← All plenary speakers

Daniel Reynolds

Professor and Thomas I. Seidman Endowed Chair in Applied Mathematics, University of Maryland Baltimore County

Flexible and High Order Time Integration for Multiphysics PDEs

Abstract

In recent years computational simulations have rapidly evolved in complexity (high order discretizations, spatial adaptivity, wide arrays of physical processes), placing ever larger strains on the time integration methods on which they rely. High spatial order necessitates comparably accurate time integration. Spatial adaptivity and multiphysics processes give rise to subsets of the solution that evolve at differing time scales, or to simulations that combine nonstiff/nonlinear processes with others that may be stiff and linear. In this talk, I will discuss recent work on time integration methods that allow the flexibility to apply different techniques to distinct physical processes. While techniques for flexible time integration have existed for some time, including additive Runge--Kutta ImEx, multirate (a.k.a. multiple time stepping), and operator-splitting methods, there have been comparably few that combine these types of flexibility into a single family, while also supporting high orders of accuracy and temporal adaptivity. In this talk, I focus on recent developments within the family of multirate ``infinitesimal'' (MRI) methods, including extensions of this family for ImEx partitions, multirate time adaptivity, and incorporation of ``super-time-stepping'' methods for efficient treatment of diffusive terms. In addition to discussing these methods themselves, I'll also discuss our progress toward implementing and releasing these as part of the open-source SUNDIALS library.

Bio

Daniel Reynolds is Professor and the Thomas I. Seidman Endowed Chair in Applied Mathematics at the University of Maryland Baltimore County. His research focuses on the development and application of robust time integrators and iterative nonlinear and linear solvers for large-scale multiphysics systems -- simulations comprised of multiple interacting physical processes. Reynolds is the lead developer of the ARKODE library, a collection of solvers for adaptive implicit-explicit and multirate time integration methods for large-scale systems of differential equations, and is a senior developer for the larger SUNDIALS suite of time integration and nonlinear solver libraries (which includes ARKODE).

Prior to joining UMBC in 2025, Reynolds was a professor at Southern Methodist University (2008-2025), held postdoctoral research positions in the Department of Mathematics and the Center for Astrophysics and Space Sciences at the University of California, San Diego (2005-2008), and in the Center for Applied Scientific Computing at Lawrence Livermore National Laboratory (2003-2005). He received a PhD in Computational and Applied Mathematics from Rice University in 2003.