An Online‐Learned Neural Network Chemical Solver for Stable Long‐Term Global Simulations of Atmospheric Chemistry
- Resource Type
- article
- Authors
- Makoto M. Kelp; Daniel J. Jacob; Haipeng Lin; Melissa P. Sulprizio
- Source
- Journal of Advances in Modeling Earth Systems, Vol 14, Iss 6, Pp n/a-n/a (2022)
- Subject
- machine learning
atmospheric chemical mechanism
chemical transport modeling
model emulation
neural network chemical solver
machine learning chemical solver
Physical geography
GB3-5030
Oceanography
GC1-1581
- Language
- English
- ISSN
- 1942-2466
Abstract A major computational barrier in global modeling of atmospheric chemistry is the numerical integration of the coupled kinetic equations describing the chemical mechanism. Machine‐learned (ML) solvers can offer order of magnitude speedup relative to conventional implicit solvers but past implementations have suffered from fast error growth and only run for short simulation times (