Pome

Proposition 19.8.134ABE18AC08D2

Proposition 19.8.13.4ABE18AC08D2 For $m\geq 1$ and the weight $u=0$,

\[ N_ m:=h^0(X_ n,L_ k^ m)=\binom {km+n-1}{n-1}, \qquad B_{m,0}=N_ m. \]

The Bergman potential is the constant $m^{-1}\log N_ m$, and the Bergman current is $k\alpha $. Moreover,

\[ \deg _\omega L_ k=2k(n-1),\qquad \lim _{m\to \infty }m^{-n}N_ m=0, \qquad (k\alpha )^ n=0, \]

whereas $\int _{X_ n}\omega ^ n=2n>0$.

Proof. A section of $L_ k^ m$ lifts to a holomorphic function $f$ on $\mathbb C^ n\setminus \{ 0\} $ satisfying

\[ f(\mathrm{e}^{-1}z)=\mathrm{e}^{-km}f(z). \]

Since $n\geq 2$, Hartogs' theorem extends $f$ holomorphically to $\mathbb C^ n$. In its Taylor series, a monomial of total degree $j$ is multiplied by $\mathrm{e}^{-j}$ under the contraction, so comparing coefficients shows that only monomials of degree $km$ occur. Conversely, every homogeneous polynomial of that degree has the required equivariance. Counting monomials gives the binomial coefficient; when $k=0$, only the constant polynomials remain.

The inner product defined by $h_ k^ m$ and $\mathrm{d}V$ is invariant under the unitary group $U(n)$, so the basis-independent function $K_{m,0}$ is $U(n)$-invariant. It is also invariant under positive radial dilations: every section is homogeneous of degree $km$, and the factor $|z|^{-2km}$ in its squared norm cancels the effect of a dilation. Unitary transformations and positive radial dilations act transitively on $\mathbb C^ n\setminus \{ 0\} $, so $K_{m,0}$ is constant. By proposition 19.8.3 its integral is $N_ m$, and $\mathrm{d}V$ has mass one. Thus $K_{m,0}=B_{m,0}=N_ m$.

The statements about the Bergman potential and current follow immediately. Since $\alpha ^ n=0$ and $\zeta ^2=0$,

\[ \int _{X_ n}k\alpha \wedge \omega ^{n-1} =k(n-1)\int _{X_ n}\alpha ^{n-1}\wedge \zeta =2k(n-1). \]

The binomial coefficient grows at most like $m^{n-1}$, which proves the normalized limit, and $\alpha ^ n=0$ proves the vanishing of the curvature mass. $\square$

Cite this page

Comments

Corrections, questions, and mathematical discussion are welcome. Sign in with GitHub to comment. Comments are public.