It seems like over a long enough time scale, if the surface temperature of a planet varies between perennially below freezing and occasionally above freezing, that all of the water on the planet would eventually have evaporated or sublimated and been deposited in a polar/perennially frozen region where it becomes trapped as ice.

Why hasn't this happened on earth? Has plate tectonics moved ice out of polar regions fast enough to melt ice that would otherwise become trapped there? Is the time scale I'm imagining large compared to the age of the earth? Or is some other process responsible for keeping earth wet?

  • $\begingroup$ I don't think it's plate tectonics. A crude estimate based on precipitation in Antarctica of about 160mm a year suggests it gets 2000 Gt a year. If it holds 24 million gigatons of ice that's an accumulation of 12000 years worth of precipitation. A plate moving 2 cm a year would cover a 2000km radius in about 100 million years, 10000 times too slow to stop ice accumulating in Antarctica. $\endgroup$
    – Cirdec
    Commented Apr 8 at 7:09
  • 2
    $\begingroup$ i think you will find that under the pressure of 3 KM ice moves a bit faster than 2CM a year.if you want to add to your question you can do so by editing it. $\endgroup$ Commented Apr 8 at 18:50

1 Answer 1


Frozen glacial ice flows downhill; even if the catchment (where the snow falls) never rises above freezing the ice there still flows towards oceans, ie does not stay where the snow fell. The edges of Antarctica and of Arctic ice experience air temperatures above freezing every Summer and ocean water warms and melts glacier ends and sea ice.

This shows the basic cycle (image originally from NASA) -

glacier, ice sheet cycle


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.