Wednesday, December 24, 2014

Only temperatures, not temperature changes, may be dangerous

A Lumo Christmas playlist

Gavin Schmidt wrote a RealClimate.ORG blog post about the difference between the temperatures and temperature anomalies – or temperature changes – and which of them is known, predicted, and important.
Absolute temperatures and relative anomalies
Just to be sure, the temperature anomaly is the difference between the temperature and the "average" temperature recorded for the same place (or region) and the same date or month or season (if applicable). The average is computed (from the data at the same place and the same date[s]) over a period, like 1951-1980 or 1980-1999 or something like that, and this base line is often being changed which makes things confusing.

If the temperature were the changing to the same values every January and every February etc. (at a given location, or globally), the temperature anomalies would be equal to zero.

The global mean temperature is something like 14.5 °C or 15 °C. No one can really determine the value at this amazing, subdegree accuracy. Different methodologies – and indeed, different detailed definitions of the global mean temperature – produce different answers.

When we talk about the temperature anomalies, the absolute part of the temperature is subtracted, so the information about the anomalies only remembers the temperature changes. They are somewhat easier to be predicted from theories or models. And they are also easier to measure – and, for this reason, they are less ambiguous, too.

Gavin Schmidt is saying pretty much the same things. Well, he uses a different accent – because he tries to sell a defect as a virtue. If we could measure and/or predict the temperatures themselves and not just the changes/anomalies, it would be better. We really can't do it and the temperature changes/anomalies are easier to access.

However, there is one critical sentence that is highlighted in Anthony Watts' blog post by Bob Tisdale
Interesting Post at RealClimate about Modeled Absolute Global Surface Temperatures
Tisdale considers the following sentence by Schmidt to be the most "interesting" one:
To be clear, no particular absolute global temperature provides a risk to society, it is the change in temperature compared to what we’ve been used to that matters.
His polite wording makes it hard to become certain whether Tisdale agrees with Schmidt. I tend to say that he does.

Well, your humble correspondent disagrees with this assertion entirely.

Instead, if something about the temperature is dangerous, it is the temperature itself, and never its change.

For example, a naked human is doing fine between +5 °C and +35 °C or so. A moderately dressed one is doing well between –10 °C and +40 °C. Inside a certain house with heating and air-conditioning, the interval may be from –80 °C to +80 °C. Each organism, plant, animal species, ecosystem, or agricultural activity has its allowed temperature range. It's never quite sharp – the extreme temperatures start to be harmful gradually – and the "ideal temperature window" and the tolerance may depend on the stage of the life, and so on, and so on.

But if you can't formulate the restriction in terms of the absolute temperature, then it shows that the temperature change cannot be relevant!

Some WUWT readers think that by saying that it's the temperature change and not the temperature itself that is a risk to the society, Gavin Schmidt seems to deny the whole foundation of the climate change hysteria. But I totally disagree with this comment, too. Why?

The whole point of the climate change hysteria has always been the temperature change – that's why the word "change" appears in "climate change" (and, effectively, in "global warming"), too. The scientists ceased to be climate scientists and they were turned into climate change "scientists". And that made a big difference.

A popular "target" has been to prevent the temperature change exceeding 2 °C. Note that this alarmist slogan is formulated in terms of the temperature change – and it has to be. They just can't say what is the "right temperature". If we talk about the "ideal temperature", it also becomes clear that this is very different for different cultures and purposes, and the "ideal temperatures" will differ by dozens of degrees, proving that it is preposterous to discuss very small changes comparable to 1 °C.

In fact, one can't even define what was the "pre-industrial global mean temperature" – the benchmark with which we are supposed to compare the future temperatures. Ice ages were pre-industrial, too, and even if we look at recent centuries, their global mean temperatures have differed by pretty much the same amounts – temperature changes comparable to 0.6 °C or 1 °C – as the 20th or 21st century changes that are supposed to make us nervous.

Because of all these reasons, the climate change hysteria has always depended on this taboo that Schmidt confirms – namely that we never talk about the temperatures but only about their changes. And this fundamental tenet of this whole ideology is completely incompatible with the basic physics of all "dangerous" processes that are caused by the temperature.


To remind you of a trivial example, let's mention saunas, for example the Finnish saunas. You first spend some time in a hot room whose air has temperature 80 °C – 110 °C. No kidding, check the Wikipedia page. Within less than 10 seconds, you jump into a swimming pool where the water may be as cold as 10 °C. The temperature surrounding your body changes by 100 °C in 10 seconds. The rate, 10 °C per second, is something like one billion times faster than the "global warming" that we are supposed to be afraid of. For some reason, the Finns (and others) survive this ritual in their sauna.

(Similar comments apply to the reverse sauna. If you have never heard of it: you spend 15 minutes in the freezer before you jump into hot baking oil.)

The climate change alarmists such as Gavin Schmidt want to claim that the temperature change – effectively, the time derivative of the temperature – is what determines whether there is a risk for the society. I said that it is more accurate to say that exceeding some safe intervals of the temperature itself is what may be dangerous.

In fact, the truth stands on the opposite side of my proposition than Schmidt's claim. The truth is close to the anti-Schmidt comment. It is the integrated temperature over some (short) period of time that defines what is dangerous – and the integration is the opposite to Schmidt's differentiation. What do I mean?

Imagine that you are facing some wind and the air in the wind changes the temperature between –100 °C and +150 °C and back ten times a second. The rate of the temperature change will be huge. But I claim that this environment will actually be just fine for you despite the fact that the rates are huge (which is irrelevant, as I have already said) and the "safe" interval for the temperature is exceeded at almost every moment. Why? Because the air heats up or cools your skin so quickly that the layers of your skin just beneath the surface are already able to keep their temperature near the comfortable +25 °C.

So the very fact that the champions of the climate panic are forced to define the hypothetical risks in terms of the temperature changes or anomalies, and not the temperatures themselves, proves that they have no evidence for any risk because by the basic laws of physics, only a wrong value of the temperature (sustained over a sufficiently long period of time) may threaten humans, animals, other organisms, or ecosystems and economies.

If the discussion about the hypothetical "climate risks" is supposed to get meaningful, one must demand the claims to be formulated in terms of the actual temperatures rather than the temperature changes. The change itself, however fast, just cannot be dangerous if the temperatures stay within a safe window at all times, and people – and everyone else – are used to the temperature changes and lots of any other changes, anyway. In the context of biology, changes are also known as life.


  1. happy Christmas Lubos!!

  2. Happy Christmas, David - and everyone!

    Already in the rhythm of Christmas rituals and parties.

  3. BTW your point about the danger of temperature extremes being in the integral over time of temperature is spot on...deaths due to hypothermia, deaths due to heat waves...both depend on how much net total heat is taken out of a body, or put into it...its the accumulated loss, or gain of that energy that eventually kills...
    not the change in temps alone...

  4. Schmidt's reasoning seems to be something like the following:

    We don't know what the absolute temperature was in the 19th century, but we know it must have been "safe." Therefore our interest is not in what the absolute temperature value is for the "unsafe" threshold, but how close we've been to that threshold.

    Some flaws in this reasoning:

    The reasoning for what makes the 19th century temperature part of a "safe interval" and the definition for the threshold for "unsafe" temperatures relative to it, is asymmetric. In reality, applying the logic symmetrically, there is in fact some "optimal" temperature, and the temperature of the 19th century is *below* that level, making it as unsafe as some temperature above the "optimal" level.

    How far we are from the optimal temperature very much depends upon what the optimal temperature is, in absolute terms. We can't "anomalize" this value, without know the actual extant temperature of the actual Earth, in absolute terms. And the alleged optimal temperature is on the order of the uncertainty in the actual temperature itself. Put another way, if someone calculates that we are 2 degrees away from the optimal temperature, they do so on the basis of finding the optimal temperature to be something like 17 degrees Celsius, based on some analysis of some kind, and assuming the current temperature in 15. They don't do so on the basis of calculating that the optimal is 2 degrees higher than whatever the level may be, regardless of what it is. It is impossible to do it that way, because various impacts depend critically on the actual temperature level.

    Of course, for a hack like Schmidt, it is crucial to keep people from realizing such things, it might cause them to question their faith.

  5. The major reason to focus on the change is that the ice core records suggest that it happens quite abruptly. At least some papers suggest the transition to ice age took place in less than a decade.
    The current 'angels on the head of a pin' discussions whether 2014 is the warmest year by a couple of hundredth of a degree seems quite contrived in that context. Climate change is ongoing, the only legitimate concern is about large discontinuities, which current models don't seem to capture well if at all.

  6. How does your argument apply to sea level?

  7. John F. HultquistDec 24, 2014, 8:19:00 PM

    Merry Christmas

    from the dry side of the Cascades in eastern Washington State


  8. John F. HultquistDec 24, 2014, 8:32:00 PM


    You wrote: “and this base line is often being changed which makes things confusing”

    I think in 1935, in Prague, weather folks held a meeting and decided on a 30 year data set to describe “normal” for reporting weather comparisons. Every 10 years the data set is advanced so folks can relate to the recent period.
    The set begins with a year ending in 1 and ends with a year ending in 0 – example: 1981-2010
    Thus, we get phrases, such as “climate normals” and “a climatological pattern” for the next week.

    All this worked well before computers and satellite daily updates.

  9. I have always thought that the "temperature anomaly", as far that temperature is a intensive variable, is just a proxy for global warming (by e.g. its correlation with the average measured rate of IR loss to space), and therefore is confusing to mix the concept of temperature (and anomaly) with "the heating" (energy) without considering the nuances.

    There is a paper of Christopher Essex and Ross McKitrick extending in this direction, and some time ago I asked to Roy Spencer in Facebook, and he told me that it was correct but that they, literally, "In my [his] view, Chris is splitting hairs." :p

    Well, f*ck the climate :D I hope you Lubos and all the people around this blog have a Merry Christmas (or winter solstice, or whatever). Cheers!

  10. "Each organism, plant, animal species, ecosystem, or agricultural
    activity has its allowed temperature range. It's never quite sharp – the
    extreme temperatures start to be harmful gradually – and the "ideal
    temperature window" and the tolerance may depend on the stage of the
    life, and so on, and so on.

    But if you can't formulate the restriction in terms of the absolute temperature, then it shows that the temperature change cannot be relevant!"

    But you can formulate such restriction in terms of absolute temperature, its just different for every organism and every ecosystem - an absolute temperature range that produces deserts is different from one that produces taigas. This is why defining one absolute allowed temperature range for the whole Earth makes little sense and its far better to speak about allowed temperature change from present situation that would not significantly disrupt current ecosystems. That is the goal - minimizing disruptions to the ecological status quo.

  11. I can think of one reason why the time derivative of temperature would be economically relevant: adaptation responses which involve modifying or replacing capital goods such as buildings are less burdensome when the expenditure is spread over longer time periods. But of course time derivatives are not the same as simple differences.

  12. I think the number 30 originated with statistics.
    With random samples of 30 or more, you can use the normal distribution for significance tests at the 5% level rather than small sample "T" tests. I don't think 30 years of weather fits that, because each new year is not an independent random sample. Weather patterns, like heavy rain and drought, tend to run in cycles rather than completely random.

  13. Merry Christmas, Luboš, and to all funster visitors to The Reference Frame.

  14. John F. HultquistDec 25, 2014, 4:13:00 AM

    I wrote Prague previously. It was
    Warsaw. Sorry.

    At the 1935 discussions [See Sec. 3 in the link below], one of the goals considered was to have a city's
    “normal” weather records be something that a person reading the
    newspaper could relate to. This idea is encompassed in these

    “(b) Having a set of normals that is
    as up-to-date as possible, in order to maximize the perceived
    relevance of those normals in the community;”


    “(f) Having an averaging period that
    the general public can relate to and appears ‘logical’; ”

    from Section 4

    This was done before computers and before “climate science” of the AGW type. Since then the
    “statistics” of the situation have been studied [Sec. 5 – of the link].

    Note the use of the word community and general public in (b) & (f).

  15. Hi! I think it doesn't apply. Rising sea levels simply devour previously dry places.

    Note that there isn't even a standardized measure of the "absolute" sea level. One would have to measure the accurate water-bounded radius of the Earth at a meter accuracy, or something like that.

  16. Merry Christmas

    from Lough Erne, Ireland

  17. happy dog

  18. And in addition, an article on rising sea leve in the Southwest Pacific posted on Anthony Watts' site made me realize that atmospheric pressure ALSO affects sea level. A change from high pressure to low pressure in an area results in higher sea levels, a change from lower to higher pressure results in lower local sea levels.

  19. Ben Santer, of the "if you don't agree with me, I will beat you
    up" style of reasoning, has a paper in Nature showing that
    volcanoes' spewing of particulates is moderating the warming trend far more than was thought...

  20. "Parties"?--for us introverts, that is a
    neologism :)
    Happy Christmas, everyone (even the
    believers ;) ) I am presently lying on the couch watching the fire, with the tree behind me and listening to the King's College Choir sing Handel's Messiah and smelling the cooking turkey--
    not bad cognitive dissonance for an atheist/agnostic, eh?

  21. For a non-human species, a small change in temperature can be very significant because it can give it a competitive advantage or disadvantage relative to other species, and that will great change its position in the local ecosystem. The result is that there will likely be less diversity for a time. Humans can adapt much more quickly, and so a wide range of temperatures are perfectly fine, with the inconvenience of local shoreline changes etc.

  22. And he's happy for he sees turtles all the way down with "the compass always pointing to Terrapin." Terrapin Station has gotta be one of the greatest rock songs ever written. Which makes it a crying shame that Rolling Stone has never listed it among the all-time greats:

  23. It's true that the absolute temperature is what matters. However its also true that humans have roughly optimized settlement patterns, construction methods etc to particular absolute temperatures. So while there may be no one optimal absolute temperature it's likely that changing temperature will invoke costs in aggregate. Doesn't mean that changing temps cant be adapted too, but your prior has to be it will have some cost.

  24. And before that he was Ben Santer of "arbitrary truncation of data for trend comparison" fame.

    One has to wonder how much of this is self deception, and how much he's bullshitting people who have no clue how impossible it is that volcanic eruptions have had enough of an effect.

  25. Merry Christmas.

  26. Too bad that you hadn't TRF Christmas thread... Therefore through this thread, Merry Christmas! :-)

  27. Dear Aleksandar, being aware that the probability is nonzero that someone is a genius writing papers misunderstood by the rest of the mankind, I still think it is not kosher to use the TRF thread to promote one's own preprints with bombshell claims like yours that, given the zero feedback by the physics community, seem to be just crackpot tirades.

    Could you please avoid promoting your preprints in the future? Thanks.

  28. Happy to Christmas to the christians on TRF.

  29. Aleksandar MikovicDec 26, 2014, 10:34:00 AM

    Dear Lubos,
    Yes, I agree that one should not use blogs to promote his own work. In this case, I just wanted to point out that there are alternative approaches to QG.

  30. Aleksandar MikovicDec 26, 2014, 11:40:00 AM

    Yes, there are new developments as far as the Regge QG is concerned, and these are Casual Dynamical Triangulations of Ambjorn, Loll and Jurkiewitz and the effective action approach by Mikovic and Vojinovic.
    As far as the BH entropy calculation is concerned, it can be done, and I expect that it will give the correct result.

  31. I am skeptical, Gordon. The shape of the temperature trend curves in Santer et. al.’s Fig. 4 doesn’t make sense to me. For instance, how would you explain the extremely rapid run-up leading to Pinatubo? The only explanation I can think of is that these people are trying to rationalize the myriad of computer models.
    It is easy to fool one’s self, particularly if one is not trying very hard.

  32. It’s about 70/30.

  33. Your last sentence reminds me of Peter Medawar's devastating review of Teilhard de Chardin's book, "The Phenomenon of Man"--
    "Yet the greater part of it, I shall show, is nonsense, tricked out with a variety of metaphysical conceits, and its
    author can be excused of dishonesty only on the grounds that before
    deceiving others he has taken great pains to deceive himself." wikiquotes

  34. dont include jesus in your holiday wishes?

  35. Hi Lubos, I know you think this warp drive thing is just bullshit, but I looked at the warpdrive 101 paper and Harold uses a higher dimensional brane model with extra dimensions and so on, I find that really interesting, maybe you could go into it in a blog post or something? Happy Christmas :)

  36. Dear Guy, happy (post)Christmas now to you, and others.

    Extra dimensions - compact or warped or otherwise - can't produce an exception from the speed-of-light limit (e.g. a new type of war drives) because the Lorentz symmetry SO(3,1) must still be a symmetry, perhaps as a subgroup of a larger symmetry group, and this symmetry is enough for all the Einstein's arguments to proceed.

  37. I think he assumes Jesus is long gone visiting other planets. Maybe some future interferometer super telescope catches a glimpse of him which is why we should support all such scientific endeavors.

  38. The hardest part of any such proof, I think, is defining what a theory must look like to qualify as QG in a sufficiently broad way. Lots of elements of the GR classical limit are well tested experimentally, but it is hard to know which of those that are not well tested actually hold in reality. If one insists that all of classical GR is matched, the limit is probably too strict.

  39. "she" assumes... :)
    We, Christians, are not "waiting for Him" or trying to" catch a glimpse of Him". He already came with a message 2014 years ago. Some other people became the jews we know today because they refused Him and try to kill Jesus. They failed, and Jesus gave them the finger by resuscitating. Cool. So now this new anti-Christ religion called judaism, the jews, are wandering and waiting for another One. Someone like Neo in the Matrix, an asshole with no personality.