Joint Program Reprint

The emergence of human influence on the ozone layer by the 1960s

Guan, J., B.D. Santer, P. Wang, Q. Fu, R.R. Garcia, Y. Li, K. Stone, D.E. Kinnison, J. Zhang, G. Chiodo, J.-F. Lamarque and S. Solomon (2026)
PNAS, 123(28), e2608286123 (doi: 10.1073/pnas.2608286123)

Abstract / Summary:

Significance

This paper examines the earliest emergence of human-caused ozone depletion: the when, the where, and the why. We employ a thought-experiment framework aimed at the physical emergence of the signal in a hypothetical world in which instruments had the same capabilities as today’s satellites. The “when” is as early as the late 1950s—about 30 y before the Antarctic ozone hole was discovered and 20 y before the Molina–Rowland theory. The “where” is the tropical upper stratosphere, where a relatively small signal stands out against even smaller noise, allowing the earliest emergence. The “why” is the use of carbon tetrachloride as a solvent decades before chlorofluorocarbons became common in refrigeration and spray cans.

Abstract

The Antarctic ozone hole was first reported in 1985, and small ozone losses at the global scale were also observed in the late 1980s. The combination of field and laboratory measurements, together with modeling, quickly established anthropogenic chlorofluorocarbons (CFCs) as the cause of both the Antarctic and global ozone depletion. However, when, where, and why the earliest ozone depletion could have been detected has not been determined. Here, we conduct a thought experiment to investigate when human-induced ozone depletion could have first been detectable, assuming the availability of accurate stratospheric ozone observations from 1950 onward. 

We find that human-caused ozone depletion was likely identifiable as early as 1957 in the tropical upper stratosphere. This region’s low internal variability enables the earliest detection of the anthropogenic signal, even though tropical ozone losses in the upper stratosphere were smaller than those in higher-latitude regions. Our results highlight the key role of considering both internal variability (“noise”) and the forced response (“signal”) in detection studies. Further, while CFCs are widely recognized as the primary drivers of current ozone depletion, we find that early ozone loss was primarily caused by human-made carbon tetrachloride (CCl4), used mainly as a solvent. 

These findings suggest that a clear human influence on the stratospheric ozone layer began nearly 70 years ago, even before substantial emissions of CFCs from spray cans or air conditioning.

Citation:

Guan, J., B.D. Santer, P. Wang, Q. Fu, R.R. Garcia, Y. Li, K. Stone, D.E. Kinnison, J. Zhang, G. Chiodo, J.-F. Lamarque and S. Solomon (2026): The emergence of human influence on the ozone layer by the 1960s. PNAS, 123(28), e2608286123 (doi: 10.1073/pnas.2608286123) (https://www.pnas.org/doi/10.1073/pnas.2608286123)