Sunshine hours writes:
I have been a bit worried about the deep deep dive in Antarctica Sea Ice Extent.
It appears to be a processing or sensor error. As of today the NSIDC data confirms it. (see image below)
In a deja vu all over again moment, I find that it isn’t just the Antarctic with wonky readings.
I agree that looks like a sensor failure of some sorts, and this NSIDC graph looks a bit odd as well. NSIDC uses a 5 day average, so the transients get smoothed out.
![S_stddev_timeseries[1]](https://wattsupwiththat.files.wordpress.com/2014/07/s_stddev_timeseries1.png?quality=75)
This Arctic graph from DMI has what appears to be a spurious element also, note the recent uptick:
![icecover_current_new[1]](https://wattsupwiththat.files.wordpress.com/2014/07/icecover_current_new1.png?quality=75)
(added: that uptick could be wind affecting the ice extent, or it could be a sensor/processing issue, we simply don’t know)
Tomorrow there will be a new paper released by former NSIDC scientist Walt Meier and others that tries to argue that some of the record sea ice extents from Antarctica recently are a victim of an adjustment in a processing algorithm that changed in 2007 from Version 1 to Version 2.
Back in 2009, in the “deja vu moment” I wrote:
In the prior thread I raised a question of why there was a large downward jump in sea ice extent on the graph presented by NSIDC’s Artic Sea Ice News page. The image below was the reason, dozens of people called my attention to it in emails and comments overnight because in the space of a weekend, a million-plus square kilometers of Arctic sea ice went missing.
Walt Meier wrote this response that he later had to eat crow for:
Thus errors do happen from time to time and one shouldn’t draw any dramatic conclusions from recent data.
I’m not sure why you think things like this are worth blogging about. Data is not perfect, especially near real-time data. That’s not news.
Now, Meier has an entire paper about such errors, and the error is far lower in magnitude than that incident where the sensor actually did fail and NSIDC was caught napping.
It makes me wonder just how good these estimates of sea ice are; what else awaits discovery?
I wish they’d exhibit the same investigative zeal when it comes to looking at Arctic sea ice record low extents, for all we know, the 2007 low extent might also be a victim of the same algorithm shift that occurred that year.
But you see, confirmation bias prevents such investigations, they expect the Arctic to be low, so they only looked at the Antarctic where there’s more ice than there is supposed to be. To paraphrase their viewpoint on it: “it didn’t look right”.
On the plus side, it looks like this means the AR4 and AR5 reports are wrong about sea ice extent values, and as we know IPCC reports are wrong about a lot of things.
We’ll have that new paper here at 6AM ET tomorrow.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Hmm, an “adjustment in a processing algorithm that changed in 2007 from Version 1 to Version 2”, huh? I’ve posted about this before — up to 2007 the ice extent charts showed a bump on the 1st July where they turned the satellite polarizer filter from “Antarctic mode” to “Arctic mode” — the problem is whether to interpret open water amidst ice as being surface water (and thus still ice-covered) or true open water. Meier’s “version 2” was their 2007 decision to twiddle the dial gradually to make the July bump disappear. This new hands-on approach rapidly degenerated to where they would turn the polarizer any way they wanted to get the desired ice extents — the one downside was that changes in the Arctic ice extent were counterbalanced by opposite changes in the Antarctic ice extent — the charts since 2008 clearly show the simultaneous opposite tweaks in the two, and this has also contributed to the increased volatility in ice extent since then. Now Meier is proposing a fix? Actually, it’s already happened — the recent fall in the Antarctic ice extent is that fix being applied, and the fix is simply that open water will henceforth be treated as true open water in both hemispheres, all the time. No more surface melt ponds allowed!
This means that today’s ice extents are no long comparable to earlier records. It will also cause a new data artifact — at the ice extent minimum, the surface melt ponds freeze over which will now be seen as a sudden jump in true ice extent, so this year’s Arctic sea ice minimum will end early, with a sudden jump in ice extent. This is what happens when you put monkeys in charge of the controls, so watch that space!
Note that my interpretation, presented here, is not from any insider knowledge about satellite operations, but just inference from years of watching the data. Heavy thumbs on the scale leave big impressions on sensitive data.
NZ Willy says:
July 21, 2014 at 12:34 pm
“This is what happens when you put monkeys in charge of the controls, so watch that space!”
ROF, two thumbs up — well said!!!
dbstealey says:
July 21, 2014 at 9:53 am
ossqss,
I have to laugh at the ridculous PIOMAS propaganda in your link above. Their Arctic chart makes it look like ice cover will hit zero any minute now.
Only if you don’t know how to read a graph. With a minimum volume in September of about 5 km^3 and a trend of -3.3 km^3/decade their data indicates about 15 years to hit zero, hardly ‘any minute now’.
But they never post a graph of the Antarctic.
Why would you expect the Pan-Arctic Ice Ocean Modeling and Assimilation System to do so?
@ur momisugly Bill Illis The Arctic sea ice is exhibiting very unusual circulation patterns this year which should lead to a substantial increase in the retention of multi-year thicker ice.
Agreed. The Arctic Ocean has had above average ice. As the summer progresses, the peripheral seas approach their minimums, and the overall Arctic sea ice area of becomes increasingly dominated by that basin’s statistics. If Arctic Basin trends continue, at the Arctic’s minimum sea ice area, there should be more ice in 2014 than there was in 2013.
The Ghost Of Big Jim Cooley says:
July 21, 2014 at 12:13 pm
MikeP, I’m with you, in that I regard ‘data’ as plural, so must be followed by ‘are’, not ‘is’. I pulled Leif up on it once, but got shot down. I argued the point, but got bored. ‘Data’ is the plural of ‘datum’.
And ‘agenda’ is the plural of ‘agendum’.
Phil. says:
…their data indicates about 15 years to hit zero, hardly ‘any minute now’.
“Any minute” was hyperbole and sarcasm. But if you want to make a wager that in 15 years the ice will be gone, I’ll fade you. Long Bets is a good place for that. $1,000.00? ☺
Phil. asks:
Why would you expect the Pan-Arctic Ice Ocean Modeling and Assimilation System to…
…show the Antarctic?
Simples: to avoid cherry-picking, and for a complete picture of global ice.
Fowler, in Modern English Usage, states, “Latin plurals sometimes become singular English words (e.g., agenda, stamina) …” As long as it’s OK to employ those words as singulars, it’s OK to do the same for “data.”
Not only is it acceptable to use “data” as a collective singular, using data as a plural word is incorrect because it throws the speaker (including those who use “data are”) into inconsistency with his habitual method of speaking, as Phillip W. pointed out [in an earlier thread]. He wrote: “‘Data’ is naturally and consistently used as a mass noun in conversation: the question is asked how much data an instrument produces, not how many; it is asked how data is archived, not how they are archived; there is talk of less data rather than fewer; and talk of data having units, saying they have a megabyte of data, …” For another example of this usage, look at the post just above this one [in an earlier thread], where the phrase “the raw data is gone” is used.
Because of this inconsistency with long-established and near-universal usage, and because, as Fowler shows, there is no real rule forbidding “data is,” “data are” will never be accepted–it will always sound odd or even affected [except in scientific journals, where it’s become a fossilized convention].
Gee, I don’t see the global insurance rate for the Northwest Passage marine transport going down. In fact it does not exist.
over a short period of years, reversion to the [mean] should be expected.
errata: reversion to the mean
“Tomorrow there will be a new paper released by former NSIDC scientist Walt Meier and others that tries to argue that some of the record sea ice extents from Antarctica recently are a victim of an adjustment in a processing algorithm that changed in 2007 from Version 1 to Version 2.”
Yeah right…
John says:
July 21, 2014 at 12:30 pm
The Ghost Of Big Jim Cooley says:
July 21, 2014 at 12:13 pm
Though you’re correct, We’re not in “old” Rome.
data |ˈdatə, ˈdātə|
noun [ treated as sing. or pl. ]
==============================================
No sir, we ARE the gladiators rebelling against Rome. To date, Spartacus still stands.
Borrow Sea Ice Webcam shows sea ice on July 21, 2014./
Here is the link: http://seaice.alaska.edu/gi/observatories/barrow_webcam
There is a fantastic new tool for observing the sea ice. “Worldview” which packages the Modis satellite pictures in a much more useable manner.
On the top right of this link, one can switch to North Pole view and then zoom-in closer and closer with the mouse wheel. At the bottom, one can pick different dates, a day at a time for example. I click backwards from the current date (looking for cloud-free pictures) and then once the picture is cached by your browser, you just move back and forth between the dates and watch how the ice moves from day to day, 2013 vs 2014 etc in animation-type mode. There are also options to download pics from any area on whatever days on wants as well as create quick-links to the view you have zoomed-into for others.
Click the “take a tour” to “don’t show again” so it doesn’t come up each time you visit. Required bookmarking for any sea ice follower.
https://earthdata.nasa.gov/labs/worldview/
In addition, you don’t have to go to the polar view and can zoom-in anywhere on Earth you want with this new highly useable website.
Arctic sea ice extents is right at a -2 std deviations below its declared average for this date. Antarctic sea ice extents right at the +2 std deviations for today’s date. BUT! Antarctic sea ice is gaining now – rising towards its yearly maxim,um between mid-September and early October. So a 2 std deviation area from an average of 13.0 Mkm^2 at latitude 62 – 61 south is MUCH more important than a -2 std deviation at latitude 78-79 up in the Arctic from a Arctic area of 3 – 4 Mkm^2 at ALL dates of the year between mid-August through early April.
Thus, in calculating reflected solar energy, for 7 months of the year, Antarctic sea ice – which for 5 years has been increasing steadily – dominates the solar energy balance. For only a short 5 months of the year does the Arctic get more energy than the Antarctic.
In fact, when the Antarctic is at its maximum in mid-September, and the Arctic sea ice at its minimum, the edge of the Antarctic sea is getting FIVE TIMES the solar energy than the edge of the Arctic sea ice. Increase the Antarctic sea ice by 1.0 million sq km’s in September? You would have to delete ALL of the Arctic sea ice to get the same apparent solar energy. But that cannot happen: There isn’t 5.0 Mkm^2 of Arctic sea ice to melt!
But that’s NOT the message that Meiers wants released to his obedient minions in the well-paid highly-funded official government propaganda machine.
At today’s rate of increase for Antarctic sea ice since 2005, the Straits of Magellan and Cape Horn (latitude 56 south) will be closed to ocean traffic within 8 – 12 years.
For example, I saved this pic for posterity from a few days ago of the Petermann glacier on the northwest of Greenland. The glacier has moved about 5 kms (as glaciers do) from the grey line where it last calved an iceberg in 2012. It will calve again in a few to several years to a decade.
http://s16.postimg.org/i5j9glgvp/Petermann_July14_14.jpg
But, show me your calculations to disprove the following:
From today’s Arctic sea ice extents, for all days after late August, the open Arctic water up past 78 – 79 north LOSES more heat energy when it is “open” (with no ice coverage) due to increased evaporation, increased convection, increased conduction and increased long-wave radiation than it gains from the few hours a day of low angle sun that is mostly reflecting off of the water at solar elevation angles of 4 – 6 – 12 degrees . From today’s limits of 3.5 million sq kilometers (2007, 2012) to 4.0 Mkm^2, ice-covered waters after late August act only as a blanket to retard heat loss to space.
Mid-summer? Different story. The sun is up higher in the sky.
But after mid-August? More Arctic sea ice loss = more heat loss from the ocean into space.
But the Antarctic sea ice is almost ALL “new” clean ice. It doesn’t get the very low albedo typically measured by Curry during the mid-summer up north. And that “clean” highly reflective Antarctic sea ice is growing up at latitude 61 – 60 – and 59 south. Down south, more sea ice = MORE solar energy reflected back into space at ALL dates of the year! .
Sea ice is nice. I like it. The only problem is that it will soon be all ice free! We must act against sea ice reducing car fumes. You have been warned.
All the quotes from my last comment are from Professor Peter ‘ice free Arctic’ Wadhams in 2016.
I wonder why a press report from NOAA regarding the average temp the past 352 months was removed . It was directly from the Canadian Press and in the context of all the the varied reports on Sea Ice from all the different agencies it might fit in because it is so opposite of what has been said about the cooling cycle we are going through I just thought there was a parallel.
Expect some acceleration in the Arctic melt, warm atmospheric circulation from the Bay of Biscay, all the way to the central Arctic
http://earth.nullschool.net/#current/wind/isobaric/1000hPa/orthographic=0.00,54.00,671
Where was that announcement removed from? What site or what board?
Racook this thread around 9.30 am July 21 copied and pasted from the Canadian Press. Re the NOAA announcement that for the last 352 months the average temp worlds wide has been increasing. I will try to get again but it’s bed time I will try tomorrow July23