Climate Deniers Deny Temperature Itself

There’s a lot of talk these days about global temperature, in news reports, policy discussions, and casual conversation. The global temperature under discussion is surface air temperature (SAT). It’s not the depths of the ocean — even though that’s really a better measure of global warming, we don’t live in the depths of the ocean. It’s not the temperature in the upper atmosphere, we don’t live in the upper atmosphere. It’s the temperature at Earth’s surface. That is, after all, where we live.

But climate deniers don’t like that, because surface air temperature has risen so dramatically these last few years. Someone proclaims “The last five years have been the hottest on record!” and the deniers want to deny it, but they can’t use actual data for SAT because the data support the statement. But there’s a sneaky way for them to try: look for some data set — ANY data set — which you can claim represents “global temperature” but contradicts the “last five years” claim. If instead of claiming it’s global temperature, you simply say nothing at all about it so readers assume so, even better. If that data set happens to be crap — best yet!


There are lots of data sets for actual SAT: the best-known are from NASA, from NOAA (National Oceanic and Atmospheric Administration), from HadCRU (the Hadley Centre/Climate Research Unit in the U.K.), Cowtan & Way (researchers from the Univ. of York in the U.K.), and the Berkeley Earth Surface Temperature project (an independent effort from researchers in the USA). Here’s the data from NASA — yearly average surface temperature for the whole planet (which is, after all, the topic):

I’ve put red circles around the five hottest years. It turns out, they’re 2015 through 2019, i.e. the last five years. Although the year 2019 isn’t over yet I’ve plotted the year-to-date average; those who know the data well know that we’ve seen enough of it already to figure out that it’s going to end up in the top five. If I left out 2019 altogether, then the top five would be 2014 through 2018, again, the last five years.

What about data from NOAA, or HadCRU, or Cowtan & Way, or Berkeley Earth?

Golly! They ALL say that the last five years are the hottest on record!


Global temperature shows a distinct trend, but it also shows plenty of fluctuations. Lots of things cause fluctuations — things like massive volcanic eruptions (which tend to cool off the planet — temporarily) and el Niño, the warm phase of the El Niño Southern Oscillation (ENSO), (which tends to warm up the planet — temporarily). But they don’t cause trend. The trend these days is from man-made climate change (i.e., global warming).

Climate deniers love the fluctuations because they make it harder to see the trend. The data set which shows the least level of fluctuation is: total heat content in the depths of the oceans. But, humans don’t live there so it hasn’t gotten a lot of attention. What does it say? This (ocean heat content to a depth of 2000 meters, the latest, and in my opinion best, estimate from the Institute for Applied Physics IAP):

The trend is quite clear! That’s because the fluctuations are much smaller (relatively speaking) than they are in surface temperature. Trend and fluctuation go together in real data, but the fluctuations don’t tell us about climate change; they’re there even when climate doesn’t change!

Fluctuations make it harder to see and to measure the trend. That’s why climate deniers love ’em — they don’t want you to know what the trend is because that does tell us about climate change. They don’t even want to know themselves; if they faced the truth, they’d either have to abandon their attempts to sabotage the rest of us by preventing us from actually doing anything about it, or admit to themselves that they are inherently evil.

Instead, they look for a data set that has the largest level of fluctuation, in order to fool themselves and especially to fool you. There’s a natural choice: temperature in the atmosphere. Not at the surface (where we live), mind you, but in thick layers of the atmosphere above Earth’s surface. The favorite choice for many climate deniers is TLT, temperature in the lower troposphere, the lowest level of the atmosphere where most of our weather happens. Lower-troposphere temperature (TLT) shows much more fluctuation than surface air temperature (SAT), mainly because it responds much more strongly to fluctuation factors like volcanic eruptions and el Niño, a fact clearly demonstrated in the peer-reviewed literature (see, for instantce, Foster & Rahmstorf 2011).

The two best-known data sets for TLT are from RSS (Remote Sensing Systems) and UAH (University of Alabama at Huntsville). They’re based on satellite measurements of microwave emissions from the bulk of the atmosphere. Climate deniers like to imply that it’s as simple as sticking a thermometer in your ear like a doctor might — but that’s a lie. It’s extraordinarily complicated, which is why there are multiple data reductions (RSS and UAH are but two), and those have been revised over the years as we’ve learned how to do it better.

They don’t start until 1979 because the satellites providing the raw data weren’t there until 1979. Here’s what they show:

The UAH data are warming more slowly than RSS data. That’s because the UAH data are crap. When we compare them to other data sets, especially from balloons which radio their data to us after release to rise through the air, the UAH data are the outlier, the lone data set showing less warming than the others (see this and this) . That’s because the scientists who created it are two climate deniers: Roy Spencer and John Christy.

Either way, they don’t show the last five years as hottest on record. That’s because the fluctuations are so much larger than they are in surface temperature or ocean heat content. You can probably see the large peak in 1998, showing how strongly that year’s el Niño affected TLT. It also affected SAT, but not by nearly as much — the biggest fluctuations are in TLT. Biggest fluctuations == least precise trend == climate denier favorite. Also, crap data == trend too low == climate denier favorite.


Case in point: David Middleton at the WUWT blog. He objects to statements by Steven Novella, so he attempts to contradict them. The first? You guessed it! Novella says “2014-2019 have been the hottest five years on record.”

Novella’s claim is mistaken because 2014-2019 is actually six years, not five. But it’s an honest mistake, and it’s not a misleading claim because whether you include the 2019 year-to-date value or not, the last five years were indeed the hottest on record.

David Middleton attempts to mock this by showing this graph:

It’s not global temperature (at least, not where we live!). It’s lower-troposphere temperature (TLT) from UAH. What a surprise! Not.

David Middleton doesn’t discuss this. Heaven forbid you know what you’re really looking at.


This blog is made possible by readers like you; join others by donating at My Wee Dragon.


15 responses to “Climate Deniers Deny Temperature Itself

  1. Does this ever stop?….
    Good work in trying to make them lesser and lesser credible, but I doubt many people, who could be convinced, i.e. are not yet deeply entrenched, read this formidable blog.

    [Response: I agree, but I’m not trying to persuade the unpersuadeable. I’m trying to alert more people to their tricksy ways.]

  2. And yet, even with all the fluctuation and the bad data, the last five years are the warmest five-year period on the graph. They are now grasping at the thinnest of straws.

  3. Yes. Actually look at the 5-year smooth of UAH, and even there, the trend is clear and the last 5 year period is clearly the warmest: http://www.woodfortrees.org/plot/uah6/mean:60

  4. This is the sequence in the never-ending “debate.”
    1. An AGW denier posts a graph of the UAH lower troposphere data and the data for average of all the climate models and claims it proves the climate models are wrong because their prediction was too warm.
    2. I respond by pointing out the UAH LT data diverges from all the other temperature datasets. When all those other datasets are used instead of UAH, they coincide with the climate model graph quite well.
    3. The denier then claims the other datasets have all been tampered with, citing one or more of Tony Heller’s blogs in which he plots corrected and uncorrected data in the same graph and then claims the difference proves fraud.
    4. I respond by explaining what Tony Heller has done, and by explaining that each of these adjustments is a correction of an error or a bias, which has been peer-reviewed and published. I challenge the denier to find ANY adjustment to ANY dataset that is incorrect or fraudulent. The only response to the challenge is more Tony Heller graphs comparing corrected with uncorrected data.
    5. The denier then claims that ALL the adjustments of ALL the datasets reduce temperatures in the first half of the 1900s and increase temperatures in the second half.
    6. I respond by showing the graph of the recent NOAA adjustment which did the opposite.
    7. The denier then tries the “satellite data is pristine” gambit.
    8. I respond that UAH is on version 6.0, and that UAH disagrees with RSS, which is also satellite data and which agrees with the ground-based datasets.
    9. The denier charges that RSS is adulterated with non-satellite data.
    10. I don’t understand that one, so I reiterate that UAH is on version 6.0 and that satellites don’t measure temperature anyway; they measure microwave intensities at different altitudes, and then a MODEL is used to convert the data to temperatures and there are significant uncertainties involved.

    The denier raises a completely different mole at that point. It could be the “MWP was warmer” mole, based on a graph of temperature obtained from an ice core from central Greenland, or it could be the “Al Gore flies a private jet” or “Norway sells oil,” (I live in Norway), but lately the escape mole is “Greta Thunberg is abused by her parents” or “AOC and Greta Thunberg are the spokespersons for AGW instead of scientists.” It never ends. I have learned that the best I can do is to force the denier to abandon his argument and switch to a different attack. But in a few weeks or a month, another denier will post the same UAH vs climate models graph and its deja vu all over again.

    • Martin Smith,
      The “adulteration with non-satellite data” of the RSS TLT data (your #9) presumably refers to the time-of-day adjustments. Unlike surface measurements which overcome the diurnal cycle by collecting max & min readings, TLT data obtains temperature measurements when the satellite is overhead which can be any time of day. This makes a big difference to the readings especially over land. So a diurnal correction is made using an inferred diurnal cycle. Both RSS & UAH do this and this diurnal cycle was in the past calculated directly from those dreadful models that denialists have such problems with. This was true for both UAH & RSS.
      The RSS v4.0 now incorporates an adjustment which re-calculates the diurnal cycle using periods where there is multiple satellite data so it is much less “adulterated with non-satellite data” than the lower-trending RSS v3.3.

      RSS have a useful info page explaining the changes introduced in v4.0.

    • I have learned that the best I can do is to force the denier to abandon his argument and switch to a different attack.

      Do not let them abandon a pint before admitting thot the previous point was wrong. That is the best way to show the audience they are ignorant and deceptive. By going along with their topic changes it looks like a debate.

      • Exactly right Victor. You can’t get distracted by all the other points they will shoot out in a scatter-gun attempt to conceal that their initial point has been proved to be utterly bogus. Keep your focus on whatever their initial point was and say you won’t answer any of their other points until they respond to their first error. 9/10 they will either explode into obscenities and conspiratorial madness or just run away. Either way, you win.

      • I used to do that, Tim, but it doesn’t work. I used to keep demanding that they address their initial point, but they never go back to their initial point, because, I think, either they know their initial point has been proved wrong, or they realize they are out of their depth and don’t have the facts (which don’t exist) to address it. So they jump to the next talking point, and they force the change. Repeated demands to address the initial point only allow them to introduce more and more obfuscation with separate points.

        So I have adopted the tactic of declaring victory on the initial point by pointing out that they are unable to defend their position on that point.

  5. The “smack-down” provided by denialist David Middleton on the remote planetoid Wattsupia is aimed not just at the global SAT record but a whole bunch of assertions. These assertions actually originate in the WMO’s ‘The Global Clmate in 2015-19’ where their derivation is set out. While the SAT data “smack-down” is easily rebutted as per the OP, the Sea Level Rise “smack-down” requires a bit more digging.
    The WMO uses this ESA SLR data with a bit more recent data added. The acceleration in SLR over the period 1993-2019 is quite evident to the casual observer. Mule-brained Middleton presents a graphic of very different-looking data that appears to show zero acceleration. The more observant will see Middleton does offer a ‘correction’ that still involves a small amount of acceleration.

    “Sea level rise has been 3.2 mm per year on average since 1993, but is 5mm per year averaged over the last five years”
    More like 3.0 and 3.4 mm/yr.

    The data Middleton uses is from NASA and a linear regression through the full 1993-2019 data and the last 5- years 2014-19 does yield the numbers presented by Middleton if you use the data uncorrected for GIA. (With GIA you get the 3.3mm/yr as per the NASA graphic and 3.8mm/yr for 2014-19.) But here lies a problem for Middleton. Using OLS over short periods of wibbly-wobbly data shows nothing more than the level of wibbly-wobbliness. Sticking with the non-GIA data used by the Mulehead, there is a lot more acceleration that he manages to miss. If you graph out the annual rise, it is obviously increasing from roughly 2.5mm/yr to 4.5mm/yr and hitting wobbly-values as high as 5mm/yr. More precise analysis, an OLS through the first half of the data gives 2.67mm/yr while through the second half 3.91mm/yr. That would suggest a rate of annual accelertation of 1.0mm/yr/decade and today’s SLR (non-GIA) thus running at 4.3mm/yr (1.3mm/yr above the average for the full period.) With GIA and its 3.3mm/yr over the full period, this 2014-19 value would be 4.6mm/yr.
    Given the WHO data is subjected to its own adjustments, the rates of dSLR wiithin the NASA data are not a million miles away and still of a significant and worrying level. But then we should never forget that on the planetoid Wattsupia, nothing has to be true if the inhabtants would rather it were not true.

  6. Not precisely on topic but:

    I think you might have some words for the people who have been judging the “Nobel” prize in economics. I am not aware of who they are, but their execrable selection criteria led to Nordhaus & Tol getting in.

    Some commentary from an economist I actually trust.

    https://evonomics.com/steve-keen-nordhaus-climate-change-economics/
    or
    https://www.nakedcapitalism.com/2019/07/the-cost-of-climate-change.html

    It may be useful to know about the rebukes here when the “Nobel prize winning” economist’s work is offered up in argument.

  7. Temperature is temperature, no matter how you measure it. The fake skeptic universe has found a “hook” that traps the gullible. Scientists, being honest, do their best to accurately and honestly collate different styles of measuring. Dishonest “brokers” have found something to complain about. The honest and the dishonest are at odds, because honesty will never quite accept the criminal intent of the illusionists. We are too tolerant of lies.

  8. The other phenological data available over impressive time-spans is the Kyoto peak cherry-blossom dates which kicked-off in AD 812 and, initially with big gaps (there are only 14 data points 812-900), it builds into an impressive archive. The individual dates range from the last week in March into early May, with average results spanning 10th-17th April. The last late peak in this ‘average’ was back in 1840 and since then the ‘average’ has become earlier, and arriving before 10th April for the first time in twelve centuries by about 1960, since which the ‘average’ has continued to arrive earlier, as of 2016 ‘averaged’ arriving about 5th April.

  9. It’s funny how you bemoan skeptics of focusing on short term data when you cite ‘unprecedented’ increases in global temperature in only five years as something we should concern ourselves is. The fact is that if you look at the temperature trend over time the warming trend is far less severe than the ‘scientists’ have said it should be. [Response: Not true.] With the amount of CO2 we’ve released the planet should be literally boiling if what they said was correct. [Response: Not just untrue, that’s a lie.] Except it isn’t and CO2 matters little in its effect on the climate.

    There’s a reason why so many ‘scientific’ papers are now laughed at in the statistical community and it’s because of their rejection of statistics. [Response: Not true.] There is no statistical significance between the CO2 trend and global warming [Response: False. I wouldn’t have guessed you were *that* ignorant, but apparently you are.], yet ‘scientists’ still peddle it as fact. The fact is that climate has always changed and always will and CO2 reduction will not change this. [Response: People have always gotten lung cancer — so you think smoking can’t be a cause, right? Have a cigarette.]

    Moreover, much of the observed temperature increases have been increasingly attributed to the urban heat island effect by actual scientists [Response: You mean perhaps those few (VERY few) who are deniers.], which has been grossly ignored by the man made climate change brigade [Response: It was climate scientists — the ones now warning us of the danger — who *identified* the urban heat island effect, and it can be taken into account (NASA data does)]. Hence their models run hot and show a greater threat of global warming than there actually is [Response: It’s not so bad that you’re getting you “information” from idiots, the real problem is that you believe them.]

    What a farce! And the sad thing is that the man made climate change brigade see themselves as activists for good when their dogma is actively harming the poor. [Response: Interesting that you accuse us of your own faults. Interesting, but not surprising.] The religion of man made climate change is driving up energy prices and reducing the developing world’s ability to progress with technology and living standards. This will impact lives, unlike CO2, and it will be the responsibility of the man made climate change brigade. If you can live with that, then fine, but I can’t/

    [Response: Somehow, the news that you can’t *live* with it, fails to disquiet me.]

    • Philippe Chantreau

      What a pathetic load of nonsense. Not only you don’t understand what the OP is about, but each and everyone of your talking points has zero grounding in reality. You would know that if you had some familiarity with the state of the science but it is obvious that you prefer propaganda. How supremely ironic to mention the poor, when the intent is to defend the continued orgy of profit by the rich. Disgusting.

  10. Tamino has already rebuked numerous falsehoods by Mark W, but let me tackle this error, which may just possibly be an honest error:

    …you cite ‘unprecedented’ increases in global temperature in only five years as something we should concern ourselves is.

    Lousy proof-reading aside, the reading for comprehension is very poor. Tamino’s assertion is that the last 5 years are the warmest in the record. That’s not about an “increase… in only five years.” It’s about a five year period compared with the entire rest of the record.

    Big difference. Think about it.