# How does the loss of sea ice affect Earth's temperature?

• stuartmacg
stuartmacg
TL;DR Summary
Earth in approx. radiative equilibrium with space (sun disk+4k elsewhere) => radiation out=in (taking account of sun angle to surface). Cold poles + fluid heat xfr => net effect of radiative coupling at poles is cooling. Loss of sea ice increases coupling at poles => should reduce earth temperature, and increase air/sea currents?
I seem to remember (physics degree 50 years ago) a kirchof law for thermal equilibrium under radiative coupling: an isolated black body will take the mean temperature of its surroundings. I have not found it on google search, so I am not sure I got the situation exactly right. Anyway using that and the sun's subtense and its surface temperature and 4k for the rest, you can get plausible figures for the equilibrium temperature of a flat surface in earth orbit, at different angles to the line to the sun. The average temp over a sphere is roughly right as you would expect.

Without heat xfr (air and sea) from the equatorial regions to the poles (and winter/summer transfer of retained heat), the winter polar regions would be 4k ... The polar regions are held above their radiative equilibrium level by heat from the rest of the planet.

If the loss of sea ice increases the radiative coupling of the polar regions to the cold space they see, I would expect this to increase their heat loss.

I am frequently told the loss of polar sea ice would heat the earth further, by TV etc. This seems implausible. If it is indeed correct, can someone please explain it to me.

PS how much does the heat flow from the earth's core raise the temperature? - I expect very little, still it is interesting to reflect that earth's radiative coupling to space reduces its surface temperature (from something which would melt rock) - we are not so much heated by the sun as cooled by the rest of the sky·

I was hoping for clear physics, not canned concept quotation :-). The wikipedia article contains no mention of radiative heat loss, which is the dominant effect at the poles, and would be increased if less reflective. The article talks of reflecting incoming heat, which is not relevant.

weirdoguy and BillTre
stuartmacg said:
I was hoping for clear physics, not canned concept quotation :-). The wikipedia article contains no mention of radiative heat loss, which is the dominant effect at the poles, and would be increased if less reflective. The article talks of reflecting incoming heat, which is not relevant.
How do you then explain why it is colder after the winter solstice?

No idea what that question is about. I seek to understand why increased thermal coupling to the cold polar sky due to ice melt has the counter intuitive effect of increasing the earth's temp.

Ice is more reflective than water or ground. So if there is less ice, more solar radiation is absorbed warming the planet.

BillTre
stuartmacg said:
TL;DR Summary: Earth in approx. radiative equilibrium with space (sun disk+4k elsewhere) => radiation out=in (taking account of sun angle to surface). Cold poles + fluid heat xfr => net effect of radiative coupling at poles is cooling. Loss of sea ice increases coupling at poles => should reduce earth temperature, and increase air/sea currents?

I seem to remember (physics degree 50 years ago) a kirchof law for thermal equilibrium under radiative coupling: an isolated black body will take the mean temperature of its surroundings. I have not found it on google search, so I am not sure I got the situation exactly right. Anyway using that and the sun's subtense and its surface temperature and 4k for the rest, you can get plausible figures for the equilibrium temperature of a flat surface in earth orbit, at different angles to the line to the sun. The average temp over a sphere is roughly right as you would expect.

Without heat xfr (air and sea) from the equatorial regions to the poles (and winter/summer transfer of retained heat), the winter polar regions would be 4k ... The polar regions are held above their radiative equilibrium level by heat from the rest of the planet.

If the loss of sea ice increases the radiative coupling of the polar regions to the cold space they see, I would expect this to increase their heat loss.

I am frequently told the loss of polar sea ice would heat the earth further, by TV etc. This seems implausible. If it is indeed correct, can someone please explain it to me.

PS how much does the heat flow from the earth's core raise the temperature? - I expect very little, still it is interesting to reflect that earth's radiative coupling to space reduces its surface temperature (from something which would melt rock) - we are not so much heated by the sun as cooled by the rest of the sky·

Just a reminder that Global Warming / CC thread starts at PF must be accompanied by mainstream technical references. It's good that you are asking questions, but please do a bit of searching to find the relevant peer-reviewed journal articles that you can post links to in order to keep this discussion going. Thanks.

Do the words radiative equilibrium mean anything to you? The earth is in quite close radiative equilibrium with the space around it. It loses heat through radiation as fast as it gains it from radiation. The poles see a colder sky and would be much colder, but for heat carried by air and sea from elsewhere. They are held above equilibrium by this transfer. They radiate more heat to their sky than they get back. Surfaces which couple more tightly with radiation will increase this net loss of heat.

stuartmacg said:
Do the words radiative equilibrium mean anything to you? The earth is in quite close radiative equilibrium with the space around it. It loses heat through radiation as fast as it gains it from radiation. The poles see a colder sky and would be much colder, but for heat carried by air and sea from elsewhere. They are held above equilibrium by this transfer. They radiate more heat to their sky than they get back. Surfaces which couple more tightly with radiation will increase this net loss of heat.
Yes it does. If the earth absorbs more solar radiation through loss of albedo, it changes the balance.

Last edited:
BillTre
stuartmacg said:
Do the words radiative equilibrium mean anything to you?
To whom? Please always use the Reply/Quote feature when responding to a particular post.

Also, can you please find and post the links I asked for? Thank you.

AlexB23
Sorry: new to this - I was replying to Frabjous.

to berkman: I did not want to wade into the subject - just hopped someone who understood it could offer an explanation, or a simple reference.

to Frabjous: As far as I remember, the reflectance of isolated bodies in thermal equilibrium has no effect to the equilibrium temperatures - it just changes the energy flow in both directions equally. I expect you can get free heat energy from a heat pump otherwise.

You do not like my responses, so I will drop out and let you interact with others.

stuartmacg said:
to berkman: I did not want to wade into the subject
I think you are missing the point. It's not a matter of whether or not you want to dig into the subject, it's a matter of anything related to climate change being a potentially contentious subject and to make sure things stay within the bounds of PF rules, we need to be talking about specific mainstream technical references.

I see you're new here, so I understand that this IS new to you but it's they way things are done here.

BillTre and berkeman
From the GW/CC rules stickied at the top of the Earth forum:

Greg Bernhardt said:
Before posting anything, please review the Physics Forums Global Guidelines.

CC/GW threads in this forum are intended for discussion of the scientific content of well-researched models of weather, climatology, and global warming that have been published in peer-reviewed journals and well-established textbooks.

(click the little up-arrow to read the whole post).

berkeman said:
From the GW/CC rules stickied at the top of the Earth forum:
(click the little up-arrow to read the whole post).
I just expected an explanation from a physicist who was familiar with the subject. I have no knowledge of the subject or opinions on it. Just on basic physics the commonly stated ice melt consequences seemed surprising to me.

No one has answered my wee question, and feathers seem ruffled, odd.

Looks like I'm stuck with not understanding this. Bye folks.

Frabjous said:
been asleep (this post came at 1am UK time). Read it now - sensible intro stuff, but does not in any way deal with my wee question.

Bystander
stuartmacg said:
I am frequently told the loss of polar sea ice would heat the earth further, by TV etc. This seems implausible. If it is indeed correct, can someone please explain it to me.
stuartmacg said:
but does not in any way deal with my wee question.
You might not like it, but “albido” is the answer to your wee question. No one is going to more in depth if you will not accept the simple answer.

BillTre
stuartmacg said:
I just expected an explanation from a physicist who was familiar with the subject. I have no knowledge of the subject or opinions on it.

We don't have any climatologist scientists on our SA staff, unfortunately. That's one reason why we have to take a mostly hands-off approach to Moderation in GW/CC threads (and why we have the rules we do). It's always best if you can find some good references on your own, and post the links to those references to help the discussion along.

One refinement you might try in your Google searching is to confine the results to Google Scholar, which usually returns better search results and more peer-reviewed papers. Let us know what you find.

Tom.G and phinds
Frabjous said:
You might not like it, but “albido” is the answer to your wee question. No one is going to more in depth if you will not accept the simple answer.
albido is a word astronomers use to predict the brightness of celestial objects as a result of sunlight. How is that a description of the error in my simple physics process understanding? I make mistakes all the time, my mother said "a man who never make a mistake never made anything", and I have made lots of good "things" and a good living from that, and lots of mistakes.

I can imagine a process like: ice covers sea in winter and retreats in summer thus increasing radiative coupling in the hotter period and reducing average net heat loss at poles. I have no idea why permanent increased polar radiative coupling to the polar skies would reduce the net radiative loss from that region. Does no one else know?

thanks - that looks like it should cover the topic. No time to look right now. Was hopping for snappier response, but know truth never pure or simple...

berkeman
stuartmacg said:
thanks - that looks like it should cover the topic. No time to look right now. Was hopping for snappier response, but know truth never pure or simple...

We're trying to hop to it!

dlgoff and phinds
looked at last (wikipedia) link - just states it happens, no mention of decreased albedo increasing radiative heat loss. Either I am wrong and changing reflectivity can increase radiation in one direction relative to the other, or an important topic is missing from that article. Happy to be told that, if true, otherwise the net heat loss increase in polar areas (as above equilibrium temp due to heat inflow) might be expected to be the dominant consequence of reduced albedo. Read the others soon, despite the hour.

Quote should have been seldom pure and never simple. Thanks for taking the trouble to look stuff up.

phinds
May be a bit too basic but, from a latent-heat point of view, the loss of the north polar ice cap is loss of the ice cubes in the drink on a hot summer's day.

Which might not sound important but, for reference, the amount of heat required to melt ice is the same as the amount of heat required to heat water up by 80degC.

stuartmacg said:
Without heat xfr (air and sea) from the equatorial regions to the poles (and winter/summer transfer of retained heat), the winter polar regions would be 4k

The Sun's colour temperature is actually closer to 5.9k Kelvin.

dextercioby
Winter polar regions do not see the sun, hence temp in simple radiative equilibrium would be space background radiation (seem to remember 4k, could be 3).

State transition heat flows are not relevant to equilibrium points.

berkeman said:
Skimmed all the articles. None touch on the effect of albedo change on the (dominant) radiative heat loss in the polar regions. The 5 facts article does mention ice cover keeping the sea warm, but not the corollary that no ice cover allows radiative heat loss from the sea.

Bystander
Well, lets try the lengthy version.

As a beginning, recall that radiation wavelength is inversely proportional to the source's temperature. The higher the temperature, the higher the energy and the shorter the wavelength.

You may be aware of how a Greenhouse for growing plants works. The incoming solar radiation has a hefty dose of Near Infrared (NIR), wavelength just beyond the Red, long wavelength, of visible light. Glass has moderate transmission at this color.

The incoming NIR hits the contents of the greenhouse warming it. The contents then radiate in the Far Infrared (FIR) (lower temperature therefore longer wavelength). The glass however is rather opaque to the longer IR radiation, trapping it in the Greenhouse and raising the interior temperature above ambient.

Our atmosphere does the same thing for the Greenhouse we call Earth. It lets in much of the NIR from the Sun but has reduced transmission for the longer wavelength, lower temperature, IR radiated by the Earth's surface.

Then we Humans came along and started burning enough stuff and putting various chemicals into the atmosphere that we reduced its transmittance of the FIR. This eventually caused the average Earth temperature to rise a little bit.

The rising temperature also raised the surface ocean temperature. The ocean currents carried some of this added heat to the polar regions, causing a reduction in ice coverage.

Here is a 2.5 minute NASA video about that:

Without actually so stating, the video implies that the polar regions were in approximate equilibrium with the Solar insolation and radiative losses to space; which makes sense if you think about it a little.

As the video pointed out, the NIR absorptivity of Sea Water is higher than that of Ice or Snow. Consequently, the Arctic waters are being warmed, and the atmosphere is still selectively blocking the outgoing FIR, further warming the Arctic regions.

-- and more trouble arises. All this warming is also thawing the Permafrost. The Permafrost has sequestered alot of organic material from dead plants. As the plant material is exposed, it decomposes. A major decomposition product is Methane, a powerful Greenhouse gas.

So overall we used to be in an equilibrium condition, now we are in a positive feedback loop until we reach a new equilibrium.

Now some people may argue that the sky area exposed to 4°K space is a great heat sink, which it is. That is about 270° below our temperature. The Sun is around 5800° above our temperature, it seems to more than make up the difference!

Hope this helps!

Cheers,
Tom

Last edited:
dextercioby and pinball1970
I will try one final time:
Last post

1) The earth's surface is cooled to close to the visible sky's mean temperature. Without radiative cooling the surface would melt rock due to core leakage.
2) The poles see a colder mean sky and are colder. This results in sea/air conveyed heat inflow from the rest of the earth. This implies that the poles are above the radiative equilibrium temperature with the sky they see. This implies that there is a net heat loss through radiation at the poles.
3) None of the above is invalidated by any thermal blanket effects in the atmosphere.
4) Increasing radiative coupling (reducing albedo) increases the net radiative energy flow in both directions equally.
5) The net flow at the poles is outward (2), so (4) strongly suggests permanent loss of polar ice would reduce the earth's surface temperature.

If this is false I would love to know why.
Perhaps someone knows of a (peer reviewed) paper analysing this properly. The links so far have been irrelevant.

Bystander
Oh, you meant Kelvin, not kilo. That was a bit confusing.

stuartmacg said:
If the loss of sea ice increases the radiative coupling of the polar regions to the cold space they see, I would expect this to increase their heat loss.

I am frequently told the loss of polar sea ice would heat the earth further, by TV etc. This seems implausible. If it is indeed correct, can someone please explain it to me.

You want verification that what you remember a TV show(s) saying is incorrect, due to radiative transfer. Or, perhaps incorrect solely in regards radiative transfer ?

Regarding the latter : if you don't mind your reductios being far, far, far into the absurdum, you can compare them.

Imagine a flat object in solar orbit : one side is a black body surface radiating at saltwater's freeze/melt point, the other totally reflective.

Using Stefan-Boltzmann Law, inverse-square law, and a bit of trig, net power in/out can be calculated for whatever angles you want to position it at.

Personally, I find "When the icecap melts it stops cooling the Earth" to be more accurate.

Last edited:
Tom.G and BillTre
The paper https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2011JD015804

### "Estimating the global radiative impact of the sea ice–albedo feedback in the Arctic"​

is the nearest I have come to something narrowly on this topic, from a google search. Only skimmed so far.
It proceeds much as one might expect to estimate the changes in solar energy taken in, and relate the estimates to experimental data.
There is no mention of the corresponding changes in radiative heat loss, which in polar regions one would expect to exceed the change in heat input.

So imagine we have a totally white planet in space and a totally black planet in space. They happen to be in identical orbitals around the sun, just always on opposite sides of the sun (they are my planets, so yes they CAN do that.)
The totally white planet reflects all the incoming solar radiation, and does not warm up.
The totally black planet absorbs a lot of the incoming radiation, warms up and radiates energy, until the incoming energy matches the radiated energy and then it is in equilibrium.
So which planet is warmer? The black planet.
Which planet is in equilibrium? Surely both planets.
So if the totally white planet then gets some black surface exposed, by scraping off a large amount of the white paint, will it get warmer or cooler?
Now apply the same to an earth that lost a lot of its white, reflective ice. Will it get warmer?

dextercioby, berkeman and BillTre
You presuppose a cold radiatively isolated body. If there is no energy exchange the word equilibrium is meaningless. The white object will stay at whatever temperature it was, hot or cold. Once coupled to the space around it it will gain or loose energy to approach equilibrium.

[Post edited by the Mentors to remove a mild insult]

Last edited by a moderator:
weirdoguy and BillTre
DrJohn said:
So imagine we have a totally white planet in space and a totally black planet in space. They happen to be in identical orbitals around the sun, just always on opposite sides of the sun (they are my planets, so yes they CAN do that.)
The totally white planet reflects all the incoming solar radiation, and does not warm up.
The totally black planet absorbs a lot of the incoming radiation, warms up and radiates energy, until the incoming energy matches the radiated energy and then it is in equilibrium.
So which planet is warmer? The black planet.
Which planet is in equilibrium? Surely both planets.
So if the totally white planet then gets some black surface exposed, by scraping off a large amount of the white paint, will it get warmer or cooler?
Now apply the same to an earth that lost a lot of its white, reflective ice. Will it get warmer?
Sorry that last comment was a bit rude. I have just about lost hope of getting any rational on-topic discussion here, and am feeling a bit grumpy.
I might as well amplify my response to you a little:

We are assuming:
1) the insulated white body is in internal equilibrium (unlike earth).
2) the hole created in the insulation to allow radiative coupling sees the average local space temperature e.g. it is not tide locked to the sun or on a spin pole orthogonal to the ecliptic.

Then thermal connection will result in loss of heat if the white body was previously above local space temperature, or gain of heat if it was below.

In the earth polar regions
1) There is an inflow of heat from the rest of the planet to the poles, via water and air, because they are colder.
2) The regions are thus hotter than local thermal equilibrium with the polar skies.
3) Increasing thermal coupling to the polar skies will therefore loose heat from the earth on balance.

stuartmacg said:
4) Increasing radiative coupling (reducing albedo) increases the net radiative energy flow in both directions equally.
What physically has to happen for the outflow to increase?

BillTre

### Similar threads

• Earth Sciences
Replies
7
Views
2K
• Earth Sciences
Replies
28
Views
4K
• Earth Sciences
Replies
36
Views
4K
• Earth Sciences
Replies
20
Views
6K
• Thermodynamics
Replies
2
Views
1K
• General Engineering
Replies
16
Views
2K
• Thermodynamics
Replies
2
Views
1K
• Earth Sciences
Replies
53
Views
12K
• Thermodynamics
Replies
101
Views
4K
• Astronomy and Astrophysics
Replies
1
Views
963