A note on \(\delta\)-equilibrium measures for rational maps (Q1118229)

Here we consider the \(\delta\)-potential \(| z-x|^{- \delta}\), \(0<\delta \leq 2\), \(z,x\in\mathbb C\), and obtain some results relating the dynamics of a rational map and equilibrium potential properties. A classical problem in potential theory is: what is the minimal state of energy for such a \(\delta\)-potential when we consider measures with support in a certain compact set? In the case this measure exists, it is called the equilibrium measure. For a fixed rational map \(f\) we ask the following question: what is the minimal state of energy for such \(\delta\)-potential when we consider measures that are invariant by \(f\)? When the parameter \(\delta\) is equal to the Hausdorff dimension of the Julia set, the Gibbs state of a uniformly expanding rational map has infinite \(\delta\)-energy integral, but it is, analogous to the situation in classical potential theory, an equilibrium measure when we restrict ourselves to the \(f\)-invariant measures (see Theorem 2 in \S4). The results of \(\delta\)-potential theory that we refer to are the ones that state that arc length (respectively normalised 2-dimensional Lebesgue measure) is the equilibrium measure in a weak sense for the value of the parameter \(\delta =1\) (respectively \(\delta =2)\), for a smooth compact curve (respectively a compact set of positive 2-dimensional Lebesgue measure) in the plane (see \textit{O. Frostman} [Skand. Mat.-Kongr., Helsinki, 86--89 (1958; Zbl 0090.08302)], \textit{L. Lithner} [Ark. Mat. 4, 31--33 (1960; Zbl 0178.46301)] and \textit{H. Wallin} [ibid., 527--549 (1963; Zbl 0114.30602)]). In this case the arc length (respectively normalised Lebesgue measure) has infinite 1-energy integral (respectively 2-energy integral), which resembles the statement above about Gibbs states. Recently, another result has appeared about Riesz transform of measures with support in the Julia set in ``Iterated function systems and the global construction of fractals'' [\textit{M. F. Barnsley} and \textit{S. Demko}, Proc. R. Soc. Lond., Ser. A 399, 243--275 (1985; Zbl 0588.28002)]. In this paper the balanced measure is considered instead of the Gibbs state.