Aashirwad N. Ballal,Vamsi P. Pingali
Abstract:We study MA-positivity, a notion of positivity relevant to a vector bundle version of the complex Monge--Ampère equation introduced in an earlier work, and show that for rank-two holomorphic bundles over complex surfaces, MA-semi-positive solutions of the vector bundle Monge--Ampère (vbMA) equation are also MA-positive. For vector bundles of rank-three and higher, over complex manifolds of dimension greater than one, we show that this positivity-preservation property need not hold for an algebraic solution of the vbMA equation treated as a purely algebraic equation at a given point. Finally, we set up a continuity path for certain classes of highly symmetric rank-two vector bundles over complex three-folds and prove a restricted version of positivity preservation which is nevertheless sufficient to prove openness along this continuity path.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is about the positivity properties of the complex Monge - Ampère equation on vector bundles (vbMA equation). Specifically, the author explores the preservation of MA - positivity under different conditions, especially for vector bundles of rank 2 on complex surfaces, as well as cases of higher ranks and dimensions. The following are the main problems and conclusions of the paper:
### Main Problems
1. **Preservation of MA - Positivity**
- For a vector bundle of rank 2 on a complex surface, is an MA - semi - positive solution also an MA - positive solution?
- For a vector bundle of rank greater than 2 on a complex manifold, is an MA - semi - positive solution also an MA - positive solution?
2. **Properties of Algebraic Solutions**
- For a vector bundle of rank greater than 2 on a complex manifold, does there exist an algebraic solution such that MA - semi - positivity holds but MA - positivity does not?
3. **Application of the Continuous Path Method**
- How to set up continuous paths on some specific high - symmetric rank - 2 vector bundles and prove the preservation of MA - positivity on these paths?
### Main Conclusions
1. **Vector Bundles of Rank 2 on Complex Surfaces**
- The author proves that for vector bundles of rank 2 on complex surfaces, an MA - semi - positive solution is indeed an MA - positive solution. That is, \( V(2,2,\eta) \cap S(2,2) \subseteq P(2,2) \).
2. **Vector Bundles of Rank Greater than 2 on Complex Manifolds**
- The author provides a counterexample, showing that for vector bundles of rank greater than 2 on complex manifolds, an MA - semi - positive solution may not preserve MA - positivity. That is, \( V(n,r,\eta) \cap S(n,r) \not\subseteq P(n,r) \) when \( n \geq 2 \) and \( r > 2 \).
3. **Restricted MA - Positivity**
- The author sets up continuous paths on some specific high - symmetric rank - 2 vector bundles and proves that on such paths, restricted MA - semi - positivity does preserve restricted MA - positivity. This provides new methods and ideas for solving related problems.
### Formula Analysis
- **vbMA Equation**
\[
(i\Theta_h)^n=\eta \otimes \text{Id}
\]
where \( \Theta_h \) is the curvature form of the Chern connection and \( \eta \) is the volume form on the manifold.
- **MA - Positivity Condition**
\[
\sum_{k = 0}^{n - 1}\text{Tr}(ia \wedge (i\Theta_h)^k \wedge a^* \wedge (i\Theta_h)^{n - 1 - k})>0
\]
for all non - zero \( a \in \Omega^{1,0}(\text{End}(V)) \).
- **MA - Semi - Positivity Condition**
\[
\sum_{k = 0}^{n - 1}\text{Tr}(ia \wedge (i\Theta_h)^k \wedge a^* \wedge (i\Theta_h)^{n - 1 - k})\geq0
\]
for all \( a \in \Omega^{1,0}(\text{End}(V)) \).
### Conclusion
Through these studies, the author not only verifies the preservation of MA - positivity under specific conditions but also reveals the complexity and challenges of the preservation of MA - positivity in cases of higher ranks and dimensions. These results provide an important theoretical basis for understanding the complex Monge - Ampère equation on vector bundles and its applications.