It is known that if $X$ is a curve over a number field $F$ equipped with a flat regular model over $O_F$ the integer ring, one can define, using a suitable ample line bundle with an Hermitian metric, the notion of height for points in $X$. In particular, there are only finitely many algebraic points with height lower than a given constant.

What happens when one studies the heights of CM points on the compactified modular curve, with respect to the ample line bundle defining modular forms of large weight (plus suitable Hermitian metric)? Is it possible to study the distribution of CM cycles through height functions? Does one have similar results as in the previous example, say bounding the number of CM points by heights?

This is also motivated by the Bogomolov conjecture. Roughly speaking, for $A$ an abelian variety over a number field, and $Z\subset A$ an irreducible closed subvariety that is not a torsion subvariety (abelian subvariety translated by a torsion point), then the "number" (or some other numerical invariants) of torsion subvarieties in $Z$ is bounded in terms of suitable constants in heights, degrees, that only depends on $Z$. It might be of interest to work out an analogue for modular curves and perhaps other moduli spaces like Shimura varieties.

Excuse me for the vague statements above and thanks in advance for comments and references.