IRREDUCIBILITY AND SMOOTHNESS OF THE MODULI SPACE OF MATHEMATICAL 5-INSTANTONS OVER ℙ<sub>3</sub> (Q4807430)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1911964
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | IRREDUCIBILITY AND SMOOTHNESS OF THE MODULI SPACE OF MATHEMATICAL 5-INSTANTONS OVER ℙ<sub>3</sub> |
scientific article; zbMATH DE number 1911964 |
Statements
IRREDUCIBILITY AND SMOOTHNESS OF THE MODULI SPACE OF MATHEMATICAL 5-INSTANTONS OVER ℙ<sub>3</sub> (English)
0 references
18 May 2003
0 references
cohomology of a monad
0 references
twistor space
0 references
restriction technique
0 references
0 references
0.8832569
0 references
0.8066526
0 references
0 references
0.7765031
0 references
0.7707701
0 references
0.7634654
0 references
Among the rank two vector bundles \({\mathcal E}\) over \(\mathbb P^3\), those with Chern classes \(c_1({\mathcal E})=0\), \(c_2({\mathcal E})=n>0\) fulfilling \(H^0({\mathcal E})=H^1({\mathcal E}(-2))=0\) (or, equivalently, stable with spectrum \(0^n\)) are of special interest due to their relation to self-dual Yang-Mills fields on \(S^4\) via the Ward correspondence in twistor theory. The moduli space MI\((n)\) of such bundles, which is an open stratum of Maruayma's moduli scheme, has been conjectured to be smooth and irreducible (``instanton conjecture''), and during the last decades there have been many attempts of proving this at least for small Chern numbers. The case of null-correlation bundles (i.e. \(n=1\)) is classical - MI\((1)\) is \(\mathbb P^5\) minus a smooth quadric. \textit{R. Hartshorne} [Math. Ann 238, 229--280 (1978; Zbl 0411.14002)] proved the conjecture for \(n=2\) and \textit{G. Ellingsrud} and \textit{S.A. Strømme} [Math. Ann. 255, 123--135 (1981; Zbl 0438.14009)] for \(n=3\). For \(n=4\), irreducibility was settled by \textit{W. Barth} [Math. Ann. 258, 81--106 (1981; Zbl 0477.14014)] and smoothness by \textit{J. Le Potier} [in: Mathématique et physique. Sémin Éc. Norm. Supér., Paris 1979-1982, Prog. Math. 37, 65--137 (1983; Zbl 0532.14003)]. Recently, \textit{P.I. Katsylo} and \textit{G.M. Ottaviani} [Transf. groups 8, 147--158 (2003; Zbl 1071.14016)] showed the smoothness of MI\((5)\). In the paper under review, the authors present a unified approach of proving the instanton conjecture, involving mathematical instantons of higher rank \(r\). Here, an mathematical \((n,r)\)-instanton is a rank \(r\) vector bundle on \(\mathbb P^3\) satisfying the conditions above and admitting a symplectic isomorphism to its dual (the latter condition is always fulfilled in the rank \(2\) case). Any such bundle is the cohomology of a monad \(H_n\otimes{\mathcal O}(-1)\rightarrow N\otimes{\mathcal O}\rightarrow H_n^*\otimes{\mathcal O}(1)\) (\(H_n\) and \(N\) are vector spaces of dimension \(n\) and \(2n+r\), respectively). As the main new tool, an induction step is introduced relating \((n,r)\)-instantons to \((n-1,r+2)\)-instantons by restricting this monad to an appropriate codimension one subspace \(H_{n-1}\subset H_n\), inducing a restriction map between the moduli spaces. The crucial step is to proof the constancy of the dimension of the fibres of this map. Here one needs to control the first cohomology of the bundle twisted by one. Along these lines, the authors recover all previous results, since by a suitable choice this cohomology vanishes for \(n\leq 4\). As the main new result, the irreducibility of MI\((5)\) is proven together with its smoothness. This involves bounding dim\(H^1(\overline{\mathcal E}(1))\leq 1\) and a clever dimension count argument. For \(n\leq 6\), this fails by now due to a lack of understanding of the cohomology. The paper is well-written and detailed, and illustrates the methods and obstructions to attack the instanton conjecture in general.
0 references