Parametric Coarea Inequality for 1-cycles
Bruno Staffa
TL;DR
The paper proves a Parametric Coarea Inequality for 1-cycles in arbitrary dimension $n\ge 4$, extending Guth-Liokumovich's results from 3-manifolds to higher dimensions. The authors develop a multi-scale coarea-cut/interpolation framework using delta-localized families, cones, and a hierarchy of PL submanifolds to control both the mass of the perturbed family and the mass of its boundary across all skeleta. They formulate and prove the interpolation machinery on rectangular domains and then extend it to piecewise smooth triangulable manifolds, establishing quantitative bounds: $\mathcal{F}(F(x),F'(x))\le\eta$, $\mathbf{M}(F'(x))\le(1+\gamma_{p})\mathbf{M}(F(x)) + c\mathrm{Vol}_{n-1}(\partial M)p^{\frac{n-2}{n-1}+\alpha}$, and $\mathbf{M}(\partial F'(x))\le c\mathrm{Vol}_{n-1}(\partial M)p^{1+\alpha}$ with $\gamma_{p}\to 0$ as $p\to\infty$. These results underpin the Weyl law for the volume spectrum of 1-cycles in any dimension and have implications for equidistribution of stationary geodesic nets. The approach blends Almgren-Pitts min-max concepts with careful geometric constructions to tame boundary contributions in high codimension.
Abstract
We prove the Parametric Coarea Inequality for $1$-cycles conjectured by Guth and Liokumovich.
