Latest News (hidethedecline)

Colder Arctic temperatures in the melt season vs. GISS temperatures.

Posted by Frank Lansner (frank) on 5th August, 2010
Latest News (hidethedecline) >>

GISS use a 1200 km radius for the temperatures measured from land temperature stations. The 1200 km radius is also applied to the coastal land/city/Airport temperature stations thus "covering" up to 1200 km over the oceans. GISS for example "covers" the Arctic ocean from land stations. 
   
Fig1
From DMI we learn, that Arctic 80N-90N temperatures in the melt season this year is colder than average. This was the case last year too, while earlier years in the DMI analysis period (1958-2010) hardly ever shows Arctic melt season temperatures this cold.
This is how DMI temperature averages for Arctic 80N-90N  melt season appears when plotted to allow compare over time:

Fig2 (When i speak of "the melt season" i refer to the period where temperatures 80N-90N are above zero Celsius. The green line above is the DMI temperature average, a little over 0,9 Celsius)
 
It seems that average Arctic temperatures 80N – 90N in melt season of the years 2004, 2009 and 2010 are araound 0,4-0,5K whereas the temperatures in 1991 and 1993 where around 1,3 K. In general DMI´s data (if correct) reveals a cooling from the mid 1990´ies till today.

The 80N-90N area of the Arctic is practically always ice covered. Therefore, the 80N-90N is perhaps not so affected by heat from the other areas of the Arctic that has been still more ice free in the period 1995-2007. Im not sure why DMI shows such a cooling trend for the 80N-90N area, but it could appear as if the ice covered areas of the Arctic has its own history of temperatures? And how should GISS data from distant land stations account for this?

Here´s how GISS temperature appears when comparing 1991 to 2009 for the Arctic Polar region:
 


Fig3
The Arctic melt season is mostly June and July. For both months the GISS Arctic temperature trend 1991 vs 2009 shows warming around 0,3-0,7K which is in contrast with the DMI trends of cooling of around 0,7K for the region.

Is it basically a convincing idea to use land/city/Airport temperatures for temperatures at sea? Give it a thought:
Imagine you stand on a boat 12 km from land…
 


Fig4
You want to know the Air temperature in 2 meters altitude. Which temperature would be most precise, the water temperature around the boat or the temperature from land ( measured at the city airport... ) ?
Now imagine the same situation, but this time you are 1150 km from land. Which temperature would you rely most on, the water temperature around the boat or the temperature from land (city/Airport) 1150 km away?
Since 1987, James Hasen, and thus GISS, har used a 1200 km radius in their global temperatures based on meteorological stations and thus extended land temperatures to cover a considerable ocean area.
Below a compare of SST with the temperatures GISS use for ocean areas. 
 
Fig5
1) Left: GISS land temperatures including land temperatures to cover ocean areas.
2) Right: As 1) but now for the ocean areas the actual SST measured by the Hadley centre are shown.
Both pictures are from june 2010. From this illustration we see, that ocean areas represented by SST are poorly reflected by GISS land temperature data and the idea of expanding land temperatures to cover ocean area appears challenged?
 


Fig6
Above I have shown first the ocean areas GISS included in their global temperature data from meteorological stations and then the corresponding SST from hadley. Again, this is June 2010 data, and its from  http://data.giss.nasa.gov/gistemp/graphs/
(In addition to land data, GISS use a few airtemperature inputs from ships in fixed position).


In the compare above, a few areas (between Sri Lanka and Singapore and around Spitsbergen and slightly more) has GISS ocean temperatures warmer than Hadley SST, but more often the GISS temperatures for oceans are warmer than the Hadley SST. In general, differences are numerous and significant, SST does not support GISS “ocean” data obtained by expanding land area temperatures to cover for ocean areas.
One cant help wonder why not MAT (Marine Air Temperatures) are not used widely in stead of both SST and GISS land temperatures?
 

Last changed: 6th August, 2010 at 00:42:40

Back

Comments

Hej Klaus! By Frank Lansner on 22nd October, 2010 at 21:58:06
Her er de årlige værdier i exceæ format. Jeg har ikke helt præcise tal idet jeg som antyddet har estimeret ud fra forstørret grafik.
http://hidethedecline.eu/media/BLANDET/DMIIS.xls

Til gengæld tror jeg at Peter Hogarth på skeptical science ligger inde med tal op til 2009.

Bedste hilsner, Frank
Thanks, please create an excel sheet with data.. By Unknown on 22nd October, 2010 at 17:38:16
Hi Frank

Thank you for your answer.

I will be pleased if you can produece an excel sheet with the data used in the graph. It is difficult to convert the graph to numbers.
I need the numbers.

Regards
KFL
CBH, again we agree on something: By Unknown on 22nd October, 2010 at 16:24:13
you write: "you could theoretically get a reasonable estimate of the temperatures of the entire NH with just 60 well-placed measure stations."

YES!
Check out the Angel and Korshover data:
http://hidethedecline.eu/pages/posts/part2-the-perplexing-temperature-data-published-1974-84-and-recent-temperature-data-183.php

THey used just 42 stations spread out on the globe and they found excellent match with the other teams with far more data.

But CBH, in this case the have a global focus, and many are stations are spread out in the sea. In the Arctic, GISS uses landstations to project over sea and ice areas with very different mechanisms. THIS is very naive indeed. And the DMI data does not at all support GISS 80N-90N in the melting season.

Apart from this, here are compares between GISS projected "SST" on top, Hadley measured SST below. For some reason, even though ANgel and Korshover get away with this, it seems that GISS projection tends to be warmer than measured SST in average. At least the match appears random at best:

http://hidethedecline.eu/media/ArcticGISS/fig6big.jpg
Hi Klaus! By Unknown on 22nd October, 2010 at 16:12:21
I estimated the period to be 66 days from the DMI graph (days in the buttom) and then i made a grid of 66 days, estimates for each day and took the average. For each yera same procedure, same period, here 2010:

http://hidethedecline.eu/media/BLANDET/dmi2010.jpg
Here I end up with an average below +0,4 C (violet line)

I think the 66 days runs from around 11 june to 16 aug.

I havent found source to original data, but perhaps if asked DMI, they might help :-)

Thankyou for the interest!

K.R. Frank Lansner
Data problem/Verify By Unknown on 21st October, 2010 at 18:25:28
Dear Frank

I have tried to verify the data on the graph and looked into the excel spredsheet with the data as indicated by you.

In the sheet, the years are not shown and the number of points in the sheet and in the graph does not match.

I will be pleased, if you may include the years and explain how you have calculated the temperatures from reading pixels.

Have you for one year taken the midpoint between min and max reading ?

How many readings have you made for one year ?

Which months/weeks are used in the calculations for each ?

Do you use the same weeks/months for each year for your reading ?

Do you know where to find original DMI data/not graphs ?

I think the graph is interesting and I will try do an statistical analysis of the data to se what the data tels.
Sigh.......... By Unknown on 21st October, 2010 at 00:32:04

Frank,

once again, you fail to understand how something which is imprecise on a fine scale makes perfectly sense on a coarse-grained scale.

Actually, as you can see here:

http://www.realclimate.org/index.php/archives/2007/07/no-man-is-an-urban-heat-island/

then if you estimate the degrees of freedom from the temperature anomaly of the Northern Hemisphere, then it turns out to only about 60 - i.e. you could theoretically get a reasonable estimate of the temperatures of the entire NH with just 60 well-placed measure stations. Of course, all stations are not well-placed, but still, there are much much more than 60 used in the GISS series. So for the global anomalies, there are very good scientific arguments not to "leave them grey". This is basically what HADCRU does - and this creates a cooling bias on the global anomaly since temperatures are rising faster in the Arctic than anywhere else.

And once again: If it were true that GISS´ extrapolation in the 80-90N made it biased and unreliable, then the DMI series should diverge from GISS on a yearly basis. And they are, as even you appear to realise, almost the same - if anything, DMI´s trend is even higher than that of GISS.

So once again-again:

"We have heard that the Arctic is warning faster than any where else, largely based on (STUPID) measurements taken from land stations."

- and this is confirmed by every single measurement series - AND BY DMI. There is simply no way you can escape this simple fact. This is what you have been repeatedly told and patiently shown by several people, including many on sceptic-friendly sites.

As said repeatedly, I do not expect you to learn something. But I am indeed amazed by the way how you (over and over again) are apparently unable to grasp quite simple factual corrections and yet able to convince yourself that you are on to some major scientific revolution. It is also striking how you can be so (excuse me) ignorant of the basics and unaffected by even politest rebuttals you receive. Do you ever start to wonder whether it might just be that you had misunderstood rather than everyone else failing to see your genius?
No offense, but you are a good example of the Dunning-Kruger effect:

http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect

"The Dunning–Kruger effect is a cognitive bias in which an unskilled person makes poor decisions and reaches erroneous conclusions, but their incompetence denies them the metacognitive ability to realize their mistakes. The unskilled therefore suffer from illusory superiority, rating their own ability as above average, much higher than it actually is.

Kruger and Dunning proposed that, for a given skill, incompetent people will:

- tend to overestimate their own level of skill;
- fail to recognize genuine skill in others;
- fail to recognize the extremity of their inadequacy;

- recognize and acknowledge their own previous lack of skill, if they can be trained to substantially improve".

Therefore, the best I could wish for you was to actually spend some time with scientists and learn something about the scientific method and process, and maybe understand some basics about climate science. I really think that would open your eyes and make you see things in a totally new light.

KR Christoffer
I agree By Unknown on 20th October, 2010 at 14:45:22
CBH, you write:
"..and you are certainly right that extrapolating from e.g Qaanaq would not the optimal way to find out the local temperature on some specific place on the sea ice near the geographical North pole.."

ohh... Yes.. and this is what i called "STUPID" in the comment you quote. So CBH, why do GISS not mark that area GREY in stead of wildly guessing on temperatures near the north pole ?????????

K.R. Frank Lansner
Another comment to put things in perspective. By Unknown on 19th October, 2010 at 19:51:43
A similar comment from Nikula can maybe highlight the problem even more. Frank wrote:

"You seem much more interested in finding some flaws in what I did rather than focus on the really really interesting here: We have heard that the Arctic is warning faster than any where else, largely based on (STUPID) measurements taken from land stations. And then both your and my reading shows this to be far from the whole picture (!)
But all you care about is your claimed “faults” by me ?"

What is really interesting here (=the big picture) is whether the Arctic does indeed warm faster than anywhere else, as claimed. And what the reading of Nikula (and that of Hogarth and I) shows, that yes, this is in fact the case. Is does not matter at all whether you look on GISS, DMI or anything else - the yearly arctic warming trend is even bigger in DMI, your preferred dataset than in GISS, and this does not even matter for which latitudes you compare.

And the fault of yours that people are complaining about is that you can only escape that conclusion by excluding 5/6 of the year, where the changes are several order of magnitudes bigger than your summer decline. This really is something that should trouble few if any high school students.

But once again, I certainly do not expect you to understand nor to actually learn anything. Which makes me feel a bit sorry for you.
GISS also does use SST By Unknown on 19th October, 2010 at 18:12:59
Besides, even though you appear to believe otherwise, GISS does in fact use SST on lower latitudes, and your example suggesting that GISS extrapolate land temperatures from South America into the Pacific is either gross ignorance or deliberate manipulation.
As Hogarth says, they just cannot use normal SST above 75N for a pretty obvious reason (the sea surface is frozen).

Again, I, of course, do not expect you to actually learn anything from this, let alone correct or even acknowledge any of your major errors.

Christoffer
No contradiction between GISS or DMI By Unknown on 19th October, 2010 at 17:58:20
Frank,

you are trying to cast doubt on the reliability of GISS. What you fail to understand is that there is no contradiction between the trends in GISS and DMI, even if one accepts every word of your post.

It is not surprising that there is little temperature change above the middle of a floating ice sheet - and you are certainly right that extrapolating from e.g Qaanaq would not the optimal way to find out the local temperature on some specific place on the sea ice near the geographical North pole, and thus it is neither surprising that individual or fine-grained measurements for regions with few stations display some variability compared to GISS, which might be dramatic over very short intervals. Here, GISS should be used with some caution, and I do not think that even James Hansen would dispute this. This is basically a problem of the same type as when the climate models perform poorly on a regional level while still performing fine on a global scale.

But if you want to claim - or simply imply - that this makes GISS trends unreliable, then you would also need to show that this matters for the increase in Arctic temperatures measured in GISS on the larger scale. This analysis is what Hogarth did, and it turns out that on this scale, DMI in fact agrees perfectly well with GISS - DMI even has a slightly higher (though obviously nonsignificant) increase over the years, even when you analyse GISS´ data from 90N down to 64C below the Arctic circle.

(By the way, this also refutes another of your old favourites about arctic warming being an artefact of including places below 70N - exactly wherefrom you pulled that number as a definition of "Arctic", I fail to understand, but in any case what you have said is clearly wrong, no matter what definition you use.)

But I digress: The main point here is, as Hogarth explains very clearly, that despite your trying to make it appear to the contrary, there is in fact no contradiction between GISS and DMI, nor between other similar temperature series, with respect to the significant Arctic warming found since 1958.

If you had a serious interest in (let alone understanding of) this, you would try to learn something from this and, take down what you wrote and apologise for the misunderstandings. But somehow, I have a feeling that you will just repeat yourself that the summer DMI data show that something is very fishy about GISS, that AGW is collapsing and that everybody will someday realise that you were a genius etc. etc., and maybe throw in a couple of distractions or new misunderstandings.

Christoffer
Hi Christoffer, long time no see :-) By Frank Lansner on 19th October, 2010 at 11:37:15
I have simply shown that DMI data shows a declining trend of melt season temperatures from +1,3 C in 1991 to + 0,3 C in 2010 for the area 80-90N. This is what DMI data shows and it totaly contradicts what GISS land based estimates for the June and July trends 1991-2010 shows.

This is true.

From there we can considder: Is it DMI data that is faulty or GISS?

And then somehow you think that because DMI data shows significant cooling for the melt season (june + july mostly) then im saying that the DMI for the whole year shows cooling. And this you and some others have "debunked". Bravo, a few readers has debunked something i never said, bravo :-)))

K.R. Frank Lansner
"How AGW is totally disproven again" (or just another classic basic mistake) By Unknown on 19th October, 2010 at 01:14:39

Frank,

I suggest that you check out Peter Hogarth´s detailed debunking of your argument. Basically, you have simply ignored the inconvenient fact that while summer temperatures show little difference, winter temperatures have risen sharply - by far outweighing the changes in summer.

http://www.skepticalscience.com/DMI-cooling-Arctic-advanced.htm

Even at the WUWT page, where people would normally be sympathetic to your point of view, Bob Tisdale and Petter Nikula had to tell you that you had used the wrong SST dataset and apparently failed to understand that most of the melting goes on in the edge of the sea ice. I quote Nikula:

"I’m really sorry, but nothing I have seen here so far can convince me that the analysis is to be taken seriously. There are a number of very basic flaws which any first year university student would be able to catch in your methodology and your reasoning. And the claims you make are completely beyond the scope of your analysis".

So no, there has been no cooling in the Arctic, neither according to GISS nor DMI.

All you need to disprove global warming is indeed good data and arguments - but somehow it turns out that everytime you think you have found a good argument, there is a very basic flaw or misunderstanding on your part. And yet, I have never seen you learning anything substantial from any exchange, no matter how grotesque your error has been or how kindly it has been pointed out to you, even by the best of researchers.

I am under no illusions that Hogarth´s rebuttal will be an exception, but I will post it here for any casual readers.

Feel free to repeat yourself about how AGW is just on the verge of collapsing, how it is just a matter of time before everybody realises it and why your climate scientific skills surpass those of Hansen, Peterson or Jones if you honestly believe this. :)

Christoffer Bugge Harder
Sad but true By Frank Lansner on 6th August, 2010 at 16:12:16
The end of the global warming movement seems to take more than good arguments and data - first when temperatures (due to the ongoing Solar low activity) reach temperatures so low than no adjustments can hide it, when the ice grows massively, then this night mare of pseudo science will pass away.

What a relief, and as you say, then we have to fight with many decades of colder times when we will remember the good old warm days. Thanks for your words!
K.R. Frank Lansner
Getting colder? By Unknown on 6th August, 2010 at 05:44:54
Thanks for the instructive post.

A steady on-going temperature drop over the next 6-24 months would do a lot to demolish the CAGW cause, and that scenerio looks more likely all the time.

Further, if solar theorists (Landscheidt, Eddy et al) are right we are heading for global cooling from 2010 to a low in 2030 and not recovering to present global levels until 2070. It will be miserable compared to our wonderful, comfortable climate of the present.

How ironic if this ends up being the case...

Add Comment