Home brew AMSR2 extent & area calculation

[Wipneus]

Status of the UH's reprocessing: latest 2012 date is now at 2012-09-20. Currently there are no missing dates. Also Antarctic data from 2012 is arriving, I modified the graphics temporary to show both AMSR2 and SSMIS 2012 data (like in the Arctic ones).

Looking at the Antarctic graphs, the SSMIS area almost perfectly matches AMSR2 area. That is good. However AMSR2 extent is clearly higher that SSMIS extent. That is strange, at the very least puzzling. I attach them for the curious, click for bigger image.

[Nick_Naylor]

2016 has already smashed the record for lowest and latest 3.125km minimum compactness.

I read this a reflection of just how pulverized the ice conditions are, and I imagine compactness will continue to be a major outlier for quite a while longer.

[Wipneus]

Update 20160821.

Extent: +24.5 (-154k vs 2015, -673k vs 2014, -672k vs 2013, +838k vs 2012)
Area: +68.2 (-295k vs 2015, -1063k vs 2014, -1001k vs 2013, +454k vs 2012)
 
You will find the updated graphs in the top post

CAB extent: +29k does the extent increase.

Same for CAB area: +63k.

Delta image is from Laptev, only there some losses can be spotted in the former fast ice.

[johnm33]

Whatever will they think of next
http://www.polarview.aq/sic/arctic/

[Wipneus]

Update 20160822.

Extent: -33.7 (-122k vs 2015, -678k vs 2014, -653k vs 2013, +796k vs 2012)
Area: -9.3 (-299k vs 2015, -1040k vs 2014, -919k vs 2013, +469k vs 2012)
 
You will find the updated graphs in the top post

No region has a noteworthy extent or area change.

Changes are of course still be found on a much finer scale, helped by the hi resolution nature of Uni Hamburgs product. Attached is the delta map of the "Barents Bite" where more little polynyas are opening.

[bbr2314]

MODIS shows an inordinate amount of very thick clouds today. Once those clear I would think the CAB flashes back to major red and we have a "false" drop of 150K+ to account for the "gains" that weren't really there.

[weatherdude88]

MODIS shows an inordinate amount of very thick clouds today. Once those clear I would think the CAB flashes back to major red and we have a "false" drop of 150K+ to account for the "gains" that weren't really there.

A key feature of these AMSR instruments is the ability to see through clouds, thereby providing an uninterrupted view of ocean measurements

http://www.remss.com/missions/amsr

Please provide scientific evidence on where AMSR instruments show "false gains" due  to cloud cover to support your claim. If you do not provide any, most in this forum including myself will conclude this is another Dunning-Kruger-Esque statement.

[Tealight]

MODIS shows an inordinate amount of very thick clouds today. Once those clear I would think the CAB flashes back to major red and we have a "false" drop of 150K+ to account for the "gains" that weren't really there.

A key feature of these AMSR instruments is the ability to see through clouds, thereby providing an uninterrupted view of ocean measurements

http://www.remss.com/missions/amsr

Please provide scientific evidence on where AMSR instruments show "false gains" due  to cloud cover to support your claim. If you do not provide any, most in this forum including myself will conclude this is another Dunning-Kruger-Esque statement.

If you ever followed concentration maps closely than you would have noticed plently of false sea ice detection.

Today there is ice detected in the Sea of Okhotsk and around northern Japan. These areas are probably masked out in August, but false detection remains in the Arctic basin.

Its possible that there are plenty of ice crystals in the clouds, but defenitly none in the sea.

[budmantis]

MODIS shows an inordinate amount of very thick clouds today. Once those clear I would think the CAB flashes back to major red and we have a "false" drop of 150K+ to account for the "gains" that weren't really there.

A key feature of these AMSR instruments is the ability to see through clouds, thereby providing an uninterrupted view of ocean measurements

http://www.remss.com/missions/amsr

 If you do not provide any, most in this forum including myself will conclude this is another Dunning-Kruger-Esque statement.

Speak for yourself Weatherdude, and the others will speak for themselves. I could be wrong, but I think most will disagree with you.

[weatherdude88]

MODIS shows an inordinate amount of very thick clouds today. Once those clear I would think the CAB flashes back to major red and we have a "false" drop of 150K+ to account for the "gains" that weren't really there.

A key feature of these AMSR instruments is the ability to see through clouds, thereby providing an uninterrupted view of ocean measurements

http://www.remss.com/missions/amsr

Please provide scientific evidence on where AMSR instruments show "false gains" due  to cloud cover to support your claim. If you do not provide any, most in this forum including myself will conclude this is another Dunning-Kruger-Esque statement.

If you ever followed concentration maps closely than you would have noticed plently of false sea ice detection.

Today there is ice detected in the Sea of Okhotsk and around northern Japan. These areas are probably masked out in August, but false detection remains in the Arctic basin.

Its possible that there are plenty of ice crystals in the clouds, but defenitly none in the sea.

this approach only detects the first appearance of open water so potential refill of ice is not accounted for.

Conversely, the better agreement with AMSR-E and IMS compared to QuikSCAT and IMS is likely due to the reduced sensitivity to transient weather events that can affect QuikSCAT (Howell et al., 2010 and Yu et al., 2009).

http://www.sciencedirect.com/science/article/pii/S0034425714000649

At variance with:

False ice concentrations can occur due to bad weather systems.

http://www.iup.uni-bremen.de:8084/amsr2/regions-amsr2.php

However, if we take a look at the other sea ice area models, and when we experienced moderate to strong loses the first several days of cyclonic weather, why did ASMR instruments not show "false gains"? Even though we had anomalously cold weather with snowfall? Why do the other sea ice metrics correlate closely with ASMR2 even though they use different algorithms?

Speak for yourself Weatherdude, and the others will speak for themselves. I could be wrong, but I think most will disagree with you.

Why would ASMR instruments show a 87.7 thousand square kilometers gain over 2 days with similar atmospheric conditions to the previous several days? It seems more likely from a scientific point of view that "false loses" were reality when areas were refreezing.   

I am still waiting for scientific proof that ASMR instruments show large scale "false gains" over multiple days above 70 degrees north.

[bbr2314]

MODIS shows an inordinate amount of very thick clouds today. Once those clear I would think the CAB flashes back to major red and we have a "false" drop of 150K+ to account for the "gains" that weren't really there.

A key feature of these AMSR instruments is the ability to see through clouds, thereby providing an uninterrupted view of ocean measurements

http://www.remss.com/missions/amsr

Please provide scientific evidence on where AMSR instruments show "false gains" due  to cloud cover to support your claim. If you do not provide any, most in this forum including myself will conclude this is another Dunning-Kruger-Esque statement.

What is wrong with you? There have been two cyclones of ~970mb the past few days, and you hurl a personal insult at me?

You will get your proof when the mega-HP arrives in the wake of the departing cyclone that's currently over the Arctic. Jeez.

[Nick_Naylor]

I am still waiting for scientific proof that ASMR instruments show large scale "false gains" over multiple days above 70 degrees north.

Since when does anyone demand "scientific proof" of anything here? That's not a very collegial tone. One of great things about this forum is how respectfully most participants are to others - even when they are wrong.

I don't know to what degree AMSR2 is affected by clouds - if at all, but it's hardly a crazy idea that it might be a factor.

[ktonine]

The self-appointed gatekeepers here are becoming tiresome.  If someone makes a factual error, fine, point it out.  But most of this is either opinion or making mountains out of molehills.

The uncertainties in the sea ice extent data are so large that it really isn't worth arguing over.  We obsess on day-to-day numbers, but they're a mirage. When you have double-digit uncertainties a 100k here or a 100k there isn't worth fretting about.

And yes, weather does affect AMSR2.  But given the large uncertainties, and actual variability, it is usually difficult to tell if differences over a short period are due to 'flash-melt,' 'flash-refreeze,' or sensor error.  We've all seen 'ice' where there isn't any.  It comes and goes.

The authoritative source for the Uni Bremen/AMSR2 extent is Sea Ice Remote Sensing Using AMSR-E 89 GHz Channels, Spreen et al, 2007

[Michael Hauber]

I've seen at least a couple instances where areas of cloud have been matched by areas of reduced concentration in the ice, which have then disappeared when the cloud moved on.  I'd say its quite possible that the large drops in area were in part due to the clouds, and that the recent small increases are partly due to the effect of the clouds reducing.  Although my suspicion s that it is not actually the clouds that make a difference but precipitation, and maybe wind blown spray or other wind effects.  Current MODIS images in 3-6-7 channel show mostly dull clouds suggesting mostly thinner lower level clouds, and very little high cloud and/or potential precipitation bearing clouds.

[Tigertown]

I have read a lot lately about the heat morphing the clouds, so that the clouds now are not the same as a couple years ago. Could play a factor. I would not read a lot into area or extent numbers right now anyhow. There is just too much breaking up and dispersion.

P.S. Increased amount of ice that is headed for doom. Warm waters to the east and Fram to the south.I am only mentioning on this thread because its an example of ice that has affected both area and extent because it has dispersed so much.
NE Greenland in bottom left

[budmantis]

MODIS shows an inordinate amount of very thick clouds today. Once those clear I would think the CAB flashes back to major red and we have a "false" drop of 150K+ to account for the "gains" that weren't really there.

A key feature of these AMSR instruments is the ability to see through clouds, thereby providing an uninterrupted view of ocean measurements

http://www.remss.com/missions/amsr

Please provide scientific evidence on where AMSR instruments show "false gains" due  to cloud cover to support your claim. If you do not provide any, most in this forum including myself will conclude this is another Dunning-Kruger-Esque statement.

Weatherdude: Being a rank amateur at the technical level, I cant supply you with the science. The knowledge I have is merely from observing the arctic over several years and reading the posts from other members who have more knowledge than I do.

My objection to your post attacking BBR was that it was interrogative and disrespectful. You could have made your point without the inflammatory remarks about the Dunning-Kruger effect.

FYI and for your own edification I included the definition of the Dunning-Kruger effect in the attachment. I think the definition points in your direction.

[6roucho]

Dunning-Kruger effect.

I thought this had to be shared:

The phenomenon was first experimentally observed in a series of experiments by David Dunning and Justin Kruger of the department of psychology at Cornell University in 1999. The study was inspired by the case of McArthur Wheeler, a man who robbed two banks after covering his face with lemon juice in the mistaken belief that, because lemon juice is usable as invisible ink, it would prevent his face from being recorded on surveillance cameras.

[Nick_Naylor]

. . . a man who robbed two banks after covering his face with lemon juice in the mistaken belief that, because lemon juice is usable as invisible ink, it would prevent his face from being recorded on surveillance cameras.

Very logical A good argument for validating your hypothesis before betting the farm on it!

[Shared Humanity]

I have read a lot lately about the heat morphing the clouds, so that the clouds now are not the same as a couple years ago. Could play a factor. I would not read a lot into area or extent numbers right now anyhow. There is just too much breaking up and dispersion.

P.S. Increased amount of ice that is headed for doom. Warm waters to the east and Fram to the south.I am only mentioning on this thread because its an example of ice that has affected both area and extent because it has dispersed so much.
NE Greenland in bottom left

I have to believe those large flows that are moving into the Fram are MYI. It is the only way they could maintain their integrity under these kinds of wind conditions.

[greatdying2]

I am still waiting for scientific proof...

There is no such thing as "scientific proof".

[weatherdude88]

Regardless,  AMSR2 from U Bremen sea ice concentration map shows a significant gain in ice area in the CAB. Our cyclone is long gone from the Canadian side of the arctic. 

[bbr2314]

... again, it is not a gain it is clouds, temps are warm and winds and waves are strong.

[budmantis]

Very logical A good argument for validating your hypothesis before betting the farm on it!

I think he would have robbed the bank anyway. (apologies to Wipneus for being off topic)

[Thawing Thunder]

... again, it is not a gain it is clouds, temps are warm and winds and waves are strong.

I second that. That's why I made occasionally an 48-hours-map from the bremen images: I wanted to get rid of the clouds and show the real extent of melt. Though in the last days I refrained from doing so, because there's too much drift even within 48 hours due to the storm (or should I say storms?).

[BornFromTheVoid]

This thread is drifting way off topic again, so I made a thread to discuss the role of cloud cover in passive microwave imagery here: https://forum.arctic-sea-ice.net/index.php/topic,1658.0.html

[Wipneus]

Updates of 2012 AMSR2 data seem to have been completed. We have complete data from the first of August to 30th September. Thanks for that.

Update 20160823.

Extent: -81.4 (-123k vs 2015, -733k vs 2014, -706k vs 2013, +784k vs 2012)
Area: -8.3 (-296k vs 2015, -976k vs 2014, -946k vs 2013, +484k vs 2012)
 
You will find the updated graphs in the top post

Regional extent declined in the CAB (-32k) and Laptev (-19k).

Regional area increased in the CAB : +53k. Area declined in Laptev (-26), Chukchi (-18k) and ESS (-16k).

The attached delta map is of the Beaufort-Chukchi-ESS corner. Lots of reds and blues confirm there is a lot of ice moving around. The detachment of the Wrangek ice progresses further.

[Wipneus]

And an animation of Laptev sea ice. The clearing of the NE Passage in progress, but needs a click to start.

[seaice.de]

Updates of 2012 AMSR2 data seem to have been completed. We have complete data from the first of August to 30th September. Thanks for that.

Thanks for your great work, Wipneus!

AMSR2 brightness temperature reprocessing is ongoing, reprocessing of version 2.2 will be completed within March 2017. 
http://gcom-w1.jaxa.jp/contents/160818_Ver2.2_release_e.pdf

[Random_Weather]

@ jdallen

You not showing the SST-Map, you showing the T2m of Arctic, SST ist this here: http://ocean.dmi.dk/satellite/plots/satsst.arc.d-00.png

[Random_Weather]

No Problem,

I see your point and see it the same way, there is no really ice increase, more its to clouds and fresh snow on ice, but its not good to make a strong text with having a wrong figure to explain the context.

[icy voyeur]

@ jdallen

You not showing the SST-Map, you showing the T2m of Arctic, SST ist this here: http://ocean.dmi.dk/satellite/plots/satsst.arc.d-00.png

Thank you for the attention to detail. Fortunately the oversight actually reinforces my argument.

I'm confused. The proper map shows all SST in the central arctic below 0C as opposed to your claim that none are below 5C. All versus none, and that 5C shift seem to more than invert your point.

In particular, if the tumbler has created enough "slush" to spread out across the surface, and that slush melts, doesn't that create a top layer of very fresh water in the slush? Now snow on that slush. I expect thin vulnerable ice.

I'm happy to be corrected and expect to be embarrassed so don't be shy.

[Lord M Vader]

Jdallen: the limit -1,8^o varies as the salt content does. The waters in the ESS and Laptev have, according to ACFNS (https://www7320.nrlssc.navy.mil/hycomARC/navo/arc_list_arcticsss.html), the lowest salinity levels and should therefore freeze at somewhat higher temps.

[weatherdude88]

AMSR2 will "flip" on a daily basis.  It fails mightily when used to assert massive changes - melt or freeze - have taken place on a day to day basis.  Its best and effective use it over longer time scales. Using it as you have, you are indulging your confirmation bias and doing bad science.
 

Update 20160820.
Regional area in the CAB increased: +31k. The Laptev regions went the other way: -23k.

Update 20160821.
Same for CAB area: +63k.

The CAB area has been increasing for a number of days now according to the Home brew AMSR2 extent & area calculation. So anomalously cold below freezing temperatures at the surface for an extended period at 80 degrees north cannot cause a several day freeze up? The adjacent water that is near its freezing point cannot have a small amount of heat removed to form ice? Why do the other regions show losses under similar weather conditions? The scenario of a temporary refreezes or clouds causing false losses are more logical than false gains. This is the scenario that makes the most sense using thermodynamics/physics.


 

As one example, consider the attached DMI 2 meter temperature map for today below.  One thing it shows pretty clearly is that we have no SST's

2 meter temperature anomaly maps are for surface temperatures and are indeed not SST's.

At a fundamental level, your posts fail to demonstrate a cogent understanding at the forces in play. 

As one example, consider the attached DMI 2 meter temperature map for today below.  One thing it shows pretty clearly is that we have no SST's currently across the central arctic below -5C. 

It is well known arctic sea water freezes at around -1.8 Celsius. (Yes, this changes with salt concentration) If SST’s were -5C we would have ice at the surface. (not arctic sea water) Please elaborate on your -5 SST Celsius claim.

[jdallen]

Jdallen: the limit -1,8^o varies as the salt content does. The waters in the ESS and Laptev have, according to ACFNS (https://www7320.nrlssc.navy.mil/hycomARC/navo/arc_list_arcticsss.html), the lowest salinity levels and should therefore freeze at somewhat higher temps.

/bemused amusement on/ 
Thank you folks for the additional details, which which I am aware of, and are relevant, but tangential to the major point I was making.  If anything, they *amplify* my point, that to make sweeping statements about the Arctic, you need a lot more than a couple days of the AMSR2 concentration map from the DMI.

[Random_Weather]

@ weatherdude88

You make it to simple, you have to look where it was cold, not only if its be cold enough, here the past 7 days resolution anomaly: http://www.karstenhaustein.com/reanalysis/gfs0p5/ANOM2m_arctic/ANOM2m_past07_arctic.html

the cold was on the side, where ice is not open and already frozen, also is to note, that winds are also play a role by refreeze and we saw often very middle to strong winds, so therefore between the ice should be small ekman-pumping

Therefore its seem very unlikly that there was any kind of refreeze

[weatherdude88]

My -5C claim is in part historical, in part from work presented elsewhere by others on these forums, and actually is shaded on the conservative side.  I've read from others here that the threshold may actually be a bit lower, at -10C.  However, I will do some additional research on it specifically.

Find an Arctic SST map from September/October with -5 Celsius water temperature. Better yet find any map where there is arctic sea water at -5 Celsius or even an example in nature. You might be looking for a while since it does not exist.

[weatherdude88]

@ weatherdude88

You make it to simple, you have to look where it was cold, not only if its be cold enough, here the past 7 days resolution anomaly: http://www.karstenhaustein.com/reanalysis/gfs0p5/ANOM2m_arctic/ANOM2m_past07_arctic.html

the cold was on the side, where ice is not open and already frozen, also is to note, that winds are also play a role by refreeze and we saw often very middle to strong winds, so therefore between the ice should be small ekman-pumping

Therefore its seem very unlikly that there was any kind of refreeze

You are proving my point. Go back and look at the sea ice area concentration maps. Some of those areas with the coldest anomalies saw an increase in concentration. Additionally, these anomalies are based on the long term mean.

[Random_Weather]

If so, then there was refreeze of melt ponds on the site to greenland, or what area do you mean?

[Lord M Vader]

Jdallen: I never wanted to amplify anything. Yes, it was a minor detail and tangential to your major point. I also see that you will hit 2000 posts very soon My early congratulations

[weatherdude88]

If so, then there was refreeze of melt ponds on the site to greenland, or what area do you mean?

The coldest regions of the arctic ocean had an extended period of anomalous cold surface temperatures based on the long term mean which is already well below freezing for 80 north and above. Here are the last three AMSR2 from U Bremen sea ice concentration map images. Look near the pole where we had anomalies. There is a significant increase in ice concentration in the last two days in this region.

[Random_Weather]

y, would not mean new ice, would mean melt pond freeze. Very normal on this time in the year.

[icy voyeur]

Typo - It should have said -5C.  Since corrected.

Keep in mind, that our starting point for freezing is -1.8, not zero C.  Between -1.8 and -5C the rate of transfer of heat out of the water isn't fast enough to support robust freezing, if indeed it is possible at all.  It doesn't take much wind to disturb the surface and generate circulation in the top few meters of water, and wind is something of which we currently have aplenty of in the Arctic.

Response here https://forum.arctic-sea-ice.net/index.php/topic,1658.msg87564.html#msg87564

[weatherdude88]

y, would not mean new ice, would mean melt pond freeze. Very normal on this time in the year.

Melt pond freeze over contributes to an increase in area. The argument here is "is CAB area increasing?" I am not going to get into an argument based on how much of the CAB increase is melt pond freeze over VS. new sea ice formation. These regions received significant increases the last several days based on AMSR2 from U Bremen sea ice concentration map images.

[weatherdude88]

Find an Arctic SST map from September/October with -5 Celsius water temperature. Better yet find any map where there is arctic sea water at -5 Celsius or even an example in nature. You might be looking for a while since it does not exist.

OK, this is exactly what I was talking about  - a Troll. 

You are trying to imply that I was suggesting water wouldn't freeze at -5C or -10C.  Nonsense.  That's nothing like what I was saying.

You appear to be posting this to accomplish two things (1) distract the audience by attacking my credibility and (2) disrupt the discussion.

Just stop.

Show me any example where the liquid water temperature in the arctic is -5 Celcius?

My -5C claim is in part historical, in part from work presented elsewhere by others on these forums, and actually is shaded on the conservative side.  I've read from others here that the threshold may actually be a bit lower, at -10C. 

Show me this historical claim. Perhaps rather than continuous ad-hominems, you can address with an example or provide some scientific evidence? 

One thing it shows pretty clearly is that we have no SST's currently across the central arctic below -5C.  That indicates pretty definitively that the transfer of heat is not yet high enough to support a refreeze. 

We have never had liquid water surface temps across the arctic at -5C.

[Random_Weather]

y, if melt ponds over freeze there is strong increase in sea ice concentration, i cant see your point, where there should be formed new ice if there is no open water?

Soory i see nothing what you mean

[epiphyte]

Find an Arctic SST map from September/October with -5 Celsius water temperature. Better yet find any map where there is arctic sea water at -5 Celsius or even an example in nature. You might be looking for a while since it does not exist.

OK, this is exactly what I was talking about  - a Troll. 

You are trying to imply that I was suggesting water wouldn't freeze at -5C or -10C.  Nonsense.  That's nothing like what I was saying.

You appear to be posting this to accomplish two things (1) distract the audience by attacking my credibility and (2) disrupt the discussion.

Just stop.

Show me any example where the liquid water temperature in the arctic is -5 Celcius?

My -5C claim is in part historical, in part from work presented elsewhere by others on these forums, and actually is shaded on the conservative side.  I've read from others here that the threshold may actually be a bit lower, at -10C. 

Show me this historical claim. Perhaps rather than continuous ad-hominems, you can address with an example or provide some scientific evidence? 

One thing it shows pretty clearly is that we have no SST's currently across the central arctic below -5C.  That indicates pretty definitively that the transfer of heat is not yet high enough to support a refreeze. 

We have never had liquid water surface temps across the arctic at -5C.

Argggh. why even do this? To re-iterate exactly the same troll beneath a quoted passage explicitly shooting it down is just going to p..s everyone off. If that was the objective, well mission accomplished, I guess - although what might motivate one to do so utterly escapes me.

[budmantis]

If everyone ignores this idiot (weatherdude88), he'll go away.

[Neven]

Even if someone is allegedly trolling, let's try and respect Wipneus' thread and take things elsewhere, 'kay?

[Reggie]

what Nevin said...ditto

[Shared Humanity]

Read the last 7 hours of comments and just realized I have wasted my time.