Fram Strait Export

[Richard Rathbone]

A commentator on Neven's site pointed out that the amount of ice going out the Fram during the melt season had gone up considerably in recent years. (As graphed by Wipneus at https://sites.google.com/site/arctischepinguin/home/piomas/grf/fram2.png )

It seems to me that there's another positive feedback pushing towards the ice free arctic here. The less of it there is, the faster its been exported Apr-Jul. The thinner it gets, the faster its moving, and the increasing speed has been pushing the export volume up faster in those months than the decreasing thickness can reduce it.

[frankendoodle]

According to NSIDC news, some of the existing multi-year ice that was left after last years melt season got transported already.  http://nsidc.org/arcticseaicenews/
The Fram & Nares Straits are super important this year because that is where the thick multi-year ice will be transported through.

[Vergent]

CT shows 550 kkm2 in the GS. It looks like almost half of that is MYI call it 200k. We started with 1.8 Mkm2 in the arctic basin, so now its about 1.6. We are trading old ice for freshly frozen leads.

[Laurent]

Fram strait is not anymore what it used to be !
The atlantic is in the arctic, it was there all the winter on the northern coast of spitsberg (svalbard), it is now appearing on the northern coast of Franz Joseph Island. I think next winter this would be more obvious !
So when you have some measurements of the export threw fram strait it is dodgy because it does melt before entering the Fram strait !

[johnm33]

I suspect that the same forces causing the early cracking have eased much of the fresher waters beneath the ice out through Fram too.

[sofouuk]

ASIB 'On the move', posted by: Artful Dodger | April 04, 2013 at 22

'... Look at the bottom image from the Apr 2, 2013 NSIDC update:

[[http://nsidc.org/arcticseaicenews/files/2013/04/Figure41.png]]

It shows clearly that about 25% of the MYI remaining on Dec 2nd was advected from the Central Arctic through Fram strait by Mar 28. This occurred in less than 4 months, and the process is ongoing.

Compare this with climate models that assume an annual figure of 10% advection of MYI, and you will understand that we are witnessing the Arctic sea ice death spiral.

The drain is Fram strait, and it is unplugged.
--
Cheers,
Lodger'

[Richard Rathbone]

I think the timing is significant.

It doesn't matter so much if export goes up during the freeze season, because what's left behind will freeze faster and compensate to some extent. (and export during that season seems more or less the constant over the last decade)

However, over the past few years it has gone up a lot during the melt season, and that must be causing volume to go down faster and lower.

While in winter feedbacks in the arctic from faster Fram export would reduce its significance, in the April-July  period, the feedbacks are going to be the other way round and an extra million km3 going out the Fram will lead to faster melting of whats left behind as well.

[ChrisReynolds]

It doesn't matter so much if export goes up during the freeze season, because what's left behind will freeze faster and compensate to some extent. (and export during that season seems more or less the constant over the last decade)

It does matter because a lot of the ice exported, as in this year, will be multi-year ice (MYI). MYI has a stabilising effect on the ice pack being less prone to melt.

[Richard Rathbone]

Why does losing MYI in the winter matter more than in the spring/early summer?

I think losing ice through the Fram at that time of year is more important than at other times whatever type of ice it is and whether or not the Fram takes an increasing proportion of its export in FYI as the coverage of FYI increases.

I've got no idea how the age distribution of ice going out the Fram has changed, but I'd have thought that it has to be getting younger if anything, just because the ice in the arctic as a whole is younger.

(Wipneus, if you are reading this and have the data that Fram export graph is plotted from posted somewhere, I'd appreciate a pointer to it. I see a lot of the data behind your other PIOMAS graphs on your site, but not that one)

[ChrisReynolds]

I didn't say winter loss was more important, I said it wasn't less important.

Check out Wipneus' graphic.
https://sites.google.com/site/arctischepinguin/home/piomas/grf/fram2.png

It's in line with published research using different methods in that flux is lower in summer than winter. The ice will have become younger and thinner. As has happened this year Fram takes a large proportion of MYI, this year it has reduced what was already a low area of MYI even further. If that ice is replaced by new growth of FYI, mainly off Siberia due to the transpolar drift, then that new ice will melt out easily.

MYI is associated with delaying the melt, the less of it that is in the Arctic the greater the risk of rapid melt back and warming of the ocean from early in the season.

[Richard Rathbone]

Thats the graph that I linked in the original post. All my comments above are based on that graph.

I see export in the winter. I don't see changes in it though, just random scatter from year to year.

What I do see, is a change in the spring to faster export.

I see a constant winter export as important to explaining why a steady state arctic has a particular age distribution of age within it. What I don't see is how a constant winter export can explain changes in that distribution over time.

Round numbers from eyeballing.
Winter 0.2 no trend
Spring gone from 0.05 to 0.15 within the last decade

A 0.1 increase is modest in the scale of overall melt, but I think it might be helping accelerate the melt because its happening at the same time as the melt.

[ChrisReynolds]

Yes I know it's the graph you started with.

Fram export is one of the factors in the decline of the pack. During the 1990s a very positive AO led to increased Fram export. However constant Fram export can still play a role in ice decline because ice aging in the Beaufort gyre has led to ice destruction as summer extent has retreated.

My point is that whatever is happening in spring in 2010/11 Fram export is typically greatest in winter and is not to be neglected. Ice transported into Fram is lost to the Arctic. Whether or not the 2010-12 late peaking of export means anything but weather is still to be seen. Synoptic scale pressure doesn't seem to me to explain the shift (monthly NCEP/NCAR), the post 2010 thinning might have a role.

[ChrisReynolds]

The most recent paper I can find is one I already have: Smedsrud, 2011, Recent wind driven high sea ice export in the Fram Strait contributes to Arctic sea ice decline.
http://www.the-cryosphere-discuss.net/5/1311/2011/tcd-5-1311-2011-print.pdf

See figure 4. That shows timeseries of ice export the start and end months aren't clear, however elsewhere in the paper they use a year as September to August, so if the last point is for August this graph doesn't show a May increase in flux. This graph is for area, and the flux is estimated using NCEP/NCAR derived geostrophic wind, which is the vector sum of the coriolis force and the pressure gradient driven wind. Oddly PIOMAS ice motion is driven by NCEP/NCAR, yet it shows an increase in mass, whereas the Smedsrud paper doesn't find and increase in area transport.

Figure 5 makes this more puzzling, although it is for Oct to April the shift shown in Wipneus' graph could be picked up. Given the above mistmatch between Smedsrud and PIOMAS one might think it can be reconciled by PIOMAS picking up the effect of thinner ice, which would move faster for a given wind driven transport. Yet figure 5 gives a scatter plot between the calculated ice velocity and that derived from satellite data, and the relationship seems to be near linear.

[Jim Williams]

Everyone talks about thinner ice and greater transport speed, but wouldn't it be a different feature of first year ice which really matters?  Isn't it flatter ice which would move faster due to less turbulence?

[Artful Dodger]

Everyone talks about thinner ice and greater transport speed, but wouldn't it be a different feature of first year ice which really matters?  Isn't it flatter ice which would move faster due to less turbulence?

Hi Jim,

Sea ice has a flat keel, or evolves toward flatness after ridging. The reason is simple thermodynamics. The deeper you go, the warmer the water temp, the softer the ice. Ice floats up, cools, refreezes to the keel, filling in thin spots.

The US NSIDC notes that "When keels erode into smooth features, they are called bummocks."



So ridging, spreading, and flattening is the process by which sea ice is able to thicken beyond the normal 2m thermodynamic limit imposed by normal Winter air temperatures.

[ChrisReynolds]

Jim,

FYI is typically flatter than ridged older MYI.

[Jim Williams]

Thanks for the explanation Lodger, but I still think that first year ice would be flatter and therefore have less turbulence -- just less so than if the MYI didn't undergo a freeze thaw cycle.  Of course, it would also offer less wind resistance, so maybe that would balance out.

[Richard Rathbone]

Thanks for the reference Chris. It looks worth digesting, but it will take me a little while to do that.

Jim,

Ice is moved by wind and water dragging on it. More turbulence means more drag means faster motion. It will only mean less motion when wind and current are acting in opposite directions and the smaller one is preferentially enhanced.

[Richard Rathbone]

An interesting but ultimately frustrating read. They end up their discussion with pretty much the same point I started out on, but don't actually present their data in any detail to support it.

"This suggests that the spring and summer (March–August) ice export should directly influence the following summer sea ice minimum (Kwok and Cunningham, 2010). We find support for such influence, and see potential for a seasonal prediction of the summer minimum using the spring
export."

That Kwok and Cunningham paper is about the Beaufort melt and I doubt its got any more in it on the Fram than the same sort of hand waving argument in Smedsrud. 

[ChrisReynolds]

Yep, we'll just have to wait for someone to publics with data we can use. I'm used to such niggles now.

[Bob Wallace]

Rather than the ice being moved through the water when blown by wind isn't it more accurate to say that the ice is moving with the water?

Wave motion is water being moved by wind.  A good blow is going to create a flow of water in the direction of the wind.  Since ice has a 10% sail and a 90% keel is it likely that the speed of ice is significantly greater than the speed of the water in which it sits?

[Artful Dodger]

Rather than the ice being moved through the water when blown by wind isn't it more accurate to say that the ice is moving with the water?

Hi Bob,

This is the subject of studies done on Ekman spiral. The theory applies for wind fields created by either High or Low pressure systems: Highs cause convergence, and Lows cause divergence.



IIRC sea ice moves at about 10% the speed of the wind in an Ekman spiral. There was much, much more discussion on this topic over on the ASI blog in Jul 2012.