientation data for the moduli stack of coherent sheaves on the local projective plane $Y = \mathrm{Tot}(\mathcal{O}_{\mathbb{P}^2}(-3))$ by proving compatibility of the canonical orientation data on quiver hearts $\mathrm{Mod}\text{-}A^n$ under the autoequivalence $\otimes \pi^*\mathcal{O}(1)$, and shows it agrees with the geometric orientation data. The result enables the definition of global motivic Donaldson-Thomas invariants on $Y$.
In this note, we show that the canonical orientation data on the quiver hearts are compatible under the autoequivalence $\_\otimesπ^*O(1)$, and hence glue to give an orientation data for the stack of coherent sheaves on local P2.
Motivation & Objective
- To establish a global orientation data for the moduli stack of coherent sheaves on the local projective plane $Y = \mathrm{Tot}(\mathcal{O}_{\mathbb{P}^2}(-3))$.
- To show that the canonical orientation data on quiver hearts $\mathrm{Mod}\text{-}A^n$ is compatible under the autoequivalence $\otimes \pi^*\mathcal{O}(1)$, enabling gluing.
- To prove that the glued orientation data from quiver hearts agrees with the geometric orientation data arising from the derived category of $Y$.
- To provide a foundational step for defining motivic Donaldson-Thomas invariants on $Y$ via a consistent orientation data.
Proposed method
- Utilizes the derived equivalence $D^b(Y) \simeq D^b(\mathrm{Mod}\text{-}A)$, where $A$ is the Jacobi algebra of a quiver with potential.
- Applies the canonical orientation data on quiver hearts $\mathrm{Mod}\text{-}A^n$ as defined in Davison [Dav].
- Proves compatibility of these orientation data under the autoequivalence $\otimes \pi^*\mathcal{O}(1)$ via canonical isomorphisms of determinant line bundles.
- Constructs a global line bundle $L_a$ on $\bigcup \mathrm{Mod}\text{-}A^n$ by gluing the $L_{A^n}$'s using the compatibility condition.
- Compares the glued quiver-based orientation data $L_a$ with the geometric orientation data $L_g$ via explicit resolution of $A^0_{\mathbb{P}^2}$ and determinant computations.
- Uses the canonical isomorphism between $\det(R\mathcal{H}om(F,F))$ and $L_a|_{\mathrm{Coh}_Y}$ to show $L_g \simeq L_a|_{\mathrm{Coh}_Y}$.
Experimental results
Research questions
- RQ1Is the canonical orientation data on quiver hearts $\mathrm{Mod}\text{-}A^n$ compatible under the autoequivalence $\otimes \pi^*\mathcal{O}(1)$?
- RQ2Can the compatible orientation data on $\mathrm{Mod}\text{-}A^n$ be glued to define a global orientation data on the stack of coherent sheaves on $Y$?
- RQ3Does the orientation data constructed from quiver hearts agree with the geometric orientation data on $Y$?
- RQ4What is the relationship between the motivic DT invariants of $Y$ and the quiver-side construction via $A$-modules?
Key findings
- The canonical orientation data on the quiver hearts $\mathrm{Mod}\text{-}A^n$ is compatible under the autoequivalence $\otimes \pi^*\mathcal{O}(1)$, as shown in Theorem 5.
- The compatible orientation data on $\mathrm{Mod}\text{-}A^n$ glues to define a global orientation data $L_a$ on the stack $\mathrm{Coh}_Y$ of coherent sheaves on $Y$.
- The geometric orientation data $L_g$ on $\mathrm{Coh}_Y$ is isomorphic to the restriction of the glued quiver-based orientation data: $L_g \simeq L_a|_{\mathrm{Coh}_Y}$, as proven in Theorem 6.
- The isomorphism $L_g \simeq L_a|_{\mathrm{Coh}_Y}$ is canonical and compatible across all $n$, ensuring consistency across the entire moduli stack.
- The construction provides a canonical, globally defined orientation data on $\mathrm{Coh}_Y$, enabling the definition of motivic Donaldson-Thomas invariants on $Y$.
- The result confirms that the quiver-side orientation data and the geometric orientation data coincide, validating the consistency of the motivic DT framework on $Y$.
Better researchstarts right now
From reading papers to final review, dramatically reduce your research time.
No credit card · Free plan available
This review was created by AI and reviewed by human editors.