Why Do Different Satellite Datasets Produce Different Global Temperature Trends?

January 6th, 2015 by Roy W. Spencer, Ph. D.

I thought it would be useful to again outline the basic reasons why different satellite global temperature datasets (say, UAH and RSS) produce somewhat different temperature trends.

They all stem from the fact that there is not a single satellite which has been operating continuously, in a stable orbit, measuring a constant layer of the atmosphere, at the same local time every day, with no instrumental calibration drifts.

Instead, what we have is multiple satellites (we use 14 of them for the UAH processing) with relatively short lifetimes (2 to 16+ years), most of which have decaying orbits which causes the local time of measurement to slowly change over the years, slightly different layers sampled by the earlier (pre-1998) MSU instruments compared to the later (post-1998) AMSU instruments, and some evidence of small calibration drifts in a few of the instruments.

An additional complication is that subsequent satellites are launched into alternating sun-synchronous orbit times, nominally 1:30 a.m. and p.m., then 7:30 a.m. and p.m., then back to 1:30 a.m. and p.m., etc. Furthermore, as the instruments scan across the Earth, the altitude in the atmosphere that is sampled changes as the Earth incidence angle of view changes.

All of these effects must be accounted for, and there is no demonstrably “best” method to handle any of them. For example, RSS uses a climate model to correct for the changing time of day the observations are made (the so-called diurnal drift problem), while we use an empirical approach. This correction is particularly difficult because it varies with geographic location, time of year, terrain altitude, etc. RSS does not use exactly the same satellites as we do, nor do they use the same formula for computing a lower tropospheric (“LT”) layer temperature from the different view angles of AMSU channel 5.

We have been working hard on producing our new Version 6 dataset, revamping virtually all of the processing steps, and it has taken much longer than expected. We have learned a lot over the years, but with only 2-3 people working part time with very little funding, progress is slow.

In just the last month, we have had what amounts to a paradigm shift on how to analyze the data. We are very hopeful that the resulting dataset will be demonstrably better than our current version. Only time will tell.


101 Responses to “Why Do Different Satellite Datasets Produce Different Global Temperature Trends?”

Toggle Trackbacks

  1. Mike O says:

    Hi Roy,
    Thanks for all the hard work.
    Is there a known tolerance for each of the datasets?

    Mike O.

  2. Patrick B says:

    What is the margin of error attributable to these sources of error?

    Giving effect to these margins of error, what is the proper ranking of the years? (I’m guessing there will be quite a few years with overlap once a proper margin of error is added to the measurements – thus numerous ties for various ranks.)

    Thank you.

  3. Margin of error depends on what space and time scales you are asking about (monthly global anomaly? yearly gridpoint anomaly? decadal global trend? etc.) Given the margin of error, 2014 global at 3rd warmest year is basically tied with the 4th and 5th place years. See my previous post on the Dec. 2014 temperature update: http://www.drroyspencer.com/2015/01/uah-global-temperature-update-for-december-2014-0-32-deg-c/

    • David Appell says:

      What is the margin of error (1-sigma) on the monthly global anomalies?
      On the yearly global anomaly>

    • David Appell says:

      Roy, what is the margin of error (1-sigma) on your published monthly global anomalies?

      For the yearly global anomaly?

      Surely you have calculated these.

    • paul says:

      roy , could we put a detector on the moon to measure the temp on earth, what calibrations would have to be done

  4. Håkan says:

    When will the v.6 be launched?

  5. Ray says:

    I am surprised that you don’t mention the lower coverage of Antarctica by RSS, versus UAH.

    • Fonzarelli says:

      Don’t be surprised… I’m sure doctor spencer has “a lot on his plate”.

    • David Appell says:

      You’re right — RSS is missing 6% of the south polar region; UAH about 1%.

      • RSS TLT omits everything within 20 degrees of the south pole, which is about 6% of the entire southern hemisphere. The UAH pole holes are limited to within 5 degrees of the poles, which means about .4% of each hemisphere.

  6. Doug Cotton    says:

     

    Roy

    What is causing temperature trends is explained in this “must read” climate change website with breakthrough new science: http://whyitsnotco2.com

     

    • JohnKl says:

      Good read, well stated. We’ve agreed for some time in regard to emission cooling of tri-atomic gas compounds. Interestingly, the alarmists at RSS readily admit tri-atomic gas compounds cool the stratosphere (except for ozone strangely enough), flip 180 degrees claim it heats the lower troposphere then spend considerable time explaining why their models reflect far too much warming in the lower troposphere. They even list a considerable number of potential explanations for the lack of warming like aerosols all in an effort to claim they’re physics remain valid.

      http://www.remss.com/research/climate

      Cognitive dissonance anyone?

      Have a great day!

      • There is a reason why ozone heats the stratosphere: It absorbs a certain range of UV.

        • JohnKl says:

          Correct! Thank you.

          Have a great day!

        • Kristian says:

          It is a common misconception (apparently (and deliberately?) promoted by NASA/GISS) that ozone warms the stratosphere by acting like an H2O/CO2 type ‘greenhouse gas’, absorbing and somehow ‘retaining’ the radiative (UV) energy from the Sun as ‘heat’.

          For instance here: http://www.giss.nasa.gov/research/features/200402_tango/

          “Ozone’s impact on climate consists primarily of changes in temperature. The more ozone in a given parcel of air, the more heat it retains. Ozone generates heat in the stratosphere, both by absorbing the sun’s ultraviolet radiation and by absorbing upwelling infrared radiation from the lower atmosphere (troposphere). Consequently, decreased ozone in the stratosphere results in lower temperatures.”

          But this is plain silly. It is the formation/destruction cycle of ozone molecules that turns the ‘dangerous’ high-energy UV radiation into kinetic energy freed into the air column, a bit like when water vapour condenses in the troposphere and turning latent heat of vaporisation from the surface into kinetic energy freed into the air column.

          The UV is simply the energetic catalyst driving the cycle.

          http://en.wikipedia.org/wiki/Ozone-oxygen_cycle

          • JohnKl says:

            UV heating of ozone in the stratosphere doesn’t seem crazy to me at all even if it appears “strange” because of relative rates of energy flow. Ozone absorbs in the UV & radiates in the infrared which has much lower energy longer wavelength radiation. This allows for the accumulation of energy in any given ozone molecule during daylight hours when UV radiation is present. Certainly much of the energy will be shared with other molecules warming the stratosphere. The case differs with CO2 which absorbs and emits largely in the infrared during the course of the day making such accumulation extremely unlikely if not impossible whether in the lower troposphere or the stratosphere!

            Have a great day!

          • Kristian says:

            JohnKl says, January 7, 2015 at 3:04 PM:

            “UV heating of ozone in the stratosphere doesn’t seem crazy to me at all even if it appears “strange” because of relative rates of energy flow. Ozone absorbs in the UV & radiates in the infrared which has much lower energy longer wavelength radiation. This allows for the accumulation of energy in any given ozone molecule during daylight hours when UV radiation is present.”

            You apparently didn’t read what I wrote just above. Or the link to the ozone-oxygen (Chapman) cycle.

            The UV doesn’t ‘heat’ the ozone molecule. The ozone molecule doesn’t absorb UV to reemit it as IR. There is no accumulation of energy in any ozone molecule from the absorption of UV.

            THIS is the misconception.

            The UV breaks the ozone molecule. The ozone molecule does absorb the UV, but is not ‘heated’ by it. It is destroyed by it, the process waste product being kinetic energy. None of the energy is retained by the ozone molecule. The energy is freed. Passed on. The ozone molecule itself does not survive the UV absorption.

            So, yes, stratospheric ozone absorbing UV radiation from the Sun provides (kinetic) energy warming the stratospheric air (which is still almost all N2 and O2), but NOT in the way most people seem to think, which is in the ‘traditional’ GHG way (as promoted by NASA/GISS).

            You should read about the Chapman cycle. A quick summary is (once again) given here:

            http://en.wikipedia.org/wiki/Ozone-oxygen_cycle

          • Bert Walker says:

            Kristian, You make a distinction between Heat and kinetic energy. What is the difference in your opinion, between “kinetic energy” and “heat”?
            I ask because I was trained to understand heat, or thermal energy is a measure of the kinetic energy of a gas. And that gas temperature is a measure of the average kinetic energy of a gas at a given pressure and volume. So heat and kinetic energy are interchangeable when pertaining to a gas, or mixture of gasses.

          • Kristian says:

            Bert Walker says, January 7, 2015 at 11:19 PM:

            “Kristian, You make a distinction between Heat and kinetic energy. What is the difference in your opinion, between “kinetic energy” and “heat”?”

            Not really. I make a distinction between IR and kinetic energy. The common misunderstanding is that ozone warms the stratosphere by absorbing UV and reemit the energy as IR. It does NOT. Read again what I write.

            “I ask because I was trained to understand heat, or thermal energy is a measure of the kinetic energy of a gas. And that gas temperature is a measure of the average kinetic energy of a gas at a given pressure and volume. So heat and kinetic energy are interchangeable when pertaining to a gas, or mixture of gasses.”

            Again, not really. You mix up ‘heat’ with ‘internal energy’:

            http://hyperphysics.phy-astr.gsu.edu/hbase/thermo/heat.html
            http://hyperphysics.phy-astr.gsu.edu/hbase/thermo/inteng.html

          • JohnKl says:

            Hi Kristian,

            Thank you for the reply. Given your admonition to read the Wikipedia link (a real fount of scientific information) I spent a little time there. It appears you may not have read carefully enough the linked information there when you claimed:

            “The ozone molecule does absorb the UV, but is not ‘heated’ by it. It is destroyed by it, the process waste product being kinetic energy. None of the energy is retained by the ozone molecule. The energy is freed. Passed on. The ozone molecule itself does not survive the UV absorption.”

            Indeed, according to the link the ozone molecule doesn’t survive the interaction with UV. However, at the end of the process there exists NO REDUCTION IN OZONE!!! No change in atmospheric composition results. Which begs the question – HOW CAN THIS PROCESS BE VALIDATED? During the ozone-oxygen cycle the ozone molecule gets broken down to O2 and a free oxygen atom (O). The free oxygen atom the links up again with an oxygen molecule (O2) and makes (you guessed it) ozone!

            “2. The ozone-oxygen cycle: the ozone molecules formed by the reaction above absorb radiation having an appropriate wavelength between UV-B and the very top end of UV-A. The triatomic ozone molecule becomes diatomic molecular oxygen plus a free oxygen atom (see figure):
            O3 + ℎν(240-310 nm) → O2 + O
            The atomic oxygen produced quickly reacts with another oxygen molecule to reform ozone:
            O + O2 → O3 + K.E.
            where “K.E.” denotes the excess energy of the reaction which is manifested as extra kinetic energy. These two reactions form the ozone-oxygen cycle, in which the chemical energy released when O and O2 combine is converted into kinetic energy of molecular motion. The overall effect is to convert penetrating UV-B light into heat, without any net loss of ozone. This cycle keeps the ozone layer in a stable balance while protecting the lower atmosphere from UV radiation, which is harmful to most living beings. It is also one of two major sources of heat in the stratosphere (the other being the kinetic energy released when O2 is photolyzed into O atoms.”

            Given that I’m incapable of directly witnessing chemical reactions at the atomic level I do not claim this process to thermalize the supposedly destructive UV impact on ozone molecules cannot take place but this process seems very theoretical to me. It would be interesting to discover what empirical (most likely lab experiments) information supposedly exists that this process even takes place.

            Thanks again and…

            Have a great day!

          • JohnKl says:

            Hi Kristian,

            Please excuse me but you also mentioned:

            “The UV doesn’t ‘heat’ the ozone molecule. The ozone molecule doesn’t absorb UV to reemit it as IR. There is no accumulation of energy in any ozone molecule from the absorption of UV.”

            This statement appears empirically FALSE. Please note that the UV band width that generates the ozone-oxygen cycle falls between (240-310 nm) as the link you provided states (see my quote from earlier post). However, the absorption spectrum of ozone in the UV region runs a continuum from 200 to 350 with a maximum near 250 nm (Hartley band). Please see my link here:

            http://igaco-o3.fmi.fi/ACSO/files/daumont_et_al_1992.pdf

            Therefore, clearly ozone will absorb UV in some bandwidths and thus warm without being destroyed and converted to oxygen (O2) and a free oxygen atom (O). MY STATEMENT WHICH YOU CLAIMED TO BE FALSE PROVES TO BE TRUE!

            Have a great day!

          • Kristian says:

            JohnKL,

            You say: “It appears you may not have read carefully enough the linked information there when you claimed:

            “The ozone molecule does absorb the UV, but is not ‘heated’ by it. It is destroyed by it, the process waste product being kinetic energy. None of the energy is retained by the ozone molecule. The energy is freed. Passed on. The ozone molecule itself does not survive the UV absorption.”

            Indeed, according to the link the ozone molecule doesn’t survive the interaction with UV. However, at the end of the process there exists NO REDUCTION IN OZONE!!!”

            I’m not sure what you’re trying to say here. Did I say anything to the effect that ozone at large is reduced by this process?

            Is it so that you doubt that the Chapman cycle in fact occurs?

            You also say: “Therefore, clearly ozone will absorb UV in some bandwidths and thus warm without being destroyed and converted to oxygen (O2) and a free oxygen atom (O). MY STATEMENT WHICH YOU CLAIMED TO BE FALSE PROVES TO BE TRUE!”

            Sorry, but no. The whole reason why the ozone molecules are destroyed when absorbing UV is that the UV is too energetic to be retained by the molecules; it rather breaks the bond between two of the oxygen atoms to create one O atom and one O2 molecule.

            Either the ozone molecule absorbs the UV and is destroyed or it doesn’t absorb it at all (like with most of the UV-A).

            https://www.youtube.com/watch?v=WE3y1Gj2dec

          • JohnKl says:

            Hi Kristian,

            Thank you for the reply. You asked:

            “I’m not sure what you’re trying to say here. Did I say anything to the effect that ozone at large is reduced by this process?

            Is it so that you doubt that the Chapman cycle in fact occurs?”

            No to the first question and as to the 2nd question I simply question how the process came to be known to occur at all. We know hydrogen and oxygen combine to form water because if one heats hydrogen in the presence of oxygen water results. If you never had a chemistry class you might not know. Humphrey-Davey along with supposedly discovering gravy (laughs optional) ran electricity through various salts including magnesium salts to discover magnesium which isn’t found to occur in elemental form in nature. We can only have a concept of time if some change occurs since all change occurs in time. Since the ozone-oxygen (Chapman) cycle results in no identifiable change in gas composition how do you know the process takes place? Please note I’m not claiming it doesn’t occur simply questioning the extent of it’s action and how one arrives at the conclusion it in-fact occurs and to what extent.

            Further, since the ozone-oxygen (Chapman) cycle maintains the presence of ozone in the atmosphere as you readily admit then apparently it can reside sometime in the atmosphere absorbing UV without disintegrating. If not then ozone should not be detectable at all in the presence of solar radiation which contains plenty of UV. As my last post indicated, only a portion of the UV absorbed by ozone causes the ozone to disintegrate leaving the remaining portion free to accumulate energy within the molecule and thus WARM it and surrounding molecules. So your following claim proves counter-factual:

            “The whole reason why the ozone molecules are destroyed when absorbing UV is that the UV is too energetic to be retained by the molecules; it rather breaks the bond between two of the oxygen atoms to create one O atom and one O2 molecule.”

            As I mentioned before, the absorption spectrum of ozone in the UV region runs a continuum from 200 to 350 with a maximum near 250 nm (Hartley band). Your own link claimed that only a substantially smaller portion of the (240-310 nm) UV bandwidth causes ozone to disintegrate. Thus the information you provided proves the falsity of your statement above and the following statement:

            “Either the ozone molecule absorbs the UV and is destroyed or it doesn’t absorb it at all (like with most of the UV-A)..”

            No, ozone will absorb UV wavelenghths from 200-<240 nm in particular and apparently from 310-350 nm without disintegrating the ozone molecule. If you have doubts, question your own data since the 240-310 nm bandwidth came from YOU!!! Since the UV bandwidth varies from 400nm to 10nm there exists plenty of UV that doesn't get absorbed at all as you suggest, and I never claimed otherwise.

            Have a great day!

          • JohnKl says:

            Hi Kristian,

            Please accept the following correction to my last post. I made the statement:

            “We know hydrogen and oxygen combine to form water because if one heats hydrogen in the presence of oxygen water results.”

            Replace the word “heats” with ignites or detonates or something similar. They will work better.

            Have a great day!

          • JohnKl says:

            310-350 nm should read >310-350nm.

    • Doug,

      Your narrative seems to be coming closer to mine as time passes.

      However, there is still one critical difference.

      One does not need to add additional energy at the top that then diffuses down to warm the surface.

      All one needs is for a portion of the heat taken up and away from the surface in convective uplift to fail to be radiated to space from within the atmosphere and of course gravitational potential energy (GPE) is not heat and does not radiate and so no energy in GPE form can radiate to space.

      Then, on descent, that GPE returns to heat (KE) at the dry adiabatic lapse rate during the descent and that warming on descent dissipates any clouds and reduces convection from the surface beneath.

      It is that reduction of clouds (letting more sunshine hit the surface) AND the restraining of convection (by reducing the lapse rate gradient) that allows the average temperature of the entire Earth to rise 33C above the purely radiative S-B calculation.

      You don’t need any downward diffusion, only dissipation of clouds and restraint of convection for the surface to become warmer than S-B.

      The longer that energy is tied up in the convective overturning cycle as GPE the higher the surface temperature will become relative to the S-B calculation.

      The more mass there is in the atmosphere the longer energy will be tied up in the convective cycle as GPE and the more energy will be so tied up.

      If one then adds GHGs they assist cooling via radiation to space from within the atmosphere but that then REDUCES convective overturning which in turn reduces radiation from the surface direct to space by exactly the same amount as radiation to space from within the atmosphere for a zero net effect on surface temperature.

      We appear to be in agreement except for your idea that downward diffusion is necessary.

      • Doug Cotton says:

        My response is in the seventh paragraph at whyitsnotco2.com.

        Give a little thought to Venus.

      • Doug Cotton says:

        The critique of the Hockey Schtick articles (right hand column here) also applies to your conjecture.

      • Doug Cotton says:

        And Stephen you wrote here “the speed of the hydrological cycle changes to a miniscule extent in order to maintain sea surface and surface air temperature equilibrium.”

        Well valid physics tells us conduction across the interface between the sea surface and the atmosphere will even out any temperature difference. That’s all.

        Valid physics tells us that some of the solar radiation gets down into the cooler regions below the ocean surface, and so we can’t count that energy as if it were all absorbed in the thin almost transparent surface layer. Most likely it just makes the lower layers less cold, but not as warm as the surface, so there’s no heat transfer back up to the surface, especially in the tropics. The energy absorbed beneath the surface in the tropics probably follows isotherms towards the poles. So the Sun’s direct radiation doesn’t have a hope of raising the surface water temperature to what is observed. Work it out.

        Valid physics tells us that the whole “Earth+atmosphere” system will act as a black body and remain in radiative balance with the Sun. It doesn’t need air mixing by hydrological cycles or wind of any form. Such mixing disturbs thermodynamic equilibrium and it is only restored in calm conditions. But mixing does not affect radiative balance significantly.

        There’s no equivalent hydrological cycles keeping Venus temperatures constant. A location on its equator cools by about 5 degrees during its 4 month long night, and warms back up on the sunlit side. How does the required energy get there Stephen, as I have asked you many times? It is not by direct solar radiation, but it is from solar radiation that has raised the temperature in regions in the upper troposphere (and above) where temperatures are under about 400K. The surface temperature at any particular location on the equator rises from 732K to 737K over the course of the 4-month-long Venus day. That needs net thermal energy input.

        Unless your hypothesis functions correctly universally for all planets it is proven wrong. It does not even function on Earth because solar radiation into the surface has a mean of 168W/m^2 or less. Work it out.

        Besides, how are you going to stop the Second Law of Thermodynamics functioning and carrying out the process described in my hypothesis?

      • Douig.Cotton  says:

        Kristian has written a critique of Stephen Wilde’s conjecture here. I quote in part …

        “Yeah, all you need to do, Stephen, is actually read the links I provide and you will see that you are clearly the confused one.

        “In fact, everyone should read those links and realise how deluded the ideas promoted by Stephen Wilde from inside his little bubble of self-invented nonsense physics really are, where, however clear and basic the evidence presented is of him being wrong, he is still right, because … he says so.”

  7. Locke says:

    Oh dear Roy, what a mess you have created for yourself. Maybe you should speak with the Texas Public Policy Foundation to help you with your funding?

  8. Thanks for the progress update on Version 6.
    Godspeed!

  9. geran says:

    “They all stem from the fact that there is not a single satellite which has been operating continuously, in a stable orbit, measuring a constant layer of the atmosphere, at the same local time every day, with no instrumental calibration drifts.”
    **************

    Thank you for your honesty. No one can claim to be a “scientist” if they do not revere honesty.

    Even with all our “advanced” technology, we do not know the temperature of our planet at this moment. And, guess what, we do not know the temperature of the planet 50 years ago.

    Go figure.

    • JohnKl says:

      “Even with all our “advanced” technology, we do not know the temperature of our planet at this moment.”

      True. Roy states:

      “there is not a single satellite which has been operating continuously, in a stable orbit, measuring a constant layer of the atmosphere, at the same local time every day, with no instrumental calibration drifts.”

      Considering that many satellites must move in and out of the desired orbital path and altitude to provide our DATA why would anyone want to make political decisions regarding the lives and property of millions based on a presumed warming anomaly of ~2-3 tenths of a degree centigrade over a period of ~37 years? Does anyone really believe without all the activity mentioned in Roy’s column involved with providing this DATA that the margin of error would only be .1 deg centigrade? Just asking…

  10. Roy,

    RSS US temperatures match raw surface data quite closely, but UAH doesn’t. That makes me believe that the RSS approach is more accurate.

    • geran says:

      Is this the same “Steve Goddard” that believes “CO2 produces warming”?

      LMAO!

    • David Appell says:

      Calculations done without correcting past raw temperatures for bias don’t represent anything physical.

      • But the devil is in the details. It’s very easy to mess up adjustments. Much easier than to get the right answer. That’s why it’s useful that independent data sets show some degree of agreement.

      • David Appell says:

        There are no “independent” surface data — they all come from the same source.

        Just because “it’s very easy to mess up adjustments” does mean they are, in fact, messed up. BEST went through all of this — they found the same results as everyone else, yet it doesn’t seem to matter one iota.

        • Why assume ‘surface data’ ?

          Why not compare sat data trends to surface or weather balloon trends, for example. Once again you want to ‘correct me’ but then end up writing something silly in your struggle to rebut anyone who disagrees with your thought bubbles.

  11. David Appell says:

    Dr. Spencer: Why shouldn’t these difficult issues with satellites, which require adjustments via your data model, be considered in the same light as the homogenization techniques for the adjustments of surface temperatures? Some posts ago you said that made surface datasets dubious…but why are these corrections any different?

    • Roy Spencer says:

      Good question. The nature of the errors are, of course, different. But when the thermometer folks make adjustments, they are dealing in whole degrees, or tenths of a degree. We instead obsess over hundredths of a degree.

      But it is true that both data sources involve adjustments as large as the global warming signal we are seeking.

      • rooter says:

        That is just plain wrong. Adjustments of thermometer-records are much smaller than the global warming signal.

        http://www.moyhu.blogspot.com.au/2014/07/ghcn-adjustments-are-much-larger-in-us.html

        • “That is just plain wrong.”

          I think I’ll stick with scientists with a proven track record rather than defer to the claims made in anonymous blog postings… But obviously people look around for confirmation of what they already believe.

          • rooter says:

            Well. Your confirmation bias has led you to one source that has no track record concerning adjustments of land based temperature records. No publications.

            Ps: Nick Stokes is just as anonymous as Roy Spencer. For land based temperature records Stokes has made more investigations than Spencer.

          • David Appell says:

            HOw about scientists like BEST, which has Nobel Laureaute on its team?

          • Aaron S says:

            David,
            It amazes me that in some augmented perception of reality you consider blogging equivalent to science. Its a classic wanna be move, again. Bloggers are not scientists and this is not to say there can not be good ideas by people that blog, but you can not seriously cite a blog with a straight face. Please inform me who this nobel laureate is relative to the citation you cite and what role they play? Once you get out away from the primary literature (which is itself biased but remains the best version of science available), then the credibility rapidly approaches zero if for no other reason then because it is not repeatable bc the method is not documented.

          • Well anonymous bloggers and appeals to authority are all they have to support their crankism, that’s why they always go down that path.

      • Fonzarelli says:

        So then, would you say that UAH is (for all practical purposes) accurate given that we’re only talking about a few hundredths of a degree?

      • RW says:

        Roy Spencer says:

        “But it is true that both data sources involve adjustments as large as the global warming signal we are seeking”

        Which is exactly why I take these sources — all of them — with only a grain of salt.

    • Frank says:

      David: UAH and RSS make sensible efforts to correct problems in their data with known causes: changing satellites, orbital decay, etc.

      Those who homogenize station temperature data are correcting apparent artifacts with an undocumented cause. A breakpoint can be caused by station maintenance that restores early observing conditions that gradually deteriorated. Such breakpoints should not be corrected. A breakpoint can also be caused by a station being moved to a new location. One might argue that such breakpoints should be corrected, but suppose the station move was prompted by gradual urbanization and the station was moved to a nearby park. Any breakpoint caused by restoring earlier observing conditions shouldn’t be corrected. There are undoubtably many breakpoints caused by a change to new observing conditions: change in time of observation, equipment changes, and moves to different observing conditions. The corrections UAH and RSS have a scientific basis and are carefully tested. Station homogenization is pure guesswork.

      • David Appell says:

        Frank: Of course such “breakpoints” need to be corrected for. And there is a scientific need for this, for which a great amount of study has taken place. Just because a station is moved to a “nearby park” doesn’t make it exactly the same as its previous locations.

        There is no way to know if breakpoints “restore earlier observing conditions” — that requires calculation, and can’t be blithely assumed.

  12. thefordprefect says:

    So you are saying that all satellite temperatures are models with a few bodges added?

    • David A says:

      “So you are saying that all satellite temperatures are models with a few bodges added?”

      Of course. The satellites aren’t measuring temperatures, they’re measuring microwaves. It takes a data model to convert those into temperatures…. He’s a description of RSS’s algorithm; it’s quite complex:

      “Climate Algorithm Theoretical Basis Document (C-ATBD)”
      RSS Version 3.3 MSU/AMSU-A Mean Layer Atmospheric Temperature
      http://images.remss.com/papers/msu/MSU_AMSU_C-ATBD.pdf

      It’s not clear to me why satellite temperatures are said to be clean and exact while surface measurements are not….

      • Objectivist says:

        @David A.

        Well, two obvious reasons “why satellite temperatures are said to be clean and exact while surface measurements are not” are:

        1) Satellites directly integrate empirical measurements for the entire surface of the globe, minus a small percentage at the poles. Surface measurement techniques have to fill in the enormous gaps where there are no actual measurements, with dubious results. You might want to check out the number of ground stations at the poles, for instance.

        2) Also, satellites are measuring temperature using a consistent set of laboratory-grade equipment versus the large range of quality for ground-based instrumentation and siting. The satellite instruments are calibrated regularly using lab grade technique, and the measurements are also verified and adjusted using balloon soundings etc.

        I hope that makes things clearer for you…

        • Franco says:

          As a thermal metrologist now for 40 years, I like to say that there is a third reason: satellites perform temperature measurements by radiometry. This technique is very sensitive to the effect of the local composition of the atmospheric column between the satellite and the ground, with a large and uncertain corrections. I am quite convinced that, at the present state-of-the-art, the uncertainty attached to the global value of +- 0.12 °C is by far under-evaluated (to the instrumental one also the errors of the model must then be added).
          Only recently we might consider to have a comparability (traceability in metrological jargon) of ground temperatures better than, say, 0,2 °C, but this uncertainty is rapidly increasing going back in time. Ambient temperature is one of the most difficult quantities to assign a reasonable correct value, even locally, even in good science.
          Only relatively recently a correlation has been tried to correct for the mismatch between satellite and ground measurements. A good European Project has been performed in the last three years on the subject of thermal metrology for metereology (http://www.meteomet.org).
          This is not to say that I do not believe in some present warming, but this is certainly not the first time for the Earth to happen. I do not believe on the prevailing anthropogenic effect due to burning fossils for energy.
          Has anybody ever tried to compute the contribution to CO2 increase due to the human respiration and to respiration of the animals grown for human food? From my first computations is huge. Do you believe that the correlation between the global warning curve and that of the human population increase in the last 30 years may be true?

        • David Appell says:

          Obj:
          1) Surface measurements have to adjust for various factors. Satellite measurements have to correct for others.
          2) Werong: As Roy has apready written, corrections need to be made as measurements go from one satellite to another. Sensors also degrade and must be corrected for heat, etc. That’s hardly a “consistent set of laboratory-grade equipment.”

          Frankly, it seems to me that measuring by surface sensors is less complex than by satellite…. Cowtan and Way have a more sophisticated method of infilling…and in any case GISS, HadCRUT4, etc aren’t measuring the temperature of the surface, they’re measuring the temperatue *changes* in their model of the surface. So, in that respect, is UAH and RSS. This is one reason why the publish anomalies and not absolute temperatures.

          I don’t think either of your points are convincing.

          • Cranks rubbish the sat’s because they don’t show the warming that they wish to be there. Cranks only reference sats when they talk about sea level rise because it’s nearly double what the land measurements tell us.

      • “The satellites aren’t measuring temperatures, they’re measuring microwaves.”

        That statement has got to win awards for stupid. Traditional thermometers don’t measure temperature either. They measure the expansion and contraction of mercury in a glass tube…

        • Franco says:

          uh ?!
          What would it measure temperature?

          • dave says:

            Franco says:

            “uh?!”

            What Will Nitschke means is that all temperature measurements are indirect. So it is misleading to single out any one method of measuring temperature as being more legitimate and fundamental than other methods.

            Changes of temperature are accompanied by certain physical changes; for example, changes in pressure, volume,electrical resistance, electromotive force, luminosity, etc., any one of which may be chosen as a basis for measurement of temperature.

            In the mercury in glass thermometer the increase of temperature is taken as proportional to the apparent expansion of mercury. (At this point, one has only measured the temperature of the mercury ITSELF. Any statement about the surroundings of the thermometer will be an inference, which may be valid or invalid.)

            Roughly, the satellites take the temperature of the originating material as proportional to the fourth root of the brightness of the luminous body in various parts of the electromagnetic spectrum. This is the traditional principle of the pyrometer – for the invention of which Josiah Wedgewood* was elected to the Royal Society in 1783.

            Of course the basic idea of the pyrometer was known to the first caveman who glanced at his fire and decided to judge how hot it was by the way it looked, or felt on his face, rather than by directly touching it.

            *Josiah Wedgewood was the grandfather of Charles Darwin. This is a little ironic for Dr Spencer.

        • Locke says:

          Dear Will,

          Before engaging in invective please familiarise yourself with the thermal equilibrium and the zeroth law of thermodynamics. Mercury thermometers are in thermal equilibrium with their surroundings, Roy’s satellites are not. Does this mean that Roy’s satellites cannot infer temperature, no, but please understand the fundamental difference between the methods.

          Kind Regards, Locke.

        • David Appell says:

          They measure the expansion and contraction of mercury in a glass tube…

          Most of today’s digital thermometers do no such thing — they use thermistors.

          In any case, the temperature expansion of mercury is very well known — as is the thermoresistance of the semiconductors used.

          I’m very sure that when you stick a thermometers in your kid’s mouth to take her temperature, you don’t then say, “well, who really knowns what the temperature is, because it all depends on the thermal properties of mercuty.”

          • That’s why I was careful to use the word “traditional” so that a silly argument like this wouldn’t be put forward. But silly people sometimes want to argue for silly reasons.

            There is also, BTW, a well understood physical relation between “brightness” and, say, the temperature of oxygen. So your second critique is just as silly as your first.

      • TedM says:

        Also consider that satellite temperature data sets correlate better with radiosonde data sets than do surface temperature data sets. Also they don’t cool the past.

  13. Alec aka Daffy Duck says:

    Just passing the along as I believe it is real new:

    Recent global warming hiatus dominated by low latitude temperature trends in surface and troposphere data†

    Abstract
    Over the last 15 years, global mean surface temperatures exhibit only weak trends. Recent studies have attempted to attribute this so called temperature hiatus to several causes, amongst them incomplete sampling of the rapidly warming Arctic region. We here examine zonal mean temperature trends in satellite-based tropospheric data sets (MSU/AMSU and GNSS Radio Occultation) and in global surface temperatures (HadCRUT4). Omission of successively larger polar regions from the global-mean temperature calculations, in both tropospheric and surface data sets, shows that data gaps at high latitudes can not explain the observed differences between the hiatus and the pre-hiatus period. Instead, the dominating causes of the global temperature hiatus are found at low latitudes. The combined use of several independent data sets, representing completely different measurement techniques and sampling characteristics, strengthens the conclusions.

    http://onlinelibrary.wiley.com/doi/10.1002/2014GL062596/abstract

  14. dave says:

    “…warming Arctic region…”

    According to the Danish Meteorological Institute, the average temperature North of the 80th Parallel ended 2014 pretty much bang on the average for 1958 – 2002, of 244 K. I do not see that whether it is 240 or 250 K in the Arctic dark makes any difference to anything, anyway.

    • Aaron S says:

      There is a pattern here… the warming is always hidden in the data gaps be it the deep ocean, by volcanic aerosols, stored as a mega el nino that is ready to unleash heat, or the highest of latitudes. Its a nice game… prolongs funding and delays lots of humble pie for a little longer. Once the Arctic & Antarctic are measured then where is it hiding? Perhaps an alternative parallel universe? Sting theorist could prob use a bump up in nsf funding!

    • David Appell says:

      “I do not see that whether it is 240 or 250 K in the Arctic dark makes any difference to anything, anyway.”

      Almost of the melt occurs in the summer months. When it refreezes, the ice is not as thick.

  15. bev says:

    “…in the Arctic dark…”

    Dark. That is kind of the whole point about the Polar regions.
    Every year the temperature inevitably sinks to much the same starting level. (At 250 K the surface will be radiating 16% more energy to space than at 240 K. and therefore it is not likely to stay that high for long.) There can be no “memory” of any temperature anomalies from earlier years.

    • David Appell says:

      “There can be no “memory” of any temperature anomalies from earlier years.”

      There can — the refreezing ice is younger and usually thinner.

      • bev says:

        “…the refreezing ice…”

        ‘Cryosat’ is the dedicated polar monitoring platform of the European Space Agency.

        Cryosat data for the Arctic ice remaining, in October, after the summer melt, for successive years:

        2011 5,300 cu. km.
        2012 5,400
        2013 8,800
        2014 7,500

        I suppose you could call that a faint memory of the recent cool summers. At least that is what Rachel Tilling from the UK’s Nerc Centre for Polar Observation and Modelling calls it.

        If you are right about the ‘memory’, Professor Peter Wadham’s confident prediction in 2011 that the Arctic summer-ice will all be gone this year, is as dead in the water as a lump of multi-year ice.

        There is no equivalent of Cryosat’s depth imaging for Antarctica, but since there is a positive anomaly there of 1,663,000 square kilometers of sea-ice cover (and similar anomalies have held now for two years) one may assume some increase in multi-year ice.

    • dave says:

      An ice-free summer Arctic Sea would – if it meant a North-West passage finally became a commercial reality – be of enormous value. It would cut the sea journey from Europe to the Far East, for big ships, from 15,000 to 8,000 miles.

      But we have been promised this many times. That Bahrens prediction in the early 1970’s actually caused quite a lot of (momentary) excitement in business circles in Canada. There were hopes that the Alaska Pipeline would become redundant.
      I was there. An old bloke said to me “Not in my life-time – and not in yours.”

      • bev says:

        The regional anomalies figures (see the data series for details of areas covered) for December have now been posted by Dr Spencer.

        North Pole Land + 0.59 C
        Ocean + 0.09

        South Pole Land – 0.51
        Ocean – 0.47

        These areas REALLY jump around: LT North Pole Ocean has dropped 2.91 C in just 10 months. Of course, these places are always more or less COLD.

        • bev says:

          “…regional anomalies…”

          Those aren’t “dashes” for the South. They are “minus signs.”

          South Pole, Land… -0.51 C

          South Pole, Ocean… -0.47 C

          • dave says:

            “[Poles are] always more or less COLD.”

            At the South Pole, the sun is shining round the clock, and it is a balmy mid-summer’s day at, -20 C. (With the wind chill, -30 C.)

            I believe a traditional instrument was used; someone stuck their nose outside.

  16. Locke says:

    Will Nitschke <—- Clinical case study for Dunning Kruger syndrome.

  17. Ron Wallenfang says:

    You dropped one satellite’s data “due to spurious warming”. I suppose warmists would argue that this is just what you (we) accuse them of doing – throwing out inconvenient data. What is the response to this?

    • If the data is rubbish it’s not uncommon in science to throw it out. No problem with that and it’s done all the time. What you don’t do is arbitrarily keep some part of the data set that supports your argument and selectively toss out the bits that don’t. This is what the “hide the decline” debacle was all about, for example.

  18. dave says:

    Although I am not an American, I understand that it is an Agency of the Federal Government of the United States, called “The National Oceanic and Atmospheric Administration,” which controls most of these satellites, is responsible for the data, and declares when an instrument is providing spurious data.

  19. bev says:

    Instruments have a finite life, and a finite time before they lose their factory settings.

    It is generally the managers of the satellites (NASA, NOAA, European Space Agency) who warn that a “characteristic failure” is occurring. Not a matter of a user of data, such as UAH or RSS, making an arbitrary call – although they might start to comment on it a few months before the instrument is retired. I remember Dr Spencer fretting, a couple of years ago, about a sudden increase in the noise of one channel – and indeed that instrument first showed an idiosyncratic drift, and then failed completely soon afterwards.

    UAH uses data from fourteen instruments at present – plenty of opportunity for cross-validation.

  20. Dan MacDonald says:

    Hi Roy,

    This was posted yesterday, 1/8/15. Found it interesting that someone else gets it.

    http://thisisindexed.com/

    Dan

  21. dave says:

    The following is from the Australian Bureau of Meteorology on January 6th 2015:

    “El Nino-like conditions in the tropical Pacific Ocean weakened in the past fortnight, after being close to or exceeding El Nino thresholds for several weeks.”

  22. Scottar says:

    How does JMA get and process it’s temperature data, especially the SST?

    • JohnKl says:

      Hi Scottar,

      You can examine their website and find out for yourself at:

      http://www.jma.go.jp/jma/indexe.html

      A little while ago I perused their site. They claimed to obtain information from various sources.

      Have a great day!

      • Scottar says:

        Yes, and it was clear as mud to me.

        The method used for Sea Surface Temperature (SST) analysis in climate monitoring at JMA was updated to an approach based on in-situ observation in March 2006. For details of the analysis, see Ishii et al. (2005).

        SST analysis involves a resolution of 1º latitude and 1º longitude. The east-west grid points run eastward from 0.5ºE to 0.5ºW, while the north-south grid points run northward from 89.5ºS to 89.5ºN. The daily analysis scheme is based on the optimum interpolation method, and the SST deviation from the normal for the previous day’s analysis is multiplied by 0.95 for use as a first guess. The inputs of daily analysis are marine meteorological data for the 7-day period centered on the day in question. Observed data averaged daily in 1.5º x 1.5º boxes are used as super-observations to save processing time.

        It appears that they are taking the SST around the equator. but basically my take is it’s like what the Hockey Stick did and that is attaching an instrumental record to a proxie record. This comment pins it down:

        https://notalotofpeopleknowthat.wordpress.com/2015/01/09/uah-confirms-2014-was-not-hottest-year/

        “Whereas surface sets, such as HADCRUT and GISS, only measure the temperature at the bottom of the atmosphere, the satellite sets of UAH and RSS measure it through the whole of the lower troposphere, and therefore must be regarded as much more comprehensive.”

        “It is also argued that UAH do not measure sea surface temperatures. However, in terms of global heat balance, the temperature of the sea at the surface is almost irrelevant when compared to that of the entire volume of sea from top to bottom. In any event, increased heat at the top of the sea will always tend to transfer into the atmosphere above it by evaporation, until equilibrium is found. While there is a short lag for this to happen, atmospheric temperature measurements do reflect these changes.”

        “It is also worth pointing out that higher sea temperatures can only affect land ones after having also affected the atmosphere. Indeed, according to theory, the global troposphere should warm 1.2 times as fast as the surface.”

        http://www.cato.org/blog/how-hot-does-it-have-be-break-record

        That and the uncertainty of surface temps from a over “a thousand thermometers scattered about the planet. Some are well-taken care of, and some are not. Some may have traces of urban warming in them. Nor is the number of readings exactly the same from year to year, or even from month to month.”

        The graph of the SST from 1890 could be questionable for the same reasons of surface temps. But it does show mostly a steady increase, even with the 2 cold periods. The pause is not very evident but the super El Nino spike is.

  23. Typhoon says:

    Whenever I see these temperature time series plots I wonder where are the [statistical + systematic] error bars?

    Given the difference between the UAH and RSS series, one would expect the systematic error estimates to be visible on the plot.

    With regards to the other terrestrial ocean/land temperature times series, the claim the one knows the global temperature to within 0.05C, in say 1900, appears to not only strain credulity, but tosses it in the blender and hits the puree button.