RATPAC – an initial look at the Global Balloon Radiosonde Temperature Series

My last post on Record Hot or Not stimulated some good discussion with the conformist warmists bringing some useful information to the debate. In particular, they drew attention  to RATPAC and how this data, in their opinion, verified the surface thermometers, confirming 2015 as the warmest year since records began. So I decided to take a look. It is indeed another great data set that I’ve heard very little about and it certainly should not be ignored.

RATPAC stands for Radiosonde Atmospheric Temperature Products for Assessing Climate – are you any the wiser? It is in fact a global temperature model based on measurements made from weather balloons. RATPAC therefore provides vertical temperature gradients through the atmosphere and offers the opportunity to compare with the surface thermometer and satellite based analyses. There are 85 recording stations with good global cover (Figure 1) and data begins in 1958.

This post is a first look at the data intended to raise questions and issues for discussion. Part 1 is simple data description and presentation with a few key (and very interesting) observations. The second part (that will hopefully follow later in the week) compares RATPAC with surface thermometers and satellites.

Figure 1 Map showing the distribution of 85 RATPAC monitoring stations. A curious thing, I could not find a copy of a RATPAC map on the web. There was however this list of stations with lat lon and my friend Luis de Sousa kindly made this map.


RATPAC is managed by NOAA and the main “portal” is here. Documentation is sparse, for example you’d think they would provide a map. We have to look at the photo caption to learn:

… a hydrogen-filled balloon that will carry a radiosonde up in the air to measure temperature, humidity, and atmospheric pressure and transmit the data back to a ground station.

And the text says…

RATPAC data come from 85 stations with near-global coverage. NCEI provides data on 13 atmospheric pressure levels: the surface, 850, 700, 500, 400, 300, 250, 200, 150, 100, 70, 50, and 30 mb. Where available, data begin in 1958 and extend through the present. Some of the 85 stations have discontinued observations since the 1990s, and not all stations have observations at all levels.

Seasoned surface temperature analysts will be familiar with issues like this…

The temporal homogeneity of many radiosonde time series is questionable due to historical changes in instruments and measurement practices.

Two products are provided RATPAC A and B. NOAA recommend using the former and I have not looked into B at all. All the data presented here are RATPAC A and this is how NOAA describes the data:

RATPAC-A contains adjusted global, hemispheric, tropical, and extratropical mean temperature anomalies. From 1958 through 1995, the bases of the data are on spatial averages of LKS adjusted 87-station temperature data. After 1995, they are based on the Integrated Global Radiosonde Archive (IGRA) station data, combined using a first difference method (Free et al. 2004). For analyses of interannual and longer-term changes in global, hemispheric, and tropical means, the team recommends use of RATPAC-A since it contains the most robust large-scale averages.

All this sounds too familiar. I fully endorse the need to apply corrections to data where the physical need is identified and understood. But I’m afraid that NOAA and NASA have undermined their own credibility by continually adjusting surface temperature records. When sceptical scientists read here that temperatures have been adjusted they will wonder by how much and why.

The data were easy to download from this link provided by NOAA. It contains three files. I have looked at “RATPAC-A-annual-levels.txt” that gives global average T anomalies for 13 levels defined on atmospheric pressure (Figure 2) and “RATPAC-A-year-to-date-layers.txt” that provides the averages for three layers – 1) the Lower Troposphere 2) the Upper Troposphere and lower Stratosphere and 3) the Stratosphere. Data for individual stations and raw temperatures are not obviously available.

So lets get straight to the point and have a look at the data:

Figure 2 There is clearly a lot going on here. The convergence of all profiles in the interval 1973 to 1993 suggests this is approximately the datum period used to calculate anomalies. And clearly something dramatic has happened post-1992. To get a better picture of what is going on the following charts show the traces for the three layers used in the merged data sets.

Before proceeding I’m sure readers will appreciate some guidance converting milli bars (mb) to height. I’m an old fashioned Brit and still understand height measured in feet better than meters and so only give feet here. This web site was used for the conversion.

850: 4,779 ft (Ben Nevis)
300: 30,053 ft (Mount Everest)
100: 51,806 ft
50:  63,367 ft

Figure 3 The 5 levels between 850 and 300 mb are highly congruent and show a familiar warming pattern. Spikes associated with the 1998 and 2010 el Ninos are clear to see. Is that a pause 1958 to 1978 and again 1998 to 2015? Or is it a single warming trend? Remember that each level is successively colder than the one below and these absolute differences are removed by the normalisation procedure. These levels encompass most of the troposphere. The surface measurements are not included because NOAA do not include the surface in their layer models and as we shall see there are issues with the surface data. Amongst other things, not all surface stations are at sea level.

Figure 4 I am skipping over the 300 to 100 layer for the moment, the reason for this will become clear when we get to it in Figure 5. The 100 to 30 mb layer is in the stratosphere and has some of the most interesting information.

The key observations:

  • There is a very distinct and large overall cooling trend of about 2.25˚C from 1958 to 2015.
  • The three large volcanic eruptions of Agung, El Chinon and Pinatubo resulted in pronounced warming spikes in the stratosphere.
  • Pre-Pinatubo these 4 layers were fairly congruent but post Pinatubo the 100 mb level defines a separate trend.
  • The Agung eruption did not warm the 100 and 30 mb levels but all 4 levels were warmed by the El Chinon and Pinatubo eruptions.
  • Post-Pinatubo there was severe cooling of levels 70, 50 and 30 that seem to have become “detached” from level 100.
  • El Nino warming in 1998 and 2010 is clearly visible in the 100 mb level but mainly absent in the higher levels.

There is a lot going on here and I will defer commenting further until the discussion.

Figure 5 The 5 levels between 300 to 100 mb present a somewhat chaotic picture. One reason for this will be the fact that the tropopause boundary between troposphere and stratosphere occurs in this interval and since the height of the tropopause is not constant across the Earth some of these layers will receive mixed messages. The 100 mb level is also included in the previous chart and is the only level to display stratospheric warming during volcanic eruptions AND tropospheric warming during el Ninos. Its possible that the stratosphere signal comes from high latitudes and the troposphere signal comes from low latitudes. RATPAC provides the latitude bands to examine this, but I have not had time to do so. Levels 300 to 150 mb all show el Nino warming but none show either warming or cooling associated with the three large volcanic eruptions.

Figure 6 

Figure 7 

Figure 8 

Figures 6, 7 and 8 provide the averages for the three layers selected by NOAA: 1) 850 to 300, 2) 300 to 100 and 3) 100 to 50 mb. The N and S hemispheres are also plotted. Figure 6 shows that the S hemisphere troposphere is warming less rapidly than the N, confirming what we already know from surface thermometers and satellites. RATPAC data indicate 1.9˚C per century in the N and 1.3 ˚C in the S. The difference in the Stratosphere is less pronounced. The N hemisphere is cooling at a whacking 3.9˚C per century and the S at 4.5˚C per century.

A curious feature of the stratosphere cooling is that it goes down a step across the El Chinon eruption and then down an even larger step after Pinatubo. Beween Agung and El Chinon the trend is relatively flat as it is post-Pinatubo. It is as if these large eruptions have left a lasting imprint on the stratosphere.

One final observation from Figure 6 is that despite assertions made by the climate science community that these large volcanic eruptions cool the troposphere I can see no evidence for it in these data. Any blips down, if they exist, have no significance compared with the year to year noise. I have been puzzled by this for many years. Earth science is empirical. We develop theories based on observations and the lack of volcanic cooling in the troposphere has puzzled me for many years. I once wrote to the UK Met Office to enquire and was told by a renowned expert that cooling must be masked by El Nino warming. Alleged volcanic cooling plays a crucial role in the overall global warming modelling argument.

Figure 7 shows that the chaotic 300 to 100 mb mixed layer averages to a flat line.

Comparison Thermometers, Satellites and Balloons

Before moving on to the discussion I want to post just one chart comparing surface, satellites and balloons to give a flavour of what this shows.

Figure 9 

What we see is that the RATPAC 850 to 300 layer shows the same high amplitude variance seen in the satellite data. RATPAC actually seems to follow the satellite trend more closely than it does the surface thermometers. But the temperature gradient through RATPAC is virtually identical to the surface thermometers. The adjusted RATPAC A and the adjusted surface thermometers are in exact agreement with each other. In 2008, RATPAC A and satellites were exactly aligned. Since 2011, the trends are blown apart.


As discussed above, the RATPAC A data for the 5 levels in the 850 to 300 mb troposphere layer are to varying degrees the same as what we already know from surface thermometers and satellites. The troposphere is warming. The exact pattern and rate of change is disputed, as is the cause. The interesting part of the troposphere debate will have to wait.

In my opinion, the really interesting data here is from the stratosphere, Figures 4 and 8. What has caused the steep stratosphere cooling trend and should we be worried about it? What causes volcanic eruption warming of the stratosphere? And do CO2 emissions have anything to do with this at all?

As stated at the beginning of this post, all these points are up for discussion. Informed commenters are invited to give their informed opinions. I find myself on unsure ground. Skating out on thin ice. Normally Google would provide instant back up, but in this case not, which I find curious. There must be a wealth of research out there. But it hasn’t jumped out at me yet.

Ozone depletion

Ozone is a molecule of oxygen. O2 is the stuff we breath. O3 is ozone. Ozone is manufactured in the stratosphere by incoming solar ultraviolet (UV) radiation from the Sun, and once it is formed it also absorbs incoming UV warming the stratosphere. That is why the stratosphere gets warmer as you go up, ozone is absorbing a large amount of the incoming UV, protecting the surface. The UV that gets through is absorbed by the surface and re-emitted as infrared (IR) that is trapped by green house gas (mainly water vapour) warming the surface.

Everyone has heard of ozone holes and ozone depletion caused by chemicals released by Man. This is a vast subject I don’t have time to go into. But a simple summary would say that stratospheric cooling is caused mainly by ozone depletion (regardless of the cause). The simple idea is that less ozone means less UV intercepted in the stratosphere that results in cooling.

What you don’t hear so often is that less ozone means more UV landing at Earth’s surface that when re-emitted means more outgoing IR that would warm the surface via the greenhouse effect. I haven’t a clue if the increased UV at surface resulting from ozone depletion is sufficient to cause significant warming of the troposphere. I’m hoping that Clive Best may be willing to provide some answers to that.


Large volcanic eruptions can inject vast quantities of sulphur dioxide into the stratosphere where it is converted to sulphuric acid which condenses to form sulphate aerosols. I am borrowing heavily from this United States Geological Survey source.

The sulphuric acid / aerosols do three things:

1. They are supposed to reflect incoming UV back to space cooling the planet. There is little observational evidence for this (see discussion of Figure 6 above).

2. They trap outgoing IR radiation warming the stratosphere. This could be described as a sulphur borne stratospheric greenhouse effect. There is clear evidence for stratosphere warming associated with large volcanic eruptions. If sulphur compound greenhouse warming is the cause, I don’t know.

3. Sulphate aerosols destroy ozone. The USGS link says this:

The sulfate aerosols also promote complex chemical reactions on their surfaces that alter chlorine and nitrogen chemical species in the stratosphere. This effect, together with increased stratospheric chlorine levels from chlorofluorocarbon (CFC) pollution, generates chlorine monoxide (ClO), which destroys ozone (O3).

There is clear evidence from the stratosphere temperature record that the large volcanic eruptions may have destroyed ozone leading to stratosphere cooling.

To sum up, large volcanic eruptions that inject large quantities of sulphur compounds into the stratosphere, are alleged by climate science to cool Earth’s surface but there is little evidence for this happening. The evidence that does exist shows quite clearly that large volcanic eruptions lead to temporary warming of the stratosphere and longer lived cooling that could be explained by destruction of ozone. The latter may lead to more UV arriving at surface that would cause warming not cooling of the troposphere.

CO2 greenhouse causes stratosphere cooling

The climate science community has gathered around the idea that the obvious cooling of the stratosphere is caused by the CO2 greenhouse induced warming of the surface. It seems so simple, less heat escaping the surface leads to cooling of higher layers. Too simple I’m afraid because the higher layers are normally warmed by incoming not outgoing radiation.

Again it’s tricky to find sources for CO2 causing the troposphere to warm and the stratosphere to cool. But here’s an old Real Climate post by Gavin Schmidt, the current director of the NASA Goddard Institute For Space Studies (NASA GISS) and one of the most pre-eminent climate scientists in the world.

Why does the stratosphere cool when the troposphere warms?

In the case of the Earth, the solar input (and therefore long wave output) are roughly constant. This implies that there is a level in the atmosphere (called the effective radiating level) that must be at the effective radiating temperature (around 252K). This is around the mid-troposphere ~ 6km. Since increasing GHGs implies an increasing temperature gradient, the temperatures must therefore ‘pivot’ around this (fixed) level. i.e. everything below that level will warm, and everything above that level will cool.

Even though the stratosphere has an opposite lapse rate to the troposphere because of the ozone absorption, the effect of increasing GHGs is the same, i.e. since it is above the effective radiating level, it will cool. The cooling will be greatest as you go higher. In the troposphere, there are important other effects that change the temperature, chiefly moist convection, and that smears out the temperature changes you expect from a pure radiative atmosphere. So while the troposphere does warm as a function of increasing GHGs, the maximum change is not at the surface, but actually in the mid-troposhere.

There’s so much to say about  this I don’t know where to begin. But I just want to focus on my understanding of how the CO2 greenhouse works on Earth that is based entirely on the writings of Clive Best.

The main IR absorption bands of CO2 are saturated at surface. Therefore increasing CO2 ppm does nothing. All the IR radiation that CO2 can absorb is already absorbed. This theoretically sets up a radiative cascade chain where tropospheric CO2 emits, absorbs and re-emits IR in the chosen bands upwards through the atmosphere until a level is reached where the density of CO2 molecules is too low to capture and re-emit IR and at that level the IR emits directly to space by passing all that overlies that critical level. The critical level is known as the emission height which lies just below the tropopause. Observations from space confirm this theory. Note that this is quite different to what Gavin Schmidt has described. I appreciate that the Real Climate post is dated 2004 and may require updating. It is in fact labelled as obsolete and points instead to this post.

As CO2 ppm increases in the atmosphere, the density of CO2 molecules at altitude also increases and this raises the emission height, which in the troposphere means a colder emission temperature. The colder emission temperature at altitude means that the surface must warm to maintain radiative energy balance. That at least is how the story goes.

One consequence of this yarn is that the stratosphere is and always has been bypassed by outgoing IR in the CO2 absorption bands. The CO2 greenhouse is diagnostic of the troposphere and its absence is diagnostic of the stratosphere. IR radiation trapped by CO2 warms the troposphere and then escapes to space. Increasing CO2 may warm the troposphere a bit more before escaping to space. According to theory, outgoing IR in the CO2 bands bypasses the stratosphere emitting directly to space and always has done.

Thus, increasing CO2 does not cool the stratosphere. Depleting the amount of ozone does.

Finally, there is the curiosity of Figure 2 that seems to show the 100 mb (52,000 ft) level belonging to the stratosphere pre-Pinatubo, but joining the troposphere post- Pinatubo. Convective storms, associated with el Ninos have clearly lifted water vapour into this level post-Pinatubo eruption. One rational explanation is that the Pinatubo eruption destroyed ozone and raised the level of the tropopause. Should this be true, understanding the exact physical reasons for this appears to be vital.

Concluding Comments

Since 1958, the troposphere has warmed and the stratosphere has cooled. These observations may be linked, not by manmade CO2 emissions but by ozone depletion triggered by the 1982 El Chinon and 1991 Pinatubo volcanic eruptions.

Ozone depletion leads to less incoming UV trapped in the stratosphere, cooling that layer, and more incoming UV reaching surface warming the troposphere by greenhouse effect. The cooling stratosphere causes a warming surface and not vice versa.

Climate science seeks to explain the observations by increased manmade CO2 emissions and a cooling effect on the surface by volcanic eruptions. There is no evidence from the data presented here that large volcanic eruptions cool the surface. And the physical science does not permit the explanation of increased CO2 cooling the stratosphere.

I have gone skating on the thin ice of science tonight. Fire away and provide greater and better insight, links to charts and data. The emphasis here is on greater, better and validated.

This entry was posted in Climate change and tagged , , , , , , , , , . Bookmark the permalink.

62 Responses to RATPAC – an initial look at the Global Balloon Radiosonde Temperature Series

  1. A C Osborn says:

    You stated “Figure 2 There is clearly a lot going on here. The convergence of all profiles in the interval 1973 to 1993 suggests this is approximately the datum period used to calculate anomalies. And clearly something dramatic has happened post-1992.”

    How much of the changes to the data post 1992 are down to the change “After 1995, they are based on the Integrated Global Radiosonde Archive (IGRA) station data” and how much to “Adjustments?
    Where is the Raw data that must exist (like the land based Thermometer data) so that we can see what and why the data needs to be Adjusted. Could it be Dataset “B”?
    Also where is the rationale for the adjustments like those that exist for land based data (even if it doesn’t match their stated values, at least we know the excuses for doing it)?

    • Euan Mearns says:

      AC, you have been following this story a long time. Have you ever come across this RATPAC data? They have 85 teams around the world releasing 2 balloons a day. Its a huge effort.

      • A C Osborn says:

        Euan, I have never actually seen any data, only other people referring to itfor comparison purposes.
        Clive Best has also been looking at the Atmosphere as well and strangely enough it refers to Yvan Dutil’s post.
        Clive has been looking at Moisture content and there is a post over at Tallbloke’s

        I find it hugely suspicious that all the data sets roughly agreed until lately, the NOAA attempts to discredit the “pause” have shown how desperate they are and to what lengths they will go to to do so.

    • Euan Mearns says:

      Here’s a comparison of the Stratosphere from UAH and RATPAC A. One problem is we (I) don’t know what levels the UAH are measuring and if this is directly comparable to RATPAC. All the features in UAH are the same, it just doesn’t fall as far as the balloon data post-Pinatubo.

      Also, if CO2 is supposed to cool the stratosphere, there is no evidence for that happening post-1995.

      AC you have a point about the change in data after 1995, but there’s no sign of that impacting the stratosphere levels.

  2. Yvan Dutil says:

    CO2 cool the stratosphere because it increase the optical thickness of the atmosphere. It acts like any insulator by increasing the temperature gradient. It is my understanding that slow down in the stratospheric cooling has been associated with an increase in water vapour countent, but I cant track the original paper.

  3. Euan Mearns says:

    And just for fun, here’s the comparison truncated at 2008. From this we see that the satellites and RATPAC are very much together and the thermometers are the odd man out (forgot to colour the lines 🙁

    • gweberbv says:

      In the end this is about climate change, right? A phenomenon that is believed to happen on a time sclale of decades. If that is true, why should we look at temperature data on a much finer time scale? No information on the phenomenon of interest will be obtained but a lot of high frequency noise (like El Nino) is showing up.
      So, I suggest to plot temperature data with a 5 year binning (or a 5-year-average). Of course, then one must skip the fun of discussing if this or that years was the most warmest – which is of total irrelevance.

  4. A C Osborn says:

    Euan here is an explanation of RATPAC A & B


  5. A C Osborn says:

    Then there is this explanation, which suggests they have some Raw Data.


  6. wehappyfew says:

    Ozone loss = stratosphere cooling.

    Ozone recovery = stratosphere warming… right?

    Ozone measurements over both poles:



    (I’m still trying to figure out the correct WordPress image tags for this blog)

    from slick EPA report:




    So it would appear from the data that ozone recovery since the late 1990’s has slowed the decline in stratospheric temps, especially in the SH, where the ozone “hole” is most intense. Recent ozone increase appears to offset part of the temp decrease caused by GHG cooling of the stratosphere.

    Of course the full picture is much more complicated than that… changes in stratospheric water vapor, nitrous oxide, solar influences, volcanoes, etc.

    Further reading:

    The role of Stratospheric Ozone in the Zonal and Seasonal Radiative Energy Balance of the Earth Troposphere System


    A comparison of model-simulated trends in stratospheric temperatures


    Shine’s paper shows that most of the action occurs above the RATPAC data – the 100mb level has a low or even positive response to GHGs, with moderate cooling due to ozone loss and water vapor.

    The 50 and 30mb levels are just getting into the zone of significant temp decrease due to GHGs.

  7. javier says:


    RATPAC is only one of the several radiosonde databases around. Others are:

    – HadAT (Thorne et al., 2005) Hadley Centre, UK Met Office
    – RAOBCORE (Haimberger, 2007) University of Vienna
    – RICH (Haimberger et al., 2008) University of Vienna
    – IUK (Sherwood, 2007; Sherwood et al., 2008) University of New South Wales

    RSS has a page claiming very good agreement with all four but curiously, RATPAC is not included.

    It even has a tool for graphing them up to 2011 that shows the following for TLT:
    1979-2011 75°N-75°S:

    RSS: 0.170 °K/decade
    HadAT: 0.195 °K/decade

    RSS: 0.180 °K/decade
    RAOBCORE: 0.199 °K/decade

    RSS: 0.180 °K/decade
    RICH: 0.206 °K/decade

    RSS: 0.202 °K/decade
    IUK: 0.175 °K/decade

    The differences in RSS are due to the poor coverage of radiosondes, so for a fair comparison RSS is sampled for same coverage.

    All the differences are within ±0.025 °K/decade.

    I would conclude that the not matching between radiosondes and satellites is mainly specific for NOAA RATPAC dataset, or more recent than 2011.

    Besides you might want to take a look at ERA interim (+ERA 40) reanalysis data, as it is available. This dataset includes surface temperature records from stations with a better coverage than GHCN, radiosondes records and satellite records. The data is feeded constantly and produces a run every few hours to produce the best weather predictions that we are capable now. Temperature is only one of the many things that ERA tracks, but displays a very prominent pause in the 21st century, and I would say that prior to 2014 it even showed a little cooling like CET.

  8. javier says:


    just a wild thought. Do you think it possible that the Stratosphere cools because the Surface and Troposphere warm, retaining a bigger share of the heat from moving up? In that case when the Surface and Troposphere would cool, the Stratosphere would warm and they will be both linked to maintain an energy balance.

    If that is the case, that the Stratosphere is not cooling in the last decades would be further proof that the surface and troposphere are not warming.

    • Euan Mearns says:

      Instinctively I don’t agree with this, but may be wrong. From the comments and the links I am very aware how little I know on this topic. I think transfer of mass and energy across the tropopause could be minimal. Obviously lots of energy goes through the tropopause but that is different to exchange across it. The structure of the stratosphere is ruled by ozone and UV. The structure of the troposphere is ruled by H2O and IR. I think it can almost be viewed as two immiscible layers of liquid. But as the link provided by Old Brew to Erl Happ points out, at the poles the tropopause disappears and we can have gravity driven flow downwards to surface.

      And so in short, the stratosphere will respond to changes in ozone and UV (solar activity) while the troposphere will respond to changes in IR and GHG (mainly H2O vapour). The energy balance between the two would be maintained by the tropopause rising and falling along with the emission heights and emission temperatures of the GHGs.

  9. Hans Erren says:

    Uah stratospheric shows a dc shift downward after every major plinian eruption. So we have to wait for the next puncture of the tropopause. Until then the stratosphere trend will remain flat.The correct name is El Chichón by the way.

  10. oldbrew says:

    Erl Happ has plenty to say about ozone and the stratosphere here:

  11. Dave Rutledge says:

    Hi Euan,

    For a comparison between balloons and satellites, I would vote for the HadAT series at


    They have developed indexes that weight the balloon temperature readings at different pressures to match the satellite radiometer channels.

    This means that you can take these HadAT indexes and weight them again to match the radiometer channel weightings in the latest UAH 6.0 data set. When I do this I get a trend of HadAT 0.14 degrees per decade from 1979 on for the lower troposphere compared with the UAH 0.11 degrees per decade. Given the poor ocean coverage of the balloons, I am not sure I would expect better agreement than this.


  12. keith harrison says:

    Noticing your commentary on ozone depletion and it possible effects of heating and cooling, let me offer links fro the University of Waterloo, Canada where Dr QB Lu published a 2013 paper where he claims he can precisely predict warming and cooling by his model. Ozone interaction with cosmic rays is the key and he is predicting cooling (as soon as 2020) as ozone layer repairs itself thanks to Montreal Protocol and its updated version in the reduction and elimination of CFCs.



    Apparently he has updated his paper and also has a book on the subject.


    I trust you find this useful.

    • Euan Mearns says:

      Keith, thanks for the links. I’ve spent quite a bit of time looking at them and trying to make some sense. The involvement of cosmic rays was interesting but the process described here appears to work in the opposite sense of empirical observations that show enhanced cosmic ray bombardment leads to surface cooling.

      • Luís says:

        Euan, the empirical observations point to increased GCR increasing cloud cover. That then this cools the surface is an inference.

        • Euan Mearns says:

          Luis, the empirical observation is that Earth cools when solar geomagnetic activity falls. This results in greater exposure to GCRs but may also alter the UV spectrum. While I’ve nothing against GCRs being the cause I favour the UV story impacting ozone production with knock on to atmospheric circulation. Someone linked to a post by Jo Nova which looked pretty interesting.

  13. Peter Shaw says:

    Euan –
    A little chemistry I’ve assembled:
    Start with methane as a marker for oxidisable materials released from the surface. In the troposphere this is oxidised by hydroxyl (OH), which limits its lifetime to some ten years. That’s long enough for significant upmixing to the stratosphere, where it’s oxidised by ozone. This is apparently the source of the (trace) stratospheric water.
    There are indications that marine life may produce low levels of halomethanes. These should track methane, except that elemental chlorine would be formed in the stratosphere naturally. Watch this space.

    The bulk of volcanic gases is carbon dioxide and steam. There are variable and (relatively) minor amounts of sulphur dioxide and some acid gases (HF and HCl). Assuming the ash segregates from these gases, the latter will produce sulphur trioxide and chlorine. Sulphur trioxide will immediately dry the surrounding air, becoming strong sulphuric acid (which I think is what is meant by “sulphate aerosols”). Such an aerosol is in dynamic equilibrium between the tendency of small droplets to evaporate water, and the high affinity of sulphuric acid for water, and can only persist at very low humidity (more below). HCl gas won’t dissolve in it.
    Strong sulphuric acid is highly refractive, so scatters light well – as in Venus’s clouds. It absorbs infra-red strongly, so I expect an aerosol to both warm itself and reduce surface cooling in addition to its effect on albedo.

    Chlorine is a chemical catalyst in ozone conversion. If it simply accumulated, there would be little ozone, so an exit must exist. The only credible mechanism I have is downmixing of stratospheric air into the wetter upper troposphere, where chlorine will partially convert back to HCl. This will form its own aerosol (“fuming hydrochloric acid”). Both aerosols can then be “washed out” in rain. This speculation suggests significant dynamic exchange between stratosphere and troposphere.

    Ozone is a high-energy molecule. It can only revert directly to oxygen where there’s a “third body” to absorb the excess heat. I suggest the surface of high-area fine volcanic dust is one such. This would both warm the dust (inhibiting settling) and sharply reduce ozone for the duration; consistent with what’s observed.

    • Euan Mearns says:

      Peter, many thanks for a fascinating comment. I think in part you have taken aim at my comment about the tropopause limiting transfer of mass and energy – fair enough. But the examples you give – CH4, Cl – could be driven by diffusion. And other processes such as phase / density changes driven by gravity.

      I would agree that large volcanic eruptions also adding water vapour, other acids and dust into the troposphere will also be important.

      The amazing thing is our atmosphere and climate are incredibly robust to all these things. Must have taken much bigger hits from Krakatoa and Tambora and a million other eruptions and bolloid impacts. But it dusts itself down and life goes on unchanged in all but the most extreme events.

      Why is ozone not rising in the stratosphere? And is there any mechanism that may cause ozone to rise a lot a trigger glaciations?

      • Peter Shaw says:

        Euan –
        I took no particular aim. My general point is that any credible view of the stratosphere must include a reason why past eruptions haven’t turned the Earth into something of a second Venus, with no ozone.

        We could both be right if the mixing I infer is localised – such as at the poles as Earl Happ claims. I note that in icecap cores, large eruptions have left their timestamps in… sulphuric acid.

        Ozone is created by one UV frequency (splitting oxygen molecules), and destroyed by a lower one, so Solar UV spectral changes (which are appreciable) matter. This might bear on your problem (or complicate it :)).

    • erl happ says:

      Re This speculation suggests significant dynamic exchange between stratosphere and troposphere.. You can see that happening here:http://www.cpc.ncep.noaa.gov/products/stratosphere/strat_a_f/
      Refer to the last table.

    • erl happ says:

      Re: Ozone is a high-energy molecule. It can only revert directly to oxygen where there’s a “third body” to absorb the excess heat. I suggest the surface of high-area fine volcanic dust is one such. This would both warm the dust (inhibiting settling) and sharply reduce ozone for the duration; consistent with what’s observed.

      The cooling that is associated with ozone reduction relates to less ozone heating in columns of descending air in the mid latitudes, reduced geopotential height, less cloud cover. There is a strong association between 500hPa geopotential height and surface temperature. https://reality348.wordpress.com/2015/12/29/3-how-the-earth-warms-and-cools-naturally/

  14. clivebest says:

    Gavin Schmidt writes:

    This implies that there is a level in the atmosphere (called the effective radiating level) that must be at the effective radiating temperature (around 252K). This is around the mid-troposphere ~ 6km.

    Since increasing GHGs implies an increasing temperature gradient, the temperatures must therefore ‘pivot’ around this (fixed) level. i.e. everything below that level will warm, and everything above that level will cool.

    Even though the stratosphere has an opposite lapse rate to the troposphere because of the ozone absorption, the effect of increasing GHGs is the same, i.e. since it is above the effective radiating level, it will cool.

    This is nonsense (IMHO). The “effective radiating level” (ERL) rises in altitude with increasing CO2 concentration. This hypothetical ERL level is not fixed but depends on greenhouse gases. If you were to instantaneously double the CO2 in the atmosphere then the ERL would suddenly move upwards to a colder level in the troposphere thereby radiating less energy to space than before. Suddenly the earth has an energy imbalance. It is receiving the same amount of solar energy as before but radiating less out to space. This is what is really meant by ‘radiative forcing’. The surface and lower troposphere must now warm up a little so as to rebalance energy by restoring the effective temperature of the new ELR to ~ 255K.

    In reality it is much more complicated than that simple picture because the ELR also depends on IR wavelength so there is not one ELR but hundreds of them. The reason why the stratosphere cools is because the strongest quantum vibrational levels for CO2 are the central lines in the 15 micron band. These lines are already saturated way up in the stratosphere where the lapse rate is inverted. As you increase CO2 so the ERL for these lines move up to a warmer level and therefore radiate more energy to space than before. The stratosphere has a greenhouse cooling effect! You can actually see this in any IR spectrum from space such as those from Nimbus. At the centre of the 15 micron band you always see an upward peak in emissions from the warmer stratosphere.

    However, I don’t believe the cooling can be as strong as that shown in figure 7 so something else is going on !

    • Euan Mearns says:

      Clive, there’s something I don’t quite get. In your various posts you have argued that CO2 eventually emits to space when the concentration density (partial pressure?) drops below a certain threshold. So I don’t see how CO2 in the stratosphere should be playing this game at all. And if it was, rising CO2 ppm, raising the emission height into the stratosphere, would result in cooling the surface would it not? Not the stratosphere. This because raising the emission height led to a warmer emission temperature CONDUCTING more heat away from the surface.

      The something else going on is volcanic eruptions that appear to have led to ozone depletion, cooling the stratosphere and presumably warming the surface as more UV makes it through.

      There’s some interesting challenges there for a physicist on vacation 😉

      • clivebest says:

        Euan, I am probably not explaining things very well.

        Heat flows up from the surface in 3 main ways.
        1. Evaporation ( latent heat) from the ocean
        2. Convection
        3. Radiation – directly to space through IR window or via absorption by green house gases
        The first two are the largest and generate a (moist/environmental) lapse rate. They also produce all the weather on earth as heat is redistributed laterally.

        In effect the atmosphere is one huge heat engine powered by the sun. Heat escapes to space only through radiation. H2O dominates the climate because it plays multiple roles from being a greenhouse gas to being a cloud/ice albedo thermostat.

        CO2 plays only a minor role but can become significant in the long term because it is well mixed and affects all layers of the atmosphere, including the stratosphere. The CO2 molecule has vibrational quantum levels which can be excited by collisions with other molecules or absorption of an IR photon. However, CO2 is in thermodynamic equilibrium with all N2, O2 molecules at the same level – i.e. they have the same temperature. The emission of IR by CO2 molecules follows Stefan Boltzmann for that temperature. The mean free path for absorption of a photon by another CO2 molecule depends on cross-section for that quantum wavelength and the density of CO2 molecules nearby. The atmosphere thins out with height so the mean free path for each wavelength increases. Photons emitted upwards whose mean free path is greater than the the top of the atmosphere will escape to space, carrying away energy. The height at which this happens is called the effective radiative level for that wavelength.

        Most quantum transitions have low cross section and the ERL is low in the troposphere. However the central peak of lines in the 15 micron band have very high cross sections and so need very few CO2 molecules above them to be absorbed. Their ERL is way up in the stratosphere where temperature now increase with height due to Ozone UV absorption (Ozone is itself a greenhouse gas which complicates things a bit). Therefore when CO2 levels increase near the surface so they will also eventually increase in the stratosphere because CO2 is ‘well-mixed’. That way the ERL for these central transition lines increase in height to a warmer level and so emit more IR photons to space than previously. That energy is taken from the stratosphere which is unaffected by convection and evaporation. It will also have a small cooling effect on the surface because it increases radiative transfer up through the atmosphere for those wavelengths.

        • Euan Mearns says:

          Thanks Clive, and I though I understood what was going on 😉 Heres’s a couple of your charts:


          The second chart shows the view from space and the central 15µ band emitting at 215-230˚K (-58 to -43˚C). Those temperatures are within the upper troposphere close to the tropopause are they not? What would we see if this were emitting from the stratosphere?

          Anyway, I don’t think the temperature profiles for the stratosphere provide any evidence for progressive cooling. There are two large step changes associated with volcanic eruptions. We are continually led on a wild goose chase by climate science that seeks to explain all observations by CO2 concocting physical processes en route.

    • robertok06 says:

      @clive best

      … on G. Schmidt explanation…

      “This is nonsense (IMHO). ”

      It could well be, after all G. Schmidt is a mathematician, good at writing computer models, but not necessarily at ease with physics and its applications to climate.

  15. ren says:

    When is growing energy troposphere tropopause is raised up. Air particles overcome the force of gravity.

    After a volcanic eruption may be in the stratosphere more solid particles which absorb thermal radiation, as the water in the clouds.

  16. ren says:

    From about 70 mbar with a temperature in the stratosphere decides to UV radiation and cosmic rays. Since the magnetic activity of the sun depends on the circulation in the stratosphere and sudden increases in temperature in winter. They are related to distort the polar vortex. The air there is very rare, so the temperature is the kinetic energy of particles.
    It is different from the temperature of the stratosphere above the equator.
    You can see a very wide belt tropopause, where the lack of ozone.

  17. Karl-Heinz Dehner says:

    It might be interesting to consider also the temperature record compiled by meteorological reanalysis.

    “Reanalysis is a scientific method for developing a comprehensive record of how weather and climate are changing over time. In it, observations and a numerical model that simulates one or more aspects of the Earth system are combined objectively to generate a synthesized estimate of the state of the system. A reanalysis typically extends over several decades or longer, and covers the entire globe from the Earth s surface to well above the stratosphere.”

    An overview is given at

    Input data not only includes synoptic observations at climate stations, but also radiosonde, satellite, buoy, aircraft and ship reports. Observations over land are blended with values from the background forecast model over sea. Both from ERA-Interim und JRA-55 time series of regional or global surface air temperature (2m above the ground) can be compiled. The time series can be downloaded at

    (unfortunately only to end of 2014 respectively 2013).

    A description of the ERA-Interim Reanalysis and it’s results can be found at

    An attached paper (Comparisons of surface temperature datasets January 2016.pdf) provides a comparison of the time series 12-month running mean until October 2015 of conventional surface temperature data sets (HadCRUT4 and NOAAGlobalTemp) and climate reanalysis ERA-Interim and JRA-55 results.

  18. erl happ says:

    Hi Euan,
    I read your post with interest. My advice to you based on exploration of the data here: http://www.esrl.noaa.gov/psd/cgi-bin/data/timeseries/timeseries1.pl

    Sea surface temperature, surface air temperature and the temperature of the atmosphere at all levels up to about 30hPa (encompassing 97% of the volume, varies primarily according to latitude and time of year.

    For a brief summary see this:https://reality348.wordpress.com/2016/01/13/7-surface-temperature-evolves-differently-according-to-latitude/ and the first part of https://reality348.wordpress.com/2016/01/13/7-surface-temperature-evolves-differently-according-to-latitude/

    When we look at global data, we don’t see the complex patterns that are lost in averaging. Its the patterns within the average that reveal modes of behaviour and modes of causation.These patterns have a much greater amplitude in their variation than the global temperature statistic.

    Co2 is well mixed. Ozone is not. Gordon Dobson who put a spectrophotometer to work to measure the total column ozone discovered that ozone mapped surface pressure. Surface pressure dictates the pattern of the winds and it is the temperature of an air mass that constitutes the environment that we surface dwellers experience as we go about our daily work. Our planet is warm in the middle and very cold at the poles. According to the direction of the compass from which it comes the air is warm or cold.

    The ozone content of the upper portion of the atmospheric column dictates surface pressure. Surface temperature in the mid and low latitudes varies directly with surface pressure.

    There are no arguments that I have seen about mysterious fiddlings with the surface pressure record. In any case, who would be worried about parts of a millibar when the variation is from about 950mb through to 1030mb.

    So, my basic message is that geography and time within the annual cycle is all important. Without an understanding of the spatial relations that manifest on an orb spinning in space, tilted in its axis of rotation to the plane of its orbit the best data and the best mathematical minds that approach from the simplistic point of view that an average is the appropriate metric will inevitably fail to engage with the problem in a meaningful fashion.

    There is an argument going on as to whether the globe has warmed or cooled since 1998. If we dip our thermometer into the waters in the tropics we will discover that for 8 of the 12 months of the year the sea is cooler in the decade up to 2014 than it was in the immediately prior decade.The remaining months are followers, not leaders. The data is contained in the last two graphs on this post: https://reality348.wordpress.com/2016/01/13/6-the-poverty-of-climatology/

    • Euan Mearns says:

      Erl, I’m with you on complexity and the shortcomings of averaging. Where I live, Aberdeen Scotland, 57˚N, the jet stream is sometimes overhead, sometimes to the N and sometimes to the S. Simply put, in winter if the jet is to the S we get cold weather and if it is to the N we get mild weather in winter.

      In the last decade the jet stream appears to have become more meandering forming deep loops. This means it can move more from E to W rather than N to S. Our winter temperatures can shift from +20˚C if the wind is blowing from Spain to -10˚C if it blows from Svalbard. What meaning does a monthly average have?

      My “belief” is that the change in behaviour (if indeed the behaviour has changed) may be down to a change in the spectrum of energy leaving the Sun. And this could link to ozone production?

      I’d certainly be interested to learn more about this:

      Gordon Dobson who put a spectrophotometer to work to measure the total column ozone discovered that ozone mapped surface pressure.

      • erl happ says:

        Yes, ozone mapped surface pressure. Unfortunately, Dobsons superior at Oxford became co chairman of the scientific assessment working group at the IPCC. Sir John Houghton was lead author of the first three IPCC reports and any thought that the stratosphere was involved in determining the direction of the wind was something that he was not prepared to entertain. Though Dobson was a private researcher he had a chair in meteorology at Oxford and perhaps he simply enjoyed teaching. Anyway, he toed the party line.

        But eventually people who seek truth begin to observe that the world is not necessarily as Sir John would have it Today, there is a school of climate science trying to make sense of the meanderings of the jet stream who talk about the inter-annual modes of surface pressure variation of to put it another way, the exchange of atmospheric mass between high and other latitudes that is observed to be the prime mode of inter-annual (year to year) variations in climate. They talk of a mysterious ‘coupling’ between the stratosphere and the troposphere that involves change occurring first high up and propagating down to the surface that they admit they do not understand.

        Ye gods. Dobson could have worked this out long ago. What differentiates the stratosphere from the troposphere is its ozone content. It is the ozone content of the atmospheric column that gives rise to surface pressure variations. We have a chicken and egg problem. There are a bunch of diehards who insist that all causation must be bottom up in nature.So, a bunch of mathematicians talk of ‘planetary waves’ and heat flux from low latitudes being responsible for warmings in the stratosphere and the atmospheric shift mob saying that the heating starts aloft and descends and a big group in the middle who don’t understand what either of these groups is on about.

        It is supposed (by Sir John and his followers) that it is short wave radiation that heats the stratosphere and that’s true enough at 10hPA and above where it is the prime mode of heating. But ozone is a greenhouse gas that absorbs at 9-10um at near the peak of the Earth’s infra-red emission. And when the sun sinks below the horizon parts of the winter atmosphere is warm…that part containing ozone. And part is cold, that part that descends from the mesosphere. NOx from the troposphere and the mesosphere destroys ozone. We need to look at the composition of the incoming air and its flow rate to work out why the temperature of the stratosphere changes. Nothing to do with CO2 in the troposphere cooling the stratosphere. The infra-red emissions from the Earth are not exhausted by absorption in the troposphere. There is no crossover between the wave lengths absorbed by CO2 and ozone. The notion of a warming stratosphere resulting in a cooling stratosphere is BS. They get away with it by lumping all the data into a single bucket called the global stratosphere. In truth there is little variation in the temperature of the stratosphere in low latitudes and one hell of a lot at the poles in winter.

        Ozone like oxygen is destroyed by photolysis by a longer wave length in the UVB. As the sun sinks low in the sky the atmospheric path gets longer and UVB is used up just as very short wave lengths that photolyze oxygen do not make it into the stratosphere. So, ozone proliferates in winter. The degree to which it proliferates determines the shifts in atmospheric mass that occur. Ozone is responsible for the generation of polar cyclones. Collectively, more intense polar cyclones lower surface pressure in high latitudes and are responsible for shifts in atmospheric mass.

        The Antarctic lost about 15mb of surface pressure between 1948 and 1998. This was accompanied by a strong warming of the stratosphere. At 10hpa temperature peaked in 1978. At 30hpa it peaked in 1998.

        There are three modes of atmospheric heating that impact differently according to location. One is through contact with a warm surface, one by the release of latent heat and another heating by ozone. The third is manifestly the most powerful.Polar cyclones regularly generate wind strengths that are equivalent to a tropical cyclone. Wind velocity increases with elevation. That indicates the location of the forces responsible.

        One more thing. There is no tropopause in high latitudes….its all stratosphere because the presence of ozone changes the lapse rate all the way down to the surface. The air at the poles is warmer than the surface and particularly so in Antarctica. If you look at the atmosphere analytically there is only one part of the Earth where there is manifestly a tropopause and that is at the equator. In mid and high latitudes high and low pressure cells do not respect this supposed ‘tropopause’ A tropopause in mid latitudes is BS. Ozone descends in high pressure cells in the mid latitudes and as it heats the air cloud cover falls away. That represents climate change. Habitually, it occurs in winter when ozone partial pressure grows at the winter pole.

        Sorry, I get a bit passionate about this….and impatient, and bad mannered too. Thanks for the question. I hope the answer is meaningful and if its not I would appreciate you telling me so I can have another go.

  19. Karl-Heinz Dehner says:

    I have just learned that the UAH global average lower tropospheric temperature (LT) anomaly for January, 2016 is +0.54°C, up from the December, 2015 value of +0.45°C (with regard to base period 1981-2010), the highest anomaly observed for January so far:
    However, it should be noted that it is still below the anomalies observed for February and April 1998. The rise from December 1997 to January 1998 was +0,23°C, whereas now amounts only to +0,1°C. Perhaps this is due to a phase shift in the El-Nino-events.

  20. robertok06 says:

    Recent testimony of Prof. Christy in front the U.S. House Committee on Science, Space & Technology, yesterday, interesting…


    • erl happ says:

      A man of uncommon good sense. That the AGW proponents can persist in with their benighted foolishness in the face of his testimony is thoroughly disconcerting.

  21. Jim says:

    Very interesting post and discussion. I have two worries that you could put to rest. Firstly the discussion is all about radiation but all engineers know that convection is more important. Greenhouses work by stopping convection. Secondly, a standard atmosphere, needing only mass of atmosphere and gravity reproduces the earth’s temperature profile of the atmosphere pretty well. I’m sure there are easy answers but I don’t know them!

    • Euan Mearns says:

      Jim, thanks for reminding us about convection. I don’t believe that convection can conceivably remain constant. And I believe we are witnessing a resetting of atmospheric circulation with a shift towards the meridional mode and this must surely affect convection / convection rate in ways beyond the imagination of climate scientists.

      As for the atmosphere, it is the properties of water that drive the IR / bottom up part of the temperature profile and CO2 may be vital to keep liquid water on the surface. Once you have liquid water it drives everything. The top down profile I believe is driven by solar activity (spectrum) and ozone. But all this is way beyond my detailed understanding.

      As an aside, if you don’t have a wood burner at home, now would be a good time to buy one. 6.8 GW of UK coal going offline in March.

  22. michael says:

    Looking at all the nice grafts the data looks to me as part of what in electronics we would say is a wave shape, perhaps not a sine wave but something similiar. Here is what I am talking about in ocean rise and fall trends. http://tidesandcurrents.noaa.gov/sltrends/global_50yr.htm?stnid=680-140. All in all it seems the data is to short in length to actually make a fact based statement aside from all the adjustments NASA may or may not have made on the data.

  23. Tom Bates says:

    Looking at all the nice grafts the data looks to me as part of what in electronics we would say is a wave shape, perhaps not a sine wave but something similiar. Here is what I am talking about in ocean rise and fall trends. http://tidesandcurrents.noaa.gov/sltrends/global_50yr.htm?stnid=680-140. All in all it seems the data is to short in length to actually make a fact based statement aside from all the adjustments NASA may or may not have made on the data

  24. Pingback: AWED Energy & Environmental Newsletter: February 15, 2016 - Master Resource

  25. Pingback: Recent Energy And Environmental News – February 15th 2016 | PA Pundits - International

  26. Hello Euan – great blog.

    This may or may not add to the collective wisdom on stratospheric cooling.

    We tend to conceive of increased GHG forcing from the surface to the ‘top of the atmosphere’ ( or perhaps the top of the troposphere ).

    But the process of absorption emission is not for just this layer.

    That is, some of the IR from the lowest 50mb of the troposphere is absorbed at 500mb, some absorbed at 300mb, and some makes it to space. This is true for all the layers.

    Thus, two variables are defined in the radiative models:
    1. the radiative flux that passes through a level and
    2. the heating(cooling) rate of a layer, bounded by two levels, determined by the net flux absorbed.

    I ran the ‘Column Radiation Model’ (CRM) on a given atmospehere, and the same with CO2 doubled.
    I acheived this plot of RF, similar to others:

    For the same circumstance, i generated this plot of Heating Rate:

    ( BTW, the change in cooling rate at 1mb is much greater than at 100mb or even 50mb )

    The layers in the stratosphere, cool very rapidly ( in comparison with 1x CO2 ) because they emit more than they receive from lower levels. This is due to CO2 opacity and also the temperature profile.

    The 1x to 2x change in cooling rates is very high ( ~ 10C per DAY! ).

    BTW, the absolute cooling rate is very high also as shown in this Clough Associates chart of cooling rates:

  27. Tom Bates says:

    I see a lot of comments about 85 balloons. from the RATPAC site that is not actually true. They say they removed 29 stations for lack of metadata as defined by them. Apparently they only have continuous data from 35 stations after 1995 but why that caused 29 stations to be thrown out is unstated. The other concern is they used one method up to 1995 to massage the data and than without changing the old method data to the new started a new method thereby making the old data incomparable with the newer data, most likely the reason their graft starts in 1996. Since Hansen got his fingers into the works around that time several things involving fraud come to mind but heck what do I know.

Comments are closed.