Why is the Arctic melting, but the Antarctic doing great? That's what the latest IPCC report says (p. 2-192 of the Physical Science Basis provides a time series), but it says nothing, unless I missed it, on the causes of such a discrepancy. It's same ice, isn't it? Why is Antarctic ice more heatproof?

enter image description here

EDIT: I've found the answer in the IPCC report (AR6 WGI), and it's even fairly intelligible (unlike much of the rest of the report). Start reading from page 7-80, if you're interested

  • 3
    $\begingroup$ It may matter that the Arctic is in direct thermal contact with the ocean, while the Antarctic isn't in the middle, only at the edges. $\endgroup$
    – pjc50
    Commented Oct 13, 2021 at 12:34
  • 1
    $\begingroup$ There is a downward trend in Antarctic temperatures, it still counts as climate change. The land is not going to melt as easily as the ice, either. $\endgroup$
    – mckenzm
    Commented Oct 13, 2021 at 22:29
  • 2
    $\begingroup$ The question is based on a false premise; that there exists a "discrepancy". This arises from the mistaken assumption that sea-ice area at the North and South poles should be directly correlated - ignoring obvious differences of geography and local climate effects. $\endgroup$ Commented Oct 14, 2021 at 8:56
  • 2
    $\begingroup$ If there's a significant difference, might it be because the Arctic is solely frozen water while the Antarctic is largely snow-and-ice over land? $\endgroup$ Commented Oct 14, 2021 at 21:35
  • $\begingroup$ IS the Antarctic doing great? Question title makes no distinction between sea ice and ice sheets - or more inclusively, to ice mass balance - but the supporting graphs make clear it is about sea ice alone. There is more to polar ice than sea ice - whether as extent (area) or quantity (mass) - which is highly variable from year to year and doesn't correlate closely to the ice loss that impacts sea level rise - ie that from mass imbalance of glacial ice sheets $\endgroup$
    – Ken Fabian
    Commented Oct 15, 2021 at 23:20

5 Answers 5


Not quite. The report actually shows an increase in extent, but not the volume, of sea-ice around Antarctica. That sounds good, but the data appendix in the report shows the Antarctic continental ice-sheet is shedding ice so fast during the summer months, and into the Autumn, that the ice can't melt or migrate away during the season. When the winter freeze comes the ice already in the water is making the ocean surface cooler and nucleating sea-ice growth farther off shore than it should be. We're seeing a larger but thinner sheet of ice around the Antarctic which can seem unaffected by climate change, but its ice is actually melting at a rate comparable to the Arctic.

  • 4
    $\begingroup$ Can you cite the source of this information as it seems to contradict another answer that says an older IPCC report said the ice was projected to thicken? $\endgroup$
    – CramerTV
    Commented Oct 13, 2021 at 17:48
  • 18
    $\begingroup$ Important context here is the vast differences among sea ice, ice shelves, and the ice caps. The Arctic is an ocean (mostly) covered with an extent and volume of sea ice that is declining over the decades; the Antarctic is an ice-covered continent with transient sea ice. It's much more important to the Arctic ecosystem and other systems how much of the Arctic Ocean is covered with ice. $\endgroup$ Commented Oct 13, 2021 at 19:59
  • 1
    $\begingroup$ "the data shows the Antarctic continental ice-sheet is shedding ice so fast during the summer months, and into the Autumn, that the ice can't melt or migrate away during the season. When the winter freeze comes the ice already in the water is making the ocean surface cooler and nucleating sea-ice growth farther off shore than it should be." This appears to be pure conjecture, do you have credible citations to back it up? Or an explanation of how the data provided shows chunks of ice breaking off and lingering near the main ice sheet for x months $\endgroup$
    – TCooper
    Commented Oct 14, 2021 at 16:54
  • $\begingroup$ @TCooper Am I talking about the data in the actual IPCC report in question, if you're going to pick and choose what you read from a report that complex skip the words and look at the data appendix first it tells the real story without the political spin. $\endgroup$
    – Ash
    Commented Oct 15, 2021 at 8:28

The Antarctic is not "doing great":

Antarctica moved from a weakly negative mass balance in the 1990s to a faster rate of mass loss at a rate of between -45 and -120 gigatonnes per year.

  • 12
    $\begingroup$ please expand your answer a bit,if this link dies you are left with The Antarctic is not "doing great": and that is not much of an answer to an important question. $\endgroup$ Commented Oct 13, 2021 at 17:34
  • 1
    $\begingroup$ @trondhansen I wrote the answer before the question focused on area coverage. The general notion that "the antarctic is doing great" stays wrong though. $\endgroup$ Commented Oct 13, 2021 at 20:12
  • 1
    $\begingroup$ @trondhansen the link text is still part of the answer that won't be affected by link rot though, it'll just lose its source. $\endgroup$
    – justhalf
    Commented Oct 15, 2021 at 8:52

The IPCC says (PDF page 4, report page 656 of an older report):

Increased melting of Arctic glaciers and the Greenland ice sheet, but thickening of the Antarctic ice sheet due to increased precipitation, were projected.

  1. Climate change is leading to increased precipitation. (Short version: Warm air picks up more humidity.) In places where it is cold enough, that precipitation is snow/ice.
  2. The Antarctic is significantly colder than the Arctic--because land can get colder than water in the first place and also because it has higher elevations. In this case, the Antarctic is cold enough that it doesn't melt so much, whereas the Arctic is warmer and melts more.
  3. Once it gets warm and starts melting, melting (and rain) makes local temperatures warmer and leads to more melting, but snow makes local temperatures colder, i.e. less melting (short version: liquid surface water makes ice which is dark-colored and radiates sunlight as heat, snow is light-colored and radiates sunlight as light).

This doesn't mean that the Antarctic is safe forever, if the global temperatures increase enough then even a locally colder temperature won't be cold enough to prevent melting...and once melting starts, it can create a feedback loop which increases the speed of melting.

You can see this kind of local variation within Antarctica as well. East Antarctica is higher (and colder, for reasons above) than West Antarctica --and the eastern side has been (slightly) gaining ice while the western side has been losing ice.

  • $\begingroup$ Uncertainties about Antarctic ice mass projections are real enough but Grace satellite data reveals significant and accelerating losses of ice over the period since AR4 (2007), with some ice sheets steady or even gaining but others showing very significant ice losses that exceed them. Between 2002-2020 losses of 150Gt pa - grace.jpl.nasa.gov/resources/31/antarctic-ice-loss-2002-2020 $\endgroup$
    – Ken Fabian
    Commented Oct 15, 2021 at 23:28
  • $\begingroup$ @KenFabian Added in a bit about that at the end. It fits in pretty well, though--the higher/colder areas are cold enough that they are benefiting from the increased precipitation, but the lower/warmer areas have warmed to the point where they are losing faster than gaining. $\endgroup$ Commented Oct 16, 2021 at 0:43

The Antarctic has a feature which the Arctic doesn't - It has the Southern Ocean. This ocean sweeps around Antarctica, to keep the hot air out and the cold air in. Although global warming increases global ocean temperature, Antarctica is less affected due to its 'shield', the Sourthern Ocean.

How it works is that the ocean moves rapidly in a circle around Antarctica, around 100km off from the land, using the wind as a 'barrier'. You can see that many hot air currents 'bounce' off.

You can see from here that the wind stops the hot air effectively.

  • $\begingroup$ That may be true, but what is the mechanism? How does the ocean keep hot air out? Can you elaborate in your answer? E.g., why would the ocean not very efficiently transfer heat by ocean currents? Please respond by editing (changing) your answer, not here in comments (without "Edit:", "Update:", or similar - the answer should appear as if it was written right now). $\endgroup$ Commented Oct 14, 2021 at 21:31
  • $\begingroup$ Think your map suggests "warm ocean currents", not air currents (though there is plenty of connection)? $\endgroup$ Commented Oct 15, 2021 at 16:40
  • $\begingroup$ This is the exact answer. Once Antarctica broke free of Australia, warm currents could not heat Antarctica the way they do say, England, which is further north than Labrador. Absent a land bridge, cold salt water circumnavigating Antarctica turn away the less dense equatorial warm currents away. Little sun because its on the poles. Warm tropical water would help, but it can't get by the gyre around Antarctica. Ice forms, albedo rises, and positive feedback does the rest. The Arctic, not having said current , or land, is just ice. Melt some and heat the water below, and ...+ve feedback again $\endgroup$
    – chiggsy
    Commented Oct 15, 2021 at 18:54

In the absence of a worldwide warming trend, there is a large scale oscillation, the thermal bipolar see-saw. When the Arctic cools, the Antarctic warms and vice versa. With a worldwide warming trend, even if both are warming, the oscillation will still cause one to race ahead of the other, then switch places.

See the article Expression of the bipolar see-saw in Antarctic climate records during the last deglaciation.

Other differences besides those mentioned in other answers:

  • 2
    $\begingroup$ As described in the paper you linked, those see-saw variations occur on the scale of millenia, not decades. We also know that there's no unaccounted-for volcanism affecting the Antarctic: climate.nasa.gov/blog/2982/… $\endgroup$ Commented Oct 13, 2021 at 17:43
  • $\begingroup$ It is not that we don't know about the volcanoes, just that we can't be sure how much heat they are transmitting to the ice. See theguardian.com/world/2017/aug/12/… $\endgroup$ Commented Oct 13, 2021 at 19:55
  • 1
    $\begingroup$ Another paper describes polar see-saw on multi-decadal timescales. nature.com/articles/srep08909 $\endgroup$ Commented Oct 13, 2021 at 19:58
  • $\begingroup$ The charts you added to your question are all about sea ice, though. $\endgroup$ Commented Oct 13, 2021 at 20:37
  • $\begingroup$ What is the order of the frequency of the oscillation? Weeks? Months? Years? Hundreds of years? $\endgroup$ Commented Oct 14, 2021 at 21:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.