English

The regularity of the boundary of a multidimensional aggregation patch

Analysis of PDEs 2016-08-30 v2

Abstract

Let d2d \geq 2 and let N(y)N(y) be the fundamental solution of the Laplace equation in RdR^d We consider the aggregation equation ρt+div(ρv)=0,v=Nρ \frac{\partial \rho}{\partial t} + \operatorname{div}(\rho v) =0, v = -\nabla N * \rho with initial data ρ(x,0)=χD0\rho(x,0) = \chi_{D_0}, where χD0\chi_{D_0} is the indicator function of a bounded domain D0Rd.D_0 \subset R^d. We now fix 0<γ<10 < \gamma < 1 and take D0D_0 to be a bounded C1+γC^{1+\gamma} domain (a domain with smooth boundary of class C1+γC^{1+\gamma}). Then we have Theorem: If D0D_0 is a C1+γC^{1 + \gamma} domain, then the initial value problem above has a solution given by ρ(x,t)=11tχDt(x),xRd,0t<1\rho(x,t) = \frac{1}{1 -t} \chi_{D_t}(x), \quad x \in R^d, \quad 0 \le t < 1 where DtD_t is a C1+γC^{1 + \gamma} domain for all 0t<10 \leq t < 1.

Keywords

Cite

@article{arxiv.1507.07831,
  title  = {The regularity of the boundary of a multidimensional aggregation patch},
  author = {Andrea Bertozzi and John Garnett and Thomas Laurent and Joan Verdera},
  journal= {arXiv preprint arXiv:1507.07831},
  year   = {2016}
}