GISS vs CRU/RSS/UAH

[nzucker]

The anomalies don't match exactly, but the fact is that as is often the case, GISS has an area of major warmth (in South America this time) that is non-existant in the AMSU map. Both maps have very similar anomaly patterns over North America, Europe, Australia, and most of Asia and Africa. But for some reason, GISS had a heatwave in South America where the satellites had nothing of the kind.

Yeah this is what I mean...there is always going to be some acceptable divergence on the magnitude of the anomalies, but GISS shouldn't have a huge heat wave where RSS has average temperatures. Where is this coming from?

[skierinvermont]

Wrong. That doesn't explains why GISS has been running warmer than HadCRU 60/60 since the early/mid 2000s.

I said it wasn't the extrapolations and you are telling me I am wrong? Well first of all, the divergence between GISS and HadCRUT 60-60 is small compared to the divergence between surface and satellite. HadCRUT and GISS are in tighter agreement than the satellites are for any periods you select. Select a period, and the surface data sources will be in closer agreement than the tropospheric sources for the same period.

Second of all, it ISNT the extrapolations causing the divergence between GISS and HadCRUT 60-60 since the early to mid 2000s.

We can perform a very simple test of this.

Remove the extrapolations from GISS 60-60 and then compare to HadCRUT. One can use the 250km version of GISS, although one must keep in mind that this is even shorter extrapolation than HadCRUT.

I performed that test from 2004-2010.

GISS 1250km extrapolation: +.0017C/year

GISS 250km extrapolation: +.0003C/year

HadCRUT: -.0040C/year

Difference between the two GISS versions: .0014C/yr

Difference between GISS 250km and HadCRUT: .0043C/yr

As you can see GISS 250km and 1250km are in much closer agreement than they are with HadCRUT. The difference between HadCRUT is over 3X greater than the difference between the two versions of GISS. If GISS offered a 350km extrapolation, which would be more equivalent to HadCRUT, I imagine the two GISS versions would agree even closer.

This once AGAIN shows that it's NOT the extrapolation. It is the DATA used and other aspects of the methodology.

Remove the extrapolation from GISS and most of the divergence with HadCRUT remains from 2004-2010. It's the raw data used, and other aspects of the method, NOT extrapolation.

[skierinvermont]

The anomalies don't match exactly, but the fact is that as is often the case, GISS has an area of major warmth (in South America this time) that is non-existant in the AMSU map. Both maps have very similar anomaly patterns over North America, Europe, Australia, and most of Asia and Africa. But for some reason, GISS had a heatwave in South America where the satellites had nothing of the kind.

Yeah this is what I mean...there is always going to be some acceptable divergence on the magnitude of the anomalies, but GISS shouldn't have a huge heat wave where RSS has average temperatures. Where is this coming from?

The difference between GISS and RSS over the region in question in S America is 1-3C. GISS is 2-4C too cool in much of Europe. You are getting confused by the color coding of GISS that paints small 1-2C +anomalies bright orange and 2-4C anomalies bright red. The error in Europe is nearly twice as large and cover over 2-3 times the area of the one in S America.

RSS has a large area of massive +4-6C anomalies in Europe which GISS is completely missing and gives only 1-2 and 2-4C anomalies.

Oh and let's not forget about the non-existent cold-spell GISS created over SE Asia Vietnam to Thailand to Singapore. GISS has it -1 to -2C even a speck of -2 to -4C.. RSS has it -.5 to +0C.

[nzucker]

The difference between GISS and RSS over the region in question in S America is 1-3C. GISS is 2-4C too cool in much of Europe. You are getting confused by the color coding of GISS that paints small 1-2C +anomalies bright orange and 2-4C anomalies bright red. The error in Europe is nearly twice as large and cover over 2-3 times the area of the one in S America.

RSS has a large area of massive +4-6C anomalies in Europe which GISS is completely missing and gives only 1-2 and 2-4C anomalies.

Oh and let's not forget about the non-existent cold-spell GISS created over SE Asia Vietnam to Thailand to Singapore. GISS has it -1 to -2C even a speck of -2 to -4C.. RSS has it -.5 to +0C.

No, but I'm mentioning the South America "heat wave" because GISS has a completely different spatial pattern/anomaly.

Both sources show Europe/Asia way above average, it's just a question of magnitude...and it might not be that much of a difference because that color on GISS is 2-4C leaving lots of room for interpretation.

RSS does show SE Asia below average, just not as much as GISS. But the spatial pattern is similar there, unlike S. America.

[skierinvermont]

No, but I'm mentioning the South America "heat wave" because GISS has a completely different spatial pattern/anomaly.

Both sources show Europe/Asia way above average, it's just a question of magnitude...and it might not be that much of a difference because that color on GISS is 2-4C leaving lots of room for interpretation.

RSS does show SE Asia below average, just not as much as GISS. But the spatial pattern is similar there, unlike S. America.

Everything is relative. The error in Europe is larger. If we just changed the baseline they would both show heat waves or coldspells wherever we wanted them too. Heat wave is a completely arbitrary term. If I wanted to I could change the baseline so that they both showed a heatwave in S. America and it was only a question of a small difference in magnitude (much bigger than the "difference in magnitude" in Europe which you are brushing aside). Or i could change the baseline and make GISS have a massive cold spell in Europe while RSS showed it as average so as to make it appear that GISS had a huge cold snap where RSS had average temperatures. And it would appear a much greater cold snap than the S. America bubble is a heat wave.

What is important is the relative error between the two. The error in Europe is much larger. GISS is too cold.

And no, RSS shows far SE Asia at nearly exactly average while GISS has it 1-2C below average. It's completely analogous to S. America.

[nzucker]

Everything is relative. The error in Europe is larger. If we just changed the baseline they would both show heat waves or coldspells wherever we wanted them too. Heat wave is a completely arbitrary term.

What is important is the relative error between the two. The error in Europe is much larger. GISS is too cold.

And no, RSS shows far SE Asia at nearly exactly average while GISS has it 1-2C below average. It's completely analogous to S. America.

It's not analogous because we are looking for places where they differ in the anomaly pattern, where one source clearly is picking up something totally different.

GISS has an area of above average temperatures in S America that doesn't show up at all on the satellites. The differences in Europe and SE Asia are just magnitude, they both show the same pattern of who was above/below average. If you look closely and tilt your monitor correctly (important!), RSS shows SE Asia a good bit below average...it's clearly colored in blue. Load the bigger map for yourself and you'll see it well.

[skierinvermont]

It's not analogous because we are looking for places where they differ in the anomaly pattern, where one source clearly is picking up something totally different.

GISS has an area of above average temperatures in S America that doesn't show up at all on the satellites. The differences in Europe and SE Asia are just magnitude, they both show the same pattern of who was above/below average. If you look closely and tilt your monitor correctly (important!), RSS shows SE Asia a good bit below average...it's clearly colored in blue. Load the bigger map for yourself and you'll see it well.

Ok if you want to set this standard, I'm just going to change the baseline to make GISS show Europe as having average temperatures, while RSS shows a massive non-existent heatwave.

Or I'll change the baseline so that both GISS and RSS show S. America as above average and it will only appear as a small difference in magnitude, and then we can completely ignore it.

And you're still wrong about SE Asia RSS clearly has southern SE Asia nearly exactly normal. Yes northern SE Asia is below normal, but Vietnam down to Singapore is normal on RSS. And it's way below normal on GISS. In fact if you look carefully there's even a hint of red for one pixel in southern Vietnam.

[BethesdaWX]

Or more likely, the satellite data is simply wrong, as argued by Fu, V&G and Zou. As well as the radiosonde data running warmer.

Before I get into how ridiculous this response is.... You need to realize that......yep.........you don't have the knowledge/background to disagree with Roy Spencer and his Refutation of the Zou paper in 2010! You're not a Climate Scientist, you're in your 20's living alone in Conneticut".

See now how ridiculous this method of argument is? I've posted the 2010 Roy Spencer Paper like 15 times already and you simply cannot get out of denial.

Anyway.....

You realize that Zou/Fu/V&G remove the error corrections made by Spencer/Christy before applying the analysis....right? The point of their study had nothing to do with "finding" satellite error.

And either way, Roy Spencer clarified defficiencies and problems with the argument in 2010, I've linked that everywhere here practically. Total Known Errors are +/- 0.05C/Decade, as in, its just as likely to be biased warm. Again this is explained.

And regardless of that:

1) The Zou paper references the trend since 1979, not AQUA beginning in 2002 (where the deviation began)

2) Surface Measurements through extrapolation have a much higher error potential AFTER exrapolation. You continuously post the same paper that was refuted by Roy Spencer regarding UAH.

This isn't really a debatable point.

Response to Andrew:

I suggest you read the 2010 paper by Roy Spencer discussing the difference between "verifyable" and "unverifyable" methods of calibration, before posting.

Then Read up on what Zou bases his error potential on, then come back.

-Not to mention he removed Spencer's/Christy's corrections first.

-Doesn't correct for satellite drift

-uses 1979 starting point, not 2002.

[skierinvermont]

Before I get into how ridiculous this response is.... You need to realize that......yep.........you don't have the knowledge/background to disagree with Roy Spencer and his Refutation of the Zou paper in 2010! You're not a Climate Scientist, you're in your 20's living alone in Conneticut".

See now how ridiculous this method of argument is? I've posted the 2010 Roy Spencer Paper like 15 times already and you simply cannot get out of denial.

Anyway.....

You realize that Zou/Fu/V&G remove the error corrections made by Spencer/Christy before applying the analysis....right? The point of their study had nothing to do with "finding" satellite error.

And either way, Roy Spencer clarified defficiencies and problems with the argument in 2010, I've linked that everywhere here practically. Total Known Errors are +/- 0.05C/Decade, as in, its just as likely to be biased warm. Again this is explained.

And regardless of that:

1) The Zou paper references the trend since 1979, not AQUA beginning in 2002 (where the deviation began)

2) Surface Measurements through extrapolation have a much higher error potential AFTER exrapolation. You continuously post the same paper that was refuted by Roy Spencer regarding UAH.

This isn't really a debatable point.

Spencer has not "refuted" Zou at all. I don't know what you read, but that's not what it said.

[nzucker]

Ok if you want to set this standard, I'm just going to change the baseline to make GISS show Europe as having average temperatures, while RSS shows a massive non-existent heatwave.

Or I'll change the baseline so that both GISS and RSS show S. America as above average and it will only appear as a small difference in magnitude, and then we can completely ignore it.

And you're still wrong about SE Asia RSS clearly has southern SE Asia nearly exactly normal. Yes northern SE Asia is below normal, but Vietnam down to Singapore is normal on RSS. And it's way below normal on GISS. In fact if you look carefully there's even a hint of red for one pixel in southern Vietnam.

How does changing the baseline affect what temperatures Europe has? You have to use the same baseline for both, which is what I did. I don't see the point here...they'd still be way above average..

GISS just seems to be further south than RSS with the cold area in SE Asia, not a big deal.

[skierinvermont]

How does changing the baseline affect what temperatures Europe has? You have to use the same baseline for both, which is what I did. I don't see the point here...they'd still be way above average..

GISS just seems to be further south than RSS with the cold area in SE Asia, not a big deal.

If one changes the baseline, then I can make Europe look average on GISS, and hot on RSS, then it will appear RSS has a massive heatwave where GISS has none.

The fact that a heatwave appears on GISS in S. America while RSS is average is entirely dependent on the baseline used. If we change the baseline, then they will both show a heatwave in S. America, and it will simply be a matter of degrees. Your whole distinction here is entirely arbitrary and dependent on the baseline used.

What is important is the relative difference between the two. The cold bias in Europe on GISS is much larger than the warm bias in S. America. End of story.

Oh GISS just magically moved the cold area south I see? Umm no.. it is falsely extrapolating from a station that got cold weather over all of Vietnam and southern SE Asia. It's exactly how a station in S. America which was warmer than the surrounding area was erroneously extrapolated. The two balance each other out. GISS erroneously extrapolated stations which were colder than the surrounding area in Europe as well.

[nzucker]

If one changes the baseline, then I can make Europe look average on GISS, and hot on RSS, then it will appear RSS has a massive heatwave where GISS has none.

The fact that a heatwave appears on GISS in S. America while RSS is average is entirely dependent on the baseline used. If we change the baseline, then they will both show a heatwave in S. America, and it will simply be a matter of degrees. Your whole distinction here is entirely arbitrary and dependent on the baseline used.

What is important is the relative difference between the two. The cold bias in Europe on GISS is much larger than the warm bias in S. America. End of story.

Oh GISS just magically moved the cold area south I see? Umm no.. it is falsely extrapolating from a station that got cold weather over all of Vietnam and southern SE Asia.

I don't understand...the two sources have to be compared using the same baseline.

If a modern baseline shows above average in Europe, isn't an older baseline that you switch to going to show more above average?

[skierinvermont]

I don't understand...the two sources have to be compared using the same baseline.

If a modern baseline shows above average in Europe, isn't an older baseline that you switch to going to show more above average?

Yes let's just use an older baseline FOR BOTH, like 1980-1985, and then all of S. America will be above average and we can dismiss the difference as simply as a matter of degrees.

Or we could use 1998 as the baseline, and then Europe will appear average on GISS, but RSS will show a mythical heat wave.

[nzucker]

Yes let's just use an older baseline FOR BOTH, like 1980-1985, and then all of S. America will be above average and we can dismiss the difference as simply as a matter of degrees.

Or we could use 1998 as the baseline, and then Europe will appear average on GISS, but RSS will show a mythical heat wave.

Won't change the weird spatial pattern on GISS that shows a bubble of extreme warmth in S America.

[skierinvermont]

Won't change the weird spatial pattern on GISS that shows a bubble of extreme warmth in S America.

Yes it will, they will both show warmth, GISS's will just be greater.

And likewise, I could change the basline so that GISS looks average in Europe and RSS has a magic bubble of hot.

Your distinction is entirely imaginary.

[nzucker]

Yes it will, they will both show warmth, GISS's will just be greater.

And likewise, I could change the basline so that GISS looks average in Europe and RSS has a magic bubble of hot.

Your distinction is entirely imaginary.

Not exactly...GISS and RSS both show Europe WAY above average...you're nitpicking over a degree. Whereas the satellites don't show any of that warm bubble in S. America. There is a spatial distinction that won't be changed by altering the base period. RSS and GISS mostly agree over the location of the warm anomalies in Eurasia, there's definitely more disagreement in South America where GISS shows a portion of the continent being much hotter than the rest whereas RSS doesn't. That wouldn't fade if you used an older base.

[skierinvermont]

Oh and BethesdaBoy...

in addition to my last post, need I remind you again that not all error is quantifiable?

+/-.05C in error is fairly large. It means that in reality the troposphere could be warming as much as .2C/decade and still fall within UAH's error bounds.

And as I said, this doesn't include all error. Some error is not quantifiable.

Even Spencer and Christy say the error may be significantly larger than .05C. You have just ignored this:

I QUOTE the ALMIGHTY SPENCER AND CHRISTY:

Error ranges of these estimates, if we do not apply information that indicates some data sets contain noticeable trend problems, are at least ±0.05°C decade^-1, which needs reduction to characterize forcing and response in the climate system accurately.

http://www.informawo...tent=a934129296

They also estimate the LT trend as .15C/decade 1979-2009 which is higher than UAH. So they're 1) saying that given all the available evidence, our best estimate is +.15C/decade 1979-2009, which is higher than UAH, and 2) even considering all the data, the error is still AT LEAST +/-.05C/decade.

[nzucker]

Oh and BethesdaBoy...

in addition to my last post, need I remind you again that not all error is quantifiable?

+/-.05C in error is fairly large. It means that in reality the troposphere could be warming as much as .2C/decade and still fall within UAH's error bounds.

And as I said, this doesn't include all error. Some error is not quantifiable.

Even Spencer and Christy say the error may be significantly larger than .05C. You have just ignored this:

I QUOTE the ALMIGHTY SPENCER AND CHRISTY:

Error ranges of these estimates, if we do not apply information that indicates some data sets contain noticeable trend problems, are at least ±0.05°C decade^-1, which needs reduction to characterize forcing and response in the climate system accurately.

http://www.informawo...tent=a934129296

GISS has also said their error for comparing global temperatures of recent years is .05C...

http://data.giss.nasa.gov/gistemp/2010november/

[skierinvermont]

Not exactly...GISS and RSS both show Europe WAY above average...you're nitpicking over a degree. Whereas the satellites don't show any of that warm bubble in S. America. There is a spatial distinction that won't be changed by altering the base period. RSS and GISS mostly agree over the location of the warm anomalies in Eurasia, there's definitely more disagreement in South America where GISS shows a portion of the continent being much hotter than the rest whereas RSS doesn't. That wouldn't fade if you used an older base.

It's not simply a degree.. the difference in Europe is much larger and occupies a larger area than that in S. America.

GISS is too cold over large areas just like it is too warm. Extrapolation doesn't create a bias. The reason for the differences between GISS and HadCRUT or the satellite data is not the methodology of extrapolations. It is the physical measurements themselves and/or other aspects of the methodology.

[skierinvermont]

GISS has also said their error for comparing global temperatures of recent years is .05C...

http://data.giss.nas...p/2010november/

Try to keep up. That was Spencer and Christy's best estimate of LT temperature trend 1979-2009, not single years. Their best estimate was .15C/decade with an error of AT LEAST +/-.05C/decade.

The GISS error over the same period is +/-.04C/decade.

Also, it is important to note that Spencer and Christy's estimate of .15C/decade for the LT 1979-2009 is warmer than UAH which is more like .13C/decade over that period. So not only are they saying our best estimate is warmer than UAH, they're also saying that the error is quite large at AT LEAST +/-.05C/decade.

[nzucker]

It's not simply a degree.. the difference in Europe is much larger and occupies a larger area than that in S. America.

GISS is too cold over large areas just like it is too warm. Extrapolation doesn't create a bias. The reason for the differences between GISS and HadCRUT or the satellite data is not the methodology of extrapolations. It is the physical measurements themselves and/or other aspects of the methodology.

It creates a bias if the stations happen to be warmer than the area they represent for one month or year. Like Greenland in March 2011.

GISS has shown more variation in the last 10 years than in the last 30...for the long term period, we have GISS at around .18C/decade, Hadley around .16C/decade, RSS also around .16C per decade, UAH at .14C/decade. The coolest source is only about 20% cooler than the warmest source.

In the short-term, we have GISS at .12C/decade, Hadley around .08C/decade, UAH also near .08C/decade, and RSS at .06C/decade. The coolest source here shows only 50% the trend of the warmest source. So something has clearly changed, and the methodologies haven't changed that much except for more accuracy with the satellites on AMSU. Also, the fact that RSS has been cooler since 1998 shows that UAH isn't just this random cold source, since it's actually shown more warming in this period to offset some of its early cold readings.

[skierinvermont]

It creates a bias if the stations happen to be warmer than the area they represent for one month or year. Like Greenland in March 2011.

GISS has shown more variation in the last 10 years than in the last 30...for the long term period, we have GISS at around .18C/decade, Hadley around .16C/decade, RSS also around .16C per decade, UAH at .14C/decade. The coolest source is only about 20% cooler than the warmest source.

In the short-term, we have GISS at .12C/decade, Hadley around .08C/decade, UAH also near .08C/decade, and RSS at .06C/decade. The coolest source here shows only 50% the trend of the warmest source. So something has clearly changed, and the methodologies haven't changed that much except for more accuracy with the satellites on AMSU. Also, the fact that RSS has been cooler since 1998 shows that UAH isn't just this random cold source, since it's actually shown more warming in this period to offset some of its early cold readings.

And there is no reason to believe that the stations have warmed faster than the areas they represent in the long run. Thus extrapolation does not inherently create a bias in the long-run. End of story.

People have been using this kind of spatial weighting for literally hundreds of years. We can keep debating mathematics which has been understood for over a century, or we can look at the real reasons GISS runs warmer than the other sources, which has nothing to do with extrapolation. As I've shown over and over again, removing the extrapolation does not remove the divergence.

1) GISS 1250km and GISS 250km have nearly the same trend 2004-2010, which is much warmer than HadCRUT's. Removing the extrapolation did not remove the warm divergence.

2. Both GISS and HadCRUT 60-60 have the same 30 year trend, which is warmer than that of UAH or RSS.

[nzucker]

Try to keep up. That was Spencer and Christy's best estimate of LT temperature trend 1979-2009, not single years. Their best estimate was .15C/decade with an error of AT LEAST +/-.05C/decade.

The GISS error over the same period is +/-.04C/decade.

Also, it is important to note that Spencer and Christy's estimate of .15C/decade for the LT 1979-2009 is warmer than UAH which is more like .13C/decade over that period. So not only are they saying our best estimate is warmer than UAH, they're also saying that the error is quite large at AT LEAST +/-.05C/decade.

"Try to keep up" sounds a little derogatory. You are so conceited it isn't even funny. Can't wait till someone knocks you down in the real world, going to be awesome.

.01C/decade in error margin isn't significant. Especially if GISS's error is going up as the divergence suggests. We can debate endlessly the merits of each source, but the fact remains that warming on GISS has outpaced the other mainstream sources since 1998. That means it may be subject to a higher error in recent times.

Spencer and Christy are just using an average of RSS and UAH which suggests about .15C/decade since 1979, nothing strange there.

[nzucker]

And there is no reason to believe that the stations have warmed faster than the areas they represent in the long run. Thus extrapolation does not inherently create a bias in the long-run. End of story.

People have been using this kind of spatial weighting for literally hundreds of years.

Do you not get it?

We are not talking about "the long run." This is a 10 year divergence, not "literally hundreds of years."

In order to determine if GISS extrapolations are causing divergence in the short term, you would have to evaluate each area that differs from the satellites, how much it was extrapolated, etc. That would be a very time-consuming project. You could get an estimate by doing what you did, but I don't know if you've yet performed this analysis for just 1998-2011 or 2001-2011.

[skierinvermont]

"Try to keep up" sounds a little derogatory. You are so conceited it isn't even funny. Can't wait till someone knocks you down in the real world, going to be awesome.

More name calling? Seriously? After the warning about being supended last night? You just can't resist insulting me can you? Should I respond to this in kind and blow the thread out of the water like last night, or is a mod going to take care of this?

It wasn't derogatory. I was simply pointing out that your reply was off-topic and unrelated to the point at hand which was error estimates for trends, not specific years. I am sorry you find being corrected derogatory. Your post was off-topic, that is all.

[nzucker]

More name calling? Seriously? After the warning last night? Should I respond to this in kind or is a mod going to take care of this?

"Try to keep up" isn't exactly above board. I'm not any worse than you...stop trying to act like the good guy. You use the same dirt all the time.

And what names did I just call you? Huh?

[tacoman25]

The difference between GISS and RSS over the region in question in S America is 1-3C. GISS is 2-4C too cool in much of Europe. You are getting confused by the color coding of GISS that paints small 1-2C +anomalies bright orange and 2-4C anomalies bright red. The error in Europe is nearly twice as large and cover over 2-3 times the area of the one in S America.

RSS has a large area of massive +4-6C anomalies in Europe which GISS is completely missing and gives only 1-2 and 2-4C anomalies.

Oh and let's not forget about the non-existent cold-spell GISS created over SE Asia Vietnam to Thailand to Singapore. GISS has it -1 to -2C even a speck of -2 to -4C.. RSS has it -.5 to +0C.

Quit being obtuse. The anomaly patterns are very similar in Europe and the vast majority of Asia. Both show Europe as very warm, both show SE Asia as cool. The one place that obviously sticks out like a sore thumb is GISS's heatwave in South America - it's in a rather large area where RSS doesn't even have warm anomalies.

Saying that GISS is 2-4C cooler than RSS in much of Europe is a blatant lie. And I love how you are focusing on the anomaly difference there but ignoring the anomaly difference just west of Greenland...

[tacoman25]

It's not simply a degree.. the difference in Europe is much larger and occupies a larger area than that in S. America.

GISS is too cold over large areas just like it is too warm. Extrapolation doesn't create a bias. The reason for the differences between GISS and HadCRUT or the satellite data is not the methodology of extrapolations. It is the physical measurements themselves and/or other aspects of the methodology.

FALSE. I don't know what map you're looking at, but that statement is completely inaccurate.

GISS has an area of +1 to +4C anomalies in South America that is about the size of western Europe and easily larger than the part of Europe RSS has 4C+ anomalies. In this smaller area where RSS has +4C to +6C anomalies in Europe, GISS has +2 to +4C. A 2C difference. In the area of South America where GISS has +2 to +4C anomalies, RSS is near average to slightly below...0 to -.5C anomalies. A 2 to 4C difference.

Regardless, we can argue over specific anomalies, but the overall point remains: the vast majority of the maps match well. Where RSS has it hot, so does GISS. Where GISS has it cold, so does RSS. The patterns are very similar. But in South America, GISS has a large area of hot where RSS has nothing of the kind. And this consistently happens. Which is why even when you put them on the same baseline, GISS consistently ends up warmer. In the case of April 2011, about .1C warmer.

[LakeEffectKing]

Wow, this debate/argument has gone on and on and.....

And this thread is boring as all hell, so I know nobody will object that I

take it upon myself to go OT to liven up this God awful thread for a few posts......

Debate this, who was hotter in their prime:

Erin Gray or Connie Sellecca???

Erin

Connie

I'm going with Connie.....her hair covers both poles!!!!

[skierinvermont]

FALSE. I don't know what map you're looking at, but that statement is completely inaccurate.

GISS has an area of +1 to +4C anomalies in South America that is about the size of western Europe and easily larger than the part of Europe RSS has 4C+ anomalies. In this smaller area where RSS has +4C to +6C anomalies in Europe, GISS has +2 to +4C. A 2C difference. In the area of South America where GISS has +2 to +4C anomalies, RSS is near average to slightly below...0 to -.5C anomalies. A 2 to 4C difference.

Regardless, we can argue over specific anomalies, but the overall point remains: the vast majority of the maps match well. Where RSS has it hot, so does GISS. Where GISS has it cold, so does RSS. The patterns are very similar. But in South America, GISS has a large area of hot where RSS has nothing of the kind. And this consistently happens. Which is why even when you put them on the same baseline, GISS consistently ends up warmer. In the case of April 2011, about .1C warmer.

I think you are vastly underestimating the area that RSS has +4C anomalies.. +4 on RSS is anything orange.. the bright yellows are +5-6C. The oranges and yellows occupy all of France, Germany, the UK, Denmark, and most of Spain Sweden and Norway. GISS has these same areas at +1-2 and +2-4. RSS has them at >+4 and plenty of +6. This is a discrepancy of 2-4C over a large area.

Again and again and again I have shown through simple tests that the extrapolations do not cause a warm bias.

1) Mathematically it is not possible over a long enough period. Any bias over 5 years caused by extrapolation will be small, over 10 years almost nonexistent. The math should be enough for anyone who understands it.

2) Even when we remove most of the extrapolation, and just look at HadCRUT 60-60 vs UAH over the last 30 years, the divergence remains. The surface is simply warming more, that is all there is to it.

3) Even when we remove the extrapolation from GISS, by using GISS 250km instead of 1250km, the divergence with HadCRUT remains from 2004-2010.

#2 and #3 empirically prove the extrapolations do NOT cause the divergence. Removing the extrapolations doesn't remove the divergence. It is the PHYSICAL data used, not the methodology. The methodology is sound.

You specifically claimed the extrapolations caused GISS to diverge from HadCRUT since the early to mid 2000s. I have performed a very straightforward test of this hypothesis which unequivocally proves it false. Removing the extrapolations from GISS since 2004 does not remove the divergence.