Before the 1950s, direct observations of the composition of the atmosphere were extremely limited. Air trapped in pockets of snow and ice has allowed the observational record for some gases to be extended back hundreds of years. But certain gases that affect climate and air quality, such as ozone, are not stable in ice or snow, limiting their records to the past few decades.

The impact of the atmosphere’s changing composition depends on both the present-day concentrations of pollutants and their concentrations before the change started. Many atmospheric changes (climate change being the most notable example) are associated with the Industrial Revolution, and so, to understand the magnitude of those effects, we need to understand the composition of the atmosphere over the past 100–200 years.

In this article by Mathew Evans from the Wolfson Atmospheric Chemistry Laboratories, Department of Chemistry, University of York, York UK explains how Measurements of atmospheric ozone levels taken during the nineteenth century cast doubt on the computational models used today to simulate the atmosphere. He further elaborates on an independent proxy of past ozone levels offers reassurance.