Stratospheric ozone drives the climate directly through local heating and further controls tropospheric chemistry and many important short-lived climate forcers. Many climate models lack the computational resources required for a fully interactive chemistry package, and hence they prescribe it with climatologies that fail to represent the interactions between circulation changes and ozone heating. The new U.S. Department of Energy Energy Exascale Earth System Model version 1 (E3SMv1) aims to improve its climate simulations via better physical representations of key climate processes and feedbacks. In E3SMv1 the computationally efficient linearized stratospheric ozone chemistry scheme, Linoz v2, was implemented, enabling stratospheric ozone to respond to dynamics; however, a prescribed climatology remained for tropospheric ozone. This chemistry led to unphysically high ozone levels and heating rates near the tropopause, potentially undermining the climate simulations.

Here we introduce a new O3v2 scheme to replace the prescribed tropospheric ozone in the E3SMv1 O3v1 scheme. With the O3v2 scheme, the ozone tendencies are better simulated around the tropopause, preserving the naturally sharp cross-tropopause gradient and heating rates. Additionally, O3v2 enables the stratosphere-troposphere exchange of ozone fluxes. Comparing to satellite observations and simulations from the University of California, Irvine chemistry transport model running the same scheme, we evaluate the key features of ozone and other closely related quantities in the O3v1 and O3v2 E3SMv1 AMIP simulations.

For the Antarctic ozone hole, O3v2 triggers a dynamic instability in the lower stratospheric polar vortex for some years that is not found in O3v1 and is presumably due to the reduced ozone heating at this location. The one free parameter in Linoz chemistry is the polar stratospheric cloud threshold temperature for chlorine activation and rapid ozone depletion. We find that a threshold temperature of 197.5 K (195-198 K as observed) allows O3v2 to reproduce much of the observed daily statistics of the ozone hole. This O3v2 configuration retains mostly the same climate and climate sensitivity as O3v1 with slight improvements. Therefore, we recommend the O3v2 scheme for future E3SM (e.g. version 2) and other climate models.