First of all, consider the source, UNSW is the same outfit that sponsored the disastrous “ship of fools” aka The Spirit of Mawson.
Secondly, Dr. Roy Spencer has been looking for this for years in the satellite data and hasn’t found it.
Thirdly, radiosonde coverage in their area of study is pretty sparse. From the University of Graz:
While many users are familiar with traditional radiosonde temperature and moisture data, the spatial and temporal coverage of radiosonde data are limited, especially over ocean and high latitude areas. Satellite remote sounding provides far greater temporal and spatial coverage of the entire planet.
Fourth, if they have really found it, where’s the picture or graph of it in the press release? You’d think that would be front and center. Instead, it isn’t shown, and they don’t even mention the title of the paper or the DOI. Essentially they are saying “trust us, no need to read the paper”. I’ve looked for the paper on the ERL website, and have yet to locate it. It is not listed in today’s ERL news feed (as of this writing) (UPDATE: located, and the abstract is posted below). It’s like Lewandowsky’s seepage paper, that had a press release over a week ago, but the paper is still not published.
Fifth, Steve Sherwood is a well known climate alarmist, and his confirmation bias seems quite strong to me. For example, see this WUWT post where we state “Professor Sherwood is inverting the scientific method”.
Sixth, by their own admission, they had to throw out data, and to do a series of adjustments to station data to find the signal they were looking for. That sounds more like a selecting process in the scope of confirmation bias than science. (added) The real question is, how many stations did they keep as they define as “good”?
Color me skeptical, I’m sure Dr. Roy Spencer will have something to say about it.
From the University of New South Wales:
New publicly available dataset confirms tropospheric hot spot and increased winds over Southern Ocean
Researchers have published results in Environmental Research Letters confirming strong warming in the upper troposphere, known colloquially as the tropospheric hotspot. The hot has been long expected as part of global warming theory and appears in many global climate models.
The inability to detect this hotspot previously has been used by those who doubt man-made global warming to suggest climate change is not occurring as a result of increasing carbon dioxide emissions.
“Using more recent data and better analysis methods we have been able to re-examine the global weather balloon network, known as radiosondes, and have found clear indications of warming in the upper troposphere,” said lead author ARC Centre of Excellence for Climate System Science Chief Investigator Prof Steve Sherwood.
“We were able to do this by producing a publicly available temperature and wind data set of the upper troposphere extending from 1958-2012, so it is there for anyone to see.”
The new dataset was the result of extending an existing data record and then removing artefacts caused by station moves and instrument changes. This revealed real changes in temperature as opposed to the artificial changes generated by alterations to the way the data was collected.
No climate models were used in the process that revealed the tropospheric hotspot. The researchers instead used observations and combined two well-known techniques — linear regression and Kriging.
“We deduced from the data what natural weather and climate variations look like, then found anomalies in the data that looked more like sudden one-off shifts from these natural variations and removed them,” said Prof Sherwood.
“All of this was done using a well established procedure developed by statisticians in 1977.”
The results show that even though there has been a slowdown in the warming of the global average temperatures on the surface of the Earth, the warming has continued strongly throughout the troposphere except for a very thin layer at around 14-15km above the surface of the Earth where it has warmed slightly less.
As well as confirming the tropospheric hotspot, the researchers also found a 10% increase in winds over the Southern Ocean. The character of this increase suggests it may be the result of ozone depletion.
“I am very interested in these wind speed increases and whether they may have also played some role in slowing down the warming at the surface of the ocean,” said Prof Sherwood.
“However, one thing this improved data set shows us is that we should no longer accept the claim that there is warming missing higher in the atmosphere. That warming is now clearly seen.”
###
UPDATE: The paper has been located. http://iopscience.iop.org/1748-9326/10/5/054007
Atmospheric changes through 2012 as shown by iteratively homogenized radiosonde temperature and wind data (IUKv2)
Steven C Sherwood and Nidhi Nishant
Letter
We present an updated version of the radiosonde dataset homogenized by Iterative Universal Kriging (IUKv2), now extended through February 2013, following the method used in the original version (Sherwood et al 2008 Robust tropospheric warming revealed by iteratively homogenized radiosonde data J. Clim.21 5336–52). This method, in effect, performs a multiple linear regression of the data onto a structural model that includes both natural variability, trends, and time-changing instrument biases, thereby avoiding estimation biases inherent in traditional homogenization methods. One modification now enables homogenized winds to be provided for the first time. This, and several other small modifications made to the original method sometimes affect results at individual stations, but do not strongly affect broad-scale temperature trends. Temperature trends in the updated data show three noteworthy features. First, tropical warming is equally strong over both the 1959–2012 and 1979–2012 periods, increasing smoothly and almost moist-adiabatically from the surface (where it is roughly 0.14 K/decade) to 300 hPa (where it is about 0.25 K/decade over both periods), a pattern very close to that in climate model predictions. This contradicts suggestions that atmospheric warming has slowed in recent decades or that it has not kept up with that at the surface. Second, as shown in previous studies, tropospheric warming does not reach quite as high in the tropics and subtropics as predicted in typical models. Third, cooling has slackened in the stratosphere such that linear trends since 1979 are about half as strong as reported earlier for shorter periods. Wind trends over the period 1979–2012 confirm a strengthening, lifting and poleward shift of both subtropical westerly jets; the Northern one shows more displacement and the southern more intensification, but these details appear sensitive to the time period analysed. There is also a trend toward more easterly winds in the middle and upper troposphere of the deep tropics.
The paper is open access: http://iopscience.iop.org/1748-9326/10/5/054007/pdf/1748-9326_10_5_054007.pdf
Here is the figure from the paper that should have been in their press release:
The SI is pretty thin, containing a single figure with no explanation: http://iopscience.iop.org/1748-9326/10/5/054007/media/erl054007_suppdata.pdf
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ thank you
Weak skepticism
“First of all, consider the source, UNSW is the same outfit that sponsored the disastrous “ship of fools” aka The Spirit of Mawson.”
1. guilt by association. This is actually the inverse of appeal to authority. You dont have to examine methods, just see what organization they belong to. Its the same logic that is used to dismiss skeptical
papers.
Secondly, Dr. Roy Spencer has been looking for this for years in the satellite data and hasn’t found it.
1 two different datasets.
2. Radiosondes are actual temperature measurements. Satellites estimates of temperature are based
on models and adjustments. hmm and they are checked against radiosondes..
Thirdly, radiosonde coverage in their area of study is pretty sparse. From the University of Graz:
1. you cant simply claim its sparse. one the reasons why Spenser ( for example ) is able to use radiosondes to cross check satellite is that the variability of temperature over spatial regions is such that
a small network can give you a good representation. In order words when the radiosonde data showed there was no hot spot, you never complained about the spatial sampling. Nobody ever said “maybe the hot spot” is smaller than the sampling can see.
‘Fourth, if they have really found it, where’s the picture or graph of it in the press release? You’d think that would be front and center. Instead, it isn’t shown, and they don’t even mention the title of the paper or the DOI. Essentially they are saying “trust us, no need to read the paper”
1. you cant judge science by what the PR department decides to DO or Not to DO. The paper is the advertisement for the science. the Press release is advertisement for the advertisement.
2. They are not saying trust us, quote their words.
3. Even IF they said trust us, that has no bearing on the facts of the matter
“Fifth, Steve Sherwood is a well known climate alarmist, and his confirmation bias seems quite strong to me.”
1. funny, nobody objected to using his previous data set.
2. The confirmation bias runs two ways. you expect bad things from him, so no doubt that you will invent them if you can find them
“Sixth, by their own admission, they had to throw out data, and to do a series of adjustments to station data to find the signal they were looking for. That sounds more like a selecting process in the scope of confirmation bias than science. (added) The real question is, how many stations did they keep as they define as “good”?”
1. here is a description of the KINDS of things people do:
“[8] We examined the data records of the 183 tropical
stations available in the Integrated Global Radiosonde
Archive (IGRA) database at the National Climatic Data
Center [Durre et al., 2005]. To use a sonde profile to
simulate LT, we required that it reach at least 100 hPa. To
include a station, we required that it have at least 180 of the
possible 312 months of data. Enforcing these criteria
reduced the number of stations to 73. Comparing the sonde
and satellite series for consistency we eliminated the 43000
block (India) as being unacceptably noisy (as in Parker et
al. [1997] and Lanzante et al. [2003]). We were left with the
58 stations shown in Figure 1 and described in Tables S1
and S21
. Of these, 29 provided observations for both day
and night, 28 for day only, and one for night only. ”
WAIT! that is how Spencer threw out data when he looked at Sondes!!
Please note that NOBODY here objected to what Spencer did with Sonde data.
yes raw observations are CRAP. Every day in business I have to deal with raw observations. The first step is always screening and cleaning. and then adjusting.
Mosher all of your points are pointless.
The bottom line is there is NO tropical hotspot and there is a negative feedback between upper atmospheric water vapor concentrations and increasing CO2.
This is why OLR emissions to space are NOT decreasing in response to an increase in CO2.
But AGW proponents will say the data is of course either wrong ,inaccurate or needs to be adjusted so it will fit in with their theory as you have just tried to do in this post.
Throw out data if it does not support the theory rather then looking at the theory to see why the data does not support it. A backwards approach and an absurd approach when it come to science.
I find Gavin’s thought on this topic interesting: The CO2 hotspot is the same as any other:
https://twitter.com/ClimateOfGavin/status/438001562766413824
Now that’s funny. That little Andrew has the same name as an important warmunist researcher in Texas.
http://en.wikipedia.org/wiki/Andrew_Dessler
Next lie.
All
method, in effect, performs a multiple linear regression of the data onto a structural model that includes both natural variability, trends, and time-changing instrument biases, thereby avoiding estimation biases inherent in traditional homogenization methods.
I’ve never heard of iterative universal kriging. Universal kriging yes, and if you’re using it then you’re probably using a search neighbourhood. This is far from ideal to begin with. Either way they are effectively solving a unique linear systems per grid point which incorporates a local structural component – the sum of the structural parts is unlikely to be the global trend (global structural modal).The issue here is that the structural trend is assumed to be absolute but it isn’t – it is in itself a statistic and a series of local ones. So it comes with its own biases etc.
I never understand why researchers just don’t use a global trend and perform simple kriging with the residuals then add the trend back in. But either way the point stands that using kriging – any type of kriging – is not a silver bullet. It is an excellent method but if you’re using a structural component that has an amplitude that is an order of magnitude greater than the variance of the residuals about this trend then you’re effectively determining the “static” in the signal. The issues lie with your structural component that gets lost in the universal kriging method (and the assumptions therein). In short we need to model the structural component first and assess this and the experimental issues there.
…need to add before someone puts me in my place. If they’re solving for each grid using the global set then the UK will be akin to just detrending using the same type of structural model and performing SK and then adding the trend back in.
Garth Paltridge did a paper looking at all the weather balloon data which was available for about 50 years and couldn’t find much evidence that as the Earth had warmed slightly that vital increase in water vapour was there. He eventually had it published, but when it was first submitted for publication, it was rejected on the basis that the message that it would send would give too much encouragement to sceptics, which really just draws attention to the need to open up the scientific process, to deal with this kind of attempt to politicise it, to suppress views that are inconvenient.
This latest claim about a hotspot is nonsense and no doubt, if and when the evidence behind the UNSW claim is independently scrutinized, it will be found to be no more credible than the 97% consensus claim. This latest claim is just all part of the overall effort to raise alarm in time for the December Paris conference.
Here is the HadAT database for the weather balloon data going back to 1958.
The Hotspot(s) that Sherwood found are at the 300 mb level or the average height that Channel 3 shows here.
Channel 3 trend is effectively Zero.
http://hadobs.metoffice.com/hadat/images/update_images/msu_timeseries.png
We can also get a more detailed latitude breakdown from RSS going back to 1987 (TTS or Channel 3 or 300 mb again).
Tropics. Nothing.
http://data.remss.com/msu/graphics/tts/plots/RSS_TS_channel_TTS_Tropics_Land_and_Sea_v03_3.png
Southern mid-latitudes. Negative.
http://data.remss.com/msu/graphics/tts/plots/RSS_TS_channel_TTS_Southern%20Mid%20Latitudes_Land_and_Sea_v03_3.png
Hence, one should be able to conclude that there is other data which completely contradicts Sherwood’s finding of the hotspot and he will need to show everyone exactly what he did in this paper or it will go into the dustbin like his previous attempts did.
Exactly!
Yes, into the dustbin for sure. BUT. There will be NO retraction of the paper and there will be constant referrals to the ‘results’ by the EcoMarxist crowd as if it were valid. sadly
I had an unsatisfactory relationship with Krige over my analysis of one of his databases. Kriging does not handle skew data very well. I believe RSA gold mines are largely a special case. Maybe one should not extrapolate to other continents with abandon. Errors of sampling in geostatistics are transferred to the answers one gets (Clark). He actually is in print defending his methodology and saying no further research need be done.
Here is what Peter Miller stated over at Jo Nova:
“Peter Miller
May 16, 2015 at 5:34 pm · Reply · Edit
Jo
I did my dissertation for my Master’s degree under Professor Krige, who at the time was a consultant for the Anglovaal mining company in South Africa.
He was a truly brilliant individual, so one day I asked him about his system of statistical analysis, which is now known as kriging. My reason was that I had just calculated the same gold reserve figure for a new mine called Deelkraal, but I knew instinctively (as I had worked on a nearby mine) that the numbers were too high.
He responded by saying it only works if you are using statistics from the same population group and if you adjust your result by a known constant.
I knew the Deelkraal drill results divided into two population groups and redid the calculation, which resulted in a much reduced grade and tonnage, which is exactly what happened when the mine was in commercial production.
As for the constant needed for adjustment, this was a figure that had been calculated for this particular orebody in adjacent mines. I cannot remember if this constant was a negative or positive figure, but I suspect it was negative.
So to return to Sherwood’s tosh. You cannot use kriging for results which are a mixture of population groups, such as over the land, over the oceans, or over the poles. I very much fear different elevations/atmospheric pressures should also not be combined.
Also, because the sun is the principal driver of our planet’s temperature, I suspect you should be dividing the kriging analysis into bands of latitude.
And you very definitely should not exclude correctly gathered data just because they contain inconvenient numbers.
As for the correcting constant, I guess you have to go to Planet Zarg, wherever that might be, to get it.
My conclusion on Sherwood’s ‘research’ paper? Mannish at best, complete BS most likely.”
http://joannenova.com.au/2015/05/desperation-who-needs-thermometers-sherwood-finds-missing-hot-spot-with-homogenized-wind-data/#comment-1711282
The key words appears to be..” He responded by saying it only works if you are using statistics from the same population group and if you adjust your result by a known constant.”
http://icecap.us/images/uploads/Crux_Flawed_Science.pdf
This is the reality which is there is probably a NEGATIVE feedback between increasing CO2 concentrations and upper atmosphere water vapor concentrations.
What the authors have done is “methods shopping”. They have applied different numerical techniques to the data until it delivered the answer they were looking for – the hotspot.
“methods shopping” is simply a form of “cherry picking”, in which the computer, not the researcher picks the cherries. as a result the researcher is often mislead into thinking there is no bias in the result.
however, if the method finally “bought” is valid, then the pattern revealed should also be visible using a wide variety of other methods. If only one method delivers the answer, then that is a strong indication that the result are simply an artifact of the method, not supported by the underlying data.
Hear hear!
“The new dataset was the result of extending an existing data record and then removing artefacts caused by station moves and instrument changes.”
Am I missing something? He’s looking at Radiosonde data. He’s looking at weather balloons. Balloons Move!
Is he tagging on some kind of adjusted ground station dataset, ala Mann?
Call me stupid, but if the oceans are giving off heat, (and I can tell because, for example, the surface across the Pacific Equatorial band is warmer than usual and all that evaporating water and heat are making for clouds…really big ones), it would stand to reason that this heat is then rising to warm the atmosphere. If it wasn’t, and just stayed in the oceans never to come out, then we would indeed not ever want to visit states and countries along the Gulf Stream, let alone the scantily clad beaches lining extravagantly clad tropical resorts.
I think what was found was that our oceans are losing heat. It’s sort of like thinking you have discovered some new orifical phenomenon only to find out it was your own ass.
Just read the abstract, will look closer into the paper later. But I may have found an autocorrelation. The authors state that, “One modification now enables homogenized winds to be provided for the first time.” I believe they mean the trades. If these indeed were a part of the modeled framework, heat would indeed invade the troposphere under calm El Nino trade wind conditions.
Yes, Pam, it is certainly not enough to find heat in the upper troposphere. It also must be caused by absorption of LWIR by CO2. Convection/thunderstorms taking heat up from the sea surface raises the temperature of the troposphere, even the top of troposphere,on its way to radiating to space, as do the trade winds and volcanoes. This is another cobbled together paper anticipating the Paris clatch. Oh the desperation at the moorage of the ship of fools, the centre of excellence indeed.
Hi Pam,
Well. If nothing else, at least the analogy in that last sentence (@9:09am) got me to pay attention. I’m still trying to figure out what kind of analogy it is, exactly.
Otherwise, yes, oceans emit heat.