Friday, March 4, 2016

101 conspiracy theories about troposphere temperature: the RSS love affair is over

Sou | 10:48 PM Go to the first of 56 comments. Add a comment
Over at WUWT there have been three articles about the lower troposphere temperatures. The first (archived here) was a Guest Post by Werner Brozek and Nick Stokes, Edited by Just The Facts with the title: "Long Satellite Pauses Ending (Now Includes January Data)".  The second article was about the February data for UAH (archived here, latest here). It had the wistful wishful title: "Global Temperature Report: Warmest Ever February 2016 driven by El Niño". The third article is the real doozy (archived here). It's by Anthony Watts so could well have some awful blunders in it. He's called his article: "The ‘Karlization’ of global temperature continues – this time RSS makes a massive upwards adjustment."

This article is a few hours late, and I'm not satisfied that I've got everything right because this is a subject on which I am feeling distinctly out of depth. Science deniers will try to tell you that there are little thermometers on satellites sending raw data to Earth and miraculously drawing temperature charts - or something like that. That's a pile of hogwash. The real story is much more complicated. Satellites come and go. Instruments change. Orbits decay. Temperature isn't measured directly, it's estimated from measurements from microwave sounding instruments (MSUs). What's reported is the result of complex calculations after adjustments and conversion to temperature. What we get are temperature trends in very thick layers in the atmosphere (kilometers thick, looking upwards into space), not the temperature of a particular spot or distinct level in the sky. Then there is "diurnal drift" - which is largely what the new paper by Carl Mears and Frank J. Wentz is all about.

Warning: this article is rather long. It explains the new RSS paper in more detail than I did in the previous article.

The lower troposphere "Pause" is officially dead

The first article was written ahead of the release of UAH and RSS data, and was in part about how the so-called "pause" will end in either February or March, depending on whether the February anomaly is 0.88 C or higher for RSS or 0.315 or higher for UAH.  It was. For both. So the RSS and UAH "pauses" are officially ended, if you take WUWT to be anything like official.

The February temperature record was driven by more than El Niño

Despite the headline that the UAH February record was "driven by El Niño", the second article had this to say (my emphasis):
By a statistically significant amount, February 2016 was the warmest month in the satellite temperature record, according to Dr. John Christy, director of the Earth System Science Center at The University of Alabama in Huntsville. Interestingly, however, that record might have as much to do with an extraordinarily warm month in the Arctic as it does with warming caused by the El Niño Pacific Ocean warming event.
Which is what was pointed out here at HW - and here.

Anthony Watts jumps up and down and sez the new RSS paper has a fatal flaw

The third article shows just how sensitive Anthony Watts is to any hint that the world really is getting hotter. He jumps on something Roy Spencer said to him, but unlike Roy, Anthony decides it means there's a "fatal flaw" in the new RSS paper. As if Anthony Watts would know!

Anthony began his article once again falsely accusing the NOAA scientists of fraud. Anthony Watts is just another wacky conspiracy nutter of the lizard men kind. He wrote:
Forget homogenization, that is so 2010. If the pause is bothering you and your belief is that there must be more warming, we only need to find it in the data, then what you need is “Karlization”, named after director of the National Climatic Data Center, (now NCEI) Tom Karl who pulled a fast one this summer trying to adjust the past down, so the present would be warmer. The sleight of hand on this was so obvious that even warm-oriented scientists such as Michael Mann and Ben Santer co-authored a rebuttal paper that said Karl was dead wrong and the pause was real. There is now a congressional investigation into Mr. Karl’s apparently political actions disguised as science.

Of course no scientists wrote anything about Karl15 being "dead wrong" or a "pause" being real in the sense that global warming stopped. Anthony is talking through his hat. (This is what he's referring to, and Graham Readfearn has a terrific article at the Guardian about it, too.)

And does Anthony Watts know that of all the main four temperature datasets, NOAA has the lowest warming trend for the period since 1971? The 1970s was the last time the trend changed, according to research published in Cahill15. Does Anthony Watts know that the trends per decade differ in only the second decimal place - by 0.01 or by 0.001 if you want to talk annual trend? Here are the trends from the different datasets from 1971 to 2015:

Figure 1 | Trend per decade for four global surface temperature datasets. Data sources: UK Met Office, NOAA, Berkeley Earth, GISS NASA

Anthony Watts has been trying to get sued for defamation for years now. He might succeed one of these days. (RSS isn't a government agency so Carl Mears could sue if he thought it worth his while.) Anthony's specialties include lying about science (as above), falsely defaming scientists (as above), looking stupid (as in Russian steampipes), acting in a cowardly fashion (slinking away from U Bristol), and being unethical.

Having set the scene, Anthony wrote about Dr Mears:
Clearly, he’s miffed. So what to do? Taking a cue from the other Karl, he publishes a paper and claims that new and improved adjustments have “found” that missing warming.

He has falsely accused the NOAA scientists of fraud, so by taking this stance he is also effectively accusing Carl Meares and Frank J. Wentz of fraud.

How does RSS v4 differ from v3.3?

Before getting into Anthony's so-called "fatal flaw", I'll try to set out how the new version 4 of RSS mid-troposphere temperature differs from v3.3. Early in their paper, Mears and Wentz explain how they have to remove the diurnally-varying component from observations to derive the long term trend in temperature.
The derivation of long-term trends in tropospheric temperature from satellite observations requires that the diurnally varying component for the observation be removed. This is because the local observation time for most of the satellites drifts over time (Christy et al. 2000; Mears and Wentz 2005), causing diurnal variations to be aliased into the long-term record.
They describe how this has been done by them previously (in v3.3) and by others:
  • In previous RSS versions, they constructed a climatology using a general circulation model. NOAA's STAR used the same approach.
  • UAH in version 5.6 used "cross-scan differences" to work out the "local diurnal slope"

They report how Po-Chedley et al showed that none of the models completely remove the effects of the diurnal cycle, and this new paper of Mears and Wentz confirms that finding. Instead of using a general circulation model, they have adopted "an harmonic method" for removing biases relating to satellite diurnal drift. This method is based on analysis of the satellite observations themselves. The authors explored three approaches, all of which gave similar results for AMSU data:
  • MIN-DRIFT - where they exclude parts of each satellite record during times of rapid drift in observation time. They couldn't adopt this approach holus bolus (a technical term) because it wouldn't work for MSU data in the early part of the record.
  • REF-SAT - where they use two satellites that didn't drift (AQUA and METOP-A) as reference satellites to adjust the drifting satellites. Again, that only works with the period in which those satellites were operational, and couldn't be applied to the earlier part of the record.
  • DIUR-OPT - where they adjusted GCM-derived diurnal cycles (ie from general circulation models, I presume) using information got from comparing satellite observations at different local times. This can be applied to the entire record and is the one that Mears and Wentz used for v4.

Oh it's all very technical. I expect some HW readers will understand it. However I'm not even a novice when it comes to satellites and microwave sensors and correcting for diurnal drift and I don't understand a lot of it, so if you can't follow any of this, blame me.

Anthony Watts issues a plea for help from denier scientists

Now Anthony Watts wouldn't know the difference between a satellite and a weather station any more than he can tell an anomaly from a baseline. So just in case his message didn't get through to Christopher Monckton (who has adopted the RSS dataset as the "gold standard" of temperature datasets), he sent a hasty missive to Roy Spencer and John Christy to see if they could find fault with the paper. Well, they'd barely had time to read it. In fact they may have just looked at the pictures without reading the text, because Roy wrote back fairly promptly. Just getting an email from Roy Spencer was sufficient for Anthony to proclaim a "fatal flaw":
Dr. Roy Spencer and Dr. John Christy have already looked into this latest “Karlization” and have found what appears to be a fatal flaw. 
It's fairly obvious that Anthony didn't understand the paper or Roy's email or what the so-called "fatal flaw" might be. He added no comment other than "Yes, yes it does", which was wise, knowing Anthony as we do. He just put up a picture from the new paper. Roy wrote back saying:
The paper is for MT, not LT…but I think we can assume that changes in one will be reflected in the other when Mears completes their analysis.
That's what I figured, too. So that part was okay, because I agree :) (However it suggests that Anthony got the wrong end of the stick and mistakenly thought the paper was about lower troposphere data.) It's the next bit that got Anthony Watts all excited.

Roy Spencer thinks he's found a problem - but has he?

In his email Roy wrote:
From what little we have looked at so far, it appears that they did not correct for spurious warming in NOAA-14 MSU relative to NOAA-15 AMSU…see their Fig. 7c.  They just leave it in.
Since this spurious warming is near the middle of the whole time period, this shifts the second half of the satellite record warmer when NOAA-14 MSU (the last in the MSU series) is handed off to NOAA-15 AMSU (the first in the AMSU series).
Why do we think NOAA-14 MSU is at fault?
1) AMSU is supposed to have a “Cadillac” calibration design (that’s the term a NASA engineer, Jim Shiue, used when describing to me the AMSU design, which he was involved in).
2) NOAA-14 MSU requires a large correction for the calibrated TB increasing with instrument temperature as the satellite drifts into a different orbit.  The NOAA-15 AMSU requires no such correction…and it wasn’t drifting during the period in question anyway. 
So, it looks like they decided to force good data to match bad data.  Sound familiar?

With his comment "They just leave it in", Roy seems to be arguing that Mears and Wentz should have removed NOAA-14 from their dataset.  See below for the big gap that would have caused. However, notice that Roy wrote "relative to NOAA-15". The authors of the RSS paper found that it was NOAA-15 that had degraded, and that's the one for which they stopped using data after December 2010.

In point 2 above, I'm guessing the TB refers to "brightness temperature". Although the letters are back to front, so I could be wrong. Whatever, section F of the paper on MSU/AMSU Difference Trends has a discussion about NOAA-14. The authors talk about differences in trends for periods when the data from different satellites overlap, that is, the older MSU overlaps with the newer AMSU (my emphasis):
One of the unexplained mysteries of the previous version of our analysis (V3.3) was a large positive trend in the MSU minus AMSU differences during the period of overlap from late 1998 until mid-2002, followed by a period of smaller negative trend from mid-2002 until late 2003. The origin of these differences is not known....We included the effects of this difference trend into the uncertainty estimate for the V3.3 dataset (Mears et al. 2011). This mystery remains in the version of the dataset even after the optimization of the diurnal adjustments for MSU and AMSU
In that section, Mears and Wentz discuss three possible reasons and exclude two of them, leaving this one: Errors in the NOAA-14 target factor. They conclude that:
We are left with explanation that the differences are caused by a calibration in either NOAA-14 or NOAA-15 (or both). Our baseline dataset will use both MSU and AMSU measurements during the overlap period. If we exclude MSU data after 1999 (implicitly assuming the error is due to NOAA-14), the long-term trend decreases by 0.019 K/decade, and if we exclude AMSU data before 2003 (implicitly assuming the error is due to NOAA-15), the long term trend increases by 0.01K/decade).  
So the authors did consider calibration problems with both NOAA-14 and NOAA-15, and they tested what would happen if they removed them. They wrote about an unresolved problem with the data, which they narrowed down to one of three things. They were able to discount two of the possibilities, leaving the possibility of "errors in the NOAA-14 target factor".  Now I wouldn't have known what that means, but Mears and Wentz explain:
The target factor is a calibration target dependent adjustment we use to account for instrument non-linearity. Because the target temperature for NOAA-14 has a strong trend during 1999-2004, an error in the target factor could lead to a spurious trend in the NOAA-14 radiances. Such an error would cause large oscillations in the MSU minus AMSU differences (See supplemental Figure S5). 
I'm not really too much wiser about that and can't find the supplementary information, which might add some clarity. It seems to me it's to do with what adjustments are made to work out actual "brightness temperature" readings so as to convert them to temperature. Maybe there's an expert who'll drop by to explain. (Mears and Wentz specifically discard the term "brightness temperature" in their paper and instead use "temperature units". They explained that the mean radiances reported are slightly different to what people usually refer to as "brightness temperature".)

NOAA-14 was the last MSU instrument, starting in 1995 and ceasing in 2004. NOAA-15 was the first satellite with AMSU instrumentation, starting in 1998 and still active in December 2014. Table 1 of the paper shows that in version 4.0 the authors left in the entire period of NOAA-14, but excluded NOAA-15 for the period from January 2010 to its end of life, December 2014.

I cannot say why NOAA-14 was left in for its entirety. I can speculate though. I'd say it's because removing it would have resulted in inconsistencies in other ways, so it probably wasn't causing errors that couldn't be corrected. Plus it would have meant they only had one satellite for a substantial period at the turn of this century, which would have made it impossible to benchmark against any other satellite.

In support of my first speculation, the authors found that their new dataset is consistent with two things:
  • For the tropics, it's consistent with the UW dataset that is "constructed using different methods, but with a similar goal of deducing the needed diurnal adjustments from the satellite measurements themselves". (That's opposed to basing diurnal adjustments on general circulation model outputs, which was the approach used for V3.3); and
  • It agrees "more closely with estimates of changes in total column water vapor than the STAR and UAH datasets".

Eliminating some records from degraded satellites: AQUA and NOAA-15

The researchers went to some effort to get the best dataset possible. They found that data from AQUA degraded from 2010, so they excluded those data, writing:
We focus first on the ocean results, which depend less on the diurnal adjustments. We note that the differences that involve AQUA all trend to large negative values starting in 2010. This corresponds to a period when increasing scan-to-scan noise suggests that this channel on AQUA is beginning to degrade (Roy Spencer, personal communication, also see Fig S1). Given the evidence for substantial instrument drift, we exclude AQUA data after December 2009. 
They also found that NOAA-15 degraded later in its life, so they eliminated it from December 2011. They noted that the impact was minor, writing:
The effect of this change on the global mean trend is fairly minor. If the post 2011 NOAA-15 data were included, the 1988-2014 AMSU trend would decrease by 0.004 K/decade. We note that if the AQUA data, which we excluded because of excess noise after December 2009, were also included the 1988-2013 AMSU trend would increase by 0.035 K/decade, more than canceling the effects of excluding the post-2011 NOAA-15 data.

I'm not an expert in the subject, I'm not even a novice, so I cannot say whether Mears and Wentz are right or whether Spencer and Christy are right, or whether they are all right or all wrong. All I can do is report what was in the paper. What I can say is that what was in the paper doesn't seem to gel with Roy's comment to Anthony.


Dr Mears had told me that, as stated in their paper, they cannot tell if the problem is NOAA-14 or NOAA-15 or both. It doesn't make a huge difference to the trend whichever way they go - if they removed NOAA-14 it would reduce the trend to Feb 2016 from 0.133 C/decade to 0.123 C/decade. He said in part:
There is no evidence that NOAA-14 drifts for time periods before 1998, because it agrees well with measurements made by other MSU satellites (NOAA-11 and NOAA-12) over the ocean, where the diurnal adjustment is not important.  The drift would have to suddenly appear in 1998.
So we cannot assume that the problem is with NOAA-14.  We instead explore the 3 cases:
  1. we assume NOAA-14 is wrong and remove it to the extent possible*
  2. we assume NOAA-15 is wrong, and remove it to the extent possible
  3. we assume that the error is shared between them, and we keep all the data.
      *we keep about a year of data during the MSU/AMSU oeverlap period -- any shorter time period would lead to excess noise in the MSU-AMSU difference we use to adjust AMSU to match MSU.
      These results are reported in the paper, and the differences between them should be viewed as part of the uncertainty in the final results. For the final dataset, we present case 3 where the errors are shared between NOAA-14 and NOAA-15, which is reasonable because we do not know the source of the problem.
      Added by Sou 5:45 am AEDT 5 March 2016

      When the satellites were active

      There was a nice chart in the paper, which I'll use to illustrate the different periods each satellite/instrument was providing data. The chart was to illustrate the first of the methods the scientists checked, but don't let that distract you. I'm just including it to give you an idea of the number of different satellites and instruments that have been used to build up the dataset from 1979 onwards. I've added some text (animated) so you can see which instruments had data that was excluded from version 4. There are only two as discussed above:
      • NOAA-15 was excluded from December 2010, and 
      • AQUA was excluded from December 2009. 

      I've also labeled NOAA-14 on the diagram. None of that data was excluded from version 4.

      Figure 2 | The time scale at the bottom shows when each of the satellites were providing data. The rest of this caption is as in the paper: Ascending local equator crossing time (LECT) for each of the satellites used. The LECT drifts over time for all satellites except AQUA, METOP-A, and METOP-B, which are maintaining at constant local time by orbit keeping maneuvers. For the drifting AMSU satellites, the thinner lines denote the portion of the missions excluded in the MIN_DRIFT analysis. (That's different to the time excluded from v4.) Source: Meares and Wentz 2016

      What if NOAA-14 was excluded?

      If NOAA-14 were pulled out, as Roy Spencer seems to be suggesting, then there'd have been an extended period in which there was only one satellite providing data. The chart below shows there was one short period early on where there was only one source of data, and I've also highlighted the extended period when there would have been only one source of data if NOAA-14 were excluded. That would be from the end of NOAA-12 in November 1998 to the start of AQUA in August 2002, a period of almost four years:

      Figure 3 | As for Figure 2, adding rectangles showing the period for which there is only one satellite, and the period for which there would only be one satellite if NOAA-14 were excluded.

      What about the UAH dataset?

      Few people would have much of a clue about what is going on with the UAH version 6.0 beta because, unlike Carl Mears and Frank Wentz, Roy Spencer and John Christy haven't published anything but a blog article. This is despite the fact that it's been out for almost a year. So while they have argued that Mears and Wentz should have applied a different correction to NOAA-14, or excluded it altogether, it can be argued that once again theirs is probably the odd one when it comes to the mid-troposphere.

      In his blog article, Roy Spencer has excluded different data than Mears and Wentz. He listed the following exclusions:
      • NOAA-17 - that wasn't included in RSS either
      • Metop (failed AMSU7) - I don't know which one. RSS has both METOP-A and METOP-B and say nothing about any failure.
      • NOAA-16 (excessive calibration drifts) - that's not in RSS either
      • NOAA-14 after July, 2001 (excessive calibration drift) - RSS uses that to December 2004
      • NOAA-9 after Feb. 1987 (failed MSU2) - RSS stopped using that on the same date.
      In addition, RSS has excluded NOAA-15 from December 2010 onwards and AQUA from December 2009, noting degradation problems. UAH haven't excluded those.

      In the chart below I've included the period NOAA-14 was operational as well as showing NOAA-15 and AQUA. The gray is when the latter two were operational but their data was excluded from RSS v4. (UAH only stopped using NOAA-14 in 2001, not altogether, and it hasn't excluded either NOAA-15 or AQUA.) I've aligned them all to a common baseline of the 1981 to 2010 mean. As always, click the chart to enlarge it.

      Figure 4 | Comparison of four mid-troposphere temperature records. I've also included the period in which NOAA-14, NOAA-15 and AQUA were operational The grey portion of the bars are where RSS v4 excluded data. Data sources: RSS v3.3 and v4; UAH v5.6 and 6.0 beta 5

      Looking at the above chart might hint at the solution to a mystery that many of us noticed back in 2011, when for the lower troposphere, RSS markedly diverged from UAH v5.6 (which was the latest at the time), and from the surface data. 2011 is around the time that AQUA and NOAA-15 were said to have degraded. Since then UAH has brought out their beta version 6.0 which got rid of the divergence between UAH and RSS. That is, until RSS v4 for TLT becomes available. (Is your head spinning yet?)

      Comparing the mid-troposphere with the surface trends

      In comparing the mid-troposphere with surface temperature, GISTemp (surface) starts off lower and ends up higher. The trend for surface temperature is greater than both mid-troposphere trends shown, but RSS v4 is closer to that of GISTemp than UAH v6.0 beta 5 is. I've aligned them all to a common baseline of the 1981 to 2010 mean.

      Figure 5 | Comparison of two mid-troposphere temperature records and GISTemp surface temperature record. Data sources: RSS v4; UAH 6.0 beta 5; GISS NASA

      The trends are:
      • GISTemp (surface) 0.16 C/decade
      • UAH version 6.0 beta 5 0.07 C/decade
      • RSS v4 0.13 C/decade.
      I'm not suggesting that the surface should warm at the same rate as the mid-troposphere. Still it's interesting that RSS v4 is much closer to the surface trend than UAH v6.0 beta 5 is.

      Unprofessional behaviour from Roy Spencer

      In keeping with the sullied reputation Roy Spencer and Anthony Watts cultivate, Anthony added an update to his article:
      UPDATE1: Given this sort of work has only two groups doing it, it is a very narrow field of scientific specialty, I asked Dr. Spencer this question:

      "I assume neither you or Christy were asked to review this paper? There aren’t many satellite temperature data experts in the world."

      He replied: "Interesting question….

      John reviewed their original paper submission to JGR, in detail, asking for additional evidence — but not advocating rejection of the paper.  The JGR editor ended up rejecting it anyway.

      Mears & Wentz then revised the paper, submitted it to J. Climate instead, and likely asked that we be excluded as reviewers."

      The Journal of Geophysical Research and Journal of Climate would hardly approve of that display of unprofessional behaviour. And Roy did not provide any evidence that Mears and Wentz asked that they be excluded as reviewers. They may have, or it may be merely conspiratorial thinking on Roy's part. (It isn't necessarily a bad thing to have people who would be expected to be very critical review your paper, I'd have thought. And Roy already said that John Christy didn't advise it be rejected.)

      From the WUWT comments

      Well, all this is very bad news for deniers at WUWT. RSS satellite data has now fallen from its hallowed spot as the ultimate in temperature data - at WUWT and other denier blogs. You can expect that now UAH v6 whatever's the latest beta will become the record of choice for deniers. That is, until Roy and John change it again.

      Not disheartened, the "thoughts" contain a multitude of reasons for ignoring or discounting the satellite data. This first lot of comments come from the article by Werner Brozek and Nick Stokes telling deniers that the pause is no more.

      dbakerber says that the El Nino blip proves "CAGW" is a sham. The reasoning, if you can call it that, is odd:
      March 2, 2016 at 7:11 am
      Well, I think an el Nino blip would be strong evidence against CAGW. Warming the sea first is the opposite of what CAGW theory says.

      Richard M suggests that the ENSO signal be removed. Now that wouldn't be a bad idea, except for the fake sceptics. That's because it would also remove the signal from that very important peak in the lower troposphere record that allowed the "pause" in the first place - the 1997-98 peak.
      March 2, 2016 at 9:20 am
      No rationalization needed. This has been predicted for months by those who understand ENSO. What’s curious is why you seem to think it is meaningful. Do you know what happens next? If the past is any guide we should see a 2-3 year La Nina. The pause may disappear for a few months just to reappear next year.
      It would be nice if Werner or Nick could use an approach like Santer et al 2014 where they remove ENSO from the signal. This would eliminate these types of meaningless discussions.

      george e. smith  is waiting for the potty peer to tell him what he should be thinking:
      March 2, 2016 at 10:39 am
      The Monckton Pause is restricted solely to the RSS data set. He makes that quite clear. He has not prohibited any person from generating their own algorithm for whatever purpose they wish.
      But the pause of 18 yrs and 8 months is limited to the RSS data set.
      Lord M of B will inform us in due time, if and when the pause is interrupted.

      emsnews is going to wait for 2017 to roll around:
      March 2, 2016 at 9:43 am
      This el Nino wasn’t as great at the peak as the 1999 one and it is collapsing rapidly now so this ‘warming’ will be short lived. The next cooling cycle may be strong.

      Werner Brozek thinks that scientists should time their dataset reviews for cold periods, or hot periods, or any period when the fake sceptics aren't shouting "pause":
      March 2, 2016 at 9:04 am
      Why in the world would anyone release a new data set now that ends the pause? The pause is ending anyway with the old data set. I believe they are just shooting themselves in the foot and opening themselves up to the criticisms directed at Karl.
      And if the pause ends on the new data set, many people would end up believing that adjustments ended the pause and not actual high temperatures.

      MattN is another one who's waiting for the next La Nina so he can draw a line from the top of the El Nino downwards (maybe):
      March 2, 2016 at 7:36 am
      Wouldn’t the coming La Nina offset the El Nino bump?

      It was the comments on the new version of RSS that brought out most of Anthony's conspiracy theorists.

      Toneb put up some charts showing that RSS v3.3 started diverging during the time of NOAA-15, not NOAA-14
      March 2, 2016 at 2:16 pm
      It’s not NOAA14 at fault.
      It’s 15.
      RSS v3.3 did not sit with calibration with RATPAC data since the new AMSU on 15.

      Tom T sez it's a conspiracy:
      March 2, 2016 at 3:43 pm
      RATPAC doesn’t suddenly veer. It didn’t exist prior to 2005. It matches that satellite data so well because it was pegged to the satellite data in development. The moment it went live it started to diverge because it was designed to. NOAA didn’t like what the weather balloons were saying so they rigged a new dataset to show warming.

      Roy Spencer was slow off the mark, but he's finally got around to submitting a paper on his new version 6 beta for publication. I must say that it would be a better look to add new, significant material to a paper before it's published, especially if it hasn't yet had peer review, rather than add various corrigenda afterwards.
      March 2, 2016 at 1:30 pm
      Our UAH v6 paper has been already been submitted for possible publication, although it still has to go through peer review. It’s a little awkward to add new, significant material to a manuscript after it’s already submitted, so John Christy and I need to discuss this.

      Since Roy and John said they reviewed another version of Mears and Wentz, Jaime innocently asks if Carl Mears will review their new paper.
      March 3, 2016 at 2:25 am
      Dr Spencer, will Carl Mears or anyone on the RSS team be peer reviewing the UAH v6 adjustments? This is all getting rather messy in my opinion. UAH adjustments decrease the trend in LT, RSS adjustments then increase the trend in MT, especially over the tropics, thereby miraculously finding the missing tropical mid-tropospheric hotspot. From the outside, for the lay-person who doesn’t understand all the intricacies of satellite tropospheric temperature retrieval, it’s coming down more to a matter of whom do we trust more to give an unbiased analysis of satellite data? Mears has demonstrated repeatedly an unscientific bias in favour of endorsing AGW theory. His recent behaviour in that regard falls far short of what one should expect from a scientist who is dispassionately surveying the evidence for and against a significant, perhaps overriding influence of man on climate. He is behaving more like an advocate than a scientist whereas yourself and John Christy have maintained a more balanced approach. So unless Mears can demonstrate an overwhelmingly good scientific case for his positive adjustments vs. the UAH negative adjustments, the balance has to be in favour of giving more credence to UAH data rather than the new ‘pause buster’ RSS data.

      Tom Halla has another conspiracy theory:
      March 2, 2016 at 1:30 pm
      RSS probably got tired of dealing with complaints from it’s funding sources. Why else “correct” to match known bad data?

      Peter Miller is quite ignorant. UAH scientists adjusted their recent records downwards, and not a whimper from WUWT. NOAA adjusted early records upwards so the long term trend declined, but deniers ignored that.
      March 2, 2016 at 2:01 pm
      Do the gatekeepers of government controlled temperature statistics ever adjust their numbers downwards?
      It would be nice to think that this was possible, even if it was only done once.

      Matt G  doesn't believe in attempts to improve datasets:
      March 2, 2016 at 2:02 pm
      I never trust data that has been adjusted to show a trend that didn’t appear before while it was in operation. Adjusting good data to bad data just because it matches the surface data closer is bad science practice.

      Village Idiot popped in with:
      March 2, 2016 at 2:08 pm
      But I thought it was UAH (Mr Spencer) that ‘adjusted’ his data (the so called Version 6) after copying RSS’s homework. Turns out Mr Mears’ answers were wrong. This leaves UAH high and dry…
      Begs the question….what will Sir Christopher Monketon make of developments?

      dbstealey adds his twopenneth worth of nonsense:
      March 2, 2016 at 2:54 pm
      Buncha nonsense. No one has ever measured AGW, so that’s merely speculation.

      rogerknights complains that fake skeptics are criticised for cherry-picking. He doesn't mention the fact that they keep shifting their "pause". His logic is quite fallacious as expected:
      March 2, 2016 at 2:21 pm
      Mears’s repeated claim that skeptics are cherry-picking the start year for the pause, despite repeated explanations that it’s calculated backwards from the end, makes him nearly as bad as anyone on his side of the fence.

      Martin Hertzberg is hopeless at arithmetic. He missed the class on averaging. I'd guess he never went to a class on biology, or climate studies.
      March 2, 2016 at 2:52 pm
      Why such a big fuss about a temperature anomaly of only few tenths of a degree C when the range of diurnal and seasonal temperatures vary by about 50 C? In any case, those variations have nothing to do with atmospheric CO3 and even less than nothing to do with human emission,

      Roy Spencer passes proof-reading, writing:
      March 2, 2016 at 2:53 pm
      carbon trioxide. I like the sound of that. :-)

      Tim mutters something about a "cool aide". Aide to whom or what - he doesn't say. A "cool dude" who has a job as an "aide"?
      March 2, 2016 at 2:54 pm
      Why am I not surprised. If a crime has similar modus operandi to other crimes it is suspected that there is a single group or individual behind the crimes.
      We are supposed to see nothing, hear nothing, and say nothing. Just believe and the cool aide will go down easier.

      FJ Shepherd thinks this will cause problems for the US Congress, at least those behind the witch hunt who touted RSS as being "perfect":
      March 2, 2016 at 4:17 pm
      Oh my. I am surprised that the folks at RSS are not less transparent. What will Congress say if one of the satellite data centres starts to “adjust” their temperature data as well. This is turning into one big mess.

      ristvan doesn't try to hide the fact that he's an utter nutter conspiracy theorist:
      March 2, 2016 at 4:55 pm
      FWIW, made essentially the same comment as this post on the previous thread. AW is correct. We now have Karlization and Mearsization. Same ization. Erase that darned pause. No matter how.

      Joel O'Bryan is another one:
      March 2, 2016 at 5:53 pm
      Did someone put the thumbscrews to Mears’ funding source?

      About the only "hootin' and hollerin'" I see is at WUWT. They are making an awful racket. Yet philincalifornia wrote:
      March 2, 2016 at 5:59 pm
      At the step back and look at the big picture level, it’s all getting a bit pathetic isn’t it? These supposed grown-ups who purport that increasing atmospheric CO2 levels pose a threat to humanity because the temperature is going up, all hootin’ and hollerin’ and high fiving because the temperature is going up. Sad people.

      D.I. is probably not too familiar with all the different UAH versions (it's already up to version 5 - and that's just of its new beta dataset), or scientific data in general. Thing is, deniers more often complain that they can't find data for old version than they complain about version numbers.
      March 2, 2016 at 6:02 pm
      Really comical that ‘Climate Science’ relies on so called ‘Version’ numbers for their Data Sets.
      I didn’t know that Science had ‘Versions’.
      Did all the great Scientists of the past have ‘Version’ numbers?

      Now I wonder which "temperature" in the troposphere talldave2 would like to see reported, and whether he would be happy with an average temperature going up several kilometres from the cool to the distinctly cold levels:
      March 3, 2016 at 7:22 am
      Once again, I urge everyone to stop reporting anomalies and start reporting temperatures.
      Chip Knappenberger seems to have looked into the NOAA-14 treatment and wrote:
      March 3, 2016 at 10:13 am
      Regarding the NOAA-14/NOAA-15 treatment, as far as I can tell, this was handled very similarly in RSS v3.3. If this is indeed the case, then, it is evidently not the source of the reported trend increase from RSS v3.3 to RSS v4, and the statement that “it looks like they decided to force good data to match bad data” in this regard is misplaced (by at least several years). Is there something I am missing regarding a new handling by RSS of the NOAA-14/NOAA-15 transition? (Also I think UAH v6 uses NOAA-14 through July 2001).

      References and further reading

      Carl A. Mears and Frank J. Wentz, 2016: "Sensitivity of satellite-derived tropospheric temperature trends to the diurnal cycle adjustment." J. Climate doi:http://dx.doi.org/10.1175/JCLI-D-15-0744.1 (subs req'd)
      Stephen Po-Chedley, Tyler J. Thorsen, and Qiang Fu, 2015: "Removing Diurnal Cycle Contamination in Satellite-Derived Tropospheric Temperatures: Understanding Tropical Tropospheric Trend Discrepancies." J. Climate, 28, 2274–2290. doi:http://dx.doi.org/10.1175/JCLI-D-13-00767.1 (pdf here)

      Mears, Carl A., and Frank J. Wentz. "The effect of diurnal correction on satellite-derived lower tropospheric temperature." Science 309, no. 5740 (2005): 1548-1551. DOI: 10.1126/science.1114772 (pdf here)

      John R. Christy, Roy W. Spencer, and William D. Braswell, 2000: "MSU Tropospheric Temperatures: Dataset Construction and Radiosonde Comparisons". J. Atmos. Oceanic Technol., 17, 1153–1170. doi: http://dx.doi.org/10.1175/1520-0426(2000)017<1153:MTTDCA>2.0.CO;2 (open access)

      Niamh Cahill, Stefan Rahmstorf and Andrew C Parnell. "Change points of global temperature". 2015 Environ. Res. Lett. 10 084002. doi:10.1088/1748-9326/10/8/084002(open access)

      Roy Spencer's article where he describes UAH version 6.0 beta

      From the HotWhopper archives


      1. The RSS is a den of warmist scientists. True skeptics (who don't let evidence sway their opinions) should listen to the respectable Mann who is a true scientist! Well listen to the bits that we like.

      2. OT but "If all else fails, we could simply grow extra food in storm proof greenhouses."

        So says master debunker Eric Worrell in response to a paper predicting increased malnutrition in a changed-climate future.

        When I say master, Eric is so good, he does not even need to read the paper before penning a rebuttal.


      3. Good post Sou,
        There is a new RSS TTT v4 out now, with the overall trend 0.172 C/decade. I think we have to wait a while for a new TLT...
        Spencer and Christy have demonstrated their scientific credibility by making important choices based on anectodes. A Nasa scientist have told them that the AMSUs have a "Cadillac calibration", which makes them trust NOAA 15 fully and not NOAA 14. I think they prefer the outcome of this choice as well..

        I believe that NOAA 15 is wrong and NOAA 14 right. The NOAA STAR team has developed a AMSU only data set, poster here (there is a paper with the same title)
        In the figure down right it is shown that the trend of AMSU channel 5 is much lower than an average of the nearby channel 4 and 6. The culprit seems to be NOAA-15, dominating with its long service, and drifting compared to the other satellites (figure upper left).
        I am also quite convinced that a validation vs radiosonde date would prove that NOAA-14 is right and NOAA-15 is wrong. NOAA-14 may have issues, but not worse than it can be fixed with IMICA intercalibration and a proper diurnal drift correction.

        1. IMICA?

          Yes, the RSS TTT is as you say, 0.172C/decade. In v3.3 TTT is 0.118C/decade and TLT is 0.126C/decade. For TMT, v3.3 0.083C/decade, v4.0 0.133C/decade.

          From the above datasets I've taken three SWAG's and end up with an estimate for v4 TLT of 0.177C/decade (0.173, 0.176, 0.181).

          Looking at the differences between versions 4.0 - 3.3, for TMT and TTT, shows residual trends that are slightly concave up (an OLS quadratic fit for TMT -> R^2 = 0.9362 and for TTT -> R^2 = 0.9364 (ludicrous precision necessary to show 1st significant digit difference)). YMMV

      4. What we get are temperature trends in very thick layers in the atmosphere (kilometers thick, looking upwards into space), not the temperature of a particular spot or distinct level in the sky.

        Surely that ain't true for the favorite dataset of the mitigation skeptical movement.

        These are the people who are comfortable with blog posts claiming that there is no such thing as a global mean temperature because it varies spatially. The temperature may well vary more within this kilometers thick layer over which the tropospheric temperatures are estimated.

      5. Thanks for this informative article. I learned quite a lot about the satellites and their interaction.

        It's especially nice to see the question of the recent apparent "divergence" of the satellite data from the surface data may have an answer. A group of deniers has been using that as evidence that the surface data is being rigged ( = conspiracy). From the new paper, it seems instead to do with AQUA and NOAA-15 getting funky (that's a technical term, I'm sure). It makes more sense that there's been mechanical problems with the orbiting weather stations that theorizing there are thousands of scientists playing a massive practical joke and keeping it secret.

        I confess too I was unaware there are so few satellites involved in the measurements. Deniers go on and on about "siting issues" with a tiny handful of the many thousands of surface stations. We don't hear (probably because some of us--meaning me--got too lazy to learn) that there are only two (2) satellites most of the time furnishing microwave data for the troposphere measurements. A tiny flaw in one of them would create massive inaccuracies, wouldn't it?

        1. "I confess too I was unaware there are so few satellites involved in the measurements."

          That surprised me when I first discovered it. I'm sure I've read denier blogs claiming there are over 100 satellites, along with claiming they are accurate to 100th of a degree. I don't know if this is the result of deliberate deception, or if they just assume that is the case.

        2. Yes, Monckton in his recent wuwt effort spoke of rss and Uah as possessing "their satellites". He actually thought that these organisations each had their own sets of data that they then proceeded to analyse.

          What a buffoon.

      6. Here are the steps required to get the RSS global mean tropospheric temperature from satellites: http://www.skepticalscience.com//pics/rss_flowchart.png . SkS has more info here: http://www.skepticalscience.com/Satellite-record-vs-thermometers.htm

      7. "I assume neither you or Christy were asked to review this paper? There aren’t many satellite temperature data experts in the world."

        He replied: "Interesting question….

        John reviewed their original paper submission to JGR, in detail, asking for additional evidence — but not advocating rejection of the paper. The JGR editor ended up rejecting it anyway.

        Mears & Wentz then revised the paper, submitted it to J. Climate instead, and likely asked that we be excluded as reviewers."

        "...likely asked that we be excluded as reviewers" is unsubstantiated speculation. Maybe they were; maybe they weren't. In the retelling, being excluded as reviewers will be retold as a fact.

        Is it an insult to be published by the Journal of Climate?

        1. They'd be comparable. J Climate has a higher impact factor:



        2. Instead of give a good faith answer, and say there are other sat temp experts that can review....

      8. classic denier tactic create a bit of FUD

      9. Spencer and Christy have responded on Spencer's blog.

        "The evidence suggests that the new RSS v4 MT dataset has spurious warming due to a lack of correction for calibration drift in the NOAA-14 MSU instrument. Somewhat smaller increases in their warming trend are due to their use of a climate model for diurnal drift adjustment, compared to our use of an empirical approach that relies upon observed diurnal drift from the satellite data themselves. While the difference in diurnal drift correction methodolgy is a more legitimate point of contention, in the final analysis independent validation with radiosonde data and most reanalysis datasets suggest better agreement with the UAH product than the RSS product."


        1. This is what Roy Spencer wrote back in 2011, before he changed his tune, or tuning :)

          my UAH cohort and boss John Christy, who does the detailed matching between satellites, is pretty convinced that the RSS data is undergoing spurious cooling because RSS is still using the old NOAA-15 satellite which has a decaying orbit, to which they are then applying a diurnal cycle drift correction based upon a climate model, which does not quite match reality. We have not used NOAA-15 for trend information in years…we use the NASA Aqua AMSU, since that satellite carries extra fuel to maintain a precise orbit.

          In their new article, they are using the trick they've done before. They've averaged radiosonde data from different datasets, which would hide the differences between them. I'd prefer to see them plot them separately. From the comments they've also excluded some key data and instead used less reliable data.

          I wonder how their peer review will go?

        2. BTW I think UAH now include NOAA-15 so they must have changed their mind somewhere along the line.

        3. This comment has been removed by the author.

        4. Yes, just saw that one, and in typical Deniersville fashion Woy shows this radiosonde chopped off at 0.75 R^2 graphic ...

          correlation != causation

          The actual differences are like ~0.05 in the total R^2 space (0 < R^2 < 1).

          Something with a LOWER r^2 just might pass a significance test at a higher level (p value) than a HIGHER scoring r^2 result. There, fixed that for me.

          M-I-C-K-E-Y M-O-U-S-E

        5. There you go! Roy blogged it so it must be true (said every denier everywhere).

      10. Great article.

        It amazes me that Anthony can attack Peter Sinclair questioning the accuracy of satellites in the same article as he says you have to exclude one of the satellite because it's showing spurious warming.

        1. The levels of cognitive dissonance must be off the charts.

      11. I tried to comment over at WUWT, but my comments didn't go through. Probably just as well, usually no one there is brave enough to address my points.

        I am glad the satellite measurements do show the El Nino peak, it does give confidence that the measurements do respond to temperature. The trend line is another matter, last time I looked over 20 years of data is required for it to show statistically significant trend.

        I am curious to see how the stratospheric measurements will respond to the El Nino.

        1. Seeing as I cannot post at WUWT, I might as well repeat my comments here.

          "There never was a "pause" of course.

          Saying the data shows "no statistically significant warming" might be syntactically correct, but it is misleading. There may be underlining warming, cooling or temps staying the same, you cannot exclude any of these possibilities with confidence if the data is not statistically significant."

        2. I wondered where you had gone Harry (Brandon too).
          I've been stirring up the pot there (Toneb)
          and also at Spencer's place.

        3. How nice to be missed. :)

          I've been giving WUWT a rest in favour of doing my own studies and writing them up on my own blog. It's really difficult (I increasingly appreciate the abilities of our host, Eli, ATTP, etc.), but I find it a more rewarding and educational challenge, and far less frustrating.

        4. I can go with that Brandon - I suppose I'm lazy in a way as I use their obvious ignorant confirmation bias to brush up my knowledge. It's the anger that drives me. Totally frustrating I know because the only vindication coming is when they don't reply. Usually you have to follow them down the "rabbit-hole" when they respond and that way lays madness. One should always deny ignorance however.

        5. There's no doubt that Watts and friends have pushed me to learn more. When I first started up over there, most were quite happy to have me and I thought many of my first discussions were quite good. Before your pal mpainter wore out his welcome at WUWT and took up residence at Spencer's, he trotted out his favourite question about the difference between the Sahara and the Amazon. Which, at the time, I thought was a perfectly reasonable and genuine question. Little did I know. Goodwill lasted about two rounds before he started quoting me out of context on other threads and declaring victory ... even going so far as to say that I "disgust" him. The man gives new meaning to the term, "gibbering twit".

          My relations with most of the other denizens soon followed suit. It long ago ceased being fun for me. Give it a few months, I might miss it. For now, whenever I chance to read a WUWT article, I have to resist the urge to hurt my computer.

      12. "They've averaged radiosonde data from different datasets, which would hide the differences between them."
        Isn't that something like the hockey stick Trick, only way more manipulative?

        1. That didn't come out quite right, since adding the modern temp record to the paleo-record isn't manipulative in the least.

        2. Averaging is fine, as long as the averaging is valid and makes sense.

          I will give a crude example: you could find the "average" weight of humans and cats by taking all the weights and averaging them. Does it make any sense to do this: no.

        3. "Isn't that something like the hockey stick Trick, only way more manipulative?"

          Not sure why you are referring to a "hockey stick Trick", unless you are indulging in rhetoric. The "trick" used in the climate change literature was not referring to the hockey stick chart exactly.

          When Mann et al calculated the hockey stick curve, they went to a lot of trouble to ensure that the calculations made sense. Mann outlines the methods they used in his book.

          There is no explanation of how the radiosonde data was averaged.

      13. To clarify about NOAA-14 and NOAA-15, Dr Mears has told me that, as stated in their paper, they cannot tell if the problem is NOAA-14 or NOAA-15 or both. It doesn't make a huge difference to the trend whichever way they go. He added:

        "These results are reported in the paper, and the differences between them should be viewed as part of the uncertainty in the final results. For the final dataset, we present case 3 where the
        errors are shared between NOAA-14 and NOAA-15, which is reasonable because we do not know the source of the problem."

        1. Dr Mears about the so far perfectly working NOAA-14:
          "The drift would have to suddenly appear in 1998"

          Yes, the drift appeared suddenly in 1998. The drift has a name, NOAA-15, the first and longest serving AMSU, that can't be verified by anything else...

        2. Olof makes a good point. These articles by tamino:


          and Nick Stokes:


          show that the satellite data have clearly been diverging from the land-ocean data from ca. 2000.

      14. @sou

        so is Dr Spencer just commenting (negitively) because he feels he HAS to comment - whether is makes sense or not

        just to give the deniers a "hook" to construct an argument on

        1. From what I can see, yes.

          As I'm aware every time RSS has made an adjustment to it's data set, UAH has followed.

          Also, every time there has been discrepancy between the satellite and surface data the scientific analysis has found in favour of the surface data.

        2. I believe that Spencer&Christy are very stressed right now. They have retreated to the "Cadillac calibration" defense line, There will probably be major difficulties to justify and publish their paper on UAH v6 beta, since its major strength, the similarity with RSS 3.3 TLT is gone now..

        3. I think Roy Spencer and John Christy get quite anxious about their data being questioned. Given the fact that they've usually only made corrections after people have pointed out what's wrong, and only then very tardily, it's not a surprise.

          This time it's not just RSS, it's the Po-Chedley paper as well. The fact the latest UAH version lowered the trend, Roy and John must be worried that they'll look foolish again. They'll be concerned that they will have to make more changes that will bring it back in line with RSS (and surface trends).

          Even though they are both science deniers, their self esteem must be tied up in getting the record right, even if the results go against their religion.

      15. I thought UAH didn't use AQUA after 2009, because of 'increasing noise' (see update 5 Oct 2012 in the Readme for version 5.6)?

        1. Henk, I'm thinking that UAH must have brought some satellites back into play with version 6.0 beta. In 2011 Roy Spencer also said that NOAA-15 had problems, but now says it doesn't.(I linked to the article about their v6.0beta up above. They process they use is quite different for v6.0 than v5.6 I believe.)

          Perhaps when their paper gets published we'll find out more about what they are currently doing.

      16. Werner Brozek's comment is especially comical. He does't realize that RSS hasn't changed the data for the lower troposphere -- it's still version 3.3 -- and still the pause ends.

        Werner is all about numerology, not science.

      17. That's part of the problem with my country cousins, always skimping on the important stuff. The AMSU calibration may indeed have been to Cadillac standard, but it obviously needed to be Lamborghini.

        1. This comment has been removed by the author.

        2. This comment has been removed by the author.

        3. Cadillac? How about Lincoln Continental?

          Some of us still remember the big ole American junkyard road boats of the 70's and 80's.

          Word association ...

          Cadillac? Junk!
          Gold Standard? Fixed!
          MSU? Weather!

          Gold standard ...

          "A gold standard is a monetary system in which the standard economic unit of account is based on a FIXED quantity of gold."


          "The $42.22 par value was made official in September 1973, long after it had been abandoned in practice. In October 1976, the government officially changed the definition of the dollar; references to gold were removed from statutes. From this point, the international monetary system was made of pure fiat money."

          Ted Cruz ...

          "Rafael Edward "Ted" Cruz (born December 22, 1970) ... "

          Someone born in 1970 using the term "gold standard" seriously needs to pick up a history book or three, then shot himself in the head.

          Someone ELSE anecdotally mentioning Cadillac and NASA, in the same sentence even, ought to shot himself in the head.

          There, fixed that for me: Part Deux.

      18. Sou,

        Maybe there's an expert who'll drop by to explain [calibration targets].

        Definitely not me, but this might be in reference to post-launch calibration which minimally involves two things:

        1) Taking a reading from deep space (the cold target)
        2) Taking a reading from an onboard emitter (the hot target)

        Some suspect, certainly not Eli, that the onboard hot targets degrade over time, or otherwise go wonky, thus introducing a bias in long-term trends.

        This has been looked at before. An older paper, Mears and Wentz (2005) is freely available on the RSS website and has this to say:

        The long-term behavior of a time series constructed from TLT is also dependent on the procedure used to merge the nine MSU satellites together into a single time series, in particular on the values of the parameters ("target factors") used to empirically remove the spurious dependence of the instrument calibration on the temperature of the hot calibration target (5,7,15) (SOM text). For the results presented below, we used exactly the same merging procedure and target factors (but different offsets) as we used when producing our results for MSU2 (26).

        What I gather from all this is that even if the onboard hot targets aren't being flukey on a single satellite, cross-satellite comparison may involve a little more art and less science than the average WHUTTer would normally stand for.

        1. What I gather from all this is that even if the onboard hot targets aren't being flukey on a single satellite, cross-satellite comparison may involve a little more art and less science than the average WHUTTer would normally stand for.

          Even Roy Spencer seems to agree - when he introduced version 6, he said:

          One might ask, Why do the satellite data have to be adjusted at all? If we had satellite instruments that (1) had rock-stable calibration, (2) lasted for many decades without any channel failures, and (3) were carried on satellites whose orbits did not change over time, then the satellite data could be processed without adjustment. But none of these things are true.


          All data adjustments required to correct for these changes involve decisions regarding methodology, and different methodologies will lead to somewhat different results. This is the unavoidable situation when dealing with less than perfect data.

        2. Let it not be said that the UAH crew are completely at sea about reality.

      19. Spencer has an update on his site that says: "while the warming in RSS v4 versus UAH v6 might be as described above, when RSS v4 is compared to RSS v3.3, the increase in warming might be mostly due to their new diurnal cycle adjustment. In other words, the NOAA-14 calibration issue was also in their v3.3, but maybe it was obscured more by diurnal drift adjustment issues."

      20. Spencer's reply looks pretty weak to me. Radiosonde data isn't particularly close to global coverage, and the uneven sampling makes a simplistic comparison to global products less than useful. The correct comparison is to a version of the satellite records subsampled to match the radiosonde coverage. When I subsample RSS TMT v4.0 and UAH v6.0b5 to match HadAT, RAOBCORE and RICH coverage, RSS is a much better fit to each than is UAH, both in terms of r^2 and, more importantly, the long-term trend.

      21. I have seen this link repeatedly posted today http://realclimatescience.com/2016/03/noaa-radiosonde-data-shows-no-warming-for-58-years/. "Steve Goddard" thinks that NOAA's balloon data shows no warming for 58 years.

        1. Goddard's just lying. Here's a fair comparison between the 1977 graph, and the same up-to-date radiosonde data he linked. Odd that he didn't bother to plot them together, since he had access to both.

        2. Thanks, MartinM. Yes, odd if one is looking to present valid information. No at all odd if one is looking to rile up one's conspiracy theory believing blog visitors.

      22. Carl Mears has written a blog article about the new paper, for anyone not yet able to access it. (I'm told the paper will be open access at some stage, not sure when.)

        His article addresses some of the questions people may have had. (I've added a link to the references under the article as well.)

        1. From the references on Mears' post:

          "University of Alabama, Huntsville Data: No relevant paper has been published."

          Subtle :D


      Instead of commenting as "Anonymous", please comment using "Name/URL" and your name, initials or pseudonym or whatever. You can leave the "URL" box blank. This isn't mandatory. You can also sign in using your Google ID, Wordpress ID etc as indicated. NOTE: Some Wordpress users are having trouble signing in. If that's you, try signing in using Name/URL. Details here.

      Click here to read the HotWhopper comment policy.