Fractional Parabolic Theory as a High-Dimensional Limit of Fractional Elliptic Theory
Blair Davey, Mariana Smit Vega Garcia
TL;DR
This work develops a high-dimensional limiting framework to derive monotonicity formulas for fractional and degenerate parabolic problems from corresponding elliptic theories. By uniting the Caffarelli–Silvestre extension for elliptic and parabolic contexts with a novel elliptic-to-parabolic transform and bridge lemmas, the authors prove Almgren-type, Weiss-type, and Alt–Caffarelli–Friedman-type monotonicity results for degenerate parabolic equations and introduce a new epiperimetric inequality for weakly a-harmonic functions. The approach combines nonhomogeneous elliptic results, careful limiting arguments, and pushforward measures to transfer elliptic insights to the parabolic setting. These results extend the toolbox for fractional and free-boundary problems in higher dimensions and illuminate the structure of solutions to degenerate parabolic equations arising in fractional diffusion models.
Abstract
This paper continues the program that was initiated in \cite{Dav18} and continued in \cite{DSVG24}, where a high-dimensional limiting technique was developed and used to prove certain parabolic theorems from their elliptic counterparts. The articles \cite{Dav18} and \cite{DSVG24} address the constant-coefficient and variable-coefficient settings, respectively. Here, we focus on fractional operators. As shown in \cite{CS07}, \cite{NS16}, \cite{ST17}, fractional operators may be associated with certain degenerate operators via extension problems, so we study the corresponding class of degenerate operators. Our high-dimensional limiting technique is demonstrated through new proofs of three theorems for degenerate parabolic equations. Specifically, we establish the monotonicity of Almgren-type, Weiss-type, and Alt-Caffarelli-Friedman-type functionals in the degenerate parabolic setting. Each new parabolic proof in this article is based on a (new) related elliptic theorem and a careful limiting argument that is reminiscent of those from \cite{Dav18} and \cite{DSVG24}. Our proof of the degenerate parabolic Weiss-type monotonicity formula additionally uses an epiperimetric inequality for weakly $a$-harmonic functions, which we also prove. To the best of our knowledge, our Alt-Caffarelli-Friedman monotonicity result is new.
