Latest PIOMAS Volume update

[Wipneus]

Updated Fram volume export graph.

[Wipneus]

Some people have good use for the updated regional data files.

daily:
https://sites.google.com/site/arctischepinguin/home/piomas/data/PIOMAS-regional.txt.gz

[oren]

Thank you Wipneus, as usual, for providing the PIOMAS data in digestible files. Really appreciated.
Here are some regional graphs.
The CAB is still rather high, though it seems the difference has shrunk a little. The CAA is playing catch-up, though still above the main group. The Beaufort is very much higher in relative terms and made relatively little progress, though it's still on a track similar to 2015 that could still take it to near-zero volume at season's end. The Chukchi continues to track relatively low, while making rather slow progress.

[Phoenix]

Oren, the black line for 2020 on the Beaufort graph terminates at end of May.

[oren]

OTOH, the ESS continues to track in uncharted lowest on record territory - 8 days before previous leader 2017 and 13 days before 2019. The Laptev is lowest on record too but the advantage is smaller. The Kara managed a reversal of fortunes and in three weeks went from momentary lowest on record to a rather mediocre 6th.
On the other side of the equation, the Greenland Sea still flirts with highest on record territory, despite relatively low export in the recent past.

[oren]

Thank you Phoenix, corrected.

[jdallen]

<snip>
The Kara managed a reversal of fortunes and in three weeks went from momentary lowest on record to a rather mediocre 6th.
<snip>

The Kara will catch up when the eastern portions near Severnaya Zemlya start to break up.  With the Laptev so hot, that's only a matter of a few days to weeks I suspect.

[oren]

From the piomas gif (in the piomas thread) by wipneus, we can see large amount of thinning by up to a meter, some spots with up to a meter and a half, and even a channel in the caa where ice has vanished when it was at least two meters thick, in two weeks time! Obviously there has been weather favourable to very favourable for melt and we are close to the insolation peak, but this was up to the 15th, while now ice losses have picked up, at least from a two dimensional perspective. If the weather in the caa maintains itself we could definitely see a much bigger clearout than last year, especially in the southern part, and while some here do not believe that the cab is at risk due to not so bad weather overhead for its ice, the rapid loss of ice recorded in the other central seas makes me think that it is in trouble, dispersion and export will be brutal this year, and the rate of thinning leads me to believe there is not much ice that is safe.

Cross-posted from the melting season thread.
I believe PIOMAS is suffering here from its assimilation of sea ice concentration data. There is so much melt ponding on the ice that the CAA channel mentioned above (is that Coronation Gulf?) appears as open water to the microwave satellites. However the reality is that the ice is still there, as can be seen on Worldview. Thus in this case PIOMAS is underestimating volume in that section.
I do agree ice is not in the greatest shape, and with the heat coming up it could further deteriorate. But it has not disappeared in 2 weeks.

[Phoenix]

We are at the portion of the year from day 165 - 210 where the top years haven't shown a great deal of variation in the CAB. Maybe 2020 can beat that pattern?

I think a beginning point of projection for the minimum might be:

CAB          4.2k km3
Greenland 0.3
Beaufort   0.1
CAA         0.1
               _________
Total        4.7k km3
               _________

Many thanks again to Wipneus, Oren and Gerontocrat for providing the information here which helps everyone to try and make sense of what's going on.

[Aluminium]

Changes from May 31.

Barents ... -41%
Laptev ... -34%
Baffin ... -31%
Hudson ... -30%
Chukchi ... -28%
CAA ... -21%
Beaufort ... -20%
Kara ... -20%
ESS ... -19%
Greenland ... -7%
CAB ... -7%

[ajouis]

Thank you Oren for grounding the model in reality, I thought that as it was showing actual open water, as opposed to black, the southern spot on the bend would be partly ice free, but apparently not, still not good for the long term though

[Phoenix]

Chart which quantifies daily Fram volume export.

Someone with the initiative and tools can take the data used to prepare these charts and construct a range of historical outcomes of this particular type of export based upon the time of year.

This can transform the discussion from being completely qualitative to at least somewhat quantitative.

Perhaps Wipneus or UW also has the information necessary to measure Barents export?

Note: I'm not debating the outlook at this point. Just suggesting how someone who is inclined to educate ASIF about measuring export ranges might proceed.

<Moved from the main thread. O>

[oren]

Despite being tired of questions posed with not enough effort invested in finding answers, I browsed back pages of this thread, and stumbled on a chart prepared by the relentless Wipneus. Click to enlarge.

[gerontocrat]

Despite being tired of questions posed with not enough effort invested in finding answers, I browsed back pages of this thread, and stumbled on a chart prepared by the relentless Wipneus.

So all you've got left to do (ha ha) is calculate the area under the curve to find the KM3 actually exported for selected time periods.

Dead simple (ha ha again),
- choose your preferred y= f(x) to fit OK the average data line & then integrate to find the  area under the curve for any period you want,
(or just sum the values of each day calculated by that formula for any period you want),

or do it the way we had to do it before PC's.
- put the graph onto a piece of A3 paper, put a vertical line for each day (i.e. transform the line graph into a column graph) graph.
- Add lots of horizontal lines,
- manually plot the values for each day.
- sum the daily values for your chosen time period.

ps:- damned if I'm going to do it.

[oren]

This chart is monthly so no need to integrate. It only goes up to June 2017 however.

[gerontocrat]

This chart is monthly so no need to integrate. It only goes up to June 2017 however.

Misread the Y axis description, thought it was the monthly rate day by day as in the graphs Wipneus puts into the current PIOMAS threads.

[Phoenix]

C'mon, not enough effort? That's not nice.

Let's review the thought process which got us here.

1) The initial observation which led to the question about the quantity of summer export export comes from your CAB volume charts. The observation that CAB volume loss is relatively stable during days 165-210 leads one to question how material and variable export during this period can be.

2) I did find Wipneus' Fram volume loss charts for the last three years and from this limited sample came up with an estimate of 25 k3 / month which I posted in the melting season thread.

3) I then asked in the stupid questions thread about efforts to quantify export and Rod posted a study which was more detailed and gave rise to a similar calculation as in #2.

4) Now you have posted another chart which seems to corroborate the numbers in #2 and # above. A typical year seems to export ~ 100 km3 through Fram in the summer.

It's a little disappointing that at no point is there any appreciation for the simple value of a good question. It was not zero effort to arrive at this point. It involved some thinking and resourcefulness and relied upon information that you are presenting in order to make the connection. 

[oren]

I do appreciate good questions and have engaged your posts from the start. Even complimented you initially on your good questions, before you started hammering your insights repetitively all over the forum, which I hope will improve as time goes by (and enough moderation).
But in this case my criticism stands - I think you should browse back, at least through this thread, over past melting seasons, see the charts, see the predictions and outcomes, play the animations, and rely less on information posted by others on the fly. Your opinions and theories will be more informed and more tempered.
You were asking for a chart that quantifies daily export and shows it historically, turns out such a chart was hiding in this thread. With some effort you would have been the one to find it and repost it, rather than just ask for it and hope someone would do it.

[Phoenix]

You were asking for a chart that quantifies daily export and shows it historically, turns out such a chart was hiding in this thread. With some effort you would have been the one to find it and repost it, rather than just ask for it and hope someone would do it.

So we can again agree to disagree on approach.

I knew what I was looking for and decided to first see if anyone was familiar with what I was looking for. My approach was to ask for help and leverage the knowledge base of the community. This yielded immediate results in the form of the research paper that Rod shared which I may never have found on my own.

You obviously prefer that people spend time digging through 3,000+ entries of a thread which may or may not yield the answer or which someone may miss from brain fatigue. There is a lot of interesting history in this thread, but it is also mind numbing to try and go through it all looking for a needle in a haystack.

[Wipneus]

PIOMAS has updated the gridded thickness data. Last day (day 182, 30th June or 1st July) the calculated volume was 12.53 [1000km³], which is the fourth lowest for day 182.

Here is the animated thickness map, which seems to start only after a click.

[Wipneus]

Volume and volume-anomaly graphs. Click for size.

[Wipneus]

Fram volume export. The monthly "normal" line has been blown up, hoping it makes it a bit easier to spot.

[Wipneus]

The thickness map for day 182, compared with previous years and their difference. All need a click for size.

[Wipneus]

Some people have use for the updated regional data files:

monthly:
https://sites.google.com/site/arctischepinguin/home/piomas/data/PIOMAS-regional-monthly.txt.gz
daily:
https://sites.google.com/site/arctischepinguin/home/piomas/data/PIOMAS-regional.txt.gz

[oren]

Thanks a lot Wipneus.
This update has been very bad for the ice. The catastrophe in the Laptev and ESS continues, with crazy record lows. The Laptev is leading by nearly two weeks compared to previous low years while also at record area low. The ESS is leading by a full week, despite dispersion that stopped its area losses temporarily.
The CAB managed to get closer to the leading pack (despite loud claims here that this never happens). CAB losses were record high for the second half of June, with 142 km3 more than 2019, 188 km3 more than 2011, and 227 km3 more than 2012.
The CAA also had record losses for the period, with 120 km3 more than 2011, 235 km3 more than 2019, and 279 km3 more than 2012.
Click to enlarge the images.

[oren]

On the other side the Beaufort stands proudly, with very high volume at 540 km3 more than 2019 and 462 km3 more than 2012. I don't expect this advantage to disappear at season's end, although it will certainly shrink.
Greenland Sea is also at near record high, with 138 km3 more than 2019 and 188 km3 more than 2012, this despite a recent lull in exports.
The other saving grace for 2020 is an advantage in the peripheral seas, similar to 2019 and 430 km3 more than 2012. However, this will not last. The chart shows Barents, Baffin and Hudson.
With the continued heat in the CAA, the HP over the CAB, ice at low albedo and export expected to pick up, it is quite obvious the first half of July will bring another bad volume update.

[Phoenix]

The CAB managed to get closer to the leading pack (despite loud claims here that this never happens). CAB losses were record high for the second half of June, with 142 km3 more than 2019, 188 km3 more than 2011, and 227 km3 more than 2012.

That's very impressive. 2020 beat a decade long trend.

[Aluminium]

Changes from June 15.

Laptev ... -64%
Hudson ... -58%
Baffin ... -46%
CAA ... -45%
Kara ... -44%
ESS ... -41%
Chukchi ... -34%
Barents ... -34%
Greenland ... -21%
CAB ... -15%
Beaufort ... -14%

[gerontocrat]

A bit late but here is the PIOMAS June tables & graphs analysed as in my posts for JAXA sea ice extent.

PIOMAS / Wipneus Data - Volume as at 01 Jul 2020 12.532 '000 km3

Projected minimum from remaining volume loss on the last 10 years 4.095 '000 km3.

If you look at the daily change graph you will see a major increase in volume loss in the days around the solstice. Extent and area losses very high in July so far. If this is replicated in the volume data......

[interstitial]

Aluminum: I like this graphic its informative. I have some suggestions take them or leave them. A background of a glacier or some other scenery might be nice. A ring to indicate average at the time might be nice. Maybe one to indicate record low. Any way keep up the good work.

[Phoenix]

If you look at the daily change graph you will see a major increase in volume loss in the days around the solstice.

Anyone have any idea what might have caused this?

edit: check out the surface temperatures back then. very high.

https://earth.nullschool.net/#2020/06/18/0600Z/wind/surface/level/overlay=temp/orthographic=-111.27,63.89,584/loc=151.837,74.394

[oren]

I am offering a reward to any reader who can actually make sense of this visualization...

The question of location of the anomalous losses bothered me enough to generate a chart of the daily volume loss by sector:
CAB
CAA
Pacific - Beaufort and Chukchi
Siberian - ESS and Laptev
Atlantic - Kara, Barents and Greenland Sea
Peripheral - all the others
Total

Due to scaling the Total is on the secondary right axis, along with a 10-year average (hat tip to Gerontocrat).
The horizontal axis is the day of the year. Day 170 this year is June 18th.

The lines are quite erratic, but it seems regions of anomalous loss were the CAA, the CAB and the Siberian sector, in that order. The CAA behavior was probably due to some grid cells appearing to be open water in NSIDC data. The CAB may have been dispersion-driven export or some other effect of the cyclone. The Siberian sector was very warm.
Click to enlarge.

[Phoenix]

Check the temperatures for the 4-5 days in the link I sent you. WAA made it to the CAB in several places.

[Phoenix]

For what it's worth gentleman, this daily regional volume fluctuation is IMO, the best possible incremental piece of information to aid in understanding what is going in the Arctic.

If we get a chance to see it for the period from July 1-15, it's going to provide a lot of clues about what drives ice loss.

The insolation in this period being so constant will mean the fluctuations in daily ice loss will likely be attributable to other factors which can be investigated and a better quantitative understanding could emerge. I think it's a rare opportunity that such an important variable remains so constant.

Just speculating, but if the powers that be ever decided to release regional PIOMAS figures on a daily basis, it would probably transform the discussion around sea ice and put the 2D measures in a lesser role.

[oren]

Remember PIOMAS is after all just a model, with its own limitations. The more one zooms into it the more one will find weird behavior, I am quite sure.
The Powers that be are The Polar Science Center, specifically Dr. Zhang. He was kind enough to increase releases (model runs, I assume) from monthly to bi-weekly at some point. I doubt he has the resources to make daily releases.

[oren]

Belatedly, here's the PSC monthly update, along with one of the figures (click to zoom):

June 2020 Monthly Update

Average Arctic sea ice volume in June 2020 was 16,600 km3. This value is 1200 km3 above the  record minimum  value of 15,400 km3 set in 2017, making it the sixth lowest on record.  Monthly  ice volume was 46% below the maximum in 1979 and 31% below the mean value for 1979-2019. May  2020 ice volume falls 0.3  standard deviations below the trend line. While the monthly average volume was only the 6th lowest on record, ice loss during June progressed much more rapidly bringing daily volume anomalies for June closer to record years of 2017 and 2019 for this month (Fig 4). Average ice thickness is in the middle of the pack for the more recent years  (Fig 5). Ice thickness anomalies for June 2020 relative to 2011-2018 (Fig 6) continue the pattern that has emerged over the winter, spring and shows relatively thin ice along the Russian Coast and thicker than normal sea ice in the Barents sea. There are some  fairly strong positive anomalies along North America and Greenland and extends further into the Barents sea. This thickness anomaly pattern is likely a combination of the recent warm temperatures along the Siberian Coast  and the sea ice drift pattern that pushed  sea ice away from the Siberian Coast and towards North American and Greenland Coast that has persistent since January (Fig 8.).  The mean circulation pattern for the first half of 2020 shows a very small Beaufort Gyre and transpolar drift stream located closer to North America than normal. The drift anomaly shows a counter clock wise pattern compared to the normal counter clockwise pattern of sea ice drift. 

[oren]

Here's a diagram showing the total number of days each year spent in record low and near-record low volume territory (the ranking is according to the current data, not the data known at the time).
2017 is 1st-4th a whopping 296 days out of the total possible 365, basically everything but Sep and Oct.
2011 is 3rd or 4th for 251 days, but has no record daily low.
Click to zoom.

[Phoenix]

I think it's well understood that PIOMAS is a model. Perhaps not so well understood are the historical steps taken to validate the model vs. physical outcomes.

At any rate, the first half of July should at least provide an opportunity for greater insight into the variables driving the model and the thinking of the experts at the Polar Science Center.

I'd be curious to know if they are only doing model runs 2x a month. I'd have guessed daily runs since they apparently are providing daily results. It's not immediately clear why it would be cheaper to run 15 days all at once, but I'm no expert in that.

It might be neat to someday have a PSC model expert conduct an AMA session (ask me anything) here at ASIF.

[interstitial]

I wonder how well Piomas models current condition in the arctic. They still run it but most of the work was done on it fifteen years ago. That is before much of the dramatic changes occurred.

[Phoenix]

I wonder how well Piomas models current condition in the arctic. They still run it but most of the work was done on it fifteen years ago. That is before much of the dramatic changes occurred.

The laws of physics and chemistry have not changed in recent years. Something like an abrupt change in atmospheric aerosol concentration might throw them off though....a 2020 wild card.

iirc - There was a notable recalibration of PIOMAS numbers ~ 7 years ago.

[oren]

In 2-3 days we should have another PIOMAS report, thanks to the tireless Wipneus. I looked at mid-July losses of past years in the CAB (days 182 to 197 to match this year
s numbering). Many years managed to lose above 1300 km3 for the period (see table below), but none managed 1400 km³. Will 2020 be the first? Considering the persistent high pressure, clear skies and low albedo, one should expect so. Will it manage 1500 km³? Who knows,
With 2020's CAB volume lagging 562 km³ behind 2019 and 340 km³ behind 2012, this is a major chance to catch up. Otherwise, a new volume record might be very hard to achieve.

2007      -1120
2008      -1224
2009      -1330
2010      -1310
2011      -1365
2012      -1384
2013      -1296
2014        -951
2015      -1355
2016      -1351
2017      -1198
2018      -1209
2019      -1343

[Phoenix]

That's awesome of you to share that data in advance Oren. I think this might be the most anticipated PIOMAS data dump of all time.

It would be equally awesome if there were a daily volume data dump like gero did for June. So much to learn.....

[Glen Koehler]

   Even though it is a short range of dates comparison between, it is surprising to see the 2007 value lower than the others (except for 2014). 

   2007 was a true blow-out year, especially for MYI ice reduction, from which the Arctic has never really recovered.  It would stand out above 2012 except the freakish GAC that pushed 2012 into new territory. 

    As for 2020, if this year's May and June melt pond conditioning and the current July roast-a-dome don't produce a 1400 km3 drop, it seems like that could only be because there is less ice to melt.

[interstitial]

I wonder how well Piomas models current condition in the arctic. They still run it but most of the work was done on it fifteen years ago. That is before much of the dramatic changes occurred.

The laws of physics and chemistry have not changed in recent years. Something like an abrupt change in atmospheric aerosol concentration might throw them off though....a 2020 wild card.

iirc - There was a notable recalibration of PIOMAS numbers ~ 7 years ago.

Yes but the model is not based on first principles(physics and chemistry). One major shortcoming is the ocean circulation is driven by forcing on the Atlantic side not first principles. The layer of water below the ice in each grid is uniform and featureless circulation is based on a fairly static model. That means not based on first principles but simplifying approximations. As the arctic has shifted from one with a large fraction of multiyear ice to less than 2% those simplifying assumptions probably don't work as well. Any changes have to applied retroactively for the comparisons between years to remain valid. Without just making stuff up because "it works". One has to figure out what has changed why that does what it does and either come up with a new approximation that works as well for older dates as it does for newer. Improving a model focuses the builder on what is the most important factors. The other option is to only work from first principles. That takes more computing power even then smaller and smaller grid size may be needed to  get good results. 

It has been updated more recently. One thing I do know is professors get busy and most of there focus is based on teaching and funded research. Professors focus on one and deal with the other. All that makes currently unfunded projects less of a priority. Creation of the model and major updates were likely funded projects. A graduate student working on it would have focused most of their time on it. Updating the model twice a month probably has more to do with minimizing the amount of work maintaining the model than anything else. The computing power is most likely a stand alone machine. Anything more would require continual funding.

Compare the to Hycom which has continual funding and full time staff. It is also run on a supercomputer I believe it is a dedicated one but I am not certain on that. When something goes wrong on a date they do not do the full 8 day forecast they just run the one day forecast for that day and move on. While resources do not equate with results. Lack of resources can substantially impede results.

[Lord M Vader]

Anyone who knows if we are going to see some mid-monthly numbers wrt the sea ice volume?

Wipneus?

[Wipneus]

PIOMAS has upgraded the gridded thickness data to day 197 (15 /16 July). Volume on day 197 was 9.04 [1000km³], which means an annual fourth lowest place for that day.

Fasten seat belts, here is the animation of July so far. Might need a click.

[Frivolousz21]

PIOMAS IS WAY OVERDONE IN THE SOUTHERN CAA.


Just wait until cryosat comes back in late September and early October. I guarantee it will be shockingly low.

[Lord M Vader]

Thanks Wipneus!🙂

Only a fourth place?! How far away are we from first place for the date?

[oren]

Distance to first (2019) has shrunk from 746 km3 to 446 km3. Distance to 2012 has shrunk from 454 km3 to 44 km3.  I would have expected higher losses though. Will wait for the regional breakdown and post some analysis.

[Wipneus]

Sorry for the delay, some trouble with the download caused a corrupted file (uice, the ice velocity data).

Here are the updated volume and volume-anomaly graphs. Worth a click for size.