A scalar version of the Caflisch-Luke paradox

Abstract

Consider an infinite cloud of hard spheres sedimenting in a Stokes flow in the whole space ${\mathbb{R}}^d$. Despite many contributions in fluid mechanics and applied mathematics, there is so far no rigorous definition of the associated effective sedimentation velocity. Calculations by Caflisch and Luke in dimension $d=3$ suggest that the effective velocity is well-defined for hard spheres distributed according to a weakly correlated and dilute point process, and that the variance of the sedimentation speed is infinite. This constitutes the Caflisch-Luke paradox. In this contribution, we consider a scalar version of this problem that displays the same difficulties in terms of interaction between the differential operator and the randomness, but is simpler in terms of PDE analysis. For a class of hardcore point processes we rigorously prove that the effective velocity is well-defined in dimensions $d>2$, and that the variance is finite in dimensions $d>4$, confirming the formal calculations by Caflisch and Luke, and opening a way to the systematic study of such problems.

arXiv comment38 pages. Update of bibliography + extended introduction

References in corpus (5)

• Quantitative stochastic homogenization and large-scale regularity
• The additive structure of elliptic homogenization
• Quantitative results on the corrector equation in stochastic homogenization
• The corrector in stochastic homogenization: optimal rates, stochastic integrability, and fluctuations
• Quantitative estimates in stochastic homogenization for correlated coefficient fields

Cited by in corpus (2)

• Corrector equations in fluid mechanics: Effective viscosity of colloidal suspensions
• A homogenized limit for the 2D Euler equations in a perforated domain