Year: 2025
Type: article
Abstract: . Coupled chemistry–climate models (CCMs) are powerful tools for investigating chemical variability in the climate system, but high computational cost limits their use for hypothesis testing and adequately sampling variability on long timescales. Here, we present the first application of a linear inverse model (LIM) to emulate a CCM. A LIM is a lowdimensional empirical model that reproduces the CCM’s statistics and dynamics at low computational cost. By linearizing the CCM’s dynamics, the LIM captures coherent modes of variability, such as the El Niño Southern Oscillation (ENSO), that describe the coupled evolution of physical and chemical fields. Deterministic seasonal forecasts of the LIM result in skillful predictions of physical and chemical variables at lead times up to a year, outperforming damped persistence models. We show that the LIM’s skill in chemical fields depends on its coupled chemistry–climate modes: forecasts without the ENSO dynamical mode show a substantial loss of skill, suggesting the importance of ENSO in driving predictable chemical variability. These results demonstrate that the LIM can efficiently emulate CCM dynamics. It offers a practical tool for testing hypotheses about the drivers of chemistry-climate interactions and may enable efficient chemical data assimilation in the future.
Institutions: University of Washington, University of California, Berkeley
Language: English
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
SDG: Climate action
Open Access status: gold
Estimated APC paid: $1,725
Funder: Schmidt Futures
Award: G7846850352
OpenAlex ID: W4415987141