Seminar

Abstract

Tropospheric ozone is a greenhouse gas, a key component of atmospheric chemistry, and is detrimental to human health and plant productivity. Hence a knowledge of historical tropospheric ozone levels is desirable.

As a reactive gas ozone does not, or is highly unlikely to, persist in bubbles in ice cores or in cylinders. Hence no current measurements of past concentrations are possible, and the historic record can only be constructed from historic measurements which have substantial uncertainties due to both bias and imprecision. The historic surface ozone record 1896-1975 has been constructed from measurements that pass the four criteria: (a) instrumentation whose ozone response can be traced to modern tropospheric ozone measurement standards, (b) samples taken when there is low probability of chemical interference (c) sampling locations, heights and times when atmospheric mixing minimises vertical gradients of ozone in the planetary boundary layer above and around the measurement location and (d) the laugh test.

Two of the commonly used early data sets do not pass all four criteria. These are the early measurements with the Schönbein filter paper technique and the potassium iodide-arsenite technique used at Montsouris for 1876-1910 The use of these data sets for trend analyses is not recommended.

In total, 58 acceptable sets of measurements are currently identified, commencing in Europe in 1896, Greenland in 1932 and globally by the late 1950's. Between 1896 and 1944 there were 21 studies (median duration 5 days) with a median mole fraction of 23 nmol mol-1 (range of study averages 15-62 nmol mol-1). Between 1950 and 1975 there were 37 studies (median duration approx. 21 months) with a median mole fraction of 22 nmol mol-1 (range of study averages 13-49 nmol mol-1), all measured under conditions likely to give ozone mole fractions similar to those in the planetary boundary layer.

These time series are matched with modern measurements from the Tropospheric Ozone Assessment Report (TOAR) Ozone Database and used to examine changes between the historic and modern observations. These historic ozone levels are higher than previously accepted for surface ozone in the late 19th early 20th Century. This historic surface ozone analysis provides a new test for historical reconstructions by Climate-Chemistry models.

---

Dr. Ian Galbally serves as Chief Research Scientist at the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Australia. During his career, he has been a pioneer in measurements of atmospheric composition and chemistry. He is a Fellow of the Australian Academy of Technological Science and Engineering and the Royal Australian Chemical Institute. He received his Ph.D. from Monash University.