The 2023 melting season

[nadir]

And this was just another “boring” season, according to some “experts”....

Though not an expert at all, I did call it boring in June. And it was... until it wasn't, and then I got excited by the beginning of August seeing what was going on.

Don’t take it personally Cid. Actually I was not referring to anyone specifically not certainly you… and I was talking more about July… that moment when there was a total disconnect between what the extent numbers seemed to say and what it could be seen in Worldview.

[jdallen]

And this was just another “boring” season, according to some “experts”....

Though not an expert at all, I did call it boring in June. And it was... until it wasn't, and then I got excited by the beginning of August seeing what was going on.

Don’t take it personally Cid. Actually I was not referring to anyone specifically not certainly you… and I was talking more about July… that moment when there was a total disconnect between what the extent numbers seemed to say and what it could be seen in Worldview.

Agree.

What we see now in Worldview reminds me a great deal of what we saw in 2013 & 2014 - large areas of extent filled with disconnected blocks of ice, very little of which is consolidated into contiguous ice coverage.

I think that to a certain degree, because of the dramatic changes that have taken place in the Arctic over the last decade, we’ve gotten used to what is I think in fact a transitional state of the pack on it’s way to a seasonal ice-free state.  I think it’s easy to lose track that prior to 2012 - perhaps 2007 would be a better break point - the end of season state of the ice would still have millions of KM2 as a solid pack, broken only occasionally by modest leads, further, with 100s of thousands of KM2 of durable MYI in excess of 3m thick.  At the END of the melt season.

So, cognitively, I think we’ve gotten used to this transitional state, and perhaps through fatigue or familiarity, have somewhat disconnected from an awareness of just how *astonishing* each season has been since 2007.

So, this season may “only” be 4th or 5th or whatever lowest, but keep in mind, that’s a “lowest” which is half the extent and one third the volume we’d see at end of season before the Arctic shifted into the current regime we see it following.  It is not normal, it is not even close to where it should be, even if it isn’t as dramatic as a sub 1 million km2 season would look.

Spend some time looking at pre-2007 worldview end-of-season images and flipping up to where we are now, and I think that should drive my point home.

[HapHazard]

Indeed, jdallen - I've been of the mind for a few years now that "rubble-fication" is [becoming] the new ice trend.

[morganism]

A Polar Vortex is starting to form in the Stratosphere over the North Pole and will impact the Weather as we head closer to Winter 2023/2024.

https://www.severe-weather.eu/global-weather/polar-vortex-forecast-winter-2023-2024-cooling-weather-impact-united-states-canada-europe-fa/

(...)
You can see the violet line, which is the latest analysis. It shows that the Polar Vortex is currently running colder than normal but at similar levels as last year.

The temperature is dropping fast in the Stratosphere at around 30km/18.5miles altitude. In the image below, you can see the temperature change forecast for late September. It shows ongoing cooling over and around the North Pole, with a strong temperature drop in the next 16 days."

(discussions of the CBO and higher chances of SSW events due to the ENSO. Expecting many cold blasts in NA this year)

[NeilT]

Spend some time looking at pre-2007 worldview end-of-season images and flipping up to where we are now, and I think that should drive my point home.

I clearly remember the 2005 "shock" and the 2007 OMG. Yet 2023 can be on top of the 2007 line and not really generate much comment other than "oh it didn't make a record".

[jdallen]

A Polar Vortex is starting to form in the Stratosphere over the North Pole and will impact the Weather as we head closer to Winter 2023/2024.

Looks like we may get to see how impervious it is (or not) to the evolving Cyclone Cannon on the US Eastern Seaboard.

[Paul]

And this was just another “boring” season, according to some “experts”....

Though not an expert at all, I did call it boring in June. And it was... until it wasn't, and then I got excited by the beginning of August seeing what was going on.

Don’t take it personally Cid. Actually I was not referring to anyone specifically not certainly you… and I was talking more about July… that moment when there was a total disconnect between what the extent numbers seemed to say and what it could be seen in Worldview.

So, this season may “only” be 4th or 5th or whatever lowest, but keep in mind, that’s a “lowest” which is half the extent and one third the volume we’d see at end of season before the Arctic shifted into the current regime we see it following.  It is not normal, it is not even close to where it should be, even if it isn’t as dramatic as a sub 1 million km2 season would look.

No doubt we are seeing an ice pack that looks more like "rubble" and there is definately a feature that causes low concentration ice to the 'right' of the pole around the 85 degree mark in low pressure set ups however we also got to remember, just because the ice is not compacted, it does not fully necessarily mean it's thin. There was no significant visible dispersion of the ice in 2020 yet if you look at worldview of the pole area towards the Atlantic, the ice looks thinner than it does now. I don't think dispersion of sea ice is a new thing, the difference now it's occuring at higher latitudes and due to a slower formation of the PV(due to climate change and increasing open water) the dispersion is lasting longer into September than it ever has done in the past.

Either way, despite some signs that extent flattered to decieve earlier in the summer(it was obvious to me we were going to be below last year's figure), I'm still surprised at how low we could get and there is little signs of the melt season being over just yet.

[oren]

Looking at AMSR2 regional area charts, courtesy of AWI, some interesting situations of the "left" side of the Arctic.
* The CAA has been hugging the historical lows since the beginning of the season, and continues to do so now. I am sticking to my prediction that season's end will see 2023 somewhere along with 2012 and 2015, and with high probability should finish below 200k.
* The Beaufort has been rather late to begin melting, and languished through the first half of the season, only to spectacularly crash to the vicinity of 2019, 2016 and of course 2012. Where it goes from here is hard to guess but it should finish below 100k.
* The Chukchi has set or record highs between mid-June and mid-July, recently crashing to the middle of the pack. A near-zero (<50k) finish is now guaranteed.
* The ESS has been seesawing between relatively high and relatively low. From this point a finish below 100k should be in hand, while a 50k is still quite possible.

Re-visiting my predictions from 6 weeks ago, it appears I aimed too high in all 4 regions, despite providing a pessimistic prediction at the time.
* The CAA continued to go low until a late stage "garlic press" import plus some apparent refreezing finally stopped the trend. 2023 did end somewhere along with 2012 and 2015, finishing at 140k, with its northern region maintaining ice cover as mostly expected.
* The Beaufort continued its late-season streak and went very low, finishing at 61k. It seems the much-reduced MYI bulge was what stopped it from going any lower.
* The Chukchi did indeed get eventually to near-zero  - finished at 3k.
* The ESS has crashed all throughout August, and finished at 3k.

I am also adding a chart of Kara, which continued to trend relatively high, but is still finishing at <37k.

[oren]

And adding another set of charts with interesting data. Note all charts show AMSR2 area with a 3km grid calculated by AWI.

* The Laptev is serving as an ice refuge this season, both thanks to import and to persistent cloudiness around the solstice. An extremely unusual result: the tentative minimum at 131k nearly double the previous record highest minimum in the AWI's dataset, AND set on Aug 8th. This could possibly still change.
* The Greenland Sea crashed to finish at 62k only to rebound immediately (91k if this value is discounted). In any case, a medium finish.
* The CAB is showing extreme behavior in the past few days, crashing to 2.778 and rebounding. Discounting this 1-day value would give a min of 2.935.
* The total Arctic area is also showing the same behavior, with a min at 3.529. Discounting this, the min would be 3.719. A medium to low finish (if finished it is, still unclear).

[Michael Hauber]

1st pic is Arctic on Sep 13 2000 - first year available in MODIS.  Atlantic fringe on right, north pole on left.  Rubble ice extends most of the way from the Atlantic front to the pole.

2nd pic is Sep 13 2001, on the Beaufort.

My model of the Arctic sea ice decline is a zone of strong ice in the middle, a zone of marginal ice/rubble around this, and then last a zone of open water.  All zones are contracting in towards the last ice zone north of Greenland over time.  The open water zone gets bigger, the strong ice zone gets smaller (and will be gone soon) and the rubble zone in between stays about the same size, but becoms a larger proportion of the total ice pack as the strong ice zone shrinks.  Both years pictures have large areas of continuous ice sheet over the majority of the Arctic, whereas in recent years this type of ice is mostly limited to a small area north of Greenalnd/Ellesmere.

[Comradez]

Cryosphere Computing finally updated, although Sept. 10th's data is missing entirely.  But the daily area for Sept. 11th was 2.635 million km^2, which officially surpasses 2020 for 3rd lowest daily area ever. 

Still a lot of strong wind and warmth coming out of the ESS and hitting the central pack for a few days, but I doubt this year can make a run for 2nd place (200k more needed). 

[NeilT]

It looks like it is beginning to solidify on the pacific side but draw back north on the Atlantic side.  I'd say it is pretty much over but there can be some surprises.

There have been late surges before but the general state of the CAB overall doesn't seem to lend itself to it.  A bit more compression maybe.  Perhaps some late melt around FJL.  But the pacific side could make up for it.

Only a week or so until we know.

[Paddy]

It looks like it is beginning to solidify on the pacific side but draw back north on the Atlantic side.  I'd say it is pretty much over but there can be some surprises.

There have been late surges before but the general state of the CAB overall doesn't seem to lend itself to it.  A bit more compression maybe.  Perhaps some late melt around FJL.  But the pacific side could make up for it.

Only a week or so until we know.

Over the last 5 years, the date of minimum extent has ranged from the 12th to 21st September so yeah, a week or maybe 10 days at the outside.

[oren]

Area is on this day 8k km2 less than the 2020 minimum, and so is guaranteed a 3rd lowest minimum behind 2016 and 2012, and I don't think sea ice area loss has quite finished yet.

For anyone who has missed this, 2023's melting season now stands only behind 2016 and 2012 in terms of NSIDC area, and not by huge margins either:
* 2012 2.25 M km2.
* 2016 2.45 M km2.
* 2023 2.67 M km2 (provisional - not over yet).

For a "boring and nondescript" year such as we had, that is quite the achievement, especially considering both leaders had a GAC in August, and 2020 had the GAAC (massive anti-cyclone and clear skies).
Extent and volume wise the situation is not as bad - the ice is uncompacted and there is an extra bit of volume in the Laptev and near the Atlantic front (pending the mid-month PIOMAS update), but I still find the situation alarming. What happens when a really strong melting season comes along again? And what if such a season comes on the heels of a previous vigorous season, as happened for 2012, 2016 and 2020? I think the result might be very bad.
Let's all hope 2024 is the coolest ever, though I wouldn't hold my breath on that one.

[Tom Stedman]

It's probably lucky we had such a cool month of May, and slow start to the melting season!

[Paul]

Area is on this day 8k km2 less than the 2020 minimum, and so is guaranteed a 3rd lowest minimum behind 2016 and 2012, and I don't think sea ice area loss has quite finished yet.

For anyone who has missed this, 2023's melting season now stands only behind 2016 and 2012 in terms of NSIDC area, and not by huge margins either:
* 2012 2.25 M km2.
* 2016 2.45 M km2.
* 2023 2.67 M km2 (provisional - not over yet).

For a "boring and nondescript" year such as we had, that is quite the achievement, especially considering both leaders had a GAC in August, and 2020 had the GAAC (massive anti-cyclone and clear skies).
Extent and volume wise the situation is not as bad - the ice is uncompacted and there is an extra bit of volume in the Laptev and near the Atlantic front (pending the mid-month PIOMAS update), but I still find the situation alarming. What happens when a really strong melting season comes along again? And what if such a season comes on the heels of a previous vigorous season, as happened for 2012, 2016 and 2020? I think the result might be very bad.
Let's all hope 2024 is the coolest ever, though I wouldn't hold my breath on that one.

However the stats suggests this summer was in fact a very warm one apart from June. I think the damage was done in July really with the dipole, I just sadly think because forum traffic is not what it was, it probably did not gather the attention that it should of.

So a persistent Beaufort ridge has been damaging for the ice and we had a strong low over towards the Laptev at times which caused the dispersion. SSTS were exceptionally high over the Beaufort also(although they have moderated in recent weeks).

Just felt like we have seen melt momentum at hand here without the headline extent drops. August should be noteworthy of the amount of ice we have loss really, must be at least the second highest I'm guessing?

[The Walrus]

2023 was really a tale of two melt seasons; the sluggish first half, followed by the enhanced second half.  The result was a rather average melt season.  Average, considering that the maximum was on the low side.  The ice melt from maximum is depicted on the followign plot, which 2023 as an open, yet-to-be-filled circle.  Of the previous 16 years, 8 displayed more ice melt and 8 had less than this year - so far. 

[oren]

As usual, judging the state of the ice and its melt just by extent and just by NSIDC leads one away from the more complex reality.

[Michael Hauber]

The gap to 2019 got down to 50k by Jaxa extent, but has opened up a little to 80k.  Comparing current Jaxa map to Sep 1 shows that Pacific side barely retreated and the majority of the losses have been Atlantic.  I did think there was more potential in the Atlantic side and wonder how much momentum may play a role.  Currents have been pushing ice in that direction all season and so when the wind changes to push ice from that side the current opposes and the retreat isn't as big perhaps. 

There may be some Pacific retreat next couple days, but looks highly unlikely that we will match 2019's very strong late season losses and I'd expect the gap to widen further from now.

[nadir]

Area is on this day 8k km2 less than the 2020 minimum, and so is guaranteed a 3rd lowest minimum behind 2016 and 2012, and I don't think sea ice area loss has quite finished yet.

For anyone who has missed this, 2023's melting season now stands only behind 2016 and 2012 in terms of NSIDC area, and not by huge margins either:
* 2012 2.25 M km2.
* 2016 2.45 M km2.
* 2023 2.67 M km2 (provisional - not over yet).

For a "boring and nondescript" year such as we had, that is quite the achievement, especially considering both leaders had a GAC in August, and 2020 had the GAAC (massive anti-cyclone and clear skies).
Extent and volume wise the situation is not as bad - the ice is uncompacted and there is an extra bit of volume in the Laptev and near the Atlantic front (pending the mid-month PIOMAS update), but I still find the situation alarming. What happens when a really strong melting season comes along again? And what if such a season comes on the heels of a previous vigorous season, as happened for 2012, 2016 and 2020? I think the result might be very bad.
Let's all hope 2024 is the coolest ever, though I wouldn't hold my breath on that one.

Fully agree, although I could argue that the silent, deceptively weak but persistent July dipole in 2023 is more pernicious than the very strong 2020 GAAC. The depletion of sea ice (some of it very thick and old) in the Western side this year has been amusing, as well as how quickly the entire Pacific sector became a soup of broken ice during July, ready to get the coup de grace from August storms.

Also, big expanses of the pack might have been thinner at the beginning of the season than what was indicated by Piomas in Chukchi/Western CAB, as suggested by SMOS/Cryosat-2. Ocean currents from the Pacific may have played a role starting in end of summer 2022 (remember those Pacific water strong ejections from Bering/Barrow canyon back then?).

[Jim Hunt]

Areas of new ice (coloured pink) are starting to appear in the nooks and crannies of the Canadian Ice Service charts of the Canadian Arctic Archipelago:

https://GreatWhiteCon.info/2023/09/facts-about-the-arctic-in-september-2023/#Sep-12

[oren]

Late season export is putting another small nail in the coffin for sea ice volume and MYI. This season has been quite destructive due to its persistent movement from Chukchi/ESS direction to Fram.
Another recent feature is a very clear "garlic press" going on in the northern CAA, exporting more precious MYI, though it may survive better there while Fram MYI is doomed.
In addition to the cumulative extent export chart and the 10-day motion map, the AWI SIC-LEADS animation shows both export motions quite well (as well as the hints of new ice in the CAA).
The full season motion map is interesting as well, though I do not have such maps for other seasons to compare with.

[kiwichick16]

@  Oren ......you are going to need a bigger graph

[jdallen]

Area is on this day 8k km2 less than the 2020 minimum, and so is guaranteed a 3rd lowest minimum behind 2016 and 2012, and I don't think sea ice area loss has quite finished yet.

For anyone who has missed this, 2023's melting season now stands only behind 2016 and 2012 in terms of NSIDC area, and not by huge margins either:
* 2012 2.25 M km2.
* 2016 2.45 M km2.
* 2023 2.67 M km2 (provisional - not over yet).

For a "boring and nondescript" year such as we had, that is quite the achievement, especially considering both leaders had a GAC in August, and 2020 had the GAAC (massive anti-cyclone and clear skies).

I was just about to comment on exactly this thing.  I'll carry it further instead.

The key thing is, the extraordinarily low numbered coupled with an indifferent melt season.  No remarkable event, nothing singularly extraordinary, save perhaps, PERHAPS the Fram export, but I don't think that's a singular defining factor in the season.

I think this drives home the conclusion I reached years ago - outcomes are all about the area.  Extent numbers are less about the systems health, and more about providing a convenience to mariners.  Extent doesn't reduce insolation uptake.  Area does.

While volume hasn't crashed, I'm not sure it presents a cheery picture either, and I suspect the actual picture is worse than current estimates might suggest.

My guesses for the extent and volume polls this year have been about right.  However, my area estimates are high by a full half million KM2.   I'd very much rather the error went the other way.

I think this may be a transitional year - like 2010 or 2011.  I'll be focusing closely on the weather leading up to and during the refreeze, where we may find ourselves looking at another low maximum like 2016.  Does the Cyclone cannon start up?  Or not? <rolls dice>

[Renerpho]

PERHAPS the Fram export, but I don't think that's a singular defining factor in the season.

I'm not so sure. Ice export has been very strange this year; not just with Fram, but the movement of the ice overall. A lot of old ice has been lost this year via Fram (maybe 30% of all the multi-year ice, I would say). If it wasn't for the old ice that made a detour into Laptev, we'd surely see the Laptev bite that we got so used to in recent years. And where would we be THEN?

[peterlvmeng]

2008(07)-2012-2016-2020-2024?
Is this the intrinsic mechanism of arctic?

[jdallen]

2008(07)-2012-2016-2020-2024?
Is this the intrinsic mechanism of arctic?

Our brains *love* organizing what we sense into patterns.  What played out over the time frame you reference is far messier, nothing intrinsic about it beyond the annual normal variations in insolation.

The change in dynamics I think actually *starts* in a serious way in the late 1990s.  That's certainly were we start seeing the accelerated decrease in end of season volume that we have now.

Renerpho's comment may be prophetic; like the bizarre weather we had in 2013 & 2014 that pretty much saved the Arctic, the difference between what we have, and a more serious run at 2012 may be exactly that unusual movement of older ice into the Laptev where it's managed to survive.

Problem is, too much heat is being imported via Atlantification, and too much heat is being retained over-winter.  Part of that is atmospheric - more CO2, more water vapor reducing re-radiation, or at least raising the bar for how much net enthalpy gets carried over year over year.
 There's also the *imported* latent heat that is arriving in increasing volume because of the increasing instability of northern hemisphere atmospheric circulation. Constant Arctic Breakouts anyone?  Flip side of that, is imported intrusions of heat from lower latitudes raising winter temperatures in the Arctic 30 degrees C.  That imported heat replaces what would normally get pulled out of the Arctic ocean.

You can see the evidence of this most clearly by way of the annual minima and maxima dates slowly creeping closer together over winter.  They're something like two weeks closer now than they were 30+ years ago.

So the problem really is, someday, over winter, the Arctic simply won't be able to dump heat and produce enough ice to survive even a relatively bland melt season.  So I watch the weather in winter now more than I do the melt season.  The heat the Arctic receives seasonally is very predictable, only affected really by Albedo.  How much heat it loses?  Not so much, at least, that which gets pulled directly from the Arctic ocean - the largest heat sink contributing the most to and driving the system's dynamics.

Like a top, slowly losing angular momentum, the transition from it spinning vertically to rolling around on the floor will be sudden, but not really a serious change in energy, just what it looks like.  The last 20 years I think are the start of the "wobbles".  Not that many left, I think.

[peterlvmeng]

2008(07)-2012-2016-2020-2024?
Is this the intrinsic mechanism of arctic?

Our brains *love* organizing what we sense into patterns.  What played out over the time frame you reference is far messier, nothing intrinsic about it beyond the annual normal variations in insolation.

The change in dynamics I think actually *starts* in a serious way in the late 1990s.  That's certainly were we start seeing the accelerated decrease in end of season volume that we have now.

Renerpho's comment may be prophetic; like the bizarre weather we had in 2013 & 2014 that pretty much saved the Arctic, the difference between what we have, and a more serious run at 2012 may be exactly that unusual movement of older ice into the Laptev where it's managed to survive.

Problem is, too much heat is being imported via Atlantification, and too much heat is being retained over-winter.  Part of that is atmospheric - more CO2, more water vapor reducing re-radiation, or at least raising the bar for how much net enthalpy gets carried over year over year.
 There's also the *imported* latent heat that is arriving in increasing volume because of the increasing instability of northern hemisphere atmospheric circulation. Constant Arctic Breakouts anyone?  Flip side of that, is imported intrusions of heat from lower latitudes raising winter temperatures in the Arctic 30 degrees C.  That imported heat replaces what would normally get pulled out of the Arctic ocean.

You can see the evidence of this most clearly by way of the annual minima and maxima dates slowly creeping closer together over winter.  They're something like two weeks closer now than they were 30+ years ago.

So the problem really is, someday, over winter, the Arctic simply won't be able to dump heat and produce enough ice to survive even a relatively bland melt season.  So I watch the weather in winter now more than I do the melt season.  The heat the Arctic receives seasonally is very predictable, only affected really by Albedo.  How much heat it loses?  Not so much, at least, that which gets pulled directly from the Arctic ocean - the largest heat sink contributing the most to and driving the system's dynamics.

Like a top, slowly losing angular momentum, the transition from it spinning vertically to rolling around on the floor will be sudden, but not really a serious change in energy, just what it looks like.  The last 20 years I think are the start of the "wobbles".  Not that many left, I think.

I agree. The melting season governs the arctic condition in the previous time. The refreezing season governs the arctic in the future time. There will be a time that the maximum extent is so low that no matter what cool summer will be, all the arctic sea ice will melt out eventually in summer.

[johnm33]

No remarkable event, nothing singularly extraordinary, save perhaps, PERHAPS the Fram export, but I don't think that's a singular defining factor in the season.

I'm thinking Fram export is a key indicator of the increasing exchange of Arctic and Atlantic/Pacific waters. The outflow is principally near surface through Fram and the ice merely a passenger in the process. That contributes two main dynamics first the loss of older ice which would otherwise pass clockwise through the pole/Fram line and be recycled through to Beaufort, second it frees up the inert rotation potential of ice moving south and losing ground in the rotating frame. This year we saw a much larger Pacific contribution which played a role in forcing ice towards N.Laptev but also presented as a different set of vortice patterns, seen most clearly in Hycom sss animations. This folowed on from new patterns of bottom melt/vortices showing up during the last two melt seasons, and speaks to a rapidly evolving subsurface dynamic. This suggests the 'old' current system has broken down and we're in a period of chaos as a new dynamic establishes, whether the chaos period lasts years or decades??

[NeilT]

Extreme years are exactly that, extreme.  But, just like the global heat of 1998 and the constant harping that "the world had cooled since", the background warming reached the "extreme" level of 1998 by 2014.

We are seeing a similar thing in Arctic melt.  2007 and 2012 were Extreme years.  2007 with heat and clear skies, 2012 with heat and incredible storms.

Now we have 2023 sitting right on top of 2007 for Extent, a touch above at the end I'd say; from a boringly mundane melt season.

There is a nino coming up and until the chart below starts going in the other direction, then this is only going to get worse and tuly Extreme years are going to see something close to, or actually, the BOE.

I'm quietly waiting for it.

[UCMiami]

jdallen - I like your choice of the 'scientific' term 'the wobbles'! :-) Because that really is what is happening in the arctic. We analyze the ice in each sea and compare it to the resent past, where is the Laptev Bite? Why didn't the Atlantic front retreat, etc. But the reality is the arctic ice is so mobile at this stage that it sloshes or wobbles around as it melts each year and while we all focus on where it is, the minimum arrives at surprisingly close to the same numbers from the previous years. The Map may look different, but the numbers tell the same story.

The change from a large unified ice mass to the rubble of the current arctic ocean that you pointed out a few pages ago had a huge impact on the mobility of the ice both overall and at the local level of the individual seas. And it also increased the influence of local winds (and weather systems) on the disposition of the ice.

[oren]

The change from a large unified ice mass to the rubble of the current arctic ocean that you pointed out a few pages ago had a huge impact on the mobility of the ice both overall and at the local level of the individual seas. And it also increased the influence of local winds (and weather systems) on the disposition of the ice.

Indeed.
And hello there UCMiami, long time.

[Paul]

Now we have 2023 sitting right on top of 2007 for Extent, a touch above at the end I'd say; from a boringly mundane melt season.

Keep seeing comments like this yet temperatures it would seem for July and August were amongst the warmest on record. I think July weather was unfavourable for the ice with high pressure and warmth from the landmass over the Beaufort and deep cyclones over the Laptev. Not sure if the set up was as extreme as it was in 2007 mind but SSTS were exceptionally high in the Beaufort and that has definately melted the ice there and possibly caused the Chukchi sea ice collapse aswell.

There was some hints whilst extent was still quite high(relatively speaking) things were not as good as it seemed. One of those was NSIDC area was tracking much lower and the other being the SMOS figures which at times during July actually hit the lowest on record.

Whilst for a while I expected this year to finish below last year, it has dropped lower than I thought it would. Also another remarkable event is going on right now, the PV is struggling to form(which is something we are seeing more and more frequently during early September), we shouldn't be see such low concentrations over the ice right now but we are because temperatures are so high. It does look like we may start seeing enough cold to allow the area numbers to improve but I'm not convinced the extent drops are over just yet.

[Michael Hauber]

Low pressure will take over the Arctic ice area in the next few days, and presumably bring dispersion and near 0 change in extent at a guess.  After that another high may move in towards about day 7.  Perhaps some compaction, but it might be cold enough for freezing to start?

[Unaffiliated]

With all these large storms taking such northernly tracks one comes to wonder,
Hurricane Lee is projected to be back over the water N of 55 monday as a tropical storm.
should we expect this to have impact to the remaining ice?
Is there any historical correlation between north bound atlantic cyclone and late season sea ice loss spikes?
if not is this modeled to change as ocean heat rises?
how far north would a TS level cyclone need to make it to have major impact on the arctic basin?
what about a cat 3?

[jdallen]

Now we have 2023 sitting right on top of 2007 for Extent, a touch above at the end I'd say; from a boringly mundane melt season.

Keep seeing comments like this yet temperatures it would seem for July and August were amongst the warmest on record. I think July weather was unfavourable for the ice with high pressure and warmth from the landmass over the Beaufort and deep cyclones over the Laptev. <snippage>

Ah - you're making the same mistaken assumption I made years ago.

Even with very high temperatures, truth be told, the direct uptake of heat from atmosphere is comparatively trivial.  The lions share of melt comes primarily from heat transferred directly from water - either melt ponds or ocean itself - which in turn is captured from insolation.

I Haven't yet been able find a tool that would let me track albedo on a 10x10 or even 25x25 km grid over time, but I think that we'd see a pretty strong correlation to higher melt and more efficient uptake of insolation (as indicated by lower albedo).

And to reiterate, it's immaterial to what I think the key season is - the refreeze - as averaged out, annual melt is fairly predictable.  Not surprising when you consider solar output over time varies at most by only 1-2%.

To lay out the dominoes at a very high level...

1) Increase in CO2 (and consequently atmospheric water vapor) increases heat retention until the increase in temperature produces enough outgoing black body radiation that the equation balances.

2) That increase in temperature reflects a net increase in total global enthalpy available to the global heat engine.

3) That heat is distributed more symmetrically because of breakdowns in circulation which would other wise prevent it from moving directly between the tropics and the pole (North, in particular) - which is what we we call "Arctic Amplification"

4) Resulting disproportionate increases in heat retained/present in the Arctic year over year reduces refreeze ice volume.

Eventually that rises to the point the typical annual uptake of melt season insolation + typical annual variability will produce a sub 1,000,000 km2 melt season.

It won't be a bang.  It'll be a sigh.

[kiwichick16]

@ jdallen  .... i guess that is the question  ...... what level of heat is going to be retained in the Arctic this autumn?                                                               

[binntho]

Even with very high temperatures, truth be told, the direct uptake of heat from atmosphere is comparatively trivial.  The lions share of melt comes primarily from heat transferred directly from water - either melt ponds or ocean itself - which in turn is captured from insolation.

Not that I am disputing this in itself - but we still lack quantification of the various inputs during melt season, and their relative contribution. Something which I assume is not easy, but a rough guess that insolation is by far the biggest contributor can hardly be far off. But then, insolation hasn't really changed during the satellite era, perhaps decreased slightly, but still the arctic has lost 12.000 km3 of average summer ice, around 55% of the total. How did that happen? Where did the extra energy come from?

The amount of ice that melts during melt season has gone from an average of just over 16.000 km3 in the 1980s to just over 18.000 km3 now. That is an increse of 12.5%, and one wonders where that extra energy came from in a sudden jump around year 2007?

As I mentioned elsewhere, the difference between the annual melt and the annual trend is very big - the trend is 280 km3 or 1.5% of the annual melt. The annual fluctuation, from one melting season to another, is at most around 2000km3 on a demi-decadal scale, or a bit more than 10% of total annual melt on average.

In an arctic where ice is on average slowly disappearing, the only definite change we have observed is the steady increase in temperature. All other hypothetical or putative changes such as increase/decrease in cloud cover, oceanic heat transfer, precipitation, storminess etc. have varying degrees of support from the scientists, but none has been quantified as to their total melt input during melt season. But perhaps cumulatively, the non-solar effects are able to cause a 10% demi-decadal variation in annual melt, given the numbers above? Maybe not very big, but definitely not trivial. Non-solar effects include anything that enhances the transfer of heat, including direcly from the sun (with melt ponds and open water playing a very significant part).

(NB these musings are only about melting season. I suspect, as do very many others, that the  refreeze is the more important side of the Arctic coin. But here the physics are harder to tease out, and the relative importance of various effects more difficult to guess. Demi-decadal fluctuation in refreeze seems to be significantly higher, at around 20%. )

[El Cid]

................
Ah - you're making the same mistaken assumption I made years ago.

Even with very high temperatures, truth be told, the direct uptake of heat from atmosphere is comparatively trivial.  The lions share of melt comes primarily from heat transferred directly from water - either melt ponds or ocean itself - which in turn is captured from insolation.
.................

That may be so, but if you look at summer temperatures in the Arctic (defined here as 70-90 N, chart: climatereanalyzer, data: ECMWF reanalysis), you can see that summer temperatures are highly correlated with the September minimum.

The 5 warmest years:

2020 (anomaly vs 1950-80 average: 1,51C)
2019 (1,25 C)
2023  !!!! (1,23C) 
2016 (1,23C)
2012 (1,22C)

If I am not mistaken these are the 5 lowest September minimum years as well based on various metrics.

There is no denying that the trend in summer Arctic temperatures is up but also, it seems that year-to-year variation ("weather") decides each seasons's fate.

[oren]

With all these large storms taking such northernly tracks one comes to wonder,
Hurricane Lee is projected to be back over the water N of 55 monday as a tropical storm.
should we expect this to have impact to the remaining ice?
Is there any historical correlation between north bound atlantic cyclone and late season sea ice loss spikes?
if not is this modeled to change as ocean heat rises?
how far north would a TS level cyclone need to make it to have major impact on the arctic basin?
what about a cat 3?

I've just approved this message by a new member. Also moved it here from the data thread.
Unaffiliated - welcome!

[kiwichick16]

Thanks for that  El Cid   ...... that graph is really quite dramatic    ....... the difference following the onset of the last strong El Nino year in 2015 , if applied to this year , could mean a new record high temperature for the Arctic next year.

[P-maker]

Oren, Thanks

Unaffiliated, nice to see your reflections. I had the same thoughts around 2004-2006, when a massive number of tropical cyclones hit the US - possibly leading to the devastating loss of MYI in the Arctic in 2007.

The actual mechanisms may be more subtle (involving melting of the Greenland icesheet), but your line of thinking is correct. The transfer of large amounts of energy from the subtropical oceans via condensation aloft over the tropical cyclones formations and the advection of this energy to the Arctic regions is by all means an important factor, which needs to be taken into account in the end game.

[IceConcerned]

Thanks for that  El Cid   ...... that graph is really quite dramatic    ....... the difference following the onset of the last strong El Nino year in 2015 , if applied to this year , could mean a new record high temperature for the Arctic next year.

Intersting to ser a 4-year period in maximums since 2013. Coïncidence or pattern ?
With El Nino in full swing WE Can assume a repeat through 2024

[NeilT]

It won't be a bang.  It'll be a sigh.

Except for the extreme events.

But once that happens then it will be a drawn out whimper....

[RoxTheGeologist]

................
Ah - you're making the same mistaken assumption I made years ago.

Even with very high temperatures, truth be told, the direct uptake of heat from atmosphere is comparatively trivial.  The lions share of melt comes primarily from heat transferred directly from water - either melt ponds or ocean itself - which in turn is captured from insolation.
.................

T
That may be so, but if you look at summer temperatures in the Arctic (defined here as 70-90 N, chart: climatereanalyzer, data: ECMWF reanalysis), you can see that summer temperatures are highly correlated with the September minimum.

The 5 warmest years:

2020 (anomaly vs 1950-80 average: 1,51C)
2019 (1,25 C)
2023  !!!! (1,23C) 
2016 (1,23C)
2012 (1,22C)

If I am not mistaken these are the 5 lowest September minimum years as well based on various metrics.

There is no denying that the trend in summer Arctic temperatures is up but also, it seems that year-to-year variation ("weather") decides each seasons's fate.

The water heats the air, not the other way around. My guess is that less ice means that the surface of the water can heat up the air above it rather than the air melting the ice and being buffered at 0°C.

[NeilT]

I always understood that rain was a far greater heat transport event than warm air.

I also understood that direct heating of the ice and the ocean below absorbed more of the solar heat than the air.  This heat then has a greater melting impact than air temp.

Even then the energy transfer at the triple point of water keeps the above ice temperatures of the air very close to freezing.  Until the ice is gone of course.

Did I get that wrong??

[RoxTheGeologist]

That's my understanding too.

Heat transport from the tropics through the oceans and through water vapor in the atmosphere, heating from the sun and the loss of heat (with varying layers of insulation) by LWIR

[KenB]

That's my understanding too.

Heat transport from the tropics through the oceans and through water vapor in the atmosphere, heating from the sun and the loss of heat (with varying layers of insulation) by LWIR

This discussion should probably migrate to: https://forum.arctic-sea-ice.net/index.php/topic,2709.0.html. "Basic questions and discussions about melting and freezing physics"

[oren]

Indeed, it started as an offshoot of the season at hand but became wider and more general.

[interstitial]

Heat flows from hot to cold no matter the medium. So if air is hotter it flows into water. the difference is the amount of energy stored in air is small compared to water so air appears to have little impact on water.