Short to Medium Term Arctic Sea Ice Conditions Discussion

[Pmt111500]

Tried to clean up and remove the clouds off tile r04c04 (resolution 1 km/px) from the area of the cyclone center, result here, http://forum.arctic-sea-ice.net/index.php?action=dlattach;topic=194.0;attach=1945;image

The image is a mess, there's a whole bunch of 'clouds' that might be 'grey ice' or the other way around, the trouble with this way of 'clearing up the skies' seems to be that some of the 'darkest ice' maybe lost if one tries too much to see what's going on under.

[Frivolousz21]

The 00z GFS break's out the Spartan Hammer(Halo reference) and start's swinging it on the Pacific side and the vortex.

It will be interesting to see how this unfold heading into peak insolation. And if the GFS is onto something with this pattern change how it effects the "churned" up ice.

While the Euro supports this. We haven't seen the Euro go for the jugular yet. The GFS beyond 180-240. Strengthens the H5 ridge and Surface HP. And centers it over the Beaufort, extending into the central arctic.

[SATire]

Nightvid Cole, there is really some loss of sea ice area in the CAB on the Russian side - as you can see also in the high res maps from Hamburg: https://sites.google.com/site/apamsr2/home/pngcby32/ -
But SIA in that upper area in the CAB (in direction of Laptev and Kara seas) is in the range of 80% - so the values of ~50% concentration from CT or hycom/CICE look very exaggerated to me. Also by Modis you see some SIA-lowering - but by far not close to 50% water.
Observations seem very uncertain in CAB these days if you find a lot of maps with large areas of 50%  water and if you measure on your own it results in 10-20% water area (80-90% ice concentration) at the same positions. What is the reason for this discrepancy? Alarmists at Bremen or just bad maps? Worst case numbers from Navy? Or bad pictures from all satellites? 

The SIA-loss on the left side of the CAB (towards Canada) still looks quite similar to water vapour artifacts - so we should wait some more days to see if those losses are real.

[Frivolousz21]

The 00z Euro is in and does a 180 on this pattern and takes the melt season into overdrive.

[ChrisReynolds]

SATire, Nightvid Cole,

Over at Neven's blog Lars Kaleschke suspects these apparent open water areas are just melt ponds.

I wait for MODIS to clear. It wouldn't surprise me if there is a lot of volume loss going on in that region right now. We're now entering the period of maximum volume loss.
http://farm4.staticflickr.com/3728/8962488978_7cc9753c53_o.jpg

[jdallen]

@ Chris - I suspect you are correct, and there is still very much potential to pass 2012.  If thickness estimates are close - ~ <= 2.25m or so across most of the basin - that matches up closely with the total melt in the Beaufort last year, if I recall.

Blowtorch plus cyclone ice grinder doesn't leave me feeling optimistic.

[ChrisReynolds]

Lars has no reposted in light of a reply. He thought the area discussed was the CAA. Now he agrees it's open water due to divergence.

[ChrisReynolds]

JDAllen,

I'm very much in two minds about this melt season. I actually think it's shaping up into a more exciting season than 2012, because in 2012 it was pretty obvious a new record was on the cards from early on. It may be worth quoting the text of an email I sent to someone the other day.

QUOTE

....But on the subject of sea ice, which is my main interest. I'm seeing mixed signals.

I know you'll be busy so I'll condense this from what's on my blog. This, for what it's worth, is my current assessment.

This year has been unusually cold in May, which has caused a late start to the melt. This is seen in Anomalies of Cryosphere Today Area (CT Area)
http://farm4.staticflickr.com/3814/8961914348_0b70bd1354_o.jpg

But the anomalies are finally showing the start of what we're calling the June Cliff, more up to date detail of the above graph.
http://farm8.staticflickr.com/7381/8994638997_bbd735c1b1_o.jpg

Although today's data shows a slight uptick for 2013. Note from the above, 2007, 2011 and 2012 all show a June Cliff, and all years were record years (2011 record in some indices only). So whatever the mechanism (I suspect melt ponding plays a role), an aggressive June cliff seems to be related to aggressive ice loss later in the season. Both a real drop in concentration caused by thinner ice after 2010, and melt ponding caused by more flat first year ice, would serve to pre-condition the pack for losses later in the season.

PIOMAS difference between May (average) 2013 and 2012 is here:


There's a large swath across the Arctic that's thinner this year, and unlike last year there isn't a band of multi-year ice (which was in the area of the low concentration region that was hit by the August Storm of 2012). But the US Navy HYCOM system is repeatedly forecasting a massive drop in thickness/concentration in Laptev. Both factors could cause a surprisingly 'enthusiastic' melt in July and August.
http://www7320.nrlssc.navy.mil/hycomARC/navo/arcticictn_nowcast_anim30d.gif
Sometimes as time proceeds HYCOM backtrack on projected loss as new forecasts of weather come in. I think it significant that this is not happening. The take home is - watch Laptev now, and expect a more rapid recession from the East Siberian sea than in 2012.

However PIOMAS volume loss after end June has been pretty invariant*, the post 2010 increase in melt season losses has happened before the June solstice.

This is also seen in the month to month differences of PIOMAS volume (monthly averages).


Note the slope to June losses, but lack of slope in July and August. Why this is happening I am unsure. The June increase in loss is likely due to transition to predominantly thinner first year ice, but the July/August lack of increased loss remains a puzzle. I suspect that typical ice condition in those months is such that the limiting factor is insolation, not ice state.

The late start seems to have eaten into the spring melt, to see a new record we would have to see melt increase after June, which hasn't happened since 2010.

I had been expecting a new crash this year, my estimate had been a pdf peaking around 1.75 to 2M km^2 CT Area, biased towards the high end with a short tail below 1.75M km^2. I now expect 2011 to be beaten, but 2012 may not be beaten.

END QUOTE

*Hat tip to Chris Randles, but that wasn't stated in the email as the recipient wouldn't know Chris.

[Peter Ellis]

I'm not seeing a lot of melt ponding in the main basin yet - even the near-shore parts of the Laptev/Kara/East Siberian seas haven't gone "blue" yet, let alone nearer the centre.  All the mass balance buoys with data are still showing substantial snow cover and temps (just) below freezing still, as are the webcams.

The lack of significant melt ponding yet across the basin implies two things:  firstly that the observed loss of ice area is genuine divergence and opening up of leads, and secondly that the June "cliff" hasn't even started yet.  When it finally does, watch for area to drop unbelievably fast for a week or so.

[Bob Wallace]

but the July/August lack of increased loss remains a puzzle

Perhaps it simply reflects the easier melting of thin, lower density ice early, leaving thicker, denser, more difficult to melt ice a larger percentage of the post-June pack.

[werther]

On the CAB holes…
Reading A-team’s comment on the blog, I can understand his interpretation of the graphic qualities of the different sensors. It is clear that there’s no graphical, reproducible method to get a good, time-lapsed series of pics out of the PAC region.
So what I’m doing on CAD is not reproducible  interpretation through visual scrutinizing of the MODIS tiles. It’s not science. It’s four years of painstaking and maybe self-illusionary sifting through the interesting parts. I’m certainly drawing my CAD lines in an arbitrary way. Anyone of you would fill in differently.
But in CAD I can enhance on unlimited scale into pixel 250x250m detail. What I’ve been counting as polynia’s is open water, for sure. Not melt ponds. They show up much different, within the floe-forms. Blueish or grey. You’ll find them on floes in the Kara Sea, the Baffin Bay, the Chukchi Sea today.
Sure, my ‘open water’ contains debris, even some slushy grey ice or nilas.
But together they characterize an enormous swath from Svalbard right up to the Chukchi/Central Basin limit.
I think the HYCOM model depicts this well in the middle part. With their resolution, I think around 15x15 km, they marginally miss these open polynia’s. Though they’re often larger (400-900 km2), they are diluted like Finnish or Minnesota woodland lakes. HYCOM gets the minimum thickness about 65 cm.  Presuming their method sets off a mean, I hold the larger floes (200 km2 mean) to still get to 1.5-2 m. The grinded rubble in between must be anything between 65 and 150 cm thick.
The ASMR-2 pixel scale is about 9x9 km. That’s just enough to pick up the broader parts of the polynia’s. That’s why today’s map shows 9 of the pixels as open water. It is a good indicator of concentration. But not of the total size of actual open water. NSIDC SIA probably is, I think the 250K anomaly in SIA  for the CAB is a good actual indicator. My guess is that the hole-polynia swath now totals between 10 and 20K open water in the CAB.
I haven’t seen this structural state first decade of June on MODIS before. The continuation of tidal and wind pull all winter into the spring must have become unsupportable for the pack ice as the temps drew closer to zero and the anomalous PAC kicked in. The question now is how fast sunny weather could get the top Ocean layer above -1.5 dC. When it does, to clear the rubble (1Mkm2) takes just days. The larger floes will then float out, to melt before the end of July.
Leaving the 1,8 Mkm2 ‘safe’ pack against Greenland and the CAA?

[jdallen]

Note the slope to June losses, but lack of slope in July and August. Why this is happening I am unsure. The June increase in loss is likely due to transition to predominantly thinner first year ice, but the July/August lack of increased loss remains a puzzle. I suspect that typical ice condition in those months is such that the limiting factor is insolation, not ice state.

Shooting from the hip - lack of slope may indicate that the "low hanging fruit" - ice most vulnerable to melt - was already cleared.  In my estimation, I expect the loss rate would tend to flatten in that case.

A thinner starting pack this year (in fact, in each successive year since ...) wouldn't necessarily mean the shape of the curve would change, but I would expect to finish with lower and lower values until the ice disappears completely. *Then* our curve would alter dramatically.

I expect we should count our blessing for the cool start.  Unfortunately, that doesn't translate into less "low hanging fruit"; rather it just delays when it melts.  Even an "average" July & August, insolation and heat-wise, may drop us to or past 2012.  Think "July Cliff" rather than June.  It will be interesting to see if my hypothesis has any merit.

[Bob Wallace]

A delayed cliff with more late summer melting than usual seems like a real possibility.  Perhaps no new annual record but a recent year record for late summer volume melting if some of the easier to melt stuff sticks for the later months.

[ChrisReynolds]

but the July/August lack of increased loss remains a puzzle

Perhaps it simply reflects the easier melting of thin, lower density ice early, leaving thicker, denser, more difficult to melt ice a larger percentage of the post-June pack.

I don't think so Bob, looking at PIOMAS thickness and the Drift Age Model, doesn't suggest that the thinnest ice in April has melted by July/August. Dr Zhang suggested it was because there was less ice to melt by those months, so melt volume loss was less. This might explain the decrease in August volume loss, but I don't see how it could explain the lack of change in August.

[Bob Wallace]

I'm just kind of throwing stuff at the wall and half asleep, Chris, but is there a chance that the thin ice of July/August is ice that thinned its way from thick earlier on?

One would expect that one meter ice would melt and two meter ice would turn into "new" one meter ice.

[SATire]

The ASMR-2 pixel scale is about 9x9 km.

Werther - the new U-Hamburg concentration maps from the ASMR-2 sensor is a bit larger than 3x3 km2 and the pixel value is proportional to SIA. So we have a very easy tool to check the Navy-concentration maps or U-Bremen/CT-maps (by very simple histogram mean function e.g. in gimp/PS/...). As mentioned above - directly measured SIA in the CAB is significantly higher (~80%) than the values given by CICE/hycom or U-Bremen/CT (~50-60 %) in some areas and possibly in agreement with ocasional observations at some spots through the clouds in MODIS (80-90%). That is realy confusing - probably something is wrong with SIA determination/prediction out there. Would be nice if someone would like to check that simple thing - otherwise I stay feeling a bit lonely.

[wanderer]

Comparing newest navy thickness with same day last year:
So we see that last year melting was ahead in several regions, on the other hand, there's less MYI and we never saw so little thickness so close to the pole over the whole last year.

[crandles]

Dr Zhang suggested it was because there was less ice to melt by those months, so melt volume loss was less. This might explain the decrease in August volume loss, but I don't see how it could explain the lack of change in August.

Why can't it explain the lack of change?

In June, lower volume at start means more volume melts out (thinner ice so higher open water formation causing albedo effect). Note that at start of June in any years not too far apart, area and perimeter are always similar.

Later in the season, the more advanced the melt gets, the less area and perimeter the ice has. Less area and perimeter means less areas where ice can melt. So shouldn't this start to counteract the lower volume causing more volume melt effect in July and even more so in August?


This year has higher area. If that continues to start of July and the thickness is low maybe this year can be used as a bit of a test: Will we get record high volume melt in July?


I suspect ice transport and ocean depth are also involved in the explanation and likely other factors as well. However, if Dr Zhang has suggested that then it may well be the dominant effect.

Does this come to dominate and cause a very long tail?

I don't think so. I think this points towards using Wipneus' trend in thickness by grid cell. But I think the tail in that is too long due to ice transport issues.

( https://sites.google.com/site/arctischepinguin/home/piomas/grf/zero.png
if anyone hasn't seen it.)

[Neven]

Vergent posted and then removed a thread that compared today's Arctic Mosaic with last year's. There's slightly less clouds today and the view is impressive:



Of course, last year was very impressive on the American side of the Arctic:

[Vergent]

r04c04:



r03c04



The Eurasian side of the CAB. This is similar to how r05c03 looked June 25 a year ago.

http://rapidfire.sci.gsfc.nasa.gov/imagery/subsets/?subset=Arctic_r05c03.2012177.terra.500m

http://www.americanwx.com/bb/index.php/topic/35176-the-pacific-side-of-the-arctic-ice-cap-has-desinigrated/

I started a topic about it at American weather. The topic was instalocked at the request of a denialist).

http://rapidfire.sci.gsfc.nasa.gov/imagery/subsets/?subset=Arctic_r05c03.2012227.terra.500m

Fifty days later r05c03 was open water. much of this was MYI.

The Eurasian side of the CAB is mostly FYI, it will be open ocean by Aug. 1.

Vergent

[ChrisReynolds]

Crandles, Bob,

I'll get back to you tomorrow. I've had a very long day and am having the night off.

Crandles,

I really do need to know if you've replied to the emails I relayed to you from Dr Schweiger. Have you replied direct or should I get back to him? If so what should I say?

[ChrisReynolds]

Vergent,

Thanks for that, I'm about to check out the evening's material and now I know where to start. Very impressive.

From what I can see it supports the interpretation that the Bremen low concentration region is the worst in the series of Bremen for June. i.e. it's not just due to difference between AMSR-2 and earlier systems.

[ChrisReynolds]

I'm just kind of throwing stuff at the wall and half asleep, Chris, but is there a chance that the thin ice of July/August is ice that thinned its way from thick earlier on?

One would expect that one meter ice would melt and two meter ice would turn into "new" one meter ice.

Bob, this is mostly the case. If you look at the PIOMAS volume breaksdown I've done it's possible to see the profile of thickness dropping throught the melt season as the thick ice thins and falls into thinner volume 'buckets'.

[ChrisReynolds]

Actually I will post some comments now...

To start with here's a plot of 1980 to 2013 PIOMAS monthly volume changes, calculated as stated month's volume average minus the preceding month's volume average.
http://farm3.staticflickr.com/2810/9034311917_9b56c89a03_o.png
I've done that as a link because the image is necessarily large.

From this the Autumn increase in volume growth can be seen, with a less marked volume growth over the early months of the year. May and June show the marked increase in volume loss in recent years. July shows little change, in fact there is a small increase in volume loss over July. August shows a marked reduction in volume loss during that month.

The following are scatter plots of average monthly CT Area as a function of PIOMAS volume change for the stated month, years 1980 to 2012. Really they'd make more sense with CT Area as the X axis, and PIOMAS volume as the Y axis, i.e Volume change as a function of area. Since it is that interpretation that I'm looking at. But I'm too tired to go ovet them again.

June


July


August


June shows a tight clustering around the trend, the volume loss increases as area reduced. July again shows that as area reduces the volume loss increases. The clustering is less marked, but if you look at the bulk of the plot there's a clear leftward shift as area reduces, i.e. volume loss increases as area reduces.

August shpws the oppositie, as area reduces volume loss reduces. The question is why, scatter plots don't help with this, but if you look at the clustering in the Autgust plot there doesn't seem to be the same tidy shift seen in June and July. In August there seems to be two regions: pre and post 2007. With the 2007 period showing a regime shift to the right. The trend for the pre-2007 period has a slope of -0.142 as against the -1.11 slope of the whole series, so most of the shift in this period is due to the regime shioft post 2007.

Here is a plot of CT Area (month average) and Violume loss for with the scales for volume loss and Area linearly adjusted to superimpose the graphs and keep any linear relationship clear.


What I'm thinking is that if area reducton were the direct cause of the reduction in volume loss in August wouldn't these two plots show similar rates of changes throughout? Or would I really need to plot ice edge? I'll ponder it.

PS, from the reply I got from Dr Zhang it seems the PIOMAS team themselves haven't researched this, the reply was just relaying a first guess.

[Bob Wallace]

This may be well off track.  I didn't know where to look for your thickness binning data or Neven's thickness graph.  I'm assuming that the ratio of thick to thin ice increases later in the year.  If that's not the case then there's no reason to read further.  (There may not be anyway.  ;o)

June ...the volume loss increases as area reduced

As area declines, volume loss accelerates?   

Volume would be leaving the pack on top, bottom and edges.  And those losses would be quickening as air and water temperatures rise.  And a lot of thin FYI is melting at an accelerating rate.

August ... as area reduces volume loss reduces

As area declines, volume loss decelerates?

The pack is relatively thicker, a higher ratio of thick/thin (that's a guess).  Which would mean less top/bottom surface melting per area.

Much of the thinner ice is gone.  A higher ratio of ice protected on all sides.

[iceman]

(per comments earlier by others)  The US Navy HYCOM system still forecasts a drop in thickness/concentration - continuing past the waning of the cyclone - from Laptev nearly to the pole.  Looking at ChrisReynolds' map of PIOMAS difference for May 2013 vs. 2012, the thinning hits the largest area where ice thickness is greater this year.  Most of the other areas of thick ice (relative to 2012) were going to melt out anyway, excepting near Banks Island and scraps here and there.  The blue area north of Franz Josef Land could have been a fastness in 2013, but now it's looking increasingly unlikely.
    Let's see what happens to the Laptev Bite thinning area after the cyclone passes.

[crandles]

Crandles,

I really do need to know if you've replied to the emails I relayed to you from Dr Schweiger. Have you replied direct or should I get back to him? If so what should I say?

Hi Chris,

My emails from Virgin.net don't seem to get through their spam filter. Sorry I haven't replied earlier. So could you forward the following reply from me:

Dr Schweiger,

Thank you for considering accommodating Chris Reynolds, Wipneus and my wishes for the data. Making June thickness data available during July would be great if possible.

I was a little surprised at how high the volume was at 31st May and I haven't done as much as I would have liked testing whether thickness distribution information is better than using Cryosphere Today area information. Nor have I got to grips with May thickness data yet.

Many thanks for making the data available.

Regards
Chris Randles

[crandles]

What I'm thinking is that if area reducton were the direct cause of the reduction in volume loss in August wouldn't these two plots show similar rates of changes throughout? Or would I really need to plot ice edge? I'll ponder it.

Interesting graphs, thank you for posting them.


Just because area reduction might be a direct cause of the reduction in volume loss, I suggest it isn't adequate to look for similar rates of changes throughout unless you can rule out other significant direct causes.

I was suggesting two major direct causes and perhaps others.

1. Reduction in (average?) ice thickness at start of month causes increased volume loss, and
2. Reduced area at start of month causes reduced volume loss.

To see whether that might be the case, don't we need to create some linear function of both direct causes and see if this can predict some of the noise in the observed trend of volume loss (the linear function being optimised to create the general shape of the volume loss).

If that doesn't work well, then there may be more direct causes or maybe the noise is just too great to tell.

There should be some way to rule out direct cause 2 or it isn't a scientific way to conclude that it probably is a factor. I guess if the remaining noise is reduced as much by a random set of random numbers as it is by using area as a cause of reduced volume loss then we should conclude that area as a cause of reduced volume loss is unlikely.

[Juan C. García]

I must ask; is it possible for this upcoming cyclone to melt the vast majority of the pack and leave very little of it for the rest of June, July and August?

fishmahboi, the cyclone is already there and no, it will not melt the vast majority of the pack. I think it will weaken it somewhat.

When I see the NSIDC and Bremen SIE graphs, I think that we are not going to have a new record this year. Even when I see the Chryosphere Today SIA graph, I believe that we will keep 2012 records. It seems that at 2013 the Arctic has more sea ice than any of the years from 2007 to 2012.


http://nsidc.org/arcticseaicenews/

http://www.iup.uni-bremen.de:8084/amsr2/extent_n_running_mean_amsr2_previous.png

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

But when I see the new AMSR2 Bremen image, I believe that at the end, the 2013 persistent cyclone is going to affect (or it has affected) an important part of the Arctic sea ice, at least the one that is between Asia and the North Pole.

http://www.iup.uni-bremen.de:8084/amsr2/arctic_AMSR2_nic.png

So I am going to wait a few days, to think if I will keep the forecast of 1.75-2 million km2 on CT SIA, or lower it even more.

[Bob Wallace]

You're only looking at two dimensions.  If you look at volume there was a bit more sea ice at the end of May 2013 than May 2011 and 2012, but less volume than May of any other year in the record.

(Third line down - looks like a deep blue to my eyes.)

There's still time for a new record or something very short of a new record.  As the climate changes it's looking like the weather is getting stranger.  If this cold persists the ice could live to float another day.  Or this lingering storm could have set it up for a massive meltout.

[wanderer]

Today you can see the blue "scratch" on the Bremen Map on Worldview:
http://earthdata.nasa.gov/labs/worldview/?map=-1497728,-476928,1643904,1773824&products=baselayers,MODIS_Terra_CorrectedReflectance_TrueColor~overlays,arctic_coastlines_3413&time=2013-06-14&switch=arctic

[jdallen]

Looking at the ice across the central arctic in Worldview, I can understand why the Navy isn't changing their HYCOM predictions.

[werther]

ASMR2 today has 38 open sea (less than 10% ice) pixels on it’s map. Coming from 6-8 a couple of days ago. The jpg-map pixel is about 80 km2.
To cross-examine, I took to MODIS. I picked this detail:


Some faint, high clouds stretch over the SW quadrant of the detail. Lots of open water. What looks like a floe (that's why I placed the ?; you can hardly call it a floe...)on the 1km resolution appears to be a cluttered collection of debris.
Reminds me of Wayne’s description of this FYI rapidly being reduced into its original pancake forms of freezing last September/October.

It looks like the briny structure of these bits and pieces is easily falling apart in the new, dynamic properties of the CAB. The role of atmospheric temps hasn’t been very important up to now. As for the top ocean layer, I suspect it is rapidly collecting insolation now. Two days ago on this Forum, I estimated the grinded rubble to be between 65 and 150 cm thick.
Now, I think it’s near the low side of that. When the warmth really kicks in, it will be gone in ten days.

[Apocalypse4Real]

Werther,

The fragmentation is obvious, thanks for your focus on measuring area and extent.

The US Climate Prediction Center is forecasting over a 40% probability of above normal temperatures in Alaska for the next 6-14 days.

Here is an example of the heat already present:

Currently in Prudhoe Bay, AK, US as of 2013-06-14 22:20 UTC:

Partly Cloudy   75°F / 24°C
Feels like 74°F / 23°C   Humidity: 40%
Wind: W 10 mph
Pressure: 30.03 in.
Dewpoint: 49°F / 9°C

It this stays fairly constant the melt in tbe Beaufort will pick up quickly.

[frankendoodle]

ASMR2 today has 38 open sea (less than 10% ice) pixels on it’s map. Coming from 6-8 a couple of days ago.

I've been keeping track of the same pixels that are north of 85N and they are the following: 6/11=8, 6/12=12, 6/13=36 (big jump yesterday)

The jpg-map pixel is about 80 km2.

I assume you're talking about the MODIS image you posted

It looks like the briny structure of these bits and pieces is easily falling apart in the new, dynamic properties of the CAB.

Agreed, most of the floes this year are smaller than 1000km2, in some cases much smaller. And lest we forget that 80% of the ice this year is salty, brittle FYI.

Now, I think it’s near the low side of that. When the warmth really kicks in, it will be gone in ten days.

Yes, SST is already increasing and a few days of clear skies will really make SIA and SIE plummet.

[jdallen]

The temperatures across Alaska and Northern Canada are indeed way up - 5 to 10 degrees above, or more in fact.  Significant portions of the CAA are being exposed to 10-15C temps.

On the Siberian side it is just ludicrous - Koeln showed temps of close to 30C within spitting distance of the Laptev sea.

I've just finished some perusal of MODIS as well.  I'll be more surprised if we don't see massive melt-out, than if we do.

[ClimateChange]

Looks like a big Arctic Alaska heatwave. Thunder was observed in Barrow yesterday... thunderstorms, although still rare, have become much more common in recent years due to global warming. From 1978 through 1999, thunder was observed twice. Since 2000, thunder has been observed six times, including the rare 2000 event in which a thunderstorm moved right over the town. Parts of Alaska also had record heat yesterday and again today, with more in the forecast.

=================================

... Thunderstorm occurred at Barrow...

High temperature of 66 degrees helped fuel
a thunderstorm at 1105 PM akdt until 1132 PM
akdt on June 13.

The last thunderstorm in Barrow was July 15 2012
with a high temperature for the day of 65 degrees.

Over the last ten years most thunderstorms occurred
during July. The year 2003 was the big year for
storms with one occurring in each month of June...
July... and August.

==================================

Wunderground Best Forecast algorithm shows intense heat for the foreseeable future.

For Fairbanks, temperatures are supposed to be in the 80s to around 90 for the next week to ten days.

Similar conditions are expected to the north in Fort Yukon, although Friday, the 21st, is expected to be very hot there. Wunderground Best Forecast is showing a high of 99F that day, just one degree shy of the all-time state record also recorded in Fort Yukon.

http://www.wunderground.com/cgi-bin/findweather/getForecast?query=fort+yukon%2C+ak

[ChrisReynolds]

CT Area Anomalies with data to yesterday.



Currently 0.974M km^2 behind 2012.

I've previously suggested that the June Cliff is due to melt pond formation. A commenter at my blog gives more detail.

Anonymous Anonymous said...
A significant portion of this drop is due to the onset of the effect of the bias error in the measurement methods. Treating fast ice that has turned blue due to some surface melting of the overlaying snow as an abrupt change to 60% coverage, when in fact it remains 100%, inflates the reported rate of change to the ice area and fabricates a "cliff" in the data. The shadows of clouds over the central arctic have a similar effect, making 80-90% ice coverage look like regions of 60% coverage.

It just the time of year the bias errors are most pronounced.

14 June 2013 12:38

All of which seems reasonable. Weather may cause the interrupted 'cliff' this year to happen again. However at present 2012 is following 2009's CT Area profile more than any other post 2007 year. There is a big difference however:

May 2009


May 2013


I think we're looking at a delayed start, not the pattern for the whole season. This does however mean that without something really rather amazing 2012 will not be beaten.

[Apocalypse4Real]

I think the snow drought in the CAA and Siberia this spring is the major contributor to the snow melt anomaly in Siberia, CAA and Alaska, that is now contributing to the heating we are observing in Alaska and Siberia, which in turn is impacting ice melt.

Attached is the UCL May 16, 2013 30 day (April-May) and 90 day (February-May) Arctic Snow Precip anomaly maps and the current snow and ice map from Nasa NEO.

The UCL drought mapping site is: http://drought.mssl.ucl.ac.uk/drought.html?map=%2Fwww%2Fdrought%2Fweb_pages%2Fdrought.map&program=%2Fcgi-bin%2Fmapserv&root=%2Fwww%2Fdrought2%2F&map_web_imagepath=%2Ftmp%2F&map_web_imageurl=%2Ftmp%2F&map_web_template=%2Fdrought.html


The NEO mapping site is: http://neo.sci.gsfc.nasa.gov/Search.html?group=50

[ChrisReynolds]

Just because area reduction might be a direct cause of the reduction in volume loss, I suggest it isn't adequate to look for similar rates of changes throughout unless you can rule out other significant direct causes.

I was suggesting two major direct causes and perhaps others.

1. Reduction in (average?) ice thickness at start of month causes increased volume loss, and
2. Reduced area at start of month causes reduced volume loss.

To see whether that might be the case, don't we need to create some linear function of both direct causes and see if this can predict some of the noise in the observed trend of volume loss (the linear function being optimised to create the general shape of the volume loss).

I'll think about it.

[wanderer]

Ice Thickness comparison for June 14th over the past four years. (sorry, 2010 is one day later)

Feel free to discuss!

[Neven]

I'm loving this melting season, by the way. It has a whole character of its own already. The Arctic just keeps on surprising and surprising. 

[SATire]

Anonymous Anonymous said...
Treating fast ice that has turned blue due to some surface melting of the overlaying snow as an abrupt change to 60% coverage, when in fact it remains 100%, inflates the reported rate of change to the ice area and fabricates a "cliff" in the data. The shadows of clouds over the central arctic have a similar effect, making 80-90% ice coverage look like regions of 60% coverage.

That would nicely explain, why in CT/U-Bremen/Navy maps we find big areas in the CAB with ~50% concentration but in the AMSR-2 maps we did find much higher concentration or mainly "weather-artifacts".
But today we can find several small regions (100 km diameter) with 60% SIA in the CAB (colored circles in the picture). A large part of the CAB between north pole and Laptev exhibits a mean concentration of 80% and I find 145 pixel with concentration < 10% - that is allready 1,400 km2 "open water" in the CAB.
The picture attached to this post is a .png of high res (3.125km pixel) concentration map of the central arctic basin.

Original image source (without the drawings): https://sites.google.com/site/apamsr2/home/pngcby32/

[ChrisReynolds]

SATire,

Not sure I think it does explain it.

Bremen SSMIS...

14/6/2012
http://www.iup.uni-bremen.de:8084/ssmisdata/asi_daygrid_swath/n6250/2012/jun/asi-SSMIS-n6250-20120614-v5_visual.png
14/6/2013
http://www.iup.uni-bremen.de:8084/ssmisdata/asi_daygrid_swath/n6250/2013/jun/asi-SSMIS17-n6250-20130614-v5_visual.png

The patterns this year are very different. I'm on the verge of doing a mid June status post, in 2012 the whole pack was warm anomalies for the first 2 weeks of June. Now in 2013 the last two weeks have shown a -4degC cold anomaly over Beaufort, with less warm across the rest of the pack.

I think the Laptev low concentration patch is real and isn't due to melt ponds. Melt ponds in Beaufort may have been retarded this year due to the cold.

The HYCOM plots Wanderer posts support what I'm wondering, if we'll see a very odd shape to the pack, with substantial recession in Laptev, but less recession in Beaufort as compared to 2012.

I'm totally unable to say how this season will work out, which makes it more exciting than a straightforward rapid melt.

[SATire]

Not sure I think it does explain it.

I am also very far from sure, that the cited explanition explaines it. It may be a possibilty to explain the observation, that CICE/hycom, U-Bremen and CT insisted on showing less concentration in the CAB than we could see. But there may be other explanations, too. If you compare the CICE concentration picture with the picture I attached above, it is quite clear that Navy is doing some strange things. But how to explain that?

I also agree with you, that melting is dominated by Laptev-CAB this year as allready mentioned a month ago due to the early and rapid snow-melt in Siberia.
 

[ChrisReynolds]

SATire,

I think there may be some misunderstanding here. As I explained in this post:
http://dosbat.blogspot.co.uk/2013/05/the-ct-area-june-crash.html

In 2012 when there was such a big drop, this was happening in the CAB region without the ice edge receding in that region. Furthermore the June Cliff happens at the same time as research shows a peak in melt pond area. So normally I see the June cliff as a result of melt ponding. I think the same applies this year but we seem to have less of an area of melt ponding due to colder condtions, hence the short drop in CT Area anomalies, followed by a levelling.

What happens next, e.g. a further crash, I don't know.

[Apocalypse4Real]

SATire,

The patterns this year are very different. I'm on the verge of doing a mid June status post, in 2012 the whole pack was warm anomalies for the first 2 weeks of June. Now in 2013 the last two weeks have shown a -4degC cold anomaly over Beaufort, with less warm across the rest of the pack.

I think the Laptev low concentration patch is real and isn't due to melt ponds. Melt ponds in Beaufort may have been retarded this year due to the cold.

The HYCOM plots Wanderer posts support what I'm wondering, if we'll see a very odd shape to the pack, with substantial recession in Laptev, but less recession in Beaufort as compared to 2012.

I'm totally unable to say how this season will work out, which makes it more exciting than a straightforward rapid melt.

Chris, what you are beginning to notice and articulate is working toward the North American Ice Service initial outlook. The link to the pdf is in the NAIS thread:

http://forum.arctic-sea-ice.net/index.php/topic,368.msg7158.html#msg7158

The pdf link is:
http://www.natice.noaa.gov/pub/special/nais_forecasts/2013/graphics/outlook/nais_seasonal_outlook_2013.pdf

I am not sure I agree with all of their forecast, but the general idea seems to be what we are seeing start playing out.

[ChrisReynolds]

A4R,

I did look at it, but couldn't think of anything to say one problem is that I don't know a lot of the locations as well as I might. The document is obviously aimed at locals with an interest in their areas.

Your comment above about lack of snow causing warming also lines up with what I'm thinking. I did a blog post on the issue a while ago.
http://dosbat.blogspot.co.uk/2013/03/eurasian-snow-cover-and-atmospheric.html

[pdjakow]

Today Alasca Forecasts:

http://meteomodel.pl/BLOG/?p=5530

[Apocalypse4Real]

Chris,

I should have been more specific:

Page 10: Western and Central Arctic

Summer temperature and ice outlook.

Average air temperatures will be slightly above normal over most locations from June to August
except near normal over the Beaufort Sea. In general, forecast ice events are expected to be
slightly earlier than normal over most locations this summer except near normal over the
southern Beaufort Sea and the Central Arctic.

Beaufort Sea south of 75°N.

Very close pack first-year ice with a trace of old ice within 90 miles of the mainland coast.
Beyond 90 miles of the coast very close pack old ice except very close pack first-year ice with 2
tenths of old ice west of 143°W.

Outlook for June…Close to very close pack ice with areas of open drift or less ice forming north
of the consolidated ice along Tuktoyaktuk Peninsula and west of Banks Island during the second
week.

Outlook for July…Close to very close pack ice with large areas of open drift or less ice north of
Tuktoyaktuk Peninsula and west of Banks Island. Ice melting completely within 90 miles north of
the mainland coast east of 145°W and west of Banks Island during the second week. Ice
becoming open drift or less ice within 90 miles of the Alaskan Coast west of 145°W at the end of
the month. Close to very close pack ice north of 72°N.

Outlook for August…Open water within 120 miles north of the mainland coast east of 145°W.
Open drift or less ice within 120 miles of the coast west of 145°W. Beyond 120 miles of the coast
close pack ice.

Outlook for September…Open water within 120 miles north of the mainland coast. Beyond 120
miles north of the coast close pack ice except open drift or less ice west of 145°W.

Page 11:  Arctic Ocean north of 75°N.

Very close pack old and first-year ice.

Outlook for July…Very close pack ice.

Outlook for August…Close to very close pack ice becoming open drift or less ice west of 145°W
during the second week.

Outlook for September…East of 140°W close to very close pack ice. West of 140°W open drift
or less ice.

The report does not cover the Russian or Nordic Arctic.