Development and Verification of a Numerical Library for Solving Global Terrestrial Multiphysics Problems

Journal Article · 04 June 2019 · Journal of Advances in Modeling Earth Systems

DOI: https://doi.org/10.1029/2018MS001560 · OSTI ID:1560160

Bisht, Gautam  ^[1]; Riley, William J.  ^[2]

---

1. Climate and Ecosystem Sciences Division Lawrence Berkeley National Laboratory Berkeley CA USA, Now at Atmospheric Sciences and Global Change Division Pacific Northwest National Laboratory Richland WA USA
2. Climate and Ecosystem Sciences Division Lawrence Berkeley National Laboratory Berkeley CA USA

Abstract Current generation Land Surface Models (LSMs) routinely simulate many nonlinear multiphysics processes. The complexity of future generation LSMs is expected to increase as critical new biophysical and biogeochemical processes are incorporated. Current generation LSMs have several shortcomings including the lack of robust numerical methods to solve discretized equations, monolithic software design that hinders testing of a process representation in isolation from other components, and absence of a flexible coupling framework to solve tightly coupled multiphysics problems. While the LSMs community vigorously evaluates the accuracy of the conceptual model to represent reality (validation), the accuracy of the numerical implementation of the conceptual model (verification) is seldom examined. Method of Manufactured Solutions is a technique for verifying complex codes when analytical solutions are unavailable. In this work, we present a stand‐alone, open source numerical library for solving global terrestrial multiphysics processes that includes a flexible coupling framework. Robust numerical solution for linear and nonlinear equations is provided via the Portable, Extensible Toolkit for Scientific Computation library. We verify the numerical library using Method of Manufactured Solutions for a range of problems comprising single and multiphysics steady state problems that are applied in one or multiple physical domains. This work provides an example of incorporating code verification as an integral part of Land Surface Model development activities.

• Sponsoring Organization: USDOE
• Grant/Contract Number: AC02-05CH11231
• OSTI ID: 1560160
• Journal Information: Journal of Advances in Modeling Earth Systems, Journal Name: Journal of Advances in Modeling Earth Systems Journal Issue: 6 Vol. 11; ISSN 1942-2466
• Publisher: American Geophysical Union (AGU)Copyright Statement
• Country of Publication: United States
• Language: English