After the Arctic has Melted

[TerryM]

A few posters have optimistically expressed indifference to the effects of an ice free Arctic. Without questioning the timing or causes of an ice free Arctic summer, it might prove interesting to explore the likely consequences.

An ice free summer predicates an albedo change from a value of >80% for snow covered ice to a value of <10% for open water - these are very conservative values from wiki. This indicates that the Arctic ocean will absorb 450% more solar radiation when ice free than when ice covered.

An ice free Arctic also predicates that latent heat of fusion will cease, and that the energy will instead become sensible heat. The energy required to melt a cubic meter of ice will raise a cubic meter of melt water to 80C (or 144F). Without considering insolation curves, simply by measuring the volume of ice melted after a certain date we can arrive at a reasonably accurate figure for the additional heat that the globe will be subjected to by this process.

Weather patterns and ocean currents will be effected, and speculations about these are what I see this thread addressing.

Is it reasonable to assume that things will only incrementally change from what we've come to see as the norm, or are conditions experienced in the PETM - often referred to as " The Great Die Off" more likely.

Will it take a considerable time for the ramifications to play out, or will things cascade out of control within a decadal time-frame.

When was the most recent period that experienced a seasonally ice free Arctic, and did it lead to a perennially ice free Arctic.

Is complex society possible without the year round moderating influence of Arctic ice.

Some still see the likelihood of the ice lasting until 2100 - or beyond, while some don't believe it can last out the decade, but what do people think will happen when it is gone?

Terry

[tacoman25]

Thank you for starting this thread, Terry.

In response to the albedo change of ice to open water, I'm not sure this has much effect beyond the short term, since that heat quickly radiates out into space as the water cools rapidly in early fall. So I do see how we can probably expect warm falls in general to continue (though months like October 2009 are still quite possible), but I'm not sure I see longer term climate consequences from this albedo change.

[TerryM]

You're welcome

I found myself getting a little snarky on the other thread and thought this might be a better place to try and get a handle on what to expect.

The water will cool in the fall, but how much is radiated into space will as I understand it be dependent on cloud cover, inversion layers and GG bouncing the heat back down. I think that AIRS measurements can be used to determine the outgoing radiation at differing frequencies, and some form of extrapolation from already heated surface waters (Beaufort Sea near the coast) might give us some figures to work with later this fall.

Terry

[tacoman25]

An ice free Arctic also predicates that latent heat of fusion will cease, and that the energy will instead become sensible heat. The energy required to melt a cubic meter of ice will raise a cubic meter of melt water to 80C (or 144F). Without considering insolation curves, simply by measuring the volume of ice melted after a certain date we can arrive at a reasonably accurate figure for the additional heat that the globe will be subjected to by this process.

As far as this goes, I wonder if it might be worth considering that 75% of the Earth is already covered with ocean water, and the Southern Hemisphere, which has a much greater percentage of water, has seen considerably less warming than the NH. Land masses have been seeing more warming than the oceans. The Arctic Ocean is relatively small, and while it may take in a lot of energy for a brief period in summer, it has also been proposed that open water will allow stronger storms and more cloudiness, which would then reduce the amount of sunlight/energy received by the water. Additionally, a warmer Arctic means more snowfall further north as well, so wouldn't that result in more energy going into melting snow every spring/early summer?

All in all, I'm just not sure how much more heat would be added to the atmosphere with open water in the Arctic a couple months of the year. I don't know if it would be significant. I'm not sure it would be much different than what we are seeing now, with an Arctic where the vast majority of ice is first year ice anyway.

How much latent heat in fusion do you calculate would be released due to less energy being used to melt ice?

[TerryM]

I'm not sure, and unfortunately have a dinner appointment coming up, The figure is 144 BTU per pound of ice so converting a cubic meter of ice into pounds we'd get about 2200# X 144 = 316,800 BTU/ cubic meter.

Probably easier to do in Joules, but most here are American.

Terry

[Vandy]

Joules should be fine. This is a science forum afterall.

My biggest interest/concern is that a general increase in Arctic temperatures could trigger the release of the stored permafrost methane that is hypothesized to occur. Also, while a human-centric viewpoint might not consider this to be a problem, I think the loss of a specific habitat is something to be concerned about. The Arctic may not be the treasure trove of life that rainforests are, but the loss of any species due to anthropogenic progress is something to mourn.

[salbers]

Thank you for starting this thread, Terry.

In response to the albedo change of ice to open water, I'm not sure this has much effect beyond the short term, since that heat quickly radiates out into space as the water cools rapidly in early fall. So I do see how we can probably expect warm falls in general to continue (though months like October 2009 are still quite possible), but I'm not sure I see longer term climate consequences from this albedo change.

Yes, the warmest Arctic anomalies are in the fall, so this would be the time of biggest effect. However I think Arctic ice maximum values are also going down, so effects might be expected year round?

Note that albedo changes can be a big deal for overall climate, think of the subtle radiation changes with the Milankovitch cycles related to how much solar radiation is being reflected of the Antarctic as precession moves things around.

[Msalgado]

Yes, the warmest Arctic anomalies are in the fall, so this would be the time of biggest effect. However I think Arctic ice maximum values are also going down, so effects might be expected year round?

Note that albedo changes can be a big deal for overall climate, think of the subtle radiation changes with the Milankovitch cycles related to how much solar radiation is being reflected of the Antarctic as precession moves things around.

While this is true, it is interesting to note that the largest climate signals occur at the 100k frequency of the Milkanovitch Cycle due to eccentricity. HOWEVER, the actual forcing at 60N is less than the shorter term orbital variations from obliquity and precession So the largest effect is felt at the smallest insolation change.

Of course its impossible to draw any conclusions from this alone, but it does point to other factors at work beyond simply insolation. Then there's also the fact that the sea ice is farther north than simply 60N.

[Msalgado]

The heat budget of the upper Arctic Ocean is examined in an ensemble of coupled climate models under idealised increasing CO₂ scenarios. All of the experiments show a strong amplification of surface air temperatures but a smaller increase in sea surface temperature than the rest of the world as heat is lost to the atmosphere as the sea-ice cover is reduced. We carry out a heat budget analysis of the Arctic Ocean in an ensemble of model runs to understand the changes that occur as the Arctic becomes ice free in summer. We find that as sea-ice retreats heat is lost from the ocean surface to the atmosphere contributing to the amplification of Arctic surface temperatures. Furthermore, heat is mixed upwards into the mixed layer as a result of increased upper ocean mixing and there is increased advection of heat into the Arctic as the ice edge retreats. Heat lost from the upper Arctic Ocean to the atmosphere is therefore replenished by mixing of warmer water from below and by increased advection of warm water from lower latitudes. The ocean is therefore able to contribute more to Arctic amplification.

http://rd.springer.c...0382-012-1454-5

Thanks for the study link. I'm not sure if its allowed to post PDFs but if so I can post it so others without access can read it as well.

I found their conclusions very interesting. They are definitely supportive of the idea that an increased exposed surface area will help release the heat efficiently. They end with a token remark about possible changes to NH atmospheric circulations but they don't provide any evidence of that as it was completely beyond the scope of their modeling.

[donsutherland1]

When was the most recent period that experienced a seasonally ice free Arctic, and did it lead to a perennially ice free Arctic.

Perhaps the closest relatively recent period that experienced near seasonally ice-free Arctic condtions was the Holocene Climate Optimum. Then, summer sea ice extent was less than half of the 2007 minimum. The situation did not lead to a permanent disappearance of summer ice.

However, several caveats are in order:

1. The past event was all natural.

2. This time around the anthropogenic forcing that is having the leading impact. That impact is continuing to grow as atmospheric CO2 is continuing to rise, even as summer ice minima are declining and fairly rapidly. Hence, were summer sea ice minima to reach that earlier level in the years ahead, the headwinds that drove that trend could remain unabated, possibly reaching the "tipping point" that some have suggested exists at a certain point, especially if Arctic temperatures reach levels beyond what occurred during the Holocene Climate Optimum.

3. Should the current rate of decline (1979-2012) be faster than what had occurred in the past and should the periodic abrupt "step-downs" continue to occur, there would possibly be greater risk of seeing things go beyond the ice situation that characterized the Holocene Climate Optimum.

In short, the answer to your question is uncertain. The Holocene Climate Optimum would lend some hope of averting such an outcome. At the same time, the current period of decline is much different than the previous one, in that it is being driven increasingly by growing anthropogenic forcing.

[Jonger]

Record low ice currently and long range models already have that surface below average 7 days out. The mid summer surface temps will probably be above normal, but the earth isnt going to become uninhabitable as you seem to think.

[TerryM]

The figure we have to deal with as far as latent heat is the additional amount of ice now being melted out on a year to year basis. I'm recalling this to be in the neighborhood of 1,000 cubic km/yr - does this seem a reasonable figure?

Terry

[TerryM]

Don

I've followed some of the driftwood studies on Ellesmere Island, 79 Glacier & the Flade Isblink area directly south of Peary Land but hadn't come across the 50% of 2007 figure. I wonder if this was a volume measurement or area/extent - and whether it led to seasonal melt out. I'd assume that if the logs were floated over on ice - and the ice melted out - the transportation wouldn't have been possible.

IIRC on Ellesmere at least, the studies found larch thought to have originated in the Tiksi region.

Terry

[donsutherland1]

Don

I've followed some of the driftwood studies on Ellesmere Island, 79 Glacier & the Flade Isblink area directly south of Peary Land but hadn't come across the 50% of 2007 figure. I wonder if this was a volume measurement or area/extent - and whether it led to seasonal melt out. I'd assume that if the logs were floated over on ice - and the ice melted out - the transportation wouldn't have been possible.

IIRC on Ellesmere at least, the studies found larch thought to have originated in the Tiksi region.

Terry

While searching for information relevant to this thread, I came across that piece. I will be looking for the corresponding study. Hopefully, it's available.

[Vergent]

The figure we have to deal with as far as latent heat is the additional amount of ice now being melted out on a year to year basis. I'm recalling this to be in the neighborhood of 1,000 cubic km/yr - does this seem a reasonable figure?

Terry

Not as simple as it sounds.If we look at the 5 year period 2006 - 2011, we have lost 4,500 km^3 from the minimums for -900 km^3/year. Melt pond volume confuses the issue though. While melt pond volume per unit area has increased, the area has decreased. Since we can't yet measure melt pond volume, we can only guess the net.

Further, the maximums have declined less than the minimums, The amplitude has increased by about 2,000 km^3.

[TerryM]

Verg

Seke Rob's new chart may provided some needed (at least by me) hard figures.

http://i137.photobucket.com/albums/q210/Sekerob/Climate/ArcticSIA-PIOMASv20Max_Min.png

Terry

[Vergent]

Verg

Seke Rob's new chart may provided some needed (at least by me) hard figures.

http://i137.photobuc...Sv20Max_Min.png

Terry

Terry,

You are right I was taking too short a trend.

Verg

Edit p.s.

Brainwashed by negative feedbacks.

[tacoman25]

The greatest warming in the Arctic has occurred from the fall into the winter since the Arctic sea ice minimum in 2007.

Fall

Winter

Spring

Summer

Just as an example of some of the extremes, October 2007 and February 2012 saw +14c and +15c anomalies in Arctic regions.

I would have to think that the anomalous high latititude blocking (-AO/-NAO) played a role in those anomalies. They certainly resulted in cooler anomalies for much of the U.S, Europe,. and Asia.

[PhillipS]

There is a discussion on this topic over at RealClimate and one of the commenters put up what I thought was a pretty good summation of possible consequences of the meltoff of arctic sea ice. From Wili, comment 28:

Meanwhile, here’s a list of possible effects compiled by “nothing-new-under-the-sun” in a comment on neven’s site.

“1. Arctic ecosystem change/habitat loss

2. Arctic (human) communities culture/infrastructure loss

3. Albedo change to global energy budget

4. Permafrost melt acceleration

5. Methane clathrates destabilisation

6. Greenland ice sheet melt acceleration

7. Geo-political tensions over Arctic resources

8. Exploitation of Arctic fossil hydrocarbon resources feedback

9. Complex effects on NH wind/weather patterns via polar jetstream effects.

10. I considered a tenth, namely, a further disruption to the global energy budget from the freeing of the latent heat energy that previously was being used to accomplish the phase transition of ice -> water, though my back of the envelope calculations suggest that this is a much smaller issue than albedo change (I’d love to see some reputable work on this topic as I’m far from any kind of expert).

Now that I think about it a little more, I can think of a further seven issues that neither Neven&Kevin nor I mentioned. Some of these I’m very tentative about (esp ##15&16).

11. The release of persistent toxins and heavy metals that had become trapped in the ice.

12. The opening up of Arctic shipping routes which (a) reduces fuel needs of global shipping by cutting distances (negative feedback) but (

brings more diesel fuel into the Arctic region, leaving black soot on glaciers (positive feedback). Not sure which is the larger effect.

13. Reconnection of marine ecosystems previously separated by ice with unpredictable ecosystem changes from invasive species. This is already occurring.

14. Opening up of Arctic fishing grounds to greater exploitation (and noise pollution).

15. Potential effects on thermohaline circulation. I haven’t seen any work on this related to seasonal sea ice loss, so I have no idea whether it is significant.

16. Potential effects on ocean acidification by increasing surface area for atmosphere-ocean gas exchange. Would this make any difference to ocean capacity to act as CO2 sink or rate of acidification? Maybe this is irrelevant. I haven’t seen it mentioned anywhere and is just an idea that came to me.

17 . Highly visual and difficult to dispute sign of climate change, representing a potential tipping point in public awareness and concern. If we are waiting for that, however, before we make any serious efforts to slash emissions (esp if it doesn’t occur until 2030 or later), we’ll already have so much warming committed that we’ll pretty much be toast. At best, therefore, this point might consolidate public support for massive rapid emissions reductions already underway. But of course, by here, we’ve moved out of the geophysical and into the political systems, and so I’ll note the above comments and cease before travelling any further…”

[PhillipS]

There is a good video out of

, of Rutgers University, giving a presentation at AGU on the link between Arctic Amplification and Extreme Weather.

[Vergent]

With an ice free arctic in June or July, The polar cell would not be the week anti-cyclone we are used to. It would have to be cyclonic. Surface air flow is from the cold to the warm. The arctic ocean is surrounded by permafrost, residual snow cover, and Greenland, the surface airflow must be to the north. We have just seen how strong and big a cyclone can be up there with a core of 0C air that is at 100% humidity. The warmer the air/water, the colder and denser the overlaying air, the closer to the pole the stronger the storm.

I have adjusted the scale from 271k (-1C) to 282 (10C)

If we compare the conditions now to the conditions just before the storm, we see that there is no reason that another storm can not form now. Further if it centers itself over the ice like it did before, it will be at the NP. This is important because away from the pole half the incoming air does not contribute much rotational energy to the storm. Basically, the cosine of the direction is proportional to the rotational energy contribution.

Further the energy contribution is greater because near the pole there is a 1:1 relationship between movement north and change in rotational axis. If a patch of air is 1,000km from the pole and is at rest with respect to the surface underneath, it is traveling 6,000km/day with the rotation of the earth. That's 250 km/hr. Is ground speed there would be 0, because the ground is moving with the same speed. Move that air down to the pole, it would have a 250 km/hr ground speed, it would have retained it angular momentum.

Since the warm, moist air is in unlimited supply from the north Atlantic, this storm at the pole, a circumpolar storm, would not peter out for a long time. Such a storm would strengthen as the low grew from the high velocity of the incoming air. Such a storm would certainly leave this arctic ice free.

There is evidence in the ice cores, for storms.

ftp://ftp.ncdc.noaa.gov/pub/data/paleo/icecore/greenland/summit/gisp2/chem/iond.txt

Sodium (Na) is the proxy for wind. Take a dive through time, the last 150k years or so.

EDIT for some reasons the attachments did not work

[TerryM]

PIOMAS recorded the volume of additional ice lost between 2010 and 2011 as 380 km cubed.

The latent heat expended in changing the phase of this ice requires about 12,7 EJ. - or a little less than 2 Krakatoa's or 9 Tohuko earthquake/tsunami's

I think this figure is conservative as the difference between 2010 & 2011 wasn't extreme, but it represents a huge summer energy pulse, and we haven't looked at albedo yet.

I'm not sure this is the right way to look at this - perhaps this needs to be divided by the amount of time that the Arctic is ice free?

Terry

[Herb@MAWS]

8/27/12 Headline: "Warm Arctic sets record for summer sea ice melt"

(Source: AP)

http://news.yahoo.com/warm-arctic-sets-record-summer-sea-ice-melt-150448268.html

[SVT450R]

Is there any actual peer review literature of possible effects?

[dabize]

PIOMAS recorded the volume of additional ice lost between 2010 and 2011 as 380 km cubed.

The latent heat expended in changing the phase of this ice requires about 12,7 EJ. - or a little less than 2 Krakatoa's or 9 Tohuko earthquake/tsunami's

I think this figure is conservative as the difference between 2010 & 2011 wasn't extreme, but it represents a huge summer energy pulse, and we haven't looked at albedo yet.

I'm not sure this is the right way to look at this - perhaps this needs to be divided by the amount of time that the Arctic is ice free?

Terry

I think the ice loss approach is a good one - it is based on actual melting, which permits one to 1) ignore things like albedo and sun angle/day length and 2) assume that the melted ice started out close to 0C, making it likely that the amount of energy needed to heat the ice to freezing is both low and a constant proportion of the ice volume melted. However, I think you have to add in a term for the temperature increase of the top 1-2 meters of water once the ice is gone to get the actual heat taken in. That requires a term for the period over which there is no ice and the change in SST over that period (assuming there is no significant heat uptake by ice when frozen). this means you DO have to integrate over the chosen period of measure for changes in SST and SIA.

I was thinking you might be better off by just measuring changes in the amount of energy coming in - by putting in a normalized summer value for cloud cover and using a proportion of maximum multiplier for each day (to adjust for sun angle and photoperiod) to calculate the total value of incoming energy integrated over the period being considered. If I recall correctly, the insolation level differences between the Circle and the Pole vary inversely, with the Circle receiving proportionately more closer tp the Autumn equinox (August) and the Pole more during June and July. So you could ignore latitude and save much trouble if you were doing a seasonal measurement, as the effects would balance out. Or you could focus on June and July, treating August as a constant (seeing as the low latitude ice is apt to be gone by then anyhow).

This would be the maximal insolation expected over that period. Then you could use the SIA number integrated over your time period and a value for albedo loss in going from ice to water to get the amount of increase in the amount of energy absorbed.

Overall, your way looks more elegant - just factor in those SST changes.

[dabize]

I'll post this here too - it's relevant enough.

Dr. Maslowski looks at the future of sea ice and how we model it

http://www.oc.nps.edu/NAME/Maslowski%20et%20al.%202012%20EPS%20Future%20of%20Arctic%20Sea%20Ice.pdf

[TerryM]

dabize

What I had thought to do was to take that one year's ice melt as a measure of latent heat entrapped in the phase change, and then add the amount of energy not being deflected back to space due to albedo changes - I haven't looked at any of the numbers yet, but I'm guessing we might have a couple of orders of magnitude more than the latent heat figure.

I'm not trying to quantify the total amount of incoming radiation that we're going to be experiencing, but rather the change that losing the ice will bring about, so I'm missing the relevancy of SST. The rate of changes we're now experiencing would be the base to which the final figures would be added.

I'm not striving for perfect accuracy, and I am trying to come up with a minimum amount where all the error bars would point to larger figures.

I'm thinking that ice area as opposed to ice extent would be the proper metric to start with when calculating changes due to albedo because area removes melt ponds and leads. Ice thickness, snow cover and age are large factors in albedo measurements (as Dr Box has shown), but to simplify things we need to come up with an average ice albedo figure. Any thoughts are welcome.

Terry

[Vergent]

Terry,

Why don't you look at how the Hudson Bay heats up after it melts?

That would incorporate all those factors plus evaporation. The only adjustments you would have to make are for insolation and angle of incidence, which probably cancel each other out.

Verg

[TerryM]

Verg

Hudson Bay may hold some answers. but what I'm striving for in this exercise is simply to quantify the additional energy that the globe will experience. Heat transferred from the ocean to the atmosphere through evaporation wouldn't be a factor at this stage. My thought is that once we have a figure for the amount of additional energy we can speculate on how it will act. Some will return to space through the greenhouse gasses in the atmosphere, but most of it is going to remain, and build with each ice free season. At some point I'd assume this would lead to a perennially ice free Arctic and well have to make a new set of calculations.

Terry

[dabize]

dabize

What I had thought to do was to take that one year's ice melt as a measure of latent heat entrapped in the phase change, and then add the amount of energy not being deflected back to space due to albedo changes - I haven't looked at any of the numbers yet, but I'm guessing we might have a couple of orders of magnitude more than the latent heat figure.

I'm not trying to quantify the total amount of incoming radiation that we're going to be experiencing, but rather the change that losing the ice will bring about, so I'm missing the relevancy of SST. The rate of changes we're now experiencing would be the base to which the final figures would be added.

I'm not striving for perfect accuracy, and I am trying to come up with a minimum amount where all the error bars would point to larger figures.

I'm thinking that ice area as opposed to ice extent would be the proper metric to start with when calculating changes due to albedo because area removes melt ponds and leads. Ice thickness, snow cover and age are large factors in albedo measurements (as Dr Box has shown), but to simplify things we need to come up with an average ice albedo figure. Any thoughts are welcome.

Terry

I thought you wanted to get to the change in total energy retained by adding the latent heat to other kinds of retained energy (such as the energy needed to heat the water once melted to the observed SSTs).

I was thinking you would need to calculate the total energy incoming and then you could use the albedo change to work out the effect of increased ice melt on the proportion of that actually absorbed.

A bit confusing...