Response to the 33rd anniversary press release from UAH.

Another UAH Misrepresentation

Looks good.

The word "observed" occurs twice in this sentence:

Moreover, scientists have observed a number of observed 'fingerprints' of human-caused warming

Also, I depsise Spencer (yes, I said "despise," and almost as much as I despise Watts and Inhofe), but the last line may be a little heavy handed.

Unfortunately, this sort of misrepresentation has become the norm for Spencer and Christy.

It sort of ends the post more like an "attack piece" rather than a clear and concise debunking.

Hi Dana,

Looks good, but I have quite a few suggestions/comments, but I hope it they can wait until tomorrow?

Sure Alby, this probably won't be published until Monday.

Just for context, have you seen Andrew Freedman's debunking of the S & C press release, on his Wash Post Weather Gang blog, which has been reposted elsewhere?

Yep Tom, I actually linked Freedman's piece in my post and used his Dessler quote.

Dana, just after you quote Dessler, it might be prudent to include a sentence or two about the MSU satellites themselves. How they actually measure the brightness of oxygen molecules in the atmosphere, and this, along with a radiation model, is used to infer actual temperature at a given height in the atmosphere. It puts Dessler's comment into context for the casual reader who won't know this stuff - a lot can go wrong in such an analysis.

In section "Predicting Lower Atmosphere Warming" point out that only the upper tropical tropospheric trend is expected to be larger and that it's not just models but basic atmospheric physics that tells us it should be. Here's a relevant paragraph from a old post at RealClimate:

"Indeed, there is a clear physical reason why this is the case – the increase in water vapour as surface air temperature rises causes a change in the moist-adiabatic lapse rate (the decrease of temperature with height) such that the surface to mid-tropospheric gradient decreases with increasing temperature (i.e. it warms faster aloft). This is something seen in many observations and over many timescales, and is not something unique to climate models."

In this way you can also avoid talking about models, which apparently drives skeptics mad.

Just reading through, Dana, and came to this paragraph:

We recently examined a paper by Foster and Rahmstorf (2011) which removed all three of these influences from the temperature record using multiple regression.  They found that, as Christy notes, volcanoes have had a slight warming effect on temperatures over the past 33 years.  However, solar activity (measured by either total solar irradiance [TSI] or sunspot number) and ENSO have both had cooling effects.  Their net effect on UAH has been approximately zero, and they have had a net cooling effect on the other major temperature records (Table 2).

I worry that the sentence I have made bold could be miscontrued by habitual practitioners of misconstruation!

Cheers - John

The presentation of uncertainty in the section "Predicting Lower Atmosphere Warming" strikes the right tone in my view, but immediately raises the question 'What do the radiosondes say'?

Rob - do you think you could come up with some text on the MSU satellites that I could insert?  That's not an area I know very much about.

Riccardo - thanks, that's very useful.  The amplification is only in the tropics?  I knew it was largest in the tropics, but didn't realize that's the only area with an expected larger tropospheric trend.

John M - how could that sentence be miscontrued?  Maybe my denialist hat isn't working today, but I'm not seeing it.

Kevin - hmm, radiosondes have their own set of issues.  I don't want to overcomplicate the discussion too much either.  I'll have to think about that.

Dana,

Below are the satellite and radiosonde products (for the mid troposphere) up until December 2010 inclusive [Source].  Note that the trend for the University of Washington (UW) analysis is much higher than the trends for RSS and especially UAH.Note how close the UW trends compare with the radiosonde (RATPAC) data-- those trends being about three times greater than Spencer's and Chrsity's UAH product.  Also, the trend for the mid-tropospheric STAR data is currently +0.128 C/decade. So who is the outlier here?  The preponderance of evidence suggests that the UAH product is and that it has a cool bias.  Why is  their press release do Christy and Spencer ignore all the other satellite based products?  Why do they assume that their product is THE gold standard?  That is grossly misleading, as it creates the very distinct impression that theirs is the only satellite product.

Spencer and Chrsity seem to think the surface data have issues-- it is clear that the issues with the surface data are much closer to being resolved than those for the AMSU data.   The fact that there is so much disparity between some of the satellite prodcts again suggests that those data just as many if not more issues than do the surface data. It is clear that there are very liekly still unresolved issues with the satellite data.

This is not all to say that one should ignore the satellite data, it is another piec eof the puzzle, but to repeatedly overstate the confidence in the UAH product and suggest that it is superior to other products is ludicrous and disingenuous.

Sorry to sound like a broken record, but I have to show this again:

"solar activity (measured by either total solar irradiance [TSI] or sunspot number) and ENSO have both had cooling effects. "

Yep, that's gonna come back on you.  Watts will scream 'SkS proclaims sun is cooling earth!'  Do you mean averaged over the long term or effects on the trend?  It's not clear as is.

Dana,

Re this statement by Christy:

“The climate models produce some aspects of the weather reasonably well, but they have yet to demonstrate an ability to confidently predict climate change in upper air temperatures."

And Re this absurd statement by Roy:

"How much of that underlying trend is due to greenhouse gases? While many scientists believe it is almost entirely due to humans, that view cannot be proved scientifically.”

Well, first Roy should understand that in science we can not prove anything definitively.  Second, as far back as AR3 the body of science has suggested that the majority of wamring is attributable to humans. See this 6-year old review paper by Barnett et al. (2005). They state:

"Recently, an independently processed version of the Microwave Sounding Unit (MSU), channel 2 recorded
temperature of the midtroposphere has become available. This new product shows global warming at a rate
that is consistent with model simulations of the warming in the midtroposphere and the differential warming between surface and midtroposphere. While it is still unclear which of the MSU channel 2 temperature datasets is more reliable, this work suggests that the differential warming between surface and lower troposphere is within present observational uncertainty. Further work is needed to assess and reduce uncertainties in estimates of observed tropospheric warming. Furthermore, observed variations in tropopause height have been shown to be consistent with model-simulated changes in tropopause height resulting from anthropogenic and natural forcing."

And

"The evidence indicates that natural drivers such as solar variability and volcanic activity are at most partially responsible for the large-scale temperature changes observed over the past century, and that a large fraction of the warming over the last 50 yr can be attributed to greenhouse gas increases. Thus, the recent research supports and strengthens the IPCC Third Assessment Report conclusion that “most of the global warming over the past 50 years is likely due to the increase in greenhouse gases."

Third, Roy and Chrsity needs to get up to speed with the latest literature (i.e., Huber and Knutti (2011) and Foster and Rahmstorf (2011).

And here I have not even yet delved into the mass of evidence presented in Chapter 9 of AR4, Hegerl et al., 2001;
Gillett et al., 2002c; Tett et al., 2002; Zwiers and Zhang, 2003; IDAG, 2005; Stone and Allen, 2005b; Stone et al., 2007a,b; Stott et al., 2006b,c; Zhang et al., 2006).  From AR4;

"They find consistent estimates for the greenhouse gas attributable warming over the century, expressed as the
difference between temperatures in the last and first decades of the century, of 0.6°C to 1.3°C (5 to 95%) offset by cooling from other anthropogenic factors associated mainly with cooling from aerosols of 0.1°C to 0.7°C and a small net contribution from natural factors over the century of –0.1°C to 0.1°C (Figure 9.9b). Scaling factors for the model response to three forcings are shown in Figure 9.9a. A similar analysis for the MIROC3.2 model (see Table 8.1 for a description) fi nds a somewhat larger warming contribution from greenhouse gases of 1.2°C to 1.5°C offset by a cooling of 0.6°C to 0.8°C from other anthropogenic factors and a very small net natural contribution (Figure 9.9b). In all cases, the fi fth percentile of the warming attributable to greenhouse gases is greater than the observed warming over the last 50 years of the 20th century (Figure 9.9c)."

And

"The detection of an anthropogenic signal is also robust to using different methods. For example, Bayesian detection
analyses (Appendix 9.A.2) robustly detect anthropogenic infl uence on near-surface temperature changes (Smith et
al., 2003; Schnur and Hasselmann, 2005; Min and Hense, 2006a,b). In these studies, Bayes Factors (ratios of posterior to prior odds) are used to assess evidence supporting competing hypotheses (Kass and Raftery, 1995; see Appendix 9.A.2). A Bayesian analysis of seven climate models (Schnur and Hasselman, 2005) and Bayesian analyses of MMD 20C3M simulations (Min and Hense, 2006a,b) fi nd decisive evidence for the infl uence of anthropogenic forcings. Lee et al. (2005), using an approach suggested by Berliner et al. (2000), evaluate the evidence for the presence of the combined greenhouse gas and sulphate aerosol (GS) signal, estimated from CGCM1 and
CGCM2 (Table 8.1; McAvaney et al., 2001), in observations for several fi ve-decade windows, beginning with 1900 to 1949 and ending with 1950 to 1999. Very strong evidence was found in support of detection of the forced response during both halves of the 20th century regardless of the choice of prior distribution."

See also section 9.4.2.3 in AR4:

"Observed trends in indices of North American continentalscale
temperature change, (including the regional mean, the
mean land-ocean temperature contrast and the annual cycle)
were found by Karoly et al. (2003) to be generally consistent with
simulated trends under historical forcing from greenhouse gases
and sulphate aerosols during the second half of the 20th century.
In contrast, they fi nd only a small likelihood of agreement with
trends driven by natural forcing only during this period."

PS:  It is interesting to note the strawman raised by Chrsity "While Earth’s climate has warmed in the last 33 years, the climb has been irregular. There was little or no warming for the first 19 years of satellite data.  Clear net warming did not occur until the El Niño Pacific Ocean “warming event of the century” in late 1997.  Since that upward jump, there has been little or no additional warming."

Duh, climate scientists have long known that, this is nothing new. From AR4:

"Global mean temperature has not increased smoothly since 1900 as would be expected if it were infl uenced only by forcing from increasing greenhouse gas concentrations (i.e., if natural variability and other forcings did not have a role; see Section 9.2.1; Chapter 2)."

Click here to access Spencer's response to Freedman's article. The comment thread to Spencer's post is also worth perusing.

Hi Dana,

Before I forget John M. and Muoncounter make a good point.  Also, Riccardo makes a good point at 21 Dec 2011, 8:49 PM21 Dec 2011.

Also, the paleo data (even conservative estimates) consistently point towards sensitivity >2 C.  A recents papers (e.g., Previdi et al. 2011 (submitted) and Rohling et al. (2011, J. Climate)) even points to > +3 C. So regardless of the models, past climate is warming us that we are heading for dramatic changes when we double CO2.

The above also refutes the claim mad ein Spencer and Christy's rebuttal:

"Suggesting that the actual climate is at odds with model projections does not sit well with those who desire that climate model output be granted high credibility."

I doubt whether there ever was an official UAH news release.

The link that Dana provides is to a posting on Pielke Sr's blog that is a reposting of an article, Thirty-three Year Temperature Update - Well Below Computer Model Predictions by Ronald Bailey, science correspondent for the Reason Foundation. It appears that Spencer and Chrtisty collaborated with Bailey to produce a faux news release.  

I've searched the UAH's Earth System Science Center's website for a news release to no avail.

Thanks John for the link,

God, they are just digging themselves deeper.

Statements that need checking/confirmation/refuting in their rebuttal.

Christy:

"With 33 years of satellite data now in the hopper (essentially a third of a century) we decided to comment on the long-term character, noting that the overall temperature trend of the bulk troposphere is less than that of the IPCC AR4 climate model projections for the same period. This has been noted in several publications, and to us is not a new or unusual statement."

"Suggesting that the actual climate is at odds with model projections does not sit well with those who desire that climate model output be granted high credibility."

He is ignoring Santer et al.  2011, Thorne et al. 2010.  These studies supercede older studies. Also, the STAR data and radiosonde data are in pretty good agreement with the models.   But the point is moot.  We do not necessarily need models to be concerned.

"I often see such posts refer to an old CCSP document (2006) which, as I’ve reported in congressional testimony, was not very accurate to begin with, but which has been superseded and contradicted by several more recent publications."

Note he does not deny what those reports said, just tries to dismiss the conclusions.  Also, he seems to be saying that a report on which he was a lead author was "was not very accurate to begin with", interesting way of dismissing inconveneint findings. Further, he does not specify which papers he is referring to-- that is quite the leap of faith.


"These publications specifically document the fact that bulk atmospheric temperatures in the climate system are warming at only 1/2 to 1/4 the rate of the IPCC AR4 model trends. Indeed actual upper air temperatures are warming the same or less than the observed surface temperatures (most obvious in the tropics) which is in clear and significant contradiction to model projections, which suggest warming should be amplified with altitude."

Gawd, the BS and the shifting of goal posts (see last sentence).  The UW UAH and RSS satellite data, STAR satellite data and RATPAC data show warming at near +0.15 C, that is pretty close to what the models are predicting.  That rate is more like 3/4 of the expected warming from models!  Re the upper-atmosphere temps warming at the same or less than surface, the warming of the troposphere at a faster rate than the surface is based on physics, not models.  Regardless, the RATPAC data are showing the troposphere warming faster than the surface. But hang on, their initial released pertained to the 33 years of data, now they goal posts have shifted to AR4 model projections since 2000.

"The blog post even indicates one of its quoted scientists, Ben Santer, agrees that the upper air is warming less than the surface – a result with which no model agrees. So, the model vs. observational issue was not presented accurately in the post. This has been addressed in the peer reviewed literature by us and others (Christy et al. 2007, 2010, 2011, McKitrick et al. 2010, Klotzbach et al. 2009, 2010.)"

I doubt that reflects Dr. Santer's position, but I could be wrong. Again, the expectation that the mid-trop will warm faster than the surface is base don physics, not models.  I also think that all those papers he lists have been refuted and/or calle dinto question by subsequent works such as Santer et al. 2001 and Thorne et al 2010.

"The Freedman blog post is completely wrong when it states that “when the problems are fixed, the trend always goes up.” Indeed, there have been a number of corrections that adjusted for spurious warming, leading to a reduction in the warming trend."

They may have a point here, but the Abraham Figure shows otherwise and does not support that statement.

"The notion in the blog post that surface temperature datasets are somehow robust and pristine is remarkable. I encourage readers to check out papers such as my examination of the Central California and East African temperature records. "

Really this is just false and juvenile, and a strawman to boot. None is claiming that the surface records are pristine, they are robust though, more so than the satellite data. This claim by him is just plain disingenuous. Church et al. (2011) and Murphy et al. (2010) also show this to be a demonstratively false statement.  He is just obfuscating.

"We also document how surface development disrupts the formation of the nocturnal boundary layer in many ways, leading to warming nighttime temperatures."

A red herring and strawman. Rural sites warm just as fast as urban sites Menne et al. showed this, so did Fall et al. (IIRC).  The affect, if any, is insignificant, but I do not have a reference off the top of my head to support that.

Spencer reply to the Washington Post article by Freedman.

Regarding the claim that"Ben Santer, agrees that the upper air is warming less than the surface".  As I suspected, Christy is misrepresenitng Samter's postion.

From Santer et al. (2011):

"We conclude from this result that there is no inconsistency between observed near-global TLT trends (in the 10- to 32-year range examined here) and model estimates of the response to anthropogenic forcing."

Re the claim tby Christy that the AR4 model runs are overerestimating the warming, from Santer et al. (2011)

"Claims that minimal warming over a single decade undermine findings of a slowly-evolving externally-forced warming signal [e.g., as in Investor’s Business Daily, 2008; Happer, 2010] are simply incorrect."

Also from Santer et al.:

"On all timescales examined here, the TLT trends in the observational satellite
datasets are not statistically unusual relative to model-based distributions of externally
forced TLT trends (Figure 6E). While this consistency is encouraging, it should
be qualified by noting that: 1) The multi-model average TLT trend is always larger
than the average observed TLT trend (Figures 6A,F);"

Regarding the last point, something that Christy chooses to ignore:

"Possible explanations for these results include the neglect of negative
forcings in many of the CMIP-3 simulations of forced climate change (see Supporting
Material, Table S1), omission of recent temporal changes in solar and volcanic forcing
[Wigley, 2010; Kaufmann et al., 2011; Vernier et al., 2011; Solomon et al., 2011],
forcing discontinuities at the ‘splice points’ between CMIP-3 simulations of 20th and
 21st century climate change [Arblaster et al., 2011], model response errors, residual
observational errors [Mears et al., 2011b], and an unusual manifestation of natural
internal variability in the observations (see Figure 7A)."

Interesting that Pielke felt the need to close with (or was it the text of the press release?):

Neither Christy nor Spencer receives any research support or funding from oil, coal or industrial companies or organizations, or from any private or special interest groups. All of their climate research funding comes from federal and state grants or contracts.

The Spencer/Christy blog post doesn't say much new.  My post already addresses most of the comments there, actually.  I'll add a couple things from there though.

And I'll add the radiosonde data and some more info on the other satellite data sets.

John H - I couldn't find the press release either, which is why I linked to Pielke.  Nobody linked to the original source, and I couldn't find it in a Google search.  Very strange.  It wasn't even on Spencer's blog.  I wonder if Christy sent it to Pielke.  I may revise to call it a press release from UAH scientists, as opposed to a UAH press release.

I'm still not seeing the problem with the quote bolded by John M.  It clearly says 'over the past 33 years'.  But I'll try to rephrase.

Pielke Sr's blog post is nothing more than a reposting of Thirty-three Year Temperature Update - Well Below Computer Model Predictions by Ronald Bailey, science correspondent for the Reason Foundation.

I suspect that Spencer and Chrtisty collaborated with Bailey to produce this faux news release.  

In a post on the comment thread to his Washington Post article, I have asked Andy Freedman to provide a link to the supposed "UAH news release."

OK, I'm almost done ;)

Dana, I think you missed this blatant distortion and misrepresentation by Christy, and IMO it should be addressed:

"There was little or no warming for the first 19 years of satellite data.  Clear net warming did not occur until the El Niño Pacific Ocean “warming event of the century” in late 1997.  Since that upward jump, there has been little or no additional warming."

This is just pure BS and may have already been addressed by SkS or Tamino or better still a journal paper.

And my thoughts on the volcano issue.  "When that cooling is subtracted, the long-term warming effect is reduced to 0.09 C (0.16° F) per decade, well below computer model estimates of how much global warming should have occurred."

I think I may have found a significant flaw in their logic. The models were in fact "aware" of the volcanic eruptions-- that is, the stratospheric aerosol data were used to drive the models.  But Chrsity did not compare the satellite data (with  the cooling from the volcanic events removed) against the predicted model temperatures with the volcanic cooling removed-- they compared against the expected temperature which included the volcanoes.  Had they done so the projected increase in model predicted temperatures would have been decreased by a similar amount as found in the satellite data. So they are not comparing apples with apples.

Interstingly, studies that used the Agung, Pintubo and El Chichon erputions to constrain climate sensitivity. Wigley et al (2007) found that:

"Comparisons of observed and modeled coolings after the eruptions of Agung, El Chicho´n,
and Pinatubo give implied climate sensitivities that are consistent with the
Intergovernmental Panel on Climate Change (IPCC) range of 1.5–4.5 C. The cooling
associated with Pinatubo appears to require a sensitivity above the IPCC lower bound of
1.5 C, and none of the observed eruption responses rules out a sensitivity above 4.5 C."

So the very volcanic eruptions they rely on to try and claim the models are wrong and overestimating climate sensitivity actually supports a median sensitivity of near 3 C! :)

Regarding the claims made by Christy in the rebuttal:

As for the corrections, he does claim that the CCSP report states that:

"Both positive and negative adjustments were listed in the CCSP report mentioned above."

However, I could not find any mention of the positive or negative adjustments in the CCSP report.  They talk about corrections, but not specifically about negative adjustments.


I would recommend that you read the report and see if there is anything in there that you can use to refute them, that is, "Christy refutes Christy" or "Christy refutes Roy".

1) Regarding the discrepancies and the culprit for them, from CCSP:

"This inconsistency between model results and observations could arise either because “real
world” amplification effects on short and long time scales are controlled by different physical
mechanisms, and models fail to capture such behavior; or because non-climatic influences
remaining in some or all of the observed tropospheric datasets lead to biased long-term
trends; or a combination of these factors. The new evidence in this Report - modelto-
model consistency of amplif ication results, the large uncertainties in observed tropospheric temperature trends, and independentphysical evidence supporting substantial tropospheric warming (such as the
increasing height of the tropopause) - favors the second explanation. However, the large observational uncertainties that currently exist make it difficult to determine whether or not models still have significant errors."

So they indicate that errors in the tropospheric obs are the most likely source of the differences, not the models.

Also,

"• Errors in observed temperature trend differences
between the surface and the troposphere
are more likely to come from errors in
tropospheric data than from errors in surface
data."

2) Re comparisons between models and satellite data:

"(3) When models are run with natural and human-
induced forcings, simulated global-average
temperature trends for individual atmospheric
layers are consistent with observations."

But there do remain discrepancies above the tropics.

3) This is a good one, especially after Roy's comment " the supposed “fingerprint” evidence of human-caused warming continues to be one of the great pseudo-scientific frauds of the global warming debate."  Well from the very report that Christy was a lead author on:

"• Fingerprint studies have identified greenhouse
gas and sulfate aerosol signals in
observed surface temperature records, a
stratospheric ozone depletion signal in
stratospheric temperatures, and the combined
effects of these forcing agents in the
vertical structure of atmospheric temperature
changes."

So Christy is then, by Ropy's own admission part of this "great pseudo-scientific fraud".


4) Regarding the claim about the poor surface temperature record:

"Systematic local biases in surface temperature
trends may exist due to changes in
station exposure and instrumentation over
land , or changes in measurement techniques
by ships and buoys in the ocean. It is likely
that these biases are largely random and
Some have expressed concern that land temperature
data might be biased due to urbanization effects. Recent
studies specifically designed to identify systematic
problems using a range of approaches have found
no detectable urban influence in large-area averages
in the data sets that have been adjusted to remove
non-climatic influences (i.e., “homogenized”).
therefore cancel out over large regions such
as the globe or tropics, the regions that are
of primary interest to this Report."

So any biases cancel out and urban influences are "no detectable". Subsequent research shows that the UHI effect is detectable but negligble.

Not being a climate science wonk like you and most of the others particpating in this discussion, I have some "dumb" questions. 

Is the UAH data base incorporated into the GISS dtabase? [I ask because NASA satellites are doing the measurements and GISS is part of NASA,]

Who else uses the UAH database and for what purposes?  

Were the NASA satellites that collect the data going into the UAH database designed to do more than measure the temperature of the lower/mid troposphere? Is temperature measurement their primary function?

Are the NASA satellites measring the temperature of the "lower" troposphere, or of the "mid" troposphere?  [I ask this question because of the information presented on slide #37 of this NOAA presentation.]

In the context of the last question, it would be helpful to SkS average readers to see a graphic of the Earth's troposhpehere showing the orbit of the NASA satellites relative to the Earth's surface and the top of the troposphere.  

John - they're NASA satellites measuring lower troposphere temps, whereas GISS measures surface temps (from surface stations).  So no, GISS doesn't use UAH data.

I don't know who else uses UAH data.  Not too many groups, except denialists arguing that surface temps are biased high.

I think the satellite instruments were designed specifically for atmospheric temp measurements.

They measure various layers of the atmosphere, but now the lower troposphere.  UAH and RSS have to process the data to try and extract lower troposphere temps.  Glenn has some posts discussing this.

Hi Dana,

Just a few more observations then I'll shut up.

1) Eric_skeptic has just made some valid comments about the Abraham Figure.  I agree that the curent caption is not accurate/correct, or rather not well wroded/explained.  That should be addressed.  See my reply to Eric.

2) I do not like the wording of this sentence:

"They found that, as Christy notes, volcanoes have had a slight warming effect on temperatures over the past 33 years."

(it reads like volcanoes cause warming), I think it can be corrected by adding a sentence that mentions that it is the timing of the cooling (early in the record) that increases the trend.  But also see my comments above aout them not comparing apples to apples.

3) Table 1 and Fig.3 .  If I recall corectly, Mears was not too happy with what we did in a recent post, specifically on how we arrived at the TLT estimates for Zou et al and Fu et al.  I think that we can still make a strong point without adding those estimates-- I'm worried that they will just give people pipoints to argue and distract from the key points of the post.  I would suggest adding the RATPAC estimates for the mid-trop (850-300 mb), which is comparable with TMT, well more or less.

4) I would add a comment about the paleo data-- rub their nose in it with those two new paleo papers that I cited above and that the volcanoe data they are using to try and discredit the models actually points to a climate median estimate of climate sensivity of ~3 C.

It may be worthwhile highlighting the contradiction betweenSpencer and Christy's press release stateing no significant warming since the 97/98 El Nino and Christy's rebutall of the Abraham graph in which he attributes  most of the rise in trends over time to rises in the data, not the adjustments.  I expound on this here.

"Further, if there had been "... little or no additional warming" since 1997 as Spencer and Christy maintained in their press release, then the trend would have declined gradually after 1997 as the years lengthened but the temperature did not increase. In other words, if their statement in the press release had been true, the trend for 2011 would have been less than that reported in 1998, with any difference being the result of adjustments. Being generous, the decline in trend should follow the result reported in 2000, meaning that if their press statement was correct, and their criticism of Abraham's graph were also true, the current UAH trend should be 0.05 C/decade or less.

This constitutes a major contradiction between their two statements, and is not reconcilable as an honest mistake. If they thought a week ago that the data indicated no increase in the trend after 1997, then one weeks data cannot have justified a switch to the position that most of the increase in trend since then has been the result of warming. If, on the other hand, the claim that the increase in trend since 1997 reflects the data rather than adjustments is their honest opinion, they cannot also have believed just one week ago that there has been "... little or no additional warming since then"."

Hello Tom,

Saw your comment at Spencer's place.  You and some others aside, God what a load of drivel he allows people to post there, and his comments are not much better.  Anyone who tries to make a serious and scientific argument or challanges hiom on something gets ridiculed, mocked and dismissed without any evidence or science.  But maybe that is how Spencer reacts when he is nervous and feeling guilty....;)

Dana et al. might be worth persuing the comments to see what Spencer has up his sleeve.

Yeah I saw your comment there Tom, good point.  Also pretty nice table from over at Eli's.

Yes,

Tom has just nailed their arses to the wall.  Well done Tom! That takes care of Christy's claim-- three of the seven adjustments were downwards but they were dwarfed by upward adjustments.  Maybe worth inlcuding in the blog post Dana.

And yes, as Eli notes, where is the code please?

PS:  I think Obscurity has been banned/censored again.

 Dana - "Rob - do you think you could come up with some text on the MSU satellites that I could insert? That's not an area I know very much about."

Well I only know the basics, but that's good enough for me. The following is a bit wordy, so abbreviate/grammar- correct it as much as you like.  

"Andrew Dessler is referring to the complex nature of constructing a global temperature series from the satellite observations. The microwave sounding units (MSU) aboard the satellites don't actually measure air temperature, but rather the intensity of microwave radiation given off by oxygen molecules in the atmosphere, and the intensity of this radiation is a proxy for air temperature. Given that the radiation reaches the satellite sensors having travelled through a warming lower atmosphere and cooling stratosphere, that bias exists between the various sensors, issues with orbital decay, and a host of other obstacles, there's a lot of careful and painstaking analysis required, and much that can go wrong. This is obvious in the continual corrections Christy and Spencer have had to make to the UAH data set over the years"  

Dana - sorry not to get back to you. I think the points in my bold sentence - volcanoes - warming; ENSO, TSI - cooling - need to be explained by elaboration. The reason for this line of thought is that it is well known that big volcanic eruptions have led to brief cooling episodes e.g. post Pinatubo; likewise we have the strongest ENSO influence being the strong 1998 warming. Deniers take such things and use them to infer self-contradiction. Never give the buggers an inch - they always take yards!

Hope that makes sense - I have a stinking cold so a head increasingly full of cottom-wool!!

Cheers - John

Climate deiers are now defending Christy-Spencer and chastizinng Dessler, Santer, Abraham, and Freedman on the comment thread to Andrew Freedman's WaPo article. I encourge you to set the record straight over there.

PS -- Freedman is ignoring my request for a link to the "UAH prews release".

I found the following on the RSS webpage, "Description of MSU and AMSU Data Products." 

Introduction

Satellite measurements of the Earth’s microwave emissions are a crucial element in the development of an accurate system for long-term monitoring of atmospheric temperature. Satellites provide global coverage at much higher densities than attainable with in situ observations. In situ observations also suffer from non-uniform temporal coverage and undocumented changes in the radiosonde instrumentation used that can lead to local biases and increased uncertainty. The Microwave Sounding Units (MSU) operating on NOAA polar-orbiting platforms have been the principal sources of satellite temperature profiles for the past two decades. The MSUs are cross-track scanners with measurements of microwave radiance in four channels ranging from 50.3 to 57.95 GHz on the lower shoulder of the Oxygen absorption band. These four channels measure the atmospheric temperature in four thick layers spanning the surface through the stratosphere.

Atmospheric temperature measurements extend for almost three decades, beginning in November 1978 and continuing through the present. A series of follow-on instruments, the Advanced Microwave Sounding Units (AMSUs), began operation in 1998 with the intent of extending microwave sounding measurements into the foreseeable future. The AMSU instruments are similar to the MSUs, but they make measurements using a larger number of channels, thus sampling the atmosphere in a larger number of layers. By using the AMSU channels that most closely match the channels in the MSU instruments, we can continue to extend our climate-quality dataset.

The MSU and AMSU instruments were intended for day to day operational use in weather forecasting and thus are not calibrated to the precision needed for climate studies. A climate quality dataset can be extracted from their measurements only by careful intercalibration of the distinct MSU and AMSU instruments.

my update on the temperature adjustments is up.

The key points:  

Readme adjustments disagree with those in peer reviewed articles, so cannot be relied on.

Contrary to the impression Christy tries to make, adjustment represent about 50% of the total current trend (ie, without adjustments S&C would understate the trend by 50%);

The 97/98 El Nino is almost inconsequential to the overall trend;

Even in the period in which the 97/98 El Nino most effected the trend, it represented only 70% of the change in trend, and was smaller than one adjustment and the same size as another made in the same period;

Spencer and Christies own adjustments (that they found themselves) are net negative, such that without the adjustments found by other teams they would be underreporting the trend by 64%.

As a side note, Christy's claims are carefully worded so that their is a way (just) in which they can be interpreted as true.  There is no way that they are not massively misleading giving the charge he is defending himself against.  Never-the-less, I think the implied contradiction is more damaging to him than his carefully strained words.

Edited to add: just did a quick check, and the 2010 El Nino also adjusted the trend by 0.008 C/decade (same as the 97/98 El Nino).  These are of course, ballpark figures and a more scientific approach will probably result in a different result, but not by much.

FYI, Christy, Spencer and Braswell detail the effect of adjustments on trends up to version D

Christy et , 2003 details the changes from version D to version 5.0.

I have been unable as yet to track down peer reviewed sources for changes after that, but note that the table from Eli is accurate to that point.

Eric (skeptic) says:

 Hi John,

I agree. I noted above that we need to carefully word the caption for the Abraham figure.  That is, not just cut and paste the original.

If we don;t, it will just give people an excuse to distract from the OP's message and impact.

I don't understand the second paragraph of Eric_Skeptic's WUWT post. Who is he criticizing? Is he playing to the denier crowd?

Too much heat?

A common theme being posted by deniers on both WUWT and WaPo is that the UAH and RSS plots are "nearly identical." Has this meme been addressed in an existing SkS article?  

John - Eric is criticizing Abraham's figure caption and whoever is defending it (i.e. Tom Curtis has here, I don't know who has on WUWT because I'd prefer not to have to read WUWT comments ;-)

Alby - I'll probably use Eli's table rather than Abraham's graphic.  It contains more information.

John again - UAH and RSS are pretty darn close, though a little less so when exogeneous factors are removed, as F&R showed.  Also see here.

John,

"Too much heat?"

Are you sure comments are closed, have you tried posting something?  If not, please try. Thanks.

I just thought I'd been censored again, but there have been only 52 comments for a while now.... 

That would be sneaky, shutting down comments at his blog and then letting the dumb, faithful and uncritical mob at WFUWT high-fiving everything he says.

Are Illis's comparisons between "observed" and "forecast" decadel trends correct?

None of the links that I have tried this afteroon work. What URL are you using?

John, Bill Illis is making the same fallacy that I noted above.  He is not comparing apples with apples.  This is a very importnat point and a very dubious lseight of hand by Christy.

Bill illis is doing something fishy, b/c his findings are at complete odds with those of Foster and Rahmstorf (2011) even though they are both doing the same thing.

Hi John,

"What URL are you uisng?"

I'm not sure I understand.  for Roy's blog?  I can access it but cannot post.  Maybe you are having network problems.  Does someone else want to try and post at Roy's plce?

"Are Illis's comparisons between "observed" and "forecast" decadel trends correct?"

Doubtful that they are.  When Foster and Rahmstorf (2011) remove the impacts of ENSO, volcanoes and solar they obtain a trend of 0.141 C/decade for UAH and 0.157 C/decade for RSS.

But why not ask Tamino to be sure?

Carrick just posted at Spencer's place...so comments are indeed open.  Maybe most of the deniers are at WUWT having fun.

Do the claims made by Ellis constitute a new denier meme that deserves a rebutall?

I cannot access Spencer's blog. What URL are you using to do so?

The UAH did in fact issue a news release. Click here to access it on Newswise. 

John,

I'm using,

http://www.drroyspencer.com/

You might want to point readers at Freedman's blog to that news release John-- good find.  They seem to think there was none issued.

Dana should also link to that in the OP instead of to Pielke's site.

John,

"Do the claims made by Ellis constitute a new denier meme that deserves a rebutall?"

I do not think so.  at least not for now.  Just point people to FR2011.

The URL does not work for me. Perhaps I have been "blocked" from accessing the site?

John H - Illis is right about volcanoes but wrong about ENSO.  This is covered in my post (and my post on Foster and Rahmstorf).

I'm still not sure that release is from UAH.  The link just goes straight to the UAH main page, not to a press release.

Try this link to Spencer's post, it works for me.

Okay guys, I made a bunch of revisions to the post, but there have been a whole lot of comments here, so I may have missed something.  Have another look and let me know.

Your link to Spencer's post doesn't work for me.  I do not recall posting anything on his blog so why/how I might be blocked from accessing it is a mystery. I've been flitting around the internet all day without any problems accessing other websites. Go figure.  

The first sentence of the of the fith paragraph in the "Modeled vs. Observed Lower Atmosphere Warming" section does not make sense to me. It reads:

Santer et al. found that the UAH trend was on average 1.55 larger than the model-expected trend (the discrepancy was even smaller when using Remote Sensing Systems [RSS] data). 

It does not square with Santer quote immediately preceeding it.

Suggest that you provide a source for Table 2.

The article would benefit from a bullet point summary of major points a la recent posts by Rob Painting. 

I have some concerns about the articles lead paragarph. It currently reads:

The end of November 2011 marked 33 years of satellite-based atmospheric temperature measurements, originated by Roy Spencer and John Christy at the University of Alabama at Huntsville (UAH).  Unfortunately, the UAH scientists marked this anniversary with a press release which propagated much of the same misinformation we keep hearing from Spencer and Christy.  As we will see, in their press release, the UAH group made a number of statements which disregard the body of scientific literature outside of their own resesarch.

The first sentence does not convey the fact that UAH massages satellite data to compute TLT. In additon, it should not be lead sentence. The 33-year annivesary is not critical. The guts of the article is a rebuttal of what Spenser and Chrsity said in their news release and in subseqeunt blog posts.

Dana,

Looks Good.  

1) Just a comment about the RATPAC data.  That 0.15 C is the anomaly for the year to date in 2011.  This is what the 2010 end-of-year summary states:

"Data collected and averaged between the 850–300 mb levels (approximately 5,000 to 30,000 feet above the surface) indicate that 1958–2010 global temperature trends in the middle troposphere are similar to trends in surface temperature; 0.13°C/decade (0.23°F/decade) for surface and 0.16°C/decade (0.29°F/decade) for mid-troposphere. Since 1976, mid-troposphere temperatures have increased at a rate of 0.18°C/decade (0.32°F/decade)."

Sorry if one of my hastily composed posts led you astray there Dana. The bolded number (0.18 C/decade for mid-troposheric wamring, 850-300 mb) is closest window to the satellite TMT data record.  I'm still not comfortable inclusing the SkS estimates for the TLT data for some groups, especially after Mears expressed some reservations.  The contrarians will probably use that as a point of dconterntion to try and distract fro the OP's message.

2) "They found that, as Christy notes, decreased volcanic activity has had a slight warming effect on temperatures over the past 33 years."

I still do not like this sentence, volcanoes cause short-term cooling.  It was the timing of that cooling that contributed to the increase in the trend.

3) "When that cooling is subtracted, the long-term warming effect is reduced to 0.09 C (0.16° F) per decade, well below computer model estimates of how much global warming should have occurred."

Sorry to harp on this, maybe I'm wrong, but they are not comparing apples to apples.  The models were forced with those eruptions, so they would have to subtract the cooling effect from the model predicted trends as well in order to compare apples with aples.  This is partculalry disingenuous on Christy's part. Now that their product is showing noteworthy warming they are bending over backwards to dismiss that warming or call it into question.  I call bullshit.

4) "While it's true that many of these 'fingerprints' are simply consistent with what we expect to see from anthropogenic warming, it's not a coincidence that they are indeed all consistent with human-caused warming."

Is the wording right here?  Did you mean to say "....some of these 'fingerprints' are simply consistent with what we expect to see from natural warming"? Actually the only one that could also be natural is "rising tropopause". The point is that one cannot explain the warming and the fingerprints without invoking GHGs.  Maybe also link to the post where that graphic appears as that post inlcude dall the citations.

5) "The two adjustments in red (made by scientists outside of UAH) alone account for 0.135"

This really needs some clarification.  Are these corrections suggested by RSS (for example) that were adopted/implemented by UAH?  Or are you referring to the University of Washington analysis of the UAH data?  The wording needs to be very clear who identified the problems and who implemented them. We cannot give them any wiggle room on this-- clarity is more important than brevity for matters such as this.

6) "There is, however, a problem when the authors of that record continually insist that there are no problems to the record, as Spencer and Christy have a history of doing, as documented by Albatross."

Thanks for the credit Dana:)  I might say that there is  "a problem when the authors of that record continually insist that there are no problems to the record, and overstating the accuracy and robustness of their data as Spencer and Christy have a history of doing".

7) Is there any chance of you addressing this nuggest from Christy's response?

"There was little or no warming for the first 19 years of satellite data.  Clear net warming did not occur until the El Niño Pacific Ocean “warming event of the century” in late 1997.  Since that upward jump, there has been little or no additional warming."

A link to this post by you would suffice to debunk this cherry picking and misinformation.  This is really poor form by Christy and he needs ot be called on it.  Again, they are bending over backwards to downplay the warming, it is quite unbelieveable.

8) "Additionally, as we've previously discussed, UAH is not the only group analyzing the satellite temperature data."

Maybe add, "...data, although one would not think so going by Spencer and Christy's  comments".  They ignore analyses of UW and STAR, the casual reader of their blogs and press releases would thnik that theirs is the only satellite product.

9) Mention that Fig. 5 is from Huber and Knutti in the caption. I thought their finding was that 74% of the warming was attributable to increases in GHGs?  Be careful not to overstate or oversell this.

10) Table 1, I'd say:

"Table 1: Lower (TLT) and mid-troposphere temperature (TMT) estimates from various groups"

Note upper case for Lower.

11)  In the summary I might also add that it is obvious that Spencer and Christy having been forced to accept that there is warming, are now bending over backwards to try and dismiss that wamring and underplay its importance.  Or something along those lines. That they have had to make corrections is perfectly fine, but it is their reluctance to do so when first pointed out to them, their continual ovserstating the robustness and accuracy of their product while exaggerrating the uncertainties of other products, their tricks to try and minimize the robust warming evident their own data, their reluctance to consider that the descrepancies between the model data and the satellite data could be because of outstanding cool biases issues with their product all show they are biased and have confirmation bias.  That is why they are getting hammered, b/c they are not being forthright and honest or critical of their own work, instead they are using their data to forward an agenda-- so that is why people are pisse doff at them, not b/c their findings are inconvenient.

That is all I can think of for now.  I have some other thoughts but I'll save those for the comments ;)  Thanks for being so accommodating Dana!

Merry Christmas and happy holidays all!

Is it the TLT should go up 1.2 times faster than the surface by the models? Exaclty why is that? The rising altitude of the 0-celsius isotherm? Something to do with water/ice transition in the clouds?

MSU O2-proxy with some undefined code to check that, great. 

(EDIT: This discrepancy makes me wonder if the normal ozone equilibrium in the stratosphere could be slightly shifted due microwawes from lower atmosphere leading to consistent low bias in sat measurements. This could possibly be checked by comparing the sat measurements from poles and tropics. Is ozone more sensitive to microwawes than thought?)

looks pretty good, though a bit technical, but it has to be so that the people in question (Christy&Spencer) get it ;-). thumb.

n/t

jyyh - it's an adiabatic lapse rate issue.  I added an explanation (from RealClimate, provided by Riccardo) to the post earlier.

thanks Dana, just jumped to the first issue that was unclear. Coming from biochemistry with some ecological background, I may have some original views on some issues that are not quite accurate.

1)  "Santer et al. found that the UAH  model-expected trend was on average 1.55 larger than the model-expected UAH trend (the discrepancy was even smaller when using Remote Sensing Systems [RSS] data)."

2)  I have significant reservations about direct comparisons of radiosonde and TMT temperature trends.  Radiosondes measure the temperature in a specific altituded band within the troposphere.  In contrast, TMT channel measures microwaves from the surface layers up into the lower layers of the stratosphere, which are slightly cooling.  Assuming the standard models are correct, the 850–300 mb levels measured by radiosondes would warm faster than both the levels below and the levels in the stratosphere above, something which would be reflected in instrument measurements in that region.  In contrast AMSU measurements of the TMT band would be expected to show a lower trend than TLT measurements because of the influence of the stratosphere.  Therefore the two products are not capable of simple comparison.

3) The reservation expressed in (2) applies with double force to treating Radiosonde data as though it were a TMT chanel to calculate a TLT trend.  First, even if that is possible for true TMT bands (and we all agree that there are significant reservations on that point), the function used assumes the measured temperatures has been cooled by incorporating a lower stratosphere component, which is a false assumption for radiosonde data.  Second, as noted above, we actually expect the trend at that altitude to by higher than the trend in the lower troposphere, so if we used a model based prediction from the radiosonde data we would predict a lower trend based on that.

Thanks Tom Curtis for clearing the radiosonde issue,  is the graph referenced (850-300mb) level more like an uneven combo of TMT and TLT channels in satellite measurements, minus the slight cooling in the low stratosphere that gets in to the TMT channel? Could there be an adjustment for the radiosonde data to get it more in line with the various estimates, i.e. what is the part of the low stratosphere in the TMT channel? It could be well advised and good to somehow include this difference in the methods of measurement in the table description.

Is "cherrypicking" really one word?

jyyh,  yes to your first question.  I suspect it would be slightly closer to a TLT channel than a TMT chanel, and I notice that Spencer and Christy in various papers treat the radiosonde data and the TLT data as strictly comparable.  I do not know whether that comparison is considered controversial.  

You probably could do an appropriate adjustment, but it would be maths heavy and hence out of my league.  It would also inevitably involve the same sort of judgement calls that make differences between UAH, RSS, and Fu et al controversial.

I think a better approach would be to show a direct comparison of several radiosonde products to RSS and UAH.  I would make a point of inlcuding HadAT2 amongst the suite of radiosonde data because Spencer and Christy have previously compared the UAH TLT product to the HadAT product without major qualification and without a need of adjustment.  As can bee seen below, such a comparison clearly shows UAH to be an outlier in the cool direction (note the trend comparison of the right):

(Source: http://www.metoffice.gov.uk/hadobs/hadat/images.html)

From Albatross I gather that RATPAC shows a trend of 0.18 C/decade over the same period, which is slightly above HadAT.  We can justifiably use the F&R adjusted surface temperature records for the surface trend, or HadCRU3v plus Gistemp.  In that way we avoid any controversy about SkS adjustments.  The higher trends found in other satellite products can be then described with a link to our previous calculations so that the point about UAH being an outlier is reinforced without putting our previous calculations front and center as a target (I agree with Albatross on that).

John Hartz, not according to the google spell checker.

Will you include the "Christy Crocks" and "Spencer Slip-Ups" buttons on this article?

As long as they exist, they should be applied in my opion.  

So Tom, you're suggesting we compare HadAT and RATPAC to TLT data?  Because NOAA compares RATPAC to TMT.

I'm going to get rid of the SkS TLT estimates based on TMT data in Table 1.  Just haven't had time before now.

John - I could include the buttons, but that may generate unwanted controversy and distract from the science.  I'm open to opinions on the question.

Hi Dana,

Yes, HadAt is probably best-- the period of observation is very siilar to the satellites.  I can always reference the RATPAC in the comments, or you could include both the RATPAC and HadAT in the Table.  If you do inlcude the RATPAC, I striong;y suggest not trying to estimate the surface rate of warming base don the mid-trop trends from the sonde data.  The NOAA site I linked to above says that the surface rate of warming from the sonde data is 0.13 C/decade since 1958, but they do not say what it is since 1978/79.  Ugh, but it is less than the warming in the 850-300mb layer.

I think I've incorporated most of the comments now.  The one thing I'm missing is Alby's comment about the pre- and post-1998 ENSO warming trend quote.  I wanted to look at the data (both UAH and FR11) and create a graphic for that, so thanks for the reminder Alby.  I've got an hour before I need to leave and do some stuff, so hopefully I can get that done first.

But in particular have a look at my TLT/radiosonde write-up and make sure it's correct.

I'll probably also break this up into a two-parter, because it's gotten really long.  I think the break will be after talking about radiosondes and other TLT data sets, before going into the pre and post-1998 trends.

Yesterday (Dec 22) Pielke Sr. posted Further Confirmation Of Klotzbach Et Al 2009. I'm not sure if anything in it bears on Dana's article, but I lack the expertise to make that call.

John,

Saw that...I think Pielke is overstating this and its significance.

Dana,

Chasing my tail today (as are you probably)-- i'll do my best to have a look!  But Tom or someone else please feel free to look too.

Dana,

This post by Dr. Joel Shore may be of some help.  It seems to corroborate Tom's findings. 

Despite all the back and forth this seems a little rushed ot me and I am nervous that we may inadverdantly give them some speaking pints to distract from the truth.  Pielke will be watching and he will be sure to defend his pals, so we might end up chasing more blimps-- I hope not.

Maybe a simplified version could be prepared for Treehugger and or the Guardian?

Just a quck note about the opening para:

"As we will see, in their press release, the UAH group made a number of statements which disregard the body of scientific literature outside of their own resesarch."

Maybe "As we will see, in their press release, the UAH group made a number of misleading statements and disregard the body of scientific literature outside of their own resesarch."

If the "Christy Crocks" and "Spencer Slip-Ups" buttons will detract from this article, it's probably time to get rid of them and the other buttons completely -- or, at least remove them from the SkS homepage.

Hello Dana,

Some more notes, sorry!

1) "As we will see, in their press release, the UAH group made a number of statements which disregard the body of scientific literature outside of their own resesarch."

Maybe "As we will see, in their press release, the UAH group made a number of misleading statements and disregard the body of scientific literature outside of their own resesarch."

2) Desller says "And, when these problems are fixed, the trend always goes up."

To be fair that is not entirely true and we should note that-- we are not cheerleaders ;).  With that noted, his general point stands.  And the radiosonde data do in fact show that the upper levels are warming faster than are the surface temps.

3) "Figure 2 clearly shows this statement is false.  The two adjustments in red (made by scientists outside of UAH) alone account for 0.135 of the 0.138°C per decade UAH TLT trend - 98%!  As Tom Curtis notes, the 1998 El Nino only accounts for approximately 0.008°C per decade (5.8%)."

This is the part that concerns me, I would err on the very conservative side.  Joel Shore finds a 50-50 split, but it is very difficult to tease out the exact contributions.  But it is safe to say his claim is at best very misleading and not supported by the data.

4) Its is dificult to say if the 850-300 mb (1.5 km -10 km above sea level) radiosonde data are equivalent tot he TLT or TMT data.  I'm not sure what to do about this.

5) "Note that the RATPAC radiosonde (weather balloon) data are in agreement"

 I would say "close agreement", they are not the same, RATPAC is higher.

6) "While removing the short-term, natural influence of volcanic eruptions is a valid analysis, it also makes us wonder, why only volcanoes?"

Am I wrong about him not comparing apples with apples?  Please do tell me if you disagree.  I have harped on this a few times now, and IMHO it is an importnat point.  If he exlcudes the influenc eof volcanoes from the obs then he must do so for the mdoels so as to make a proper comparison.

7) "which was co-authored by John Christy himself"

I would say he was a "lead author" that is what the reports lists him as.

8) "The aforementioned CCSP document also contradicts Spencer here"

Maybe "The same CCSP report that Christy was a lead author on also...".  I think it important to make is very clear that Spencer is essentially accusing his colleague of being part of a "pseudo-scientific fraud".

1) Interesting that NOAA compares TMT and RATPAC.  As Albatross notes, there is not "correct" comparison because TLT includes to great a weight near the surface, while TMT includes a significant weight in the stratosphere.  Of the two I would guess that the additional weight from the lower troposphere in TLT is less distorting.  

IF  (A) the models are correct, the lower troposphere will cool slower than the upper troposphere, so inclusion of it will bias TLT low compared to the radiosonde data, but not by as much as including stratospheric data.  If (B) the models are incorrect, and the lapse rate has remained constant or increased, including lower tropospheric data will bias TLT high compared to radiosondes, but not by as much as the stratospheric data biases TMT low.  I draw that conclusion from the fact that TMT has a consistently lower trend than TLT and nobody finds in that a discrepancy with the models, which they would if the stratospheric bias was small in effect. 

In case (A), based on the models we would expect TLT trends to be slightly lower than radiosonde trends, but higher than surface trends.  In case (B) both radiosonde and TLT trends should be higher than surface trends.  In either case having radiosonde trends higher than the surface trends (without F&R adjustment) while the TLT trends are lower than surface trends suggests the measurement problems are in the TLT analysis. Therefore comparing radiosondes and TLT trends directly justifies calling the TLT trends, and in particular the UAH trend into question.  There is a greater discordance between UAH and radiosondes than there is between radiosondes and the surface records.

As a side note, given that the factors which have lowered apparent surface trends over the last decade will also have lowered apparent radiosonde midle tropospheric trends, an F&R style analysis would raise the trend of the radiosonde data substantially, with the result that there may not be any disagreement between radiosonde and surface data at all (depending on data set and the final results of such an analysis).  At the moment that remains pure speculation, however, and should not be mentioned in any post.

2)  I had a quick look at Pielke's post on Klotzbach et al (where he and Christy are among the et al).  Basically he is saying that even after comments by Gavin Schmidt are taken into account, Klotzbach et al's results still stand.  That is probably true regarding the adjsuted data for Klotzbach (they had issues with landmasks).  However their conclusion is a bit of a hatchet job and unsupported by their analysis.  In particular they compare surface and satellite trends.  They then discuss reasons to distrust the surface data, but include no comparable discussion of satellite data.  That bias alone would have led me to reject the paper had I been the editor.  They then conclude that the satellite data and radiosonde data give reason to believe their are problems with the surface record.  First, they never discuss the radiosonde data so mentioning it in the conclusion as bolstering their case is very bad form.  Second, because they never discussed issues with the satellite data, they cannot reasonably draw any conclusion about which data set is in error.

So, to John Hatz's question, yes it is relevant, but no, there is no need to specifically discuss either the post or paper.

3)  I am a little unhappy with the wording, "The two adjustments in red (made by scientists outside of UAH) ... "  Other scientists found the discrepancies that required adjustment, but Spencer and Christy made the actual adjustments in their program (even if duplicating suggested adjustments by others).  I don't think the distinction is important enough for the extra wording - but I'm flagging it in case anybody else feels differently.

Tom,

Re #3, I agree (I think I flagged it earlier too)-- we need to be fair and accurate as to who discovered the error and who came up with the correct solution. Did S&C do the adjustment properly though?  I recall a comment by Mears adn Wentz in Science that shows S&C applied the correction incorrectly. See the post by Chris here.

We do not wish to give them easy points to nit pick at to distract from the overall message and their errors.

We are heading out now, and I will not have time look at this anymore this weekend.

With regard to 3), perhaps the wording "The two adjustments in red (made to correct for errors found by scientists outside UAH) ..."

would resolve the issue.  Not sure the point is worth the extra 5 words though.  If deniers make an issue of the wording, they merely make a point of the fact that Spencer and Christy did not make the errors.  They will also make it prominent that the second error was an incorrect sign in the equation (a very basic error) which had different parts of the atmosphere reacting differently to the day night cycle, so was obviously in error from the get go.   For most people the mathematical arguments regarding the proper adjustments to make to produce a TLT product are pretty abstract.  But everybody gets that using the wrong sign is a pretty clutzy thing to do.

Tom - I've always had the same issues with the Klotzbach paper.  They spend a bunch of time discussing problems they see with the surface temp record, but have nary a bad word to say about the satellites.  And as we've discussed here, there are a lot of reasons to question the accuracy of the satellite data.  But the 'skeptics' have a major blind spot when it comes to the satellite record.

You're right, the other groups didn't make the adjustments to UAH, they identified the errors, and the adjustments were then made by UAH.  It's worth rewording to get that right.  I'll make the other suggested changes now, and finish with the pre- and post-1998 trend graphics (I've got Foster and Rahmstorf in there, just need to finish the equivalent unadjusted UAH graphic).

Okay here we go, now a two parter:

• UAH Misrepresentation Anniversary, Part 1 - Overconfidence
• UAH Misrepresentation Anniversary, Part 2 - Of Cherries and Volcanoes

Dana:

Your lead paragraph as currently written:

The end of November 2011 marked 33 years of satellite-based atmospheric temperature measurements, originated by Roy Spencer and John Christy at the University of Alabama at Huntsville (UAH).  Unfortunately, the UAH scientists marked this anniversary with a press release which propagated much of the same misinformation we keep hearing from Spencer and Christy.  As we will see, in their press release, the UAH group made a number of misleading statements and disregard the body of scientific literature outside of their own resesarch.

Suggested revised paragpraph:

November 2011 marked the 33rd consecutive year that Roy Spencer and John Christy at the University of Alabama (UAH) have computed global atmospheric temperatures using data collected by NASA weather satellites. They marked this anniversary with a press release propagating much of the same misinformation about global climate change as they have throughout their careers at UAH. Spencer and Christy not only made a number of misleading statements in the UAH press release and in subsequent blog posts about it, they also ignored a body of scientific literature that contradicts their views on global climate change.

I like John's suggestion, except maybe say "Unfortunately, they marked this anniversary..."

But I think this is now good to go Dana-- great work!  This can be a nice Boxing day pressie for Roy and John amd their pals;)

[Update]:  Hmm, I'm really not comfy with ose buttons.  I know what John H. means when he says why have them if we do not use them.  But I think having them up front and centre like will probably only give people like people a crutch and speaking points. Certainly link to them though!  We can get tough in the comments....

I don't like the GIFs in Part 2.

First, seeing we are criticizing Spencer and Christy's press release and related comments, the UAH GIF should extend to the last UAH data point at the time of their press release.  Ending it earlier at the peak of an El Nino smacks of cherry picking.  I know you ended it earlier to have it end at the same time as F&R's analysis, but IMO that is insufficient reason.

Second, starting the second period after the 97/98 El Nino is dubious.  This is partly because that El Nino is part of the temperature record and should be included in the GIF.  It is also because starting in 1999 starts in a year with a low relative temperature immedieately followed by a La Nina year.  In such a short record, starting with a La Nina and ending with an El Nino is guarenteed to give a strong trend, but it is dubious the strong trend is significant (let alone statistically significant).  It would be better either to include the 97/98 El Nino in both sections, or else to make the cut off point the month with peak temperature in that El Nino (giving a 1 month overlap).

I understant that if you make both of these corrections, the contrast between early and late trends will be diminished, and may even reverse.  However, if it is diminished enough as to not support your point, the point is not worth making.

Given that we so frequently, and justly, criciticize deniers for their use of short term trends, we should be hyper-vigilant ourselves when using short term trends to not do anything unjustifiable.  Even a whif of cherry picking in our use of short term trends would be damaging, IMO, and not worth it to make a simple point when so much else is available for criticism.

Sorry to give you such an unpleasant gift for Christmas, but I wished to avoid not getting the message to you before publication because of being caught up in holiday festivities (which I am now turning to). 

Happy Christmas

Oops,

I missed that Tom.  Good catch.  AAARGH!

dana:

1)  More of a pain, but you could possibly also extend FR11's analysis to current, although that depends on their being current data for AOD which may well not be the case.

2) FR11 do filter out El Nino, but so far as is possible it is best to keep the time periods the same.  To that end, the modification of the first GIF might be simply to add a final fram shoing the trend 199? to current.  That way you keep the direct comparison with FR11 without any danger of percieved cherry picking.

3)  I understand your points, but still disagree.  In particular, "no warming since 1998" is clearly a euphamism for "no warming since 1997 and provided you include the largest El Nino on record in the data set".  I am certain Spencer and Christy are inconsistent enough to criticize us for excluding 97/98 despite their actual words. 

The solution may be again to extend the GIF,  by showing trends with overlap, but that probably makes the GIF to complex.  Depending on what others think, It might be an idea instead to simply have the trends with overlap already calculated (and plotted) ready for the inevitable denier challenge.  On refection, the best overlap would be to take the early period to Dec 98, and start the later period in Jan 97 which would be fairly bullet proof from criticism, but should still show a 97-2011 trend greater than 0.14 per decade (I think) thus rebutting Christy's claim.

Must go, Christmas breakfast calling.

I can't believe that we are working on this on Christmas eve and Christmas (for Tom).

I think I have a potential solution to the 1998 cherry-pick.  Dana why not just show your animated gif that was such a hit-- you know the one features in going down the up escalator post, and just say how cehrry-picking short-periods is meaningless.  I can't remember if you link to the "continued lower atmos. warmring", but that also deals well with this. Showing that was effective at silencing people defending Christy's cherry pick on the Freedman Washington Post thread.  It also avoid the whole starting point issue (what does he mean by "since 1998" is that 1998 inlcusive or after 1998 (i.e., 1999). 

Additionally you could point out the marked increase in 0-700 m and 0-200 m OHC has increased since 1998 (inclusive to be conservative, b/c there was a blip that year). That will also be a nice jab at Pielke et al who claim that OHC is more importnat than temperature.  Actually-- this suggests that these pals are contradicting each other.  Pielke says we should use only OHC (and that has increased a lot since 1998), while the pals that he routinely defends (S&C) say that the satellite data are the best metric.  Maybe this is an opportunity to drive a wedge between them and show how they cherry-pick those data that are convenient, and contradict each other in terms of what is the mnost suitable metric.  OHC since 2003-2004, or satellite data since 1998-- whatever hides the long-term warming.

Dana, please write the probability of the claim being incorrect down in the paper, even if it's not quite 95%. if it's 80%:ish  you might say they have 1/5 possibility in being correct. I know IPCC has specific definitions for using vocabulary, but there's no need in SkS to be so unspecific :-). Sorry for the extra work this may entail.

Dana,

In order to make your article (s) more undestandable to the average person, you may wnat to incorporate the following "background information" contained in the UAH release into the article. If doing so is not feasible, this information should be incoporated into a separate article and referenced,

NOTE: According to the below, Spencer & Christy have been working as a team at UA for only 20 plus years. Is that when work on converting the MSU data into global temperatures began? 

When the first MSU went into orbit in 1978, it wasn’t designed for monitoring long-term changes in the climate. Instead, it was built to give meteorologists two temperature readings a day over about 96 percent of the planet to provide input into computerized weather prediction models, the forerunners of climate models.

“All of the satellite instruments but one were designed to measure day-to-day weather changes, not long-term climate,” said Spencer. “It has been a challenge to make the necessary corrections to the data so we can use the instruments for long-term climate monitoring.”

While the satellite data record is shorter than the surface thermometer record, it has several strengths. It has the greatest global coverage: With 96 percent coverage of the globe (except for small areas around the north and south poles), the satellite sensors cover more than twice as much of Earth’s surface as do thermometers.

It is also less likely than surface-based thermometers to be influenced by local development, Spencer said. Urbanization typically contributes to local warming due to the asphalt effect, when paving and buildings absorb and convert into heat sunlight that would naturally have been reflected back into space.

While that heat can raise temperatures recorded by thermometers at surface weather stations, the effect on the atmosphere is so local and so shallow that it dissipates before it can heat the deep atmosphere above it. As a result, satellite measurements have shown no indication of an urban contamination effect, Spencer said.

Another strength is that the microwave sensors gather temperature data for a deep layer of the atmosphere, rather than just at the surface.

“What we look at is a bulk measurement of the atmosphere’s heat content,” Christy said. “That is the physical quantity you want to measure to best monitor changes in the climate. Plus, it’s consistent. You can take a single satellite ‘thermometer’ and measure the temperature of the whole Earth, rather than just at a single spot.”

While the satellite dataset has its strengths, unlike thermometers and temperature probes used on weather balloons the Microwave Sounding Units were new, largely untested tools when they were put into space. Spencer, Christy and other scientists have had to develop small corrections that they use every month to reduce errors caused by the satellites losing altitude or drifting in their orbits.

While year-to-year temperature variations measured by the satellite sensors closely match those measured by both surface thermometers and weather balloons, it is the long-term warming trend on which the satellites and the surface thermometers disagree, Spencer said, with the surface warming faster than the deep layer of the atmosphere.

If both instruments are accurate, that means something unexpected is happening in the atmosphere.

“The satellites should have shown more deep-atmosphere warming than the surface, not less” he said. “Whatever warming or cooling there is should be magnified with height. We believe this is telling us something significant about exactly why the climate system has not warmed as much as expected in recent decades.”

The following is an "add-on" to the article, "Scientists mark 1/3 of a century of satellite climate data" by Doyle Rice, UAA Today, Dec 19, 2011.

For additional perspective on this, I requested a comment from meteorologist Jeff Masters of the Weather Underground, who adds this:

"The UAH satellite data shows the Earth has warmed by .254 degrees F per decade since 1979, which is the .82 degrees F that Dr. Christy quotes for the past 33 years. An independent satellite data set maintained by Remote Sensing Systems, Inc. puts this warming at .268 degrees F per decade.

The three major surface temperature data sets (National Climatic Data Center, NASA Goddard Institute for Space Studies, and Hadley Climate Research Unit) all give 0.281 - 0.301 degrees F per decade warming.

Satellite data is affected much more strongly than surface data by volcanic eruptions and the El Nino phenomena, so it is best to correct the data sets to remove these natural variations, which are not part of the global warming signal we are trying to see.

When we remove these variations, the five data sets range between 0.254 - 0.308 F per decade warming, with the warmest years in the past 33 years being 2010 and 2009. This is well within the range of warming predicted by climate models used in the 2007 Intergovernmental Panel on Climate Change report.

These models predicted that global warming should average near 0.360 F per decade between now and 2030, so the observed trend is 15 - 30% below the prediction. This is close enough that we should have confidence that models' predictions that a dangerous warming in the 3 - 7 degree F range by 2100 will occur -- particularly when you consider that we set a new record for emissions of heat-trapping gases last year, and emissions show no sign of slowing down.

A change in Earth's temperature of a few degrees may not seem like much, but remember that the difference in temperature between the last Ice Age and the current climate is 9 degrees F. A few degrees can make a huge difference."

John H - I agree, a simple version of the UAH history would be very informative, but that article is one big deception. Where's the mention that it measures the brightness of atmospheric oxygen for instance? It's the same old bullshit from Christy and Spencer - trash-talk the surface temperature record and conveniently ignore all the issues with the satellite record.

John, Spencer and Chrisity first released the UAH data set in March 1990 with Spencer RW, Christy JR. Precise monitoring of globaltemperature trends from satellites. Science 1990

In March 1990, Spencer and Christy worked at NASA’s Marshall Space Flight Center. Spencer resigned from NASA in 2001 and became director of the Earth System Science Center (ESSC) at UAH. Is that when the ESSC was created?  Do we know anything about how it is funded? (Its website is dated and contains very little detail about its mission and history.)

Note: I might be able to garner more information about the ESSC from Spencer's website if I weren't blocked from accsessing it.

I obviously need hlep in separating the wheat form the chaff re UAH and what it actually does or does not do. 

I would recommend the following ammendments to the second post:

1)

"This claim simply has no basis in reality.  Apart from the physical absurdity of claiming that El Niño causes temperature step changes, t The UAH data directly contradicts the claims in this quote."

2) Figure 1 modified to include an additional step showing the trend from Jan 1999 to Nov 2011.  If it does not do so already, the first section of should show the trend from Dec 1978 to Nov 1997 to exactly coincide with the period in the S&C quote.

3) Additional paragraph immediately after figure 1:

"Of course, Spencer and Christy may well have meant, not 'since that upward jump' but, 'including that upward jump'.  Indeed, from January 1998 to November 2011 the trend has a low value of 0.059 C/decade.  It is, however, very easy to find low trends if you cherry pick your starting points.  Spencer and Christy owe us, but do not provide some physical reason why we should expect a 'step change' as a result of the 97/98 El Nino, and not (for example) from the 1999-2000 La Nina.  The obvious explanation as to why no physical reason is given is that such step changes from ENSO events are absurd.  Appealing to them, however, sounds like an explanation, even though no explanation is actually given.  It acts, in other words, as a nice form of words to defuse any tendency to critical thought. "

4)

"We see much the same thing  If there was a step change in temperature after the 97/98 El Nino, it would show up clearly as elevated temperatures after that event when we first filter out short-term influences on TLT, as Foster and Rahmstorf (2011) did (more on this paper below).  Clearly such an elevated jump in temperatures is absent from the record.  What is more,   Tthe 1979 to 1999 trend in this case is 0.10°C per decade.  From 1999 to 2010 it's 0.21°C per decade, which again easily exceeds the trend for the entire record of 0.14°C per decade (Figure 2).  This clearly shows that Spencer and Christy's "upward jump" is an artifact of cherry picked starting points for a period short enough to be dominated by short term fluctuations in temperature that show no long term trend.

Obviously bold shows additions, and strike through shows deletions.  The reason for the suggested ammendments is that unless we directly confront Spencer and Christy's implicit apeal to step changes, we must either show trends that appear to support their claim (1998-2011) or leave ourselves open to a charge of cherry picking start points.  With the suggested ammendments the charge of cherry picking cannot stick, for the GIF shows the period specified by Spencer and Christy, and we directly acknowledge the low trend for the period they are really interested in.  But they do not get to float the argument for a step change past us, for it is explicitly adressed, with a further link to more detailed discussion.

If they choose to respond, they will then face the invidious choice of having to choose to defend step changes as physically based, despite strong contradictory evidence and no theoretical basis.  Or they will have to effectively admit to cherry picking.

Okay guys, Part1 is published and Part 2 is revised to incorporate all the comments so far:

UAH Misrepresentation Anniversary, Part 2 - Of Cherries and Volcanoes

Tom and Alby, you've put a lot of effort into this.  I'd be happy to add you as co-authors on Part 2 if you'd like.  Just let me know, and also let me know if there are any additional comments.

Hi Dana,

I only have a minute to  spare here.  Thanks for the generous offer to include my name, but I personally do not feel I did much wrt the second part, maybe more so for the first part?  I second including Tom as co author for the second part, he made some good catches and good calls.

I can't look at teh latest draft right now, family waiting for me, but i'll trust that you and Tom and the rest of the SkS team will get it right :)

Dana,

I tried to flag this for you yesterday. The lead sentence may not be accurate.

According to Tom Curtis, Spencer and Chrisity first released the UAH data set in March 1990. At that time, they both worked at the at NASA’s Marshall Space Flight Center. When they started working there is unkown to me. It is therefore questionable whether they have crunched the satellite data for 33 consecutive years. 

dana, no need to include me as co-author.  My effort has been minor compared to yours (and your productivity on SkS continues to amaze me).

john, where is the claim that they have worked together for 33 years made?  I cannot find it in either the press release, or either of dana's articles so I cannot see the relevance?  What is claimed is that they have crunched 33 years of data, which is true.

Tom,

Dana saw my note and promptly changed the lead sentence -- all before you saw the article and my note. 

Having now read the last revision , definitely a thumbs up from me.

John, Dana is too quick of the mark.