Evaluating the relationship between interannual variations in the Antarctic ozone hole and Southern Hemisphere surface climate in chemistry–climate models

• Published in: Journal of Climate
• Other Authors: Gillett, Zoe E. (author), Arblaster, Julie M. (author), Dittus, Andrea J. (author), Deushi, Makoto (author), Jöckel, Patrick (author), Kinnison, Douglas E. (author), Morgenstern, Olaf (author), Plummer, David A. (author), Revell, Laura E. (author), Rozanov, Eugene (author), Schofield, Robyn (author), Stenke, Andrea (author), Stone, Kane A. (author), Tilmes, Simone (author)
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2019
• Subjects: Antarctic; The Antarctic; Antarc*
• Online Access: https://doi.org/10.1175/JCLI-D-18-0273.1

Studies have recently reported statistically significant relationships between observed year-to-year spring Antarctic ozone variability and the Southern Hemisphere annular mode and surface temperatures in spring-summer. This study investigates whether current chemistry-climate models (CCMs) can capture these relationships, in particular, the connection between November total column ozone (TCO) and Australian summer surface temperatures, where years with anomalously high TCO over the Antarctic polar cap tend to be followed by warmer summers. The interannual ozone-temperature teleconnection is examined over the historical period in the observations and simulations from the Whole Atmosphere Community Climate Model (WACCM) and nine other models participating in the Chemistry-Climate Model Initiative (CCMI). There is a systematic difference between the WACCM experiments forced with prescribed observed sea surface temperatures (SSTs) and those with an interactive ocean. Strong correlations between TCO and Australian temperatures are only obtained for the uncoupled experiment, suggesting that the SSTs could be important for driving both variations in Australian temperatures and the ozone hole, with no causal link between the two. Other CCMI models also tend to capture this relationship with more fidelity when driven by observed SSTs, although additional research and targeted modeling experiments are required to determine causality and further explore the role of model biases and observational uncertainty. The results indicate that CCMs can reproduce the relationship between spring ozone and summer Australian climate reported in observational studies, suggesting that incorporating ozone variability could improve seasonal predictions; however, more work is required to understand the difference between the coupled and uncoupled simulations.