A wetter and warmer Arctic

Later this week I'll be posting more analysis of the current, very interesting melting season, but here's something that popped up in my mail box via Google Alerts.

It's an article from Alaska Dispatch News on a scientific paper by Linette Boisvert from the University of Maryland and the NSIDC's Julienne Stroeve, based on AIRS (Atmospheric Infrared Sounder) satellite data for the 2003-2013 period, which reveal that the Arctic has become warmer and wetter, lengthening the melting season. I had a short chat with Dr. Boisvert during a poster presentation at this year's EGU General Assembly. This is one of the few papers analyzing AIRS data in relation to changes in the Arctic.

Here's the article:

As sea ice shrinks, the Arctic gets warmer and wetter, study finds

The Arctic is getting warmer and wetter, changes linked to melting sea ice and more persistent open-water conditions - NASA Earth Observatory

The Arctic became warmer and wetter since the beginning of the 21st century, a self-reinforcing trend likely to continue because it is linked to sea-ice melt and more persistent open-water conditions in the world’s northern ocean, a newly published study concludes.

Data from NASA shows that average surface temperatures across the Arctic Ocean increased an average of 0.16 degrees Celsius per year from 2003 to 2013, and air temperatures rose 0.09 degrees Celsius annually over the same period, says the study, published online in Geophysical Research Letters.

The changes weren't evenly distributed, though. They were dominated by large increases in the November-to-April period,during which Arctic-wide surface temperatures rose 2.5 degrees Celsius and air temperatures rose 1.5 degrees Celsius from 2003 to 2013.

November saw the biggest increases in “skin temperature” (defined as temperature at the Earth’s surface), and air temperature, with an average annual rise of 0.42 degrees Celsius on the surface and 0.32 degrees Celsius in the air, said the study, by Linette Boisvert of the NASA-affiliated Earth System Science Interdisciplinary Center at the University of Maryland and Julienne Stroeve, a research scientist with the National Snow and Ice Data Center in Colorado.

The NASA data also showed that moisture flux -- the transport of water vapor from the ocean to the air -- increased in nearly all the months, but grew most dramatically during months of maximum open water. The biggest increase in moisture flux over the time period came in October, followed by increases in September and August, the data showed.

The changes are linked to the sea-ice trends, Stroeve said.

Moisture and temperature changes match up with expanded annual Arctic ice melt, she said.

“The autumn is when we see the largest responses to summer ice loss. The later the freeze-up the more the temperature/water vapor feedbacks will extend into autumn,” Stroeve wrote in an email. “In October there is still quite a bit of open water, but the air temperatures are dropping and so we start to get large feedbacks of amplified warming and more moisture into the atmosphere that extends into November.”

The study is the first to summarize changes in the Arctic using data from a relatively new NASA weather-monitoring system. The data is collected by what NASA calls the Atmospheric Infrared Sounder, or AIRS, which is attached to the agency’s Aqua satellite that was launched into space in 2002.

Read the rest of the article here.

Comments

[Sam]

I have wondered for some time now whether this set of observations combined with the loss of deltaT between the equator and the pole shutting down the jet streams as we know them might offer an explanation of the equable climate quandary.

As heats builds up and the ice melts leaving warmer conditions through the arctic night, it would seem to make sense to me that it may lead to a warm Arctic Ocean in the summer that results in evaporation and cloud formation that acts as a heavy blanket over the arctic through the winter. As summer breaks and the sun returns to the arctic, the clouds warm and disappear. The ocean is then heated through the summer. Winter returns, the atmosphere cools, and condensation clouds return once again to blanket the arctic holding the heat in.

I am not a 'cloud guy', so I am out of my depth here. But it does seem like a plausible logical progression.

Sam

[John Christensen]

From the article:

"The study found that moisture flux timed to ice melt also correlated with increases in heat-holding cloud cover in late fall and mid-winter, reduced cloud cover over the Beaufort Sea in summer along with earlier onset of seasonal melt and later onset of seasonal refreeze, the study found."

It find this item a bit spurious and perhaps a result of analyzing data only for a very short period.

Following the argument of the article and the paper, it should follow that after the record melt in the summer of 2012, the late fall of 2012 and winter months 2012/13 should have higher than normal cloud cover, which you can detect in the air pressure level.
However, as seen here (http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/month_ao_index.shtml), the weather patterns decided to provide mainly high pressure and cloud-free skies in that winter following the record low ice cover, which then allowed ice to rebuild and the summer of 2013 to become a rebound year.

Looking at the AO index going back to 1950, it is very difficult to identify a trend that winters are becoming more prone to low pressures and that summers are becoming more prone to high pressures.
Overall, it seems like we are moving to a pattern with higher incidence of high pressure across the Artic region (See here: http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/JFM_season_ao_index.shtml), which as we know will have opposite impact on the ice, depending on if the high pressure is present during summer or winter months.

[AbbottisGone]

@ sam,

..clouds remain the biggest unknown quantity, it is true!

[bobcobb]

Neven,
Does this mean more methane emissions? I remember it being said that more emissions would come out as methane instead of co2 if the Arctic got warmer and wetter.

[Neven]

Does this mean more methane emissions?

Yes, probably, but not end-of-the-world levels.

[OldLeatherneck]

Neven,

Thanks for posting this. I'd often wondered if anyone had quantified the obvious increasing temps in the Arctic during the dark months. It will be interesting to see how much the current El Nino will impact temps in the Arctic this next winter. If so, will these temp increases have a noticeable impact on ice regrowth in any, or all, of the metrics we use. Might lead to another exciting melt year in 2016.

[bobcobb]

Neven,
Thanks

[UltraVerified]

Off topic here, but this graph seems to indicate a deviation from the "fits and starts" of the total area. That line of descent is remarkably smooth, and the steepness of that descent seems to be stronger than other years. That -1.6M sq.km anomaly is lower than any in the past two years.

Shouldn't it be slowing down a bit by now?

Anyone else notice this trend?

http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.recent.arctic.png

[Jim Hunt]

UV - There's a whole thread dedicated to that sort of question over on the Arctic Sea Ice Forum, including a "prediction" of forthcoming CT area numbers:

http://forum.arctic-sea-ice.net/index.php/topic,1112.msg60480.html#msg60480

[John Christian Lønningdal]

Looks like the Arctic sea ice area is getting very close to last years minimum, 1 whole month before!

http://arctic.atmos.uiuc.edu/cryosphere/arctic.sea.ice.interactive.html

No doubt on track to be one of the 5th lowest areas on record this year.

[Chris Reynolds]

Thanks Neven,

A useful confirmation of a significant factor.

[opensheart]

So just to run some numbers.
If the arctic winter is warming by 2.5 degrees in 10 years, that is .25 degrees C per year.

And if my eyeball, approximation of the Temp north of 80 degrees latitude suggests the average winter temp north of 80 degrees latitude is -30 degrees.

Admitting there is a problem comparing an average warming rate of the whole arctic to such an approximation of just a part of it... but just to do the math.

And assuming the warming rate computed from 10 years would remain constant over a 10 times longer time period...

30 / .25 = 120

And assuming this warming started with the start of the observed window of 2003.

This might poorly suggest that after 2123 the average winter temp of at least the area north of 80 degrees latitude would be above freezing.

Ignoring the possibility that Greenland and/or the arctic landmasses might still generate sub freezing winter temps, one might use this single statistic to make the wildly extrapolated claim that by or after 2123 winter would be no more in the northern hemisphere.

???

[AbbottisGone]

...I hope your figures aren't good !!

[Kris]

Albeit not 'officially' declared as such, the daily Uni-Bremen chart shows Amundsen's route is cleared and free now.

OTOH, the NW-passage is getting more and more blocked by shelves pushed into the McClure Strait [into direction Baffin Bay]. So, it's fair to assume tne NW-passage won't open this year.

[Remko Kampen]

"No doubt on track to be one of the 5th lowest areas on record this year."

3rd, no doubt.
1st, very well possible indeed.

[Kris]

Let's avoid at all costs sheer speculations - it only feeds the deniers' lusts.

The day before yesterday [5.662.981] minus yesterday [5.618.256] gives minus 44.725 square km, far to less to even think about the “1st”

Bottom line:

1st: impossible
2nd: unlikely
3th: slightly possible
4th: possible
5th: about sure.

[Bill Fothergill]

Thanks for this informative article Neven.

At a broad-brush level, this is basically what one would expect from the Clausius-Clapeyron* relationship. However, as well as validating the theory, the "peaks and troughs" detail really did need to be supplied by empirical study - and this represents a good step along the way. (*Spell-check didn't like either of those names.)

One thing I would cavil about is the usage of the term "skin temperature". This is used with an entirely different meaning in David Archer's "Understanding the Forecast". (Archer is a professor in the Geophysical Sciences department of the University of Chicago.)

In Prof Archer's book, "Skin Temperature" is used as the effective surface temperature that the Earth would have in the absence of any atmosphere - in other words, it is basically the solution of the Stefan Boltzman equation for a bare lump of rock with Earth's dimensions experiencing the same so-called Solar Constant.

A corresponding concept, known as the "Skin Altitude" is also introduced in the book, and this roughly equates to the altitude in Earth's atmosphere (aggregated across all latitudes, and all diurnal and seasonal cycles) at which one would expect to experience the "Skin Temperature".

This, in turn, results in the relationship (using Archer's terminology)...

Temp(skin) = Temp(ground) - (Skin Altitude) X Lapse Rate

Life is simpler when we use terminology in a consistent way.

[Remko Kampen]

"Let's avoid at all costs sheer speculations - it only feeds the deniers' lusts."

I don't give a damn about the thugs or their lusts. I will NOT change anything in my life or my speech for them.

SIA per yesterday down to #3 position. The state of what's left in large part to rotten to evade a #1 melt-out year imo - things are moving very fast. NW passage open within days. Don't speculate, see it happen.

Btw, the melting season ends second half of September (in principal), not by the end of the first week of the month like we were used to.

[MCz]

Offtop:
Anyone knows what this lighter/greener color of waters of the Barents sea is? Some microorganisms blossoming due to anomalously warm waters? It started somewhere around mid July (based on Nasa Worldview). Is this a regular phenomenon?

https://earthdata.nasa.gov/labs/worldview/?p=arctic&l=MODIS_Aqua_CorrectedReflectance_TrueColor(hidden),MODIS_Terra_CorrectedReflectance_TrueColor,Reference_Labels(hidden),Reference_Features(hidden),Coastlines&t=2015-08-04&v=1050375.9744836448,-385121.2432147707,2524935.9744836446,416670.7567852293

[Bill Fothergill]

@ MCz

Cyanobacterial bloom perhaps?

@ Remko,

As you rightly observe, Arctic Sea Ice Area on Day 224 (3.587 million sq kms) was indeed 3rd lowest (for that day) in the CT dataset. (Behind 2012 & 2011)

I try not to get too worked up about exact positions, and tend to regard the CT SIA annual minimums as falling into 4 "discrete-ish" blocks.

The main cluster contains everything before 2007. This has a sort of "glass ceiling" - although glass floor might be more apt - at about the 4 millions sq km mark.

Next, clustered around the 3.5 million sq km mark we have 2009, 2013 and 2014.

Near the bottom, straddling the 3 million mark, there is 2007, 2008, 2010 and 2011.

Finally, in splendid isolation - for the moment - there is the jaw dropping 2012, coming in at around the 2 and a quarter mark.

Short of some sort of Divine Intervention, 2015 will kiss goodbye to the 3 and a half cluster in the very near future. (Although if one looks at the forum link provided by Jim Hunt, tomorrow's CT SIA drop will be single digit, and 2007 will return to 3rd lowest as at Day 225.)

I strongly suspect 2012 will not be reached, far less surpassed, this year. However, it is still very much up in the air how 2015 will fare in comparison with that second lowest cluster.

[Remko Kampen]

"I try not to get too worked up about exact positions" - We shouldn't, because margin of error with those numbers is considerable and in part even unknown. But its what we have and what we compare years/days with.
Main thing that can be said is that 2015 is one of the worst, though (not yet) one of the very worst years, and that there is not the tiniest sign the 'Arctic Death Spiral' is in 'hiatus' let alone rebound.

[Jim Hunt]

Kris,

That may be what the Bremen visualisation suggests, but MODIS revealed this yesterday:

whilst the Canadian Ice Service revealed this:

Click the images for a closer look.

Hence my conclusion that:

NO, the Northwest Passage is not open yet. In all the circumstances that answer might change quite quickly though!

[Kris]

Jim wrote:

That may be what the Bremen visualisation suggests, but MODIS revealed this yesterday ...

Any particular reason why MODIS should to be considered as more reliable as Uni-Bremen [or vice-versa]?

Do keep in mind it's [already] 14th August now. So, there still might be a few centuries to come till the end of the month, but that will be [has been] it. Al the more as reanalyser.org doesn't foresee any special event in the next week.

And sure, the melting proces might continue till the end of September, but only at a 5000 @ 15000 square km rate, hence peanuts (as it has been each year hitherto].

I'm sorry, but by making wild assumptions we only are feeding deniers'lusts, and yes, to predict now, on 14 august 2015 that 2015 would be "the first" is a plain wild assumption.

[Jim Hunt]

Kris,

I was commenting on your "Amundsen's Route" post. Did you click through and read my article?

Whether you did or not, please explain how I'm "making wild assumptions" and "feeding deniers' lusts".

To answer your first question, in the absence of clouds MODIS has much better resolution than the Uni-Bremen visualisations, and the CIS look at RadarSat 2 as well.

If you have broadband us mere mortals can look at Sentinel SAR instead:

http://www.polarview.aq/images/106_S1jpgsmall/S1A_EW_GRDM_1SDH_20150813T133306_4EC8_N_1.jpg

[MCz]

Thanks Bill,

Indeed phytoplankton bloom. Happening normally every year. My bad - should have googled it first before asking.

[Bill Fothergill]

NASA have just updated their Gistemp LOTI numbers with the July value. The anomaly for that month is (provisionally) shown as +0.75C. Should this stand, it will be yet another monthly record.

The positions for each of the last 12 months (highest = 1st) in the LOTI dataset now read...

Aug 1st, Sep 1st, Oct 1st, Nov 9th, Dec 1st=, Jan 2nd
Feb 2nd, Mar 2nd=, Apr 4th, May 2nd, Jun 1st, Jul 1st

The rolling 12-month anomaly has now reached +0.8C, edging out the +0.79C equivalent figure first reached in March, then repeated in June this year.

By way of comparison, the highest of the rolling 12's prior to 2015 ended in May 2010 at +0.74C Each of the seven 12-month periods completed so far this year have been above this value.

We are now 2/3 of the way through the 2015 meteorological year, and the anomaly is presently +0.11C in excess of the current highest value of 0.73C, shared by 2010 and 2014.

The Nino 3.4 figures should be updated by NOAA in the very near future: I suspect this could be interesting reading.

[G man]

@ Bill. Interesting you bring up the newest GISS-LOTI temps. The latest UAH has only .18C anomaly for July which further illustrates the ever expanding gap between both satellite/radiosonde temps data sets and all of the surface data sets. According to the laws of physics and the IPCC the lower tropo should be warming 20% faster yet this latest GISS-LOTI shows the surface warming 400% faster!

[Werther]

Oh, please...
Not Christy and Spencer again.

[G man]

Do you also have a problem with Carl Mears?

[Neven]

It's clear that it's the satellite measurements that have the most problems. No wonder, as so many things can go wrong.

And now back to the Arctic, please, where July was megahot too. My post on that is going live within the next hour.

[Bill Fothergill]

@ G man & Werther

The RSS TLT figures for July are also pretty divergent from the non-satellite based measurements, so it's not just Spencer and Christy.

Many readers will recognise the names Carl Mears and Frank Wentz, both from Remote Sensing System. Mears and Wentz were largely instrumental (pun intended) in showing up some of the original deficiencies in the work being done by Spencer and Christy.

For an insider's viewpoint on the "hiatus" - be it real or otherwise - there is an interesting blog piece from Sept 2014 by Carl Mears on the RSS website...
http://www.remss.com/blog/recent-slowing-rise-global-temperatures

One will often get clowns suggesting that the satellite record must inherently be more reliable owing to the fact that they - satellites - utilize highly accurate "platinum thermometers".

For an indication of just how gross an over-simplification this represents, RSS also have an overview of Brightness Temperature that should be mandatory reading for anyone ranting about the accuracy of the satellite temperature record.
http://www.remss.com/measurements/brightness-temperature

[G man]

[Thanks for your concern, but there's too much going on in the Arctic right now to tolerate trolling diversions; N.

Note to Bill and others: DFTT]

[NeilT]

there are now some quite significant outlying areas which have resisted melting out. I'm guessing that a lot of the area final figures will depend on what they do.

Whether we drop below 3m this year could heavily depend on these outlying areas and how long the melt season goes on.

My main concern about this is that those figures for that outlying ice will belie the very real impact happening to the CAB and the CAA today when looked at as a statistic rather than as an indicator.

[D_C_S]

G man:

UAH anomalies and GISS anomalies use different baselines. GISS uses 1951-1980. Satellites hadn't even been invented in 1951. The UAH baseline starts in about 1979, close to when the GISS baseline ends. Directly comparing the UAH anomaly to the GISS anomaly is entirely improper

[P-maker]

More wood on the fire:

New Japanese study is out (abstract only): http://www.sciencedirect.com/science/article/pii/S1873965215300050

Using a high resolution (60 km grid) global model, observed SSTs and a robust SST ensemble for the future under an A1B scenario, they find that enhanced advection of moisture to the Arctic is going to increase even further.

It says in the abstract, that: “The increases in PAVE, SDII, and R5d can be partly attributed to an increase in water vapor associated with increasing temperatures, and to an increase in the horizontal transport of water vapor from low to high latitudes associated with transient eddies.”

As is clear by now, the early autumns (Aug-Oct) in the Arctic will be warmer and warmer and if the tropics at the same time get colder and colder due to more intense tropical cyclones, we may soon see the equable climate conditions unfold.

[John Christensen]

From the abstract of the article and as mentioned above:

"..evaporation rates (i.e., moisture flux) increased between August and October.. increasing the water vapor feedback and cloud cover."
"..warming skin temperatures and radiative responses to increased water vapor and cloud cover in autumn delay freeze-up."

As I mentioned in my entry above, this argument does not seem to be well supported by data, which indicates that the chance of cloud cover in the Arctic region is governed by the atmospheric circulations as expressed in the AO index.

I have used the monthly AO index values from the NOAA site and calculated the averages for each calendar month for the whole period as well as per decade to spot any trends (http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/monthly.ao.index.b50.current.ascii.table)
NOTE: With random weather each decade should have a chance of having the most extreme value (positive or negative AO) in more than three months of the year (3.42 to be exact out of 24 occurrences)

January:
Average (1950-2015) -0.39
Average (1986-1995) 0.70
Average (2006-2015) -0.15
Average (1996-2005) -0.37
Average (1950-1959) -0.47
Average (1966-1975) -0.65
Average (1956-1965) -0.89
Average (1976-1985) -0.99
Comment: January is typically dominated by high-pressure (HP) and clear skies (-0.39).
The period of 1986-95 was extreme with a very positive AO index value (0.70) indicating overall pervasive low-pressure (LP) across the Arctic region in this decade.
2006-15 had overall clear skies, but less so than the long-term average.

February:
Average (1950-2015) -0.43
Average (1986-1995) 0.22
Average (1996-2005) 0.14
Average (2006-2015) -0.38
Average (1950-1959) -0.67
Average (1966-1975) -0.75
Average (1976-1985) -0.79
Average (1956-1965) -0.96
Comment: Again a month dominated by HP with 1986-95 as the main decade deviating, followed by 1996-2005, while 2006-15 was very close to long-term average and had overall HP across the Arctic.

March:
Average (1950-2015) -0.16
Average (1986-1995) 0.80
Average (2006-2015) 0.22
Average (1966-1975) -0.22
Average (1996-2005) -0.35
Average (1976-1985) -0.38
Average (1950-1959) -0.65
Average (1956-1965) -0.75
Comment: Less pronounced HP in long-term average, again 1986-1995 as extreme LP decade, but also 2006-15 as mainly LP.

April:
Average (1950-2015) 0.12
Average (2006-2015) 0.57
Average (1966-1975) 0.09
Average (1976-1985) 0.07
Average (1956-1965) 0.06
Average (1986-1995) 0.04
Average (1996-2005) -0.02
Average (1950-1959) -0.11

Comment: April is the month of the whole year with overall lowest air pressure in the Arctic region.
2006-15 saw stronger LP values than any other decade, but not as strong as 1986-95 for the months of January and March.

May:
Average (1950-2015) 0.01
Average (2006-2015) 0.20
Average (1996-2005) 0.15
Average (1956-1965) 0.09
Average (1986-1995) 0.09
Average (1966-1975) 0.06
Average (1976-1985) -0.16
Average (1950-1959) -0.38

Comment: A month on the balance, but again with 2006-15 with more pronounced LP areas than in other decades. This month had very small deviations between decades.

June:
Average (1950-2015) -0.01
Average (1966-1975) 0.14
Average (1986-1995) 0.11
Average (1976-1985) 0.07
Average (1996-2005) 0.01
Average (1956-1965) -0.09
Average (2006-2015) -0.20
Average (1950-1959) -0.30
Comment: Another month on the balance with 1966-75 having stronger LP areas than other decades and 1950-59 having the most pronounced clear skies.

July:
Average (1950-2015) -0.13
Average (1966-1975) -0.01
Average (1986-1995) -0.04
Average (1996-2005) -0.04
Average (1950-1959) -0.08
Average (1976-1985) -0.17
Average (1956-1965) -0.32
Average (2006-2015) -0.36
Comment: After April (mainly LP areas) and May and June in neutral positions, July typically is dominated by HP areas, where 2006-2015 had the highest value, closely followed by 1956-65.

August:
Average (1950-2015) -0.14
Average (1986-1995) 0.13
Average (1996-2005) -0.03
Average (1976-1985) -0.04
Average (2006-2015) -0.20
Average (1966-1975) -0.23
Average (1950-1959) -0.26
Average (1956-1965) -0.50
Comment: August on average is similar to July and saw 1986-95 as the only decade dominated by LP areas (as also seen for many other months).
2006-15 is right in the middle near the long-term average.

September:
Average (1950-2015) -0.02
Average (2006-2015) 0.17
Average (1986-1995) 0.00
Average (1996-2005) -0.02
Average (1966-1975) -0.04
Average (1976-1985) -0.06
Average (1950-1959) -0.10
Average (1956-1965) -0.19
Comment: A month on the balance as seen for May and June also. 2006-15 had the strongest LP areas of any decade. While all of the decades are very close together, maybe this is the indicator showing the increased moisture flux? If so, the signal is not strong compared to deviations seen for other decades in other months.

October:
Average (1950-2015) 0.00
Average (1956-1965) 0.34
Average (1986-1995) 0.21
Average (1950-1959) 0.20
Average (1966-1975) 0.04
Average (1976-1985) -0.09
Average (1996-2005) -0.18
Average (2006-2015) -0.28
Comment: A fourth month on the balance and interestingly had 1956-65 with strongest LP areas.
Also very interesting is that 2006-15 saw the strongest HP areas/clear skies, which seems to fully contradict the argument from the article about increased cloud cover in autumn months..

November:
Average (1950-2015) -0.10
Average (1986-1995) 0.40
Average (2006-2015) 0.34
Average (1976-1985) -0.03
Average (1966-1975) -0.08
Average (1996-2005) -0.20
Average (1956-1965) -0.66
Average (1950-1959) -0.67
Comment: November turns back trending toward clear skies, but with strong deviations between decades. 1986-95 reappears with strong LP areas followed by 2006-15.

December:
Average (1950-2015) -0.19
Average (1986-1995) 0.37
Average (2006-2015) 0.01
Average (1966-1975) -0.11
Average (1950-1959) -0.20
Average (1976-1985) -0.36
Average (1956-1965) -0.49
Average (1996-2005) -0.53
Comment: Winter is approaching, so December has slightly stronger HP areas in the long-term average. 1986-95 again showing strong LP areas with 2006-15 on the balance.

All Months:
Average (1950-2015) -0.12
Average (1986-1995) 0.25
Average (2006-2015) -0.01
Average (1996-2005) -0.12
Average (1966-1975) -0.15
Average (1976-1985) -0.24
Average (1950-1959) -0.31
Average (1956-1965) -0.36
Comment: For all calendar months the long-term average shows a tendency towards HP areas, where 1986-95 deviates most significantly with overall strong LP areas and cloud cover. 2006-15 is seen as overall on the balance between negative and positive values.
Overall, it seems like atmospheric patterns and multi-decadal oscillations are the main forces to consider, when determining the chance of cloud cover or clear skies in the Arctic. The decline in the Arctic sea ice area and extent cannot explain why the decade of 1986-95 had such strong and persistent periods of low pressure across the Arctic region.
Secondly, it would seem to be clear that beyond the short-term significant impact of clear skies and cloud cover on the ice (Compare weather in July 2015 to July 2013) that on an annual scale the ice will fare better, when the atmospheric patterns and oscillations provide high pressure areas across the Arctic region and that the longer term trend in the AO index therefore will either assist in reducing the Arctic sea ice cover or keep the ice with us a bit longer..