Last lesson ended in 1975 with the chemical industry's objection: there was no hard data. Molina and Rowland had already named what would sink them. If the total amount of ozone overhead did not fall as CFCs built up, they were wrong. So somebody had to measure it.
Remember the three-millimetre layer from last lesson. Scientists measure it in Dobson units. One Dobson unit is a hundredth of a millimetre, so a three-millimetre layer is 300 Dobson units, and a reading of 150 means half as much ozone overhead.
Two very different instruments were measuring it over Antarctica.
| On the ground | In orbit | |
|---|---|---|
| Instrument | A Dobson spectrophotometer, designed by the Oxford physicist Gordon Dobson in the 1920s | The Total Ozone Mapping Spectrometer, on NASA's Nimbus-7 satellite |
| Where | Halley, a British research station on an ice shelf in Antarctica | In orbit, looking down |
| Measuring since | 1956 | October 1978 |
| How it works | Points at the sun and compares two kinds of ultraviolet light: one that ozone absorbs strongly, one it barely touches | Measures the ultraviolet sunlight scattered back up from the atmosphere |
| What it covers | The sky above one spot | The whole globe |
| What comes out | A dial reading, turned into an ozone figure using numbers worked out in advance for that particular instrument | Measurements, turned into ozone figures by a computer program |
Neither instrument shows you ozone. Both give you a number that somebody, or something, has had to work out.

Jon Shanklin in outdoor clothing, using Dobson spectrophotometer outside, Halley, 22 Jan 1982. Source
<aside> 🔗
If you want to look further:
✍️ Write here, on your own, in one sentence: if these two instruments disagreed about the ozone over Antarctica, which would you trust, and why?
You work on the NASA team that turns the satellite's measurements into ozone figures. This is what your team knows.
| What you know | |
|---|---|
| The program | It turns each measurement into an ozone figure by comparing it with a set of standard ozone profiles, built from years of earlier measurements by balloons, ground instruments and rockets. The profiles run from 200 to 650 Dobson units. A measurement far outside that range cannot be turned into a reliable figure. |
| The rule | Any reading below 180 Dobson units is flagged. A flagged reading is not deleted. It is held back and passed to people on the team to check. |
| The satellite | It was designed to last one year. Within that year, contamination built up on the optical surfaces of its ultraviolet instruments, and a panel concluded they could no longer be relied on to detect small, slow changes in ozone. |
| The science | The best models say CFCs will cut total ozone across the globe by about 1 per cent a decade. Ozone naturally rises and falls by far more than that, so nobody expects to see the change in the data for ten years or more. |
| The date | Computers are slow and processing runs about nine months behind. It is June 1984, and you are only now working through October 1983. |
✍️ Write here, on your own: is the 180 rule a sensible rule? Give one good reason for it and one risk it carries.
The program works through the October 1983 measurements over Antarctica. A large number of them come back flagged.
✍️ Write here, on your own: list every explanation you can think of for the flagged readings, most likely first. Then say what you would check.