The authors write in this plain language summary:
Changes in the output of the Sun are thought to influence surface weather and climate through a set of processes initiated by the enhancement of upper stratosphere (32‐48 km) ozone. In order to understand and assess the solar impact on the climate system, it is important that models reproduce the observed solar signal. However, the recommended dataset for comparison with climate models remains disputed.
We use newly improved observed ozone composites to determine both why there is disagreement between composites, and which is most likely to be correct. We find that artefact‐corrected composites represent the response better than those based on SBUV data alone. Further, we identify a U‐shaped spatial structure with lobes emanating from the tropics to high altitudes at mid‐latitudes.
An idealised chemistry climate model experiment, and simulations considering historical meteorological conditions, both support this conclusion. The results are of benefit to satellite‐instrument scientists, and to those engaged in atmospheric and climate research using both observations and climate models.
The results will be important for assessing the solar signal in currently active and future assessments of chemistry climate models (e.g. CCMI). We recommend recommend the BASICv2 ozone composite to best represent historical upper stratospheric variability.
The upper stratospheric solar cycle ozone response
W. T. Ball E. Rozanov J. Alsing D. R. Marsh F. Tummon D. J. Mortlock D. E. Kinnison J. D. Haigh
First published: 31 January 2019, Geophysical Research Letters
The solar cycle (SC) stratospheric ozone response is thought to influence surface weather and climate. To understand the chain of processes and ensure climate models adequately represent them, it is important to detect and quantify an accurate SC ozone response from observations. Chemistry climate models (CCMs) and observations display a range of upper stratosphere (1–10 hPa) zonally averaged spatial responses; this and the recommended dataset for comparison remains disputed. Recent data‐merging advancements have led to more robust observational data. Using these data, we show that the observed SC signal exhibits an upper stratosphere U‐shaped spatial structure with lobes emanating from the tropics (5–10 hPa) to high altitudes at mid‐latitudes (1–3 hPa). We confirm this using two independent CCMs in specified dynamics mode and an idealised timeslice experiment. We recommend the BASICv2 ozone composite to best represent historical upper stratospheric solar variability, and that those based on SBUV alone should not be used.