http://www.skepticalscience.com/DMI-data-on-Arctic-temperatures-Intermediate.html
regarding the article:
http://wattsupwiththat.com/2010/08/05/dmi-polar-data-shows-cooler-arctic-temperature-since-1958/
I can see that skepticalscience appears satisfied with the DMI data when you use the full year data – so what causes the summer temperature mismatch north of 80N between GISS data and DMI data?
Let’s refresh our memories:
A few days after the WUWT article, the DMI “melt season” was over and the final version of updated DMI 80-90N DMI summer (melt season) temperatures appears as follows:
Fig 1.
– Yes, the DMI melt season temperatures 80-90N in 2010 hit an all time low temperature record of just near +0,34 Celsius thus once again confirming the cold trend that started around 1991.
DMI trend summer 1991-2009: COOLING
GISS 80-90N temperatures june and july mostly projected up to 1200 km.
Fig 2
GISS june, july trend 1991-2009: WARMING
This does not make the GISS temperature projection method look good.
I can’t see how the writing at Skepticalscience.com should change that. I also showed other examples of problems with the GISS temperatures projected 1200 km over the ocean not really addressed in the skepticalscience article.
Normally when examining ice extent, believers of the global warming hypothesis mostly focus on the summer melt period. But now when a data source (the best data source for 80N-90N) shows temperatures for the melt period to be cooling of the area 80-90N, then we should look at the whole year. OK, lets then focus on the FULL year ice extent for the FULL globe based on Cryosphere data:
Fig 3
The 2010 column (an early prognosis) so far comes in number five since year 2000. That is, the fifth smallest global sea ice extent since year 2000.
So to begin with, the anomalies of global sea ice extent for 2008-10 appears to be just 0,3-5 mio sq km under normal.
However, Cryosphere in January 2007 made a Correction/reduction in Arctic sea ice data:
Fig 4
Here we see that the whole level of Arctic sea ice after year 2000 has been corrected down by Cryosphere with around 0,3 – 1,0 mio sq km. So this correction itself is perhaps large enough to fully account for the “missing” sea ice extent 2008-10. The strong La Nina cooling 1999-2001 is clearly reflected in the CT 2006 data, but not easy seen in the CT 2010 data.
So, without the Cryosphere correction done in January 2007, the sea ice anomalies 2008-2010 would have been zero or positive.
In my archives, I found this compare of arctic summer ice extents showing, that CT´s Arctic summer ice decline is over 1 mio sq km larger in 2007 than other data sources:
(Im not sure who collected these data.)
This indicates that the essential Cryosphere Jan 2007 correction may be an outlier.
Similar to the uncorrected CT data are the gridded NSIDC data presented by Jeff Id:
Fig 6
Again, the years 2008-2010 is not really supporting any downward trend, although the entire period 1978-2009 shows decline using a banal flat trend.
For both CT data and Jeff Id´s NSIDC data presentation we see that its in fact it is mostly the years 2005, 2006 and 2007 that shows a large dip in global sea ice extent. Take away those years, and where is the decadal declining trend?
When Jeff Id Zooms in on the years after 1995, it becomes clear, that the 3 years (2005-7) is responsible for downward trends if we use the banal flat trend argumentation for global ice extent:
Fig 7
Link to Jeff Id´s article:
http://noconsensus.wordpress.com/2009/12/09/sea-ice-copenhagen-update/
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I have been going back and forth with Tamino on sea ice:
http://tamino.wordpress.com/2010/10/15/go-ice-go-going-going-gone/#comment-45030
and since NSIDC does not offer a Global Monthly Sea Ice Extent Anomalies Chart, like they do for the Northern Hemisphere;
http://nsidc.org/data/seaice_index/images/n_plot_hires.png
and Southern Hemisphere;
http://nsidc.org/data/seaice_index/images/s_plot_hires.png
I figured that I’d take a shot at it. I used NSIDC’s corresponding data;
ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/Sep/N_09_area.txt
ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/Sep/S_09_area.txt
and first recreated the NSIDC Northern and Southern Sea Ice Extent Anomalies Sept 2010 charts:
http://i54.tinypic.com/30uby9d.jpg
http://i53.tinypic.com/nbxfs6.jpg
to check the data and methodology. I then merged the two data sets and created a chart for Global Sea Ice Extent Anomalies Sept 2010:
http://oi55.tinypic.com/2lmq4qw.jpg
Based on my calculations, the slopes of the trendlines are as follows:
Northern Sea Ice Extent Anomalies Sept 2010: Slope = -11.3% per decade
Southern Sea Ice Extent Anomalies Sept 2010: Slope = 0.8% per decade
Global Sea Ice Extent Anomalies Sept 2010: Slope = -2.6% per decade
If anyone would care to check my math, it would be most welcome. Also if Walt or Julienne are passing by, I really think you should add a Global Sea Ice Extent Anomaly chart to the right side of this page:
http://nsidc.org/data/seaice_index/
Jakers,
I wrote:
“Why do DMI make a green average line if we should not get the message that 2010 is significantly under average and 1991 significantly over average? ….
Why should anyone bother looking at the DMI data anymore if even such a huge dive in data means “nothing”? ”
You wrote:
“I think it is informational, as it is not a real data set or a continuous observational record. There is probably a good reason why DMI does not do that analysis with their model output.”
So in the summer when we daily checkout DMI temperatures for 80-90N, we should not care if they are far under average or far above average????
Well, you are entitled to you opninion, its a free world. Would you have said the same if this years DMI melting period temperatures where very high?
I live in Denmark, and these DMI data used to be taken quite seriously. In previous years when for example the melting period started a little early or late, this is something we heard about in the medias! This DMI melting period is indeed something that used to be taken serious. But this year, when the temperatures of the melting period is even record low, suddenly we have heard nothing about it in the Danish media.
ThomasJ, you write:
“The [Swedish] MSM, including the so called ‘public service’ [TV & radio] are so much more DDR/North Korea than anyone on this blog could ever imagine! I’m sorry, my Country hurts me!”
My country “hurts me” too. Its really grotesk how people with journalist background just eats even the most obvious warmist propaganda raw and sends it out to scare the poor Danes, Swedes etc etc. Its definetely my motivation: Peoble has their life qualitu reduced by this [SNIIIP] media coverage. Some dare hardly get children, its really so terrible. But it has the effect, that as a sceptic, everytime I hear this TV-propaganda i get up from my chair and start writing a new article 🙂 I simply cant live with this.
K.R. Frank
KR
I compared summer temperature trends from DMI (which DO originate from the 80N-90N area) with GISS temperature trends (thats just projected land temperatures!):
DMI trend 1991-2009 80-90N summer:
http://hidethedecline.eu/media/GlobalIceExtend/fig1.jpg
with
GISS trend 1991-2009 80-90N summer:
http://hidethedecline.eu/media/GlobalIceExtend/fig2.jpg
There IS a huge mismatch, and it IS 100% FAIR to raise this issue!! Nothing Appaling in this at all, please!
IF i had said that these DMI summer trends where indicator of the general (yearly) trend in the 80N-90N THEN it would be appalling. I used the example to show a clear mismatch between methods. I also in my original article showed a mismatch example from South America and the rest of the world how extremely poor GISS land temperature projection matches other sources of SST data than DMI, that is: i also showed that GISS ocean temperatures from land matches awfully with Hadcrut SST!
Now, unless your own stand here should be called “appalling” i think you should start considdering the 3 scientific issues actually raised here:
1) WHY is there such a mismatch between DMI and GISS for summer temperatures 80-90N ?
2) WHY do DMI summer temperature (melt season) go from far above average to far below average from 1991-2010 at the same time that the ice cover has been reduced?
3) And then from this last writing if you could please answer: WHY does Cryosphere from year 2000 strongly reduce their Arctic ice cover data??
Please try to answer these core questions in stead of claiming i say things i dont say, and then call it appaling. Please focus on what I actually write.
K.R. Frank
Hi Just the Facts!
Nice work!
Try to see fig 6 and 7 in the article above, the ones from Jeff Ids article.
Hes doing a little like what you do, just more on day to day data it seems. I think your idea to use such an alternative global ice indicator on WUWT is fine.
K.R. Frank
It says right on their page:
Calculation of the Arctic Mean Temperature
The daily mean temperature of the Arctic area north of the 80th northern parallel is estimated from the average of the 00z and 12z analysis for all model grid points inside that area.
The ERA40 reanalysis data set from ECMWF, has been applied to calculate daily mean temperatures for the period from 1958 to 2002,
from 2002 to 2006 data from the global NWP model T511 is used and
from 2006 to present the T799 model data are used.
Earlier info on DMI discussed the changes in data put in the models.
In light of all those temporal changes in data and methods, I don’t see how the graph in Fig. 1 can be seen as a true reflection of the actual temperature changes through time, especially in decimal degrees.
Frank Lansner says:
October 19, 2010 at 12:40 am
“HR says:
“October 18, 2010 at 10:37 pm
Why would you compare the 80-90oN temp with the full globe polar ice?”
Hi mr Ewing (?)
In my original article i simply pointed out that DMI´s melt-season over the Arctic is becoming colder since 1991. It really significant from + 1,3 Celsius to now just + 0,3 Celsius. This happens at the same time Sea ice is reduced and might indicate that other factors than melting have caused the ice retreat too. Never mind what the reason is, I think it is 100% scientifically fair to point out this interesting phenomena, that melting seasons are colder when the ice extend is reduced.
I cant say I have the full explanation.
The DMI data is based on (modelled) data from bouys actually in the ocean rather than GISS data that are sitet far away on land. Therefore, the GISS data to me appears weakest.
So, “Why would you compare the 80-90oN temp with the full globe polar ice?”
Its because Skeptical science when confronted with the 80-90N summer DMI data showing cooling, they dont want to find out how this can be, what is the exciting scientific explanation. They in stead want to look at something else, the 80-90N FULL year data which is warming. To this kind of escape from any cooling dataset I say: Ok, if you want to look at FULL year for some reason, why not go all the way and also look at the FULL globe??”
An explanation for the behavior at summer time temperatures at 80-90N was given at Skeptical Science through a communication from the DMI:
“From the link to WUWT, that you’ve attached below[*], it seems that a cooling temperature trend in the Arctic summer is present, throughout the past approximately 10 years. Where ‘summer’ is defined as the period where the +80N mean temperature is above 273K.
However, I very much doubt that a simple conclusion can be drawn from that, as there are complicating aspects to that analysis, e.g.:
1) The surface in the +80N area is more or less fully snow and ice covered all year, so the temperature is strongly controlled by the melting temperature of the surface. I.e. the +80N temperature is bound to be very close to the melt point of the surface snow and ice (273K) and the variability is therefore very small, less than 0.5K. I am sure you will find a much clearer warming trend in the same analysis applied to the winter period. The winter period is more crucial for the state of the Arctic sea ice, as this is the period where the ice is produced and the colder the winter the thicker and more robust the sea ice will become.
2) The +80N temperature data after 2002 are based on the operational global deterministic models at ECMWF, at any given time. Before 2002 the ERA 40 reanalysis is used. I.e. the +80N temperatures are based on 4 different models, the model used for the ERA 40 data set and the operational models T511, T799 and T1279. The point is that there can be a temperature bias in one or more of the models, that can cause the lower temperature level since approximately 2002, where the shift between the ERA40 data and the operational model data occur in the WUWT-plot from the link below.”
So focusing on this particular corner of the data is not really instructive and significant. Why would one want to do that? Because they want to get away from the idea that warming of the Arctic is an important and significant phenomenon.
Frank also writes:
“If we have GLOBAL warming, and ICE-extend for some reason is used to show this, then its fair to sometimes also examine the GLOBAL ice extend 🙂
Im not comparing 80-90N summer DMI data with any global data, im just showing the GLOBAL trends as well.
AND : Then i think that the Cryosphere-correction needed more attention, its not mentioned much on the net, so here it is graphically illustrated not to be forgotten 🙂
K.R. Frank”
The truth in science is not determined by some false equivalence or fairness. The summer ice extent is an important driver of global warming. Looking at the global ice extent, you are balancing the winter sea ice in the Southern Hemisphere, which reflects little sunlight because there is little sun in the winter, with the summer sea ice extent in the Northern Hemisphere, which does have an impact on the reflection of sunlight. Summer ice pretty much disappears in the Southern Hemisphere so that is not a factor in global warming at the present time. That is the reason attention is appropriately focused on summer sea ice in the Arctic.
I am surprised that someone who blogs as much as you do on climate wouldn’t understand that, but it is probably a matter of avoiding cognitive dissonance.
Jakers,
so you dont think that a trend going from the warmest ever anomaly for DMI melt temperatures 80N-90N in 1991 to the coldest ever in 2010 indicates anything at all. Why do DMI publish them if the biggest change possible indicates nothing?
Yes, the DMI data is stitched – so when DMI data gives a huge cold trend that you did not expect (?) you simply assume DMI doesn’t know how do a stitch?
But a “DMI-cant-stitch-when-data-are-not-what-I-expected” approach fails too because around 80% of the dive 1991-2010 happened before the stitch, 2002.
So, you HAVE to find another way to get rid of data not in line with the global warming mantra 🙂
There must be a way, right?
K.R. Frank
Eadler,
When focussing at summer conditions for sea ice as you recommend, im not sure why you would think its useless to also focus on the summer temperatures in the Arctic as I did to begin with in the article.
But since some readers are very sure that only all seasons, full year data can be of any interest what so ever, then I looked at full year data too.
your argument is that Summer ice is more important for the Earths Albedo than winter ice and thus that Arctic ice is more important than global sea ice.
This I find this to be oversimplified, but my first comment is:
One of the most important reasons to talk about sea ice in the climate debate is that the sea ice was supposed to back the argument that the earth is warming alarmingly.
If the global sea ice does not really show an alarming development, this fundamental argument becomes weak. Therefore it is relevant to talk about global sea ice.
Then a from a single point of view you claim that albedo from northern hemisphere is much much more important than the albedo effect on the southern hemisphere. I understand fully that the Arctic ocean –I if it was really empty for ice for example as in 2007 or worse, and for longer time perhaps, then you have a point. This does not happen in SH, obviously.
But…
A huge area around Antarctic is either water or ice. This has a HUGE effect on albedo. (And since the sun melts the ice in the spring time, the sun obviously reaches the areas enough to say that this albedo is important too, especially due to the huge areas we are talking about.
The corresponding area around the Arctic is land area and changes in Sea ice has no say.
And sea ice in SH reaches up to around 60S, whereas sea ice in NH rarely go that far south. So im not sure that you have a point, really.
But as I said, if sea ice, globally does not really confirm something dramatic, then its hard to use sea ice as argument for a warming globe. You can only use it to say that oscillations perhaps move heat from south to north. A little like the AMO indicates.
K.R Frank
I’m saying I don’t think you can use the “data” in Fig. 1 for any temporal analysis at all. It is not scientifically valid. It wasn’t generated correctly to do that, and DMI knows that, which is why they don’t do it. Sources change through time, models change through time, and there is no so-called “stitching” mentioned by DMI anywhere as they don’t intend the figures to be turned into numbers and used as a long-term data set.
KR Frank wrote:
“But…
A huge area around Antarctic is either water or ice. This has a HUGE effect on albedo. (And since the sun melts the ice in the spring time, the sun obviously reaches the areas enough to say that this albedo is important too, especially due to the huge areas we are talking about.
The corresponding area around the Arctic is land area and changes in Sea ice has no say.
And sea ice in SH reaches up to around 60S, whereas sea ice in NH rarely go that far south. So im not sure that you have a point, really.
But as I said, if sea ice, globally does not really confirm something dramatic, then its hard to use sea ice as argument for a warming globe. You can only use it to say that oscillations perhaps move heat from south to north. A little like the AMO indicates.
K.R Frank”
It seems that you still don’t understand what is happening, and why the focus is on the Arctic Sea Ice extent, and the Antarctic is ignored.
The Southern Hemisphere sea ice has always totally disappeared in the Summer time, so its contribution to absorption of sunlight, is a constant factor in time. This means that it is not affecting time dependence of global absorption of the sun’s light. The Northern Hemisphere Ice is decreasing its area/extent with time, in recent years, thus amplifying global warming. This is why it is of interest.
eadler says: October 20, 2010 at 7:56 pm
“The Southern Hemisphere sea ice has always totally disappeared in the Summer time”
No it doesn’t, note where the charts bottom out each year:
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.area.antarctic.png
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/seaice.recent.antarctic.png
Eadler, come on.
You write”The Southern Hemisphere sea ice has always totally disappeared in the Summer time, so its contribution to absorption of sunlight, is a constant factor in time. ”
You are only adressing conditions concerning the minimum… how about the maximum???? Thats what i have used time to write you about, please read my last answer again, and take your time before answering like this.
K.R. Frank