Weighted projective embeddings, stability of orbifolds, and constant scalar curvature Kähler metrics (Q661934)

Recently, there has been some important progress on the problem of existence of a constant scalar curvature Kähler metric over polarised complex manifolds. In this paper, the goal of the authors is to extend Donaldson's theory (of balanced metrics and K-stability) to the case of polarized orbifolds. They actually consider a more restrictive case of orbifolds with cyclic quotient singularities, giving some hints to extend their theory to the most general case. There are obviously different problems that appear in this context to extend Donaldson's work. The first main issue is to consider the right embeddings of the manifold. Actually one needs to embed the orbifold into weighted projective spaces. Another problem is coming from the fact that one needs to consider weighted Bergman kernels, and prove new asymptotics results, see the companion paper of the authors [J. Differ. Geom. 88, No. 1, 87--107 (2011; Zbl 1244.32002)]. Also, it is necessary to obtain a new form of the Riemann-Roch formula adapted to this context. As one can guess, it is not a straigtforward generalization of the manifold case since new tools and new definitions need to be developped with care. The main result of the authors is the following. If \((X,L)\) is a polarised orbifold with cyclic quotient singularities and if \(c_1(L)\) admits an orbifold Kähler metric of constant scalar curvature, then \((X,L)\) is K-semistable. The authors provide also the right definition of balanced metric and balanced embedding for orbi-ample line bundles. They prove that the limit of a balanced metric, if it exists, is an orbifold Kähler metric with constant scalar curvature. Finally, they generalize Donaldson's definition of K-stability to orbifolds. Various applications of these interesting results are given to the case of orbifold ruled surfaces or slope stability of Riemann surfaces. It is likely that, in a long range, this paper will be very useful to define and understand stability for pairs \((X,D)\) where \(D\) is a divisor in relationship with existence of cuspidal metrics on \(X\setminus D\) with constant scalar curvature. The paper is very well written and opens new perspectives on the topic of the Yau-Tian-Donaldson conjecture on non smooth varieties.