• Tim Wildey, Sandia National Labs
• Graham Harper

Multiphysics systems are typically strongly coupled, highly nonlinear and characterized by multiple physical phenomena that span a large range of length- and time-scales. Performing direct numerical simulation of such systems that resolves all the relevant length- and time-scales is often prohibitive, even on the modern leadership-class computing platforms. Nevertheless, these fine-scale variations often impact the behavior of the system on a much larger scale in non-negligible ways. Thus, one often seeks to model or approximate the subgrid scale phenomena to accurately and efficiently capturing their impact on the coarse scale solution. Recent advances in data science, including scientific machine learning, have opened up new possibilities for developing data-informed or data-driven multiscale modeling strategies. This includes, but is not limited to, discovery of multiscale systems and solution of multiscale systems via neural networks or physics-informed neural networks. In addition, recent developments in modern computational architectures has spurred growth in the field of multiscale multiphysics systems, with efficiency and accuracy increasing as new approaches are considered.

The goal of this mini-symposium is to provide an opportunity for researchers to present recent work and exchange ideas on novel methods for multiscale systems, integrating data with multiscale models and applications to challenging multiphysics applications.

We appreciate contributions on the following topics:
- Multiscale approaches for multiphysics applications
- Spatial and/or temporal multiscale methods
- Concurrent or sequential multiscale approaches
- Data driven and machine learning approaches for multiscale applications
- Efficient algorithms for multiscale modeling