Geometric rigidity of \(\times m\) invariant measures (Q713957): Difference between revisions

This paper gives news insights into the mechanism behind abelian measure rigidity. Fix a measure \(\mu\) on \([0,1]\) invariant under the map \(T_a:x\mapsto ax\) modulo \(1\) (there are of course a great diversity of such measures with diverse other properties). \textit{D. J. Rudolph} [Ergodic Theory Dyn. Syst. 10, No. 2, 395--406 (1990; Zbl 0709.28013)] and \textit{A. S. A. Johnson} [Isr. J. Math. 77, No. 1--2, 211--240 (1992; Zbl 0790.28012)] showed that if each ergodic component of \(\mu\) has entropy in \((0,\log a)\) then \(\mu\) cannot be \(T_b\)-invariant for any \(b\) with the property that \(a\) and \(b\) are not both powers of the same integer; among many later developments \textit{A. Johnson} and \textit{D. J. Rudolph} [Adv. Math. 115, No. 1, 117--140 (1995; Zbl 0934.28010)] showed that if \(\nu\) is \(T_b\)-invariant for such a \(b\) and has intermediate local dimension (that is, local dimension in \((0,1)\) almost everywhere, eliminating the possibility of both atoms and Lebesgue components), then \(\mu\perp\nu\). Here the main result is that if \(\mu\) is ergodic for \(T_a\) and has intermediate local dimension then there is some \(n\in\mathbb N\) so that if \(f:\mathbb R\to\mathbb R\) is a diffeomorphism and \((f\mu)|_E\) is equivalent to \(\mu|_E\) then \(f'(x)\) lies in \(\{\pm a^{k/n}: k\in\mathbb Z\}\) for \(\mu\)-almost every \(x\in f^{-1}E\), and if \(\nu\) is another \(T_a\)-ergodic measure with \((f\mu)|_E\) equivalent to \(\nu|_E\) then there is some \(t\in\mathbb R\) with the property that \(f'(x)\) lies in \(\{\pm ta^{k/n}: k\in\mathbb Z\}\) for \(\mu\)-almost every \(x\in f^{-1}E\). A corollary is that if \(\mu\) is a \(T_a\)-ergodic measure with intermediate dimension and global support, then any \(C^1\) map preserving \(\mu\) has the form \(T_b\) for some \(b\) with the property that \(a\) and \(b\) are powers of the same integer. These results give a generalisation of the work of Rudoplh and Johnson, showing that essentially their conclusion holds for any pair of maps \(C^2\) conjugate to \(T_a\), \(T_b\).