Support the Arctic Sea Ice Forum and Blog

Author Topic: Latest PIOMAS Volume update  (Read 1896412 times)

Ned W

  • Guest
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2150 on: September 18, 2017, 05:51:27 PM »
That is quite the amazing turnaround from expectations at the start of the season.  Wow.

Richard Rathbone

  • Nilas ice
  • Posts: 1730
    • View Profile
  • Liked: 387
  • Likes Given: 24
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2151 on: September 18, 2017, 05:55:23 PM »
Is there anything (e.g. changepoint analysis) behind choosing 1997 as the point to start the linear fit at?

The long term anomaly trend is -3.1 per decade, which is quite a lot shallower than the -4.5 you get (and outside the error range for that trend too), and 1997 looks like its chosen to give an overly steep trend.

On the other hand assuming 2012 type deviations below trend are still possible before 2082, the anomaly trend of -3.1 brings virtually ice-free conditions in range around 2023 and makes them extremely likely before 2030 which isn't that different from the 2025 you get projecting at -4.5.

At the conservative end, the anomaly trend is quoted with a +-1.0 error range and if its only -2.1, that pushes the possible first blue minimum out to 2030, and the time by which its extremely likely to have happened to 2040.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2152 on: September 18, 2017, 06:57:25 PM »
Is there anything (e.g. changepoint analysis) behind choosing 1997 as the point to start the linear fit at?


Something like that, but I never fully went into that path. IIRC I looked for a midpoint such that the acceleration evident in the residuals using a linear fit over the whole range would disappear in the noise.
That was BTW in 2010, more years and an other version of PIOMAS. The acceleration is much less now but still as can be seen in one of the other graphs.

slow wing

  • Grease ice
  • Posts: 823
    • View Profile
  • Liked: 155
  • Likes Given: 546
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2153 on: September 18, 2017, 11:43:23 PM »
Thanks Wipneus, very interesting!

So this season must have had one of the lowest melt volumes for a while then?

Just eye-balling the graph, it appears that 2014 was the most recent year with a similarly small volume loss, then we need to go back to 2006 for another such year.
« Last Edit: September 18, 2017, 11:48:28 PM by slow wing »

slow wing

  • Grease ice
  • Posts: 823
    • View Profile
  • Liked: 155
  • Likes Given: 546
Re: Latest PIOMAS update (mid July update)
« Reply #2154 on: September 19, 2017, 12:01:35 AM »
Thanks Wipneus, always interesting!

So the current ice thickness contours are fairly simple in shape: nearly triangular at 1.75m, 1.50m, 1.25m and weighted to the Atlantic side; and with the Pacific side filled out to form more of a trapezoid at the 1.00m contour.

At this time of year, would we expect it to melt out to around the 1.25m triangular contour?

If so then the Pacific side is going to look very blue come September, as in the past years 2007, 2012 and 2016.

Just eye-balling it, call the 1.25m triangle as having a 2600 km base and 2000 km height, so half-base-times-height would be an end-of-melt-season extent of roughly 2.6 million square kilometers. That is somewhere close to the 2012 record of 3.18 in those units if we add in some extra bits e.g. in the CAA. (Very rough estimate only! It just says it's plausible we could have another low year - perhaps a record or close to a record. Whether that happens or not is still up to the weather and the ocean currents.)
Nope! Looking back 2 months, not the 1.25m contour but instead the 1.00m contour back then was a very good predictor of the melt region perimeter at minimum.
« Last Edit: September 19, 2017, 12:19:54 AM by slow wing »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2155 on: September 19, 2017, 11:57:49 PM »
Here is the original animation of Piomas data 31 Aug to 15 Sept below the ESRL ice thickness out to the 18th in the Piomas palette followed by ESRL ice thickness in a more effective continuous ColorBrewer2 palette, reversed CB_RdYlBl.cpt. See explanation over in the Dev Corner animation section for scripting these out of REB .nc files at the ESRL archive.

The ice thickness forecast is shown from the 18th out to the 28th in 6 hour increments in the CB palette.
« Last Edit: September 20, 2017, 12:59:15 AM by A-Team »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2156 on: September 20, 2017, 01:10:13 AM »
Very little change is foreseen by ESRL in ice thickness averaged over latitude over the next ten days. This latitudinal averaging capability of Panoply is probably of most interest to variable like temperature, vis-a-vis less nuanced products that lump everything over 80ºN into one bin.

« Last Edit: September 20, 2017, 11:26:20 PM by A-Team »

ghoti

  • Grease ice
  • Posts: 767
    • View Profile
  • Liked: 12
  • Likes Given: 15
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2157 on: September 20, 2017, 05:56:39 PM »
Mean thickness by latitude seems to be a very odd metric given what we know about the distribution of thick ice. Thickest ice tends to pile up along the north shore of the CAA but that latitude most of the way around the arctic is ice free part of the year.

Averaging therefore would hide information rather than provide clarity. The mean thickness along latitude of different sectors defined by longitude reflecting familiar seas might work.

Shared Humanity

  • Guest
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2158 on: September 20, 2017, 06:03:11 PM »
Could this measure provide more clarity during the winter as it would perhaps highlight the impact of a weak freeze season?

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2159 on: September 20, 2017, 11:45:26 PM »
Quote
mean thickness along latitude of different sectors defined by longitude reflecting familiar seas might work.
Right -- and that's doable as Panoply's geolocated grid has an excel-compatible export option.
Quote
weak freeze season. For a fixed latitudinal ring (land is excluded), the CAA kicks in a constant whereas the animation looks at change.
We definitely need to go all-in on monitoring the strength of this coming freezing season since that is the essence of Arctic amplification. This new ESRL site offers many resources for doing just that, indeed 49 of them. However which of these are most useful and accurate? There is no coverage of previous years so what best provides historical context for a 'weak freezing season'?

Extreme cyclone events in the Arctic: Wintertime variability and trends
A Rinke et al 2017 free full
http://iopscience.iop.org/article/10.1088/1748-9326/aa7def/meta;jsessionid=F01EB98E1055DA2CFDECBCD17C8AD495.c4.iopscience.cld.iop.org

Enhanced wintertime greenhouse effect reinforcing Arctic amplification and initial sea-ice melting
Y Cao et al free full
https://www.nature.com/articles/s41598-017-08545-2

How much should we believe correlations between Arctic cyclones
and sea ice extent?
JGL Rae et al 2017 free full
https://www.the-cryosphere-discuss.net/tc-2017-140/tc-2017-140.pdf

A weekly Arctic sea-ice thickness data record from merged
CryoSat-2 and SMOS satellite data
R Ricker et al 2017 free full
https://www.the-cryosphere.net/11/1607/2017/tc-11-1607-2017.pdf



aice       sea ice area fraction
hi         sea ice thickness
hs         surface snow thickness
snow ai    lwe snowfall rate
rain ai    rainfall rate
meltl      lateral ice melt
meltb      basil ice melt
meltt      top ice melt
sss        sea water salinity
sst        sea water temperature
Tair       air temperature
Tsfc       surface temperature where sea ice
strength   compressive strength of sea ice
vocn       northward sea water velocity
uocn       eastward sea water velocity
vvel       northward sea ice velocity
uvel       eastward sea ice velocity   
divu       divergence of sea ice velocity
hi_h   grid cell mean ice thickness
hs_h   grid cell mean snow thickness
Tsfc_h   snow/ice surface temperature
aice_h   ice area  (aggregate)
uvel_h   ice velocity (x)
vvel_h   ice velocity (y)
uatm_h   atm velocity (x)
vatm_h   atm velocity (y)
Tair_h   air temperature
Arctic1    RASM-ESRL vs GFS ice area and snow depth
Arctic2    RASM-ESRL vs GFS 2m temp and surface pressure
Arctic3    ice thickness thermo + dynamics + snow melt + ice melt
Arctic4    RASM-ESRL vs GFS 850 hPa temp and precip
Arctic5    ice and snow thickness contoured
Arctic9    RASM-ESRL vs GFS 500-1000 hPa thickness and precip
Arctic10   bottom ice growth + lateral melt + snow melt + top ice melt
Arctic11   longwave + solar flux + shortwave + albedo
Arctic12   RASM-ESRL vs GFS surface temp and LWP
Arctic13   RASM-ESRL vs GFS surface pressure + 850 hPa height
Arctic14   RASM-ESRL energy flux + LWP + surface temp + IWP
Arctic15   melt pond fraction + SST + heat flux + wind speed
Arctic16   RASM-ESRL ice speed
Arctic19   RASM-ESRL 500hPa height and wind vectors
Arctic20   310K potential vorticity and Surface theta PVU
Arctic22   RASM-ESRL vs GFS longwave flux + shortwave flux
Arctic23   RASM-ESRL energy flux + LWP + temp + IWP [duplicate of Arctic14?]
Arctic24   RASM-ESRL vs GFS surface wind + energy flux
AlaskaRegion2   RASM-ESRL vs GFS 2m temp and surface pressure
AlaskaRegion4   RASM-ESRL vs GFS 850 hPa temp and precip
AlaskaRegion5   ice and snow thickness
AlaskaRegion12  RASM-ESRL vs GFS surface temp and LWP
« Last Edit: September 21, 2017, 09:44:20 AM by A-Team »

Shared Humanity

  • Guest
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2160 on: September 21, 2017, 04:42:18 AM »


Enhanced wintertime greenhouse effect reinforcing Arctic amplification and initial sea-ice melting
Y Cao et al free full
https://www.nature.com/articles/s41598-017-08545-2



This article serves to confirm my own unsupported biases. Not sure if this is a good thing.

Anyone who lives in northern latitudes understands that cloudy winter nights do not get as cold as cloudless nights because the clouds reduce the amount of heat that can escape into the upper atmosphere. In the long Arctic night, clouds should have a similar effect and Arctic winters are increasingly cloudy.

Fairbanksnchill

  • Guest
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2161 on: September 21, 2017, 05:07:07 AM »


Enhanced wintertime greenhouse effect reinforcing Arctic amplification and initial sea-ice melting
Y Cao et al free full
https://www.nature.com/articles/s41598-017-08545-2



This article serves to confirm my own unsupported biases. Not sure if this is a good thing.

Anyone who lives in northern latitudes understands that cloudy winter nights do not get as cold as cloudless nights because the clouds reduce the amount of heat that can escape into the upper atmosphere. In the long Arctic night, clouds should have a similar effect and Arctic winters are increasingly cloudy.

It snowed here in Fairbanks in March and the temperature warmed from -20F to 10F in an hour. The snow was coming down hard, but since the front was coming from the south it increased the air temperature outside by 30F. I have read way to many comments referring to how hot or cold it is instead of thinking scientifically about energy. Consider that the people evacuating Irma were doing it in 110F weather. That was a significant system of collecting ocean heat through evaporation and then dumping that heat onto south Florida.

slow wing

  • Grease ice
  • Posts: 823
    • View Profile
  • Liked: 155
  • Likes Given: 546
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2162 on: September 21, 2017, 05:32:36 AM »
There is no coverage of previous years so what best provides historical context for a 'weak freezing season'?
Imho a map over the Arctic Basin of freezing-degree-days should show which freezing seasons are relatively weak, and in which regions. That data exists for previous years as we have their temperature maps.

Sterks

  • Guest
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2163 on: September 21, 2017, 08:24:45 AM »
There is no coverage of previous years so what best provides historical context for a 'weak freezing season'?
Imho a map over the Arctic Basin of freezing-degree-days should show which freezing seasons are relatively weak, and in which regions. That data exists for previous years as we have their temperature maps.
The 80N FDD, a simple scalar per day, has proven an excellent indicator of weakness of the past two freezing seasons.That and the amount of Fram export can help assess the state of the ice in May (plus some thickness and ice age products to know the ice distribution, and we have had those since a long time ago. Pre-Panoply times)

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2164 on: September 21, 2017, 11:00:12 AM »
Quote
80N FDD, a simple scalar per day, has proven an excellent indicator of weakness of the past  ... FDD shows which freezing seasons are relatively weak, and in which regions.

Really? Please tell us how you measure "excellence of weakness" and where this was "proven". Air temperature is only one of many components contributing to refreeze season outcome and FDD is only a crude summary of it. The days of just eyeballing the Arctic are long gone.

Clinging to the simplicity of rough measures does not add value because there's no extra compute time in replacing them with something much more refined. That's why FDD and 80ºN are little used in climate science modeling of the Arctic Ocean. They don't have the spatial or temporal resolution commensurate with other model components, mooting precision improvements there and so introducing gratuitous error.

Panoply has nothing whatsoever to do with RASM_ESRL forecasting. It is simply a visualization tool for climate science data sets archived in the standard GRIB, HDF or netCDF formats, for example UH AMSR2. These standardized data formats, unlike ad hoc storage, enable seamless processing by everything from cell phones to desktops to supercomputers.

Panoply does allow easy quantitative comparisons, so if you can post your FDD data for us in one of these formats along with your associated computation of daily lateral, top and bottom freeze to day 5 like ESRL, we can assess its adequacy. The days of just intuiting explanations for the Arctic's state or trends are long gone.

What's unprecedented with ESRL-PSD is the depth of the daily product release. We have gotten very accustomed to accepting complex climate products that cannot be replicated, trusting the final graphic because none of the underlying data (or code) used to make it is specified or made available.

There's an unexpected collateral benefit to the ESRL archive because their web site can carry only a small fraction of the possible product combinations. No one here as even looked yet at other data like compressive strength of the ice pack. We cannot stay in the game with a couple of back-of-the-napkin items like FDD and 80ºN.

The vexatious problem of replication (methods specification) in climate science is discussed in depth here: http://journals.ametsoc.org/doi/full/10.1175/BAMS-D-15-00010.1
« Last Edit: September 21, 2017, 11:16:04 AM by A-Team »

slow wing

  • Grease ice
  • Posts: 823
    • View Profile
  • Liked: 155
  • Likes Given: 546
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2165 on: September 21, 2017, 12:43:17 PM »
Again imho, the best 1-parameter indicator of a 'weak freezing season' is the ice volume gain over the Arctic basin. We do have estimates for that from previous years.

What beyond that would make one characterize a freezing season as 'weak'? It could be said to be weak in some regions. An ice thickness gain map could therefore be of interest - up to and ignoring ice position displacements. A freezing season could be said to be weak if it left the Arctic vulnerable to a melt-out in the following melt season. Parameters such as the ice temperature and the quality of the ice (e.g. "rotten ice") could be considered.

But imo the ice volume gain captures much of what we might call a 'weak freezing season'.



Sterks

  • Guest
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2166 on: September 21, 2017, 04:05:36 PM »
Really? Please tell us how you measure "excellence of weakness" and where this was "proven". Air temperature is only one of many components contributing to refreeze season outcome and FDD is only a crude summary of it. The days of just eyeballing the Arctic are long gone.

Clinging to the simplicity of rough measures does not add value because there's no extra compute time in replacing them with something much more refined. That's why FDD and 80ºN are little used in climate science modeling of the Arctic Ocean. They don't have the spatial or temporal resolution commensurate with other model components, mooting precision improvements there and so introducing gratuitous error.

Thanks.
In fact I can't prove such excellence. I just meant to note that the FDD of 2015 and 2016 were extremely anomalous, and it showed in 2016 and 2017 state of ice prior to summer, nothing else.
Lately I tend to be overly sarcastic on occasions, my excuses. Actually my opinion is that your work with ESRL data is really excellent, I wish I had the time and resources to get hands-on with Panoply and produce material for the forum as well.
So I am just another guy with an opinion, and I'll try to stay quiet :)
« Last Edit: September 21, 2017, 06:31:16 PM by Sterks »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2167 on: September 29, 2017, 08:21:37 AM »
PIOMAS gice (gridded sea ice thickness distribution) was updated to the 25th.
As mentioned before, thickness/volume can be calculated from gice, but not exactly. Differences may be as big as 30 km3.

With that in mind here are the September (1-25) volume numbers (times 1000 km3):
array([ 4.65,  4.63,  4.62,  4.63,  4.63,  4.62,  4.6 ,  4.57,  4.54,
        4.52,  4.51,  4.52,  4.55,  4.59,  4.63,  4.64,  4.67,  4.69,
        4.67,  4.65,  4.64,  4.64,  4.65,  4.67,  4.7 ])

From it, 2017 regrowth after reaching minimum on the 11th is relatively slow. It is now 4th lowest (went below 2010) and very near to 2011 and 2016.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (September mid month update, includes 2017 minimum volume)
« Reply #2168 on: September 29, 2017, 09:08:08 AM »
With gice of September available I can fill in the area covered at the 10th of September (taking it as minimum area). Some time ago I noticed the similarity of the area at the 10th of September (any ice thickness) with the area during July covered with ice categories >=0.71m, or at least since 2007.

Area btw is PIOMAS area and very similar to NSIDC NT area.

The updated graph is attached. Predicting the minimum area from the gice-0.71m at the 15th of July would have been quite accurate. In most other years (2007 and later) that would have been true as well except for 2008, 2012 and 2016 when the area turned out much lower.

There is more information of course (other cat's, other days) so it might be possible to squeeze an even better prediction out of gice data. Nice project for the coming freezing season.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October update)
« Reply #2169 on: October 03, 2017, 08:25:46 AM »
PIOMAS has updated, only the gridded data are available yet.

Since the volume data are still missing, I use volume calculated from the thickness data.
The anomaly graph is attached, as can be seen 2017 is now fourth lowest occasionally changing place with 2011, currently (30 Sep) fifth lowest.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October update)
« Reply #2170 on: October 03, 2017, 08:31:15 AM »
Fram export has been none last month.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October update)
« Reply #2171 on: October 03, 2017, 09:37:01 AM »
September 2017 animation. There is some spurious ice south of Greenland.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0

Blizzard92

  • New ice
  • Posts: 80
    • View Profile
    • Personal Research Website
  • Liked: 43
  • Likes Given: 0
Re: Latest PIOMAS update (October update)
« Reply #2173 on: October 03, 2017, 10:54:59 PM »
My PIOMAS plots have been updated for September 2017 at http://sites.uci.edu/zlabe/arctic-sea-ice-volumethickness/
Currently: Postdoctoral Research Associate - Princeton University & NOAA GFDL - Atmospheric and Oceanic Sciences
UC Irvine - Earth System Science Ph.D., M.Sc.
Cornell University - Atmospheric Sciences B.Sc.

Twitter: @ZLabe
Mastodon: https://fediscience.org/web/@ZLabe
Website: https://zacklabe.com

Shared Humanity

  • Guest
Re: Latest PIOMAS update (October update)
« Reply #2174 on: October 04, 2017, 02:21:29 PM »
Will an anomalously warm freeze season, similar to 2016, result in a ridiculously low volume entering next years melt season?

I hope not.

Shared Humanity

  • Guest
Re: Latest PIOMAS update (October update)
« Reply #2175 on: October 04, 2017, 02:23:21 PM »
Wipneus...

What do you make of Fram export? Is this unprecedented or simple variation? If unusual, what are the likely causes?

It would be one thing if the Atlantic side of the Arctic had aggressive melt in the just finished melt season but the opposite has been the case with far more ice on the Atlantic side than in recent years.
« Last Edit: October 04, 2017, 02:34:52 PM by Shared Humanity »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Latest PIOMAS update (October update)
« Reply #2176 on: October 04, 2017, 02:54:28 PM »
Quote
Zack's ZIOMAS plot update for September 2017 at http://sites.uci.edu/zlabe/arctic-sea-ice-volumethickness/
Maybe use the 'fit to data'  and 'histogram equalization' options for palette? There is no ice thicker than 3.2m on the plot (pink). The bins from 2.6-3.2 are scarcely used (orange). Two reasons for using 'perceptual palettes': visually distinguishable colors and 3D extrusion tools that drape off CIE luminance. Rather than over-write precious data, lat-lon option might be turned off or left to land and periphery as viewers are familiar with them already.

RASM-ESRL has a somewhat different take on ice thickness, both in absolute values and in distribution pattern. The effect of palette squeezing in 'parula' is shown below (from PanoplyCL). Forum software is now assigning pure black to residual image alpha channel; it used to let forum background through. The bottom animation isolates the thicker ice in a continuous palette and its binned version.

« Last Edit: October 04, 2017, 04:04:31 PM by A-Team »

Shared Humanity

  • Guest
Re: Latest PIOMAS update (October update)
« Reply #2177 on: October 04, 2017, 03:12:15 PM »
There is no ice thicker than 3.2m on the plot (pink). The bins from 2.6-3.2 are scarcely used (orange).

Frightening.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October update)
« Reply #2178 on: October 04, 2017, 04:14:27 PM »
Wipneus...

What do you make of Fram export? Is this unprecedented or simple variation?

Good question. I will have a look, but not today.

Quote
If unusual, what are the likely causes?

High atmospheric pressure/low pressure configuration was not favorable for export. Why that is so would be wonderful to know.


FishOutofWater

  • Nilas ice
  • Posts: 1088
    • View Profile
  • Liked: 696
  • Likes Given: 332
Re: Latest PIOMAS update (October update)
« Reply #2179 on: October 04, 2017, 04:25:04 PM »
There's always going to be a trade off between making an individual image with the best visual clarity with having the best system for intercomparison  of images. There are many scientific images on the net with variable baselines that I find very frustrating because they make it difficult if not impossible to visualize changes over time.

Zach's graphs do a good job of showing how thin all of the ice is now. That's the most important, and most frightening, information in the image. A-Team's adjustments make the details more visible but many of those details are model dependent. The lack of reproducibility of specific details by different models is good reason to question whether those details are real.

Richard Rathbone

  • Nilas ice
  • Posts: 1730
    • View Profile
  • Liked: 387
  • Likes Given: 24
Re: Latest PIOMAS update (October update)
« Reply #2180 on: October 04, 2017, 04:35:05 PM »
Wipneus...

What do you make of Fram export? Is this unprecedented or simple variation?

Good question. I will have a look, but not today.

Quote
If unusual, what are the likely causes?

High atmospheric pressure/low pressure configuration was not favorable for export. Why that is so would be wonderful to know.

Fram export shows seasonal variation, normally being much lower in the summer than other seasons. Significant export in the summer correlates with high melt, but I think its more to do with the pressure pattern that is good for melt also being good for export rather than the higher export itself directly reducing ice by a significant amount during the summer.

The best explanation I know of for why the summer pressure patterns were generally good for melt in 2007-2012 rather than before or since is the paper discussed earlier this year here
https://forum.arctic-sea-ice.net/index.php/topic,1920.0.html
 

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Latest PIOMAS update (October update)
« Reply #2181 on: October 04, 2017, 06:16:22 PM »
Quote
adjustments make the details more visible but many of those details are model dependent. The lack of reproducibility of specific details by different models.
The models and data are fixed. These are just variations in their visualization. Doing that most effectively is in itself a large, distinct and important scientific research area (eg T1 brain scans, cosmic background anisotropy, subtropical water vapor advection).

Here the issue was best treatment of outliers (via pre-determination of std deviation, skew, kurtosis and radial variance of the data to be visualized, thus constraining -- indeed automating -- quasi-optimal palette choices).

It's part of a wider rebellion against the rainbow palette which is on the verge of being banned from journal use. And that in turn is about not putting a two dollar saddle on a forty dollar mule.

Far too often, we just copy over a climate graphic off the internet but the data used to make that graphic is not provided. It is all but impossible to compare two such graphics -- even to earlier iterations -- especially if one is dithered to millions of colors, disconnected from the supposed legend, overlain with text debris, and presented at low resolution (eg Hycom ice thickness).

In this instance (recalling we have maybe ~48 parameters to compare in addition to thickness), it is only when both Piomas and RASM-ESRL are both provided in a standard climate science format like Grib, HDF or netCDF that it becomes practical to quantitatively compare them in an effective graphic. No one has time to delve into hundreds of artisanal archival silos. Not when AWS is nearly free.

The other issue is data integration (multiplexing), trying to get multiple related variables displayed simultaneously (eg ice thickness, snow thickness, adjacent water temperature). That inevitably makes for a more complex graphic but one that arguably improves on trying to synchronize and relate to variables spread out across multiple animations.

Rough example below, triple scripted, of snow thickness color draped on an extruded perceptual (luminance) palette of ice thickness. Water temperature or salinity could have been added for open water, air temperatures for land, and wind direction vectors for all over but that's about it for what can be accommodated, unless the Gimp xcf tile file is furnished that allows all the combinatorics of all the variables just by single-click layer choices. Which is where we're headed over at the Dev Corner graphics forum.
« Last Edit: October 04, 2017, 06:43:46 PM by A-Team »

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October update)
« Reply #2182 on: October 04, 2017, 07:18:47 PM »
Re: Fram pause

It is true that low export in Jul-Aug is always low, but here the line seems essentially flat. Also some years see firm recovery of the export in September.
 
Depending what you think "flat" means, it is not rare, but did not happen very recently, except perhaps a less than one month pause in 2010.
Before that 2001,2002,2003 and 2004 look similar to 2017.

(the daily thickness data does not start until 2000)

2phil4u

  • New ice
  • Posts: 27
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: Latest PIOMAS update (October update)
« Reply #2183 on: October 08, 2017, 04:38:51 PM »
How did 2016 manage not to grow in volume from sep to Dez ?
I was not at the spot at this date. I just cant imagine how this should work, also i cant detect a superbig Frametransport there.

Richard Rathbone

  • Nilas ice
  • Posts: 1730
    • View Profile
  • Liked: 387
  • Likes Given: 24
Re: Latest PIOMAS update (October update)
« Reply #2184 on: October 08, 2017, 06:53:10 PM »
How did 2016 manage not to grow in volume from sep to Dez ?
I was not at the spot at this date. I just cant imagine how this should work, also i cant detect a superbig Frametransport there.

What are you looking at? Volume nearly tripled over that time span.

Shared Humanity

  • Guest
Re: Latest PIOMAS update (October update)
« Reply #2185 on: October 08, 2017, 07:42:16 PM »
Re: Fram pause

It is true that low export in Jul-Aug is always low, but here the line seems essentially flat. Also some years see firm recovery of the export in September.
 
Depending what you think "flat" means, it is not rare, but did not happen very recently, except perhaps a less than one month pause in 2010.
Before that 2001,2002,2003 and 2004 look similar to 2017.

(the daily thickness data does not start until 2000)

Thank you for this animation. What is clear is the dramatic drop in export variability in this time period. This can be seen in almost all of the seasons. What might cause this? Could it be related to ice melt which results in less ice available to move in either direction through the Fram?

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October, mid month update)
« Reply #2186 on: October 18, 2017, 08:59:41 AM »
Mid month (1-15 Oct) daily gridded thickness data is available.

Volume calculated from this thickness is 5.96 {102km3] on 15th October, fourth lowest for the day after 2012, 2016 and 2011.

Volume 1-15 Oct:
 [1] 5.09 5.13 5.14 5.16 5.17 5.21 5.28 5.36 5.45 5.54 5.64 5.73 5.80 5.88 5.96

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October, mid month update)
« Reply #2187 on: October 18, 2017, 09:00:40 AM »
Animation for October thus far.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0

Neven

  • Administrator
  • First-year ice
  • Posts: 9470
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 1333
  • Likes Given: 617
Re: Latest PIOMAS update (October, mid month update)
« Reply #2189 on: October 18, 2017, 11:48:29 AM »
Thanks, Wipneus.
The enemy is within
Don't confuse me with him

E. Smith

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October, mid month update)
« Reply #2190 on: October 18, 2017, 01:40:34 PM »
Fram export: a little more than nothing.

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 4220
    • View Profile
    • Arctische Pinguin
  • Liked: 1025
  • Likes Given: 0
Re: Latest PIOMAS update (October, mid month update)
« Reply #2191 on: October 18, 2017, 01:51:17 PM »
Volume and volume-anomaly graphs.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Latest PIOMAS update (October, mid month update)
« Reply #2192 on: October 19, 2017, 12:08:43 AM »
AGU17 Search “piomas”
5/6 RESULTS
https://agu.confex.com/agu/fm17/meetingapp.cgi/SearchResults/0

GC43J-04 PIOMAS-20C: Variability of Arctic sea ice thickness and volume over the 20th Century.
Axel J B Schweiger UW-APL

Changes in Arctic sea ice are a fingerprint of natural and anthropogenic climate change. The dominant signal in sea ice variability from 1979 to the present is the reduction of sea ice extent, area and thickness. Prior to 1979, the state of our knowledge about sea ice variability is limited to information about sea ice extent and concentration assembled mostly from shipping logs and very little is known about the variability of sea ice thickness and total sea ice volume.

Here we use the Panarctic Ice and Ocean Modelling and Assimilation system (PIOMAS) to generate a sea ice reanalysis from 1900 to 2010 (PIOMAS-20C). PIOMAS-20C is generated by forcing PIOMAS with atmospheric reanalysis data from the ERA-20C project. We present initial results that include validation of atmospheric forcing parameters over sea ice from the ERA20C project and sea ice thickness from PIOMAS-20C.

The PIOMAS-20C sea ice thickness is generally in good agreement with available observations before and after 1979. We specifically investigate patterns of sea ice thickness and volume variability in the early 20th century and compare them with changes over the more recent period.


C33C-1205 Seasonal evolution of the Arctic marginal ice zone and its power-law obeying floe size distribution
Jinlun Zhang  UW-APL

A thickness, floe size, and enthalpy distribution (TFED) sea ice model, implemented numerically into the Pan-arctic Ice–Ocean Modeling and Assimilation System (PIOMAS), is used to investigate the seasonal evolution of the Arctic marginal ice zone (MIZ) and its floe size distribution.

The TFED sea ice model, by coupling the Zhang et al. [2015] sea ice floe size distribution (FSD) theory with the Thorndike et al. [1975] ice thickness distribution (ITD) theory, simulates 12-category FSD and ITD explicitly and jointly. A range of ice thickness and floe size observations were used for model calibration and validation. The model creates FSDs that generally obey a power law or upper truncated power law, as observed by satellites and aerial surveys.

In this study, we will examine the role of ice fragmentation and lateral melting in altering FSDs in the Arctic MIZ. We will also investigate how changes in FSD impact the seasonal evolution of the MIZ by modifying the thermodynamic processes.


C21B-1119 Winter Arctic sea ice growth: current variability and projections for the coming decades
Alek Petty

Arctic sea ice increases in both extent and thickness during the cold winter months (~October to May). Winter sea ice growth is an important factor controlling ocean ventilation and winter water/deep water formation, as well as determining the state and vulnerability of the sea ice pack before the melt season begins. Key questions for the Arctic community thus include: (i) what is the current magnitude and variability of winter Arctic sea ice growth and (ii) how might this change in a warming Arctic climate?

To address (i), our current best guess of pan-Arctic sea ice thickness, and thus volume, comes from satellite altimetry observations, e.g. from ESA's CryoSat-2 satellite. A significant source of uncertainty in these data come from poor knowledge of the overlying snow depth.

Here we present new estimates of winter sea ice thickness from CryoSat-2 using snow depths from a simple snow model forced by reanalyses and satellite-derived ice drift estimates, combined with snow depth estimates from NASA's Operation IceBridge.

To address (ii), we use data from the Community Earth System Model's Large Ensemble Project, to explore sea ice volume and growth variability, and how this variability might change over the coming decades. We compare and contrast the model simulations to observations and the PIOMAS ice-ocean model (over recent years/decades). The combination of model and observational analysis provide novel insight into Arctic sea ice volume variability

C33B-1201 The Impact of Moisture Intrusions from Lower Latitudes on Arctic Net Surface Radiative Fluxes and Sea Ice Growth in Fall and Winter
Bradley M Hegyi

The fall and winter seasons mark an important period in the evolution of Arctic sea ice, where energy is transferred away from the surface to facilitate the cooling of the surface and the growth of Arctic sea ice extent and thickness.

Climatologically, these seasons are characterized by distinct periods of increased and reduced surface cooling and sea ice growth. Periods of reduced sea ice growth and surface cooling are associated with cloudy conditions and the transport of warm and moist air from lower latitudes, termed moisture intrusions.

In the research presented, we explore the regional and Arctic-wide impact of moisture intrusions on the surface net radiative fluxes and sea ice growth for each fall and winter season from 2000/01-2015/16, utilizing MERRA2 reanalysis data, PIOMAS sea ice thickness data, and daily CERES radiative flux data.

Consistent with previous studies, we find that positive anomalies in downwelling longwave surface flux are associated with increased temperature and water vapor content in the atmospheric column contained within the moisture intrusions.

Interestingly, there are periods of increased downwelling LW flux anomalies that persist for one week or longer (i.e. longer than synoptic timescales) that are associated with persistent poleward flux of warm, moist air from lower latitudes. These persistent anomalies significantly reduce the regional growth of Arctic sea ice, and may in part explain the inter-annual variability of fall and winter Arctic sea ice growth.


C21D-1144: Anomalous circulation in the Pacific sector of the Arctic Ocean in July-December 2008
Gleb Panteleev

Variability of the mean summer-fall ocean state in the Pacific Sector of the Arctic Ocean (PSAO) is studied using a dynamically constrained synthesis (4Dvar) of historical in situ observations collected during 1972 to 2008. Specifically, the oceanic response to the cyclonic (1989-1996) and anticyclonic (1972-1978, 1997-2006) phases o f the Arctic Ocean Oscillation (AOO) is assessed for the purpose of quantitatively comparing the 2008 circulation pattern that followed the 2007 ice cover minimum.

It is shown that the PSAO circulation during July-December of 2008 was characterized by a pronounced negative Sea Surface Height (SSH) anomaly along theEurasian shelf break, which caused a significant decline of the transport in the Atlantic Water (AW) inflow region into the PSAO and increased the sea level difference between the Bering and Chukchi Seas. This anomaly could be one of the reasons for the observed amplification of the Bering Strait transport carrying fresh Pacific Waters into the PSAO.

Lagrangian analysis of the optimized solution suggests that the freshwater (FW) accumulation in the Beaufort Gyre has a negligible contribution from the East Siberian Sea and is likely caused by the enhanced FW export from the region north of the Canadian Archipelago/Greenland.The inverse modeling results are confirmed by validation against independent altimetry observations and in situ velocity data from NABOS moorings. It is also shown that presented results are in significantly better agreement with the data than the output of the PIOMAS model run utilized as a first guess solution for the 4dVar analysis.
« Last Edit: October 19, 2017, 12:50:41 AM by A-Team »

charles_oil

  • Frazil ice
  • Posts: 337
    • View Profile
  • Liked: 38
  • Likes Given: 74
Re: Latest PIOMAS update (October, mid month update)
« Reply #2193 on: October 25, 2017, 11:43:04 AM »

Salty Snow ... could the figures we have been watching be incorrect? 


CALGARY, Alberta, Oct 24 (Reuters) – Arctic sea ice may be thinning faster than predicted because salty snow on the surface of the ice skews the accuracy of satellite measurements, a new study from the University of Calgary said on Tuesday.The report from the Canadian university’s Cryosphere Climate Research Group published in the academic journal Geophysical Research Letters found satellite estimates for the thickness of seasonal sea ice have been overestimated by up to 25 percent.
http://gcaptain.com/arctic-sea-ice-may-be-declining-faster-than-expected-study/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+Gcaptain+%28gCaptain.com%29&goal=0_f50174ef03-8c7e0a568f-139828445&mc_cid=8c7e0a568f&mc_eid=6f655bf7d8
Reuters:
https://uk.reuters.com/article/uk-canada-arctic-research/arctic-sea-ice-may-be-declining-faster-than-expected-study-idUKKBN1CT2K5
It doesn't seem to have a date linked so I am not sure how recent this is.... anyone have the text?


Richard Rathbone

  • Nilas ice
  • Posts: 1730
    • View Profile
  • Liked: 387
  • Likes Given: 24
Re: Latest PIOMAS update (October, mid month update)
« Reply #2194 on: October 25, 2017, 01:00:20 PM »


It doesn't seem to have a date linked so I am not sure how recent this is.... anyone have the text?



http://onlinelibrary.wiley.com/doi/10.1002/2017GL074506/full

Its paywalled so I can't be sure what the article itself says, but it looks like a journalist has misunderstood the abstract and has got the story wrong.

The abstract says taking salinity into account could reduce error in Cryosat thickness by up to 25%, which is not the same thing at all as it being overestimated by up to 25%.

Phil.

  • Grease ice
  • Posts: 540
    • View Profile
  • Liked: 76
  • Likes Given: 11
Re: Latest PIOMAS update (October, mid month update)
« Reply #2195 on: October 25, 2017, 03:16:58 PM »


It doesn't seem to have a date linked so I am not sure how recent this is.... anyone have the text?



http://onlinelibrary.wiley.com/doi/10.1002/2017GL074506/full

Its paywalled so I can't be sure what the article itself says, but it looks like a journalist has misunderstood the abstract and has got the story wrong.

The abstract says taking salinity into account could reduce error in Cryosat thickness by up to 25%, which is not the same thing at all as it being overestimated by up to 25%.

The abstract states that the salinity causes an increase in thickness of ~7cm.
From the conclusions of the paper:
<em>"Our corrected FYI thickness estimates using the snow salinity correction factor demonstrate reductions in CS-2 relative errors. The reductions are considerable at FYI thicknesses <1 m. At 0.95 m, the relative error reduces by ~11% and at 0.7 m the error reduces by ~25%. These reductions also help to close the uncertainty gap between SMOS and CS-2 thin ice thickness retrievals by 0.25 m. We find that current retrieval methods are likely more suited to very cold, low snow salinity FYI, which limits their scope. We also find that FYI with warm, highly saline snow has the potential to produce the highest retrieval errors. To increase confidence in CS-2 error analyses in all seasons, subsequent research should focus on using in situ FYI thickness data for validation, to quantify the error objectively. With the recent and rapid decline of MYI, and its replacement by FYI, the role of snow salinity should be considered whenever FYI freeboard is estimated using CryoSat-2 on local to pan-Arctic scales."</em>

Shared Humanity

  • Guest
Re: Latest PIOMAS update (October, mid month update)
« Reply #2196 on: October 25, 2017, 04:34:28 PM »
Very interesting study but one problem is to understand just how much high salinity snow exists on FYI. I don't know of any method for determining this.

charles_oil

  • Frazil ice
  • Posts: 337
    • View Profile
  • Liked: 38
  • Likes Given: 74
Re: Latest PIOMAS update (October, mid month update)
« Reply #2197 on: October 25, 2017, 11:14:11 PM »

Thanks - I was suspicious.


Big difference between 25% over estimate and an ~11 to 25% reduction in error in some cases.  I like that really the best thing is in situ FYI thickness data - what a surprise !

Richard Rathbone

  • Nilas ice
  • Posts: 1730
    • View Profile
  • Liked: 387
  • Likes Given: 24
Re: Latest PIOMAS update (October, mid month update)
« Reply #2198 on: October 26, 2017, 12:10:06 PM »
Very interesting study but one problem is to understand just how much high salinity snow exists on FYI. I don't know of any method for determining this.

It will be referenced in the paper, they have previous publications on it, but the abstracts aren't informative enough to tell just what the method is or how robust their measurements are.

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Latest PIOMAS update (October, mid month update)
« Reply #2199 on: October 27, 2017, 11:44:03 PM »
AGU17 Search “piomas”
5/6 RESULTS
https://agu.confex.com/agu/fm17/meetingapp.cgi/SearchResults/0

GC43J-04 PIOMAS-20C: Variability of Arctic sea ice thickness and volume over the 20th Century.
Axel J B Schweiger UW-APL
C33C-1205 Seasonal evolution of the Arctic marginal ice zone and its power-law obeying floe size distribution
Jinlun Zhang  UW-APL
C21B-1119 Winter Arctic sea ice growth: current variability and projections for the coming decades
Alek Petty
C33B-1201 The Impact of Moisture Intrusions from Lower Latitudes on Arctic Net Surface Radiative Fluxes and Sea Ice Growth in Fall and Winter
Bradley M Hegyi
C21D-1144: Anomalous circulation in the Pacific sector of the Arctic Ocean in July-December 2008
Gleb Panteleev
A series of papers that marvelously quantify what we've been observing the last few years, most dramatically the last 2 1/2 or so.  I look forward to more.
This space for Rent.