sea ice · Antarctica
Antarctic (circumpolar)
February minimum sea-ice extent
Modelled demonstration series · not observed data
For thirty years this was the chart sceptics pointed at, and fairly so: it was not doing what the Arctic was doing. Then it stepped off a cliff. The honest reading is that Antarctic sea ice is poorly understood, changed regime recently, and is now producing values far outside anything in its own record.
75 years, modelled
Same reading, two normals
| Normal | Mean | σ | Pct | Return |
|---|---|---|---|---|
| 1951–1980 | 3.10 million km² | −4.40σ | <0.001th | 1 in 180,000 years |
| 1991–2020 | 2.85 million km² | −4.03σ | 0.003th | 1 in 36,000 years |
Today would be the lowest value in the entire 75-year modelled series. Past the series' standing low of 2.28 million km² (2014). Nothing about the reading changes between rows — only the yardstick.
The season here
Antarctic sea ice behaved very differently from Arctic ice for decades — flat or slightly growing — and then dropped abruptly.
Where this series comes from
MODELLED SERIES — not observed. Parameters set from published Antarctic February minimum extent climatology, including its long flat period and the abrupt recent decline. The 75 annual values are synthesised by the seeded generator.
Every number on this page comes from a modelled series: 75 annual values synthesised by a seeded generator from published climatological parameters for Antarctic (circumpolar). It shows how anomaly reasoning works. It is not a measurement and must not be cited as one.
How to read σ
σ counts standard deviations from this place's own normal. The sigma guide · return periods.
More sea ice stations
Open Antarctic (circumpolar) on the field →