Hot Spot 2 – Tropospheric Warming Continues

Hotspot Found!

… by Nova. She just didn’t realise it.

Often you will hear Joanne Nova comment about the “missing hotspot” as if  this were evidence against global warming. It’s not. This has been addressed numerous times (1, 2, 3, 4, 5, 6, 7, 8).

Here we examine each of her posts (tagged as Hot Spot) and explain why they are incorrect or misleading.

2008 Oct. The missing hotspot

In this post Nova makes a repeat of her partner’s (David Evans) assertions that the tropospheric atmosphere is not warming. Jo’s view of the hotspot is based only a small subset of available data, and she ignories the documented problems with long term radiosonde analysis. Thrown for good measure is Nova’s typically deceptive behaviour.

40 Year Expectation vs 20 Years Observation

Nova’s first deceptive move was to compare oranges and half-oranges, in this instance the images she compares are not of the same timeframe. The image for the model shows warming expected from 1958-1999. The radiosonde observation are from 1979-1999; 21 years less – not quite half of the time shown for the model projection. even if the models were perfect, this comparison would not show the same amount of warming.

How do we know she is being deliberate in her deception? Because the image for the observations was taken from page 116 of the US CCSP
2006 report (as Nova cites) and in that report, immediately above the observations, are expectations from four different models, all using the same timeframe as the observations. Nova chose the IPCC report image because, A, it was a longer timeframe and showed more expected warming, and B, because three models did not expect as much warming as the fourth. All four models from the US CCSP 2006 report are shown below.

The four models compared to observations.

Signature of Warming vs Signature of Greenhouse Gas

The hotspot is not the signature of greenhouse gas, it’s a signature of any surface warming and would be created by any force that warms the surface. Realclimate showed this best by modelling the expected warming from an additional 2% solar output vs that of doubled CO2. The hotspot appears in both cases, not just from greenhouse gases. It is not a signature of greenhouse gas, but an expected result of any surface warming. The major difference between the two images is the stratospheric cooling, the purple cold patch in the top of the left hand image.

Source

Stratospheric Cooling

The real “signature” of greenhouse gas related warming is the cooling of the stratosphere (the purple cold patch in the top of the left hand image), and that is cooling, although the effect due to greenhouse gases is masked by the effect from ozone depletion. Read more here if you’re so inclined.

The Hot Spot Does Exist

Jo’s claim that weather balloons find NO SIGN is more than a little exaggerated. The hotspot shows up on seasonal, annual timescales and more recently in decadal timescales. It is surprising Nova doesn’t mention this, because it appears in the same paper that her observations are sourced.

The US CCSP state …

On short timescales (month-tomonth and year-to-year variations in temperature), the estimated tropospheric amplification of surface temperature changes was in good agreement in all model and observational data sets considered, and was in accord with basic theory.

Other well-known papers also find evidence for the tropospheric warming.

Allen & Sherwood (2008). A unique method that uses estimates based upon the wind shear of the radiosonde data, which does not suffer the same problems temperature data does, also finds tropospheric warming matching that expected in models.

Warming patterns are consistent with model predictions except for small discrepancies close to the tropopause. Our findings are inconsistent with the trends derived from radiosonde temperature datasets and from NCEP reanalyses of temperature and wind fields. The agreement with models increases confidence in current model-based predictions of future climate change.

Johnson & Xie (2010) used yet another method and conclude …

We conclude that, in contrast with some observational indications, the tropical troposphere has warmed in a way that is consistent with moist-adiabatic adjustment, in agreement with global climate model simulations.

Dessler (2010) …

The five reanalyses analyzed here … unanimously agree that specific humidity generally increases in response to short‐term climate variations (e.g., El Niño). In response to decadal climate fluctuations, the NCEP/NCAR reanalysis is unique in showing decreases in tropical mid and upper tropospheric specific humidity as the climate warms. All of the other reanalyses show that decadal warming is accompanied by increases in mid and upper tropospheric specific humidity.

Thorne (2010) looks at all papers on the topic and conclude …

It is concluded that there is no reasonable evidence of a fundamental disagreement between tropospheric temperature trends from models and observations when uncertainties in both are treated comprehensively.

Santer et al (2012) …

The multidecadal cooling of the stratosphere and warming of the troposphere, which is evident in all satellite datasets and simulations of forced climate change examined here, cannot be explained by solar or volcanic forcing, or by any known mode of internal variability.

But of course Nova will only consider data that supports her biased position, even if the data has been shown to be unreliable.

Radiosonde Data Problems

Why, when using radiosonde data, has the hotspot appeared on short seasonal/yearly timescales but not on decadal timescales?

Part of the problem is that the observational data for the troposphere is obtained from radiosonde data stitched together from datasets that have been gathered using different equipment and reporting techniques. Using the raw data is unlikely to produce accurate results.

The radiosonde data was found to be unreliable by Elliott et al 1991 who say …

Changes in biases in the data are highlighted, as these can lead to misinterpretation of climate change. We conclude that the upper-air data record for the United States is not homogenous, especially before 1973. Because of problems with the humidity data in cold, dry conditions, the water vapor climatology in the upper troposphere, nominally above the 550-mb level, is not well known.

Titchner et al. (2009) also studied the radiosonde data and found …

… tropical tropospheric trends in the unadjusted daytime radiosonde observations, and in many current upper-air datasets, are biased cold, but the degree of this bias cannot be robustly quantified. Therefore, remaining biases in the radiosonde temperature record may account for the apparent tropical lapse rate discrepancy between radiosonde data and climate models.

Satellite data suffers from similar problems with adjustments required for orbital decay and homogenisation of different satellite sources. Over time a number of corrections have been made (Mears et al 2003, Mears et al 2005, Po-Chedley & Fu 2012), and yet more may be required.

2009 Jan. Reply to Deltoid

Joanne Nova confuses fingerprints

In this post Joanne Nova confirms her ignorance of the fingerprinting of greenhouse gases (GHG). Joanne says that the stratospheric cooling could be caused by ozone, but if we look at the model expectations, GHG, they look quite different to Ozone. The images below show the response to (A) GHG only, (C) Ozone & (F) All forces combined.

US CCSP 2006, Chapter 1 (as used by Nova)

Observations show that the stratospheric cooling (large purple patch at the top) matches the models more closely for all forcings, including GHGs.

Observations from 1979 – 1999 match more closely with GHG forcings.

2009 Jan. Even gurus of warming

In this post Nova says …

Big names like Santer, Sherwood, and Schmidt admit that the models predict more warming 10 km above the equator than what the weather balloons could find. Each time they announce that they’ve resolved the differences, they have to start by admitting there are differences to resolve.

Correct! And, as explained above, the experts go even further to show that the weather balloon data is unreliable and that when more reliable analysis is performed using multiple sources of data, tropospheric warming exists. Something Nova somehow “neglects” to mention.

2010 Apr. No, Dr Glikson etc…

Nova, lead clown of the circus and children’s science presenter, attempts to correct Dr Andrew Glikson, an Earth and Paleo-climate Scientist with a vast list of publications.

In her argument with Dr Glikson, Nova brought nothing new to the table, however she did conceed that warming in the troposphere does exist …

Glikson apparently doesn’t understand that the upper tropospheric graph is supposed to show a higher rate of warming than the surface graph. Instead it’s about the same.

Nova concurs that both the surface and troposphere are warming, however, in her haste to point out that the lack of a higher rate, forgot about uncertainty in both the models and the data. Oh well, it’s a start.

During the back and forth Nova also makes reference to “science” by the NIPCC, a few scientists that create their own non-peer-reviewed unpublished research. In this case Nova references NIPCC “research” located (her link is now broken) on the Heartland website. The Heartland Institute is a political group for hire, not a journal for scientific research.

2010 Jun. How John Cook …

This time Nova repeats the usual arguments and still refuses to look at any data except the flawed raw radiosonde data. She repeats the comparison of data that are on different timescales, and has yet to understand that the hotspot would be caused by any surface warming, not just greenhouse gases.

Rather than scientifically rebutting the Allen & Sherwood paper, Nova calls the paper “desperate”. She offers no quantitative reason as to why the method would be invalid, and nothing to counter the fact that it avoids the homogenous problems that temperature data faces.

But Nova does suggest that the research is bad, because the government has funded it.

It’s the kind of reasoning you get when a government pours $79 billion into supporting a theory and none into finding holes in it.

Nova can’t actually fault the work, so instead plays the usual “Merchant of doubt”, “Big Government” card and then repeats science using the un-corrected radiosonde data.

2010 Jul. Sherwood 2008

By combining the wind shear data and temperature data Sherwood find the model expectations and observation match more closely, however it’s Sherwood’s use of red that has Nova’s nostrils flaring. Of a previous graph she also remarks “even if pink is a questionable choice for the ultra cold”.

Nova repeats her previous incorrect comparison of data from different timescales and repeats her mantra “the balloons (that have homogenisation problems) don’t show the hotspot” – surprise!

Sherwoods work reduces the amount of error with the radiosonde data and finds stronger tropospheric warming where better radiosonde data is available. Nova still prefers the raw unhomogenised data because it agrees with her ideology – that’s not science Nova!

Nova says “There is no justification in the paper for saying that the adjusted sonde data now finally agrees “broadly” with the models.”, but of course that’s her interpretation, not a statistical analysis of the paper. The “ill-coloured” graph Nova disliked was the evidence, but perhaps Nova was too sidetracked to notice?

Nova also says “and almost all the reanalyzing seems to be in a non-random model-friendly direction”, reaffirming her predisposition that models could never be correct no matter what the science says.

2010 July. The Unskeptical Guide

Nova continues her reliance on raw data from weather balloons (that are known to have data homogenisation issues) and continues to ignore other evidence for the tropospheric warming.

Nova again spends considerable time chatting (to herself?) about whether the fingerprint is unique or not.

2010 Aug. The models are wrong

Here Nova introduces a new paper, by economic statisticians, hoping to blur the picture by making comparisons against the “model mean”. Why is this going to result in the wrong answer? Because it’s asking the wrong question.

Here’s an analogy. 3 runners run 100 meters races many times over. The stats are (fictional values only):

• Runner A. Average time 10±2 seconds (between 8 & 12 seconds)
• Runner B. Average time 11±2 seconds (between 9 & 13 seconds)
• Runner C. Average time 15±2 seconds (between 13 & 17 seconds)
• Runner Ensemble: 13±2 seconds.

Using only the “ensemble mean” of 13±2 seconds, we would be incorrect to say that no runner can run faster than 11 seconds; we know that both Runner A and B had faster times. By only using the average, we come to a very wrong conclusion. So does Nova.

But it gets worse. A single climate model will produce different results over short timeframes due to “weather” events. Only by running the same climate model multiple times will you achieve their true average. For most of the models MMH 2010 only make one single run and accept that to be representative of that model’s mean – which it’s not.

But again it gets worse, the models are a variety and not all of them were using the same forcings. Ozone depletion, solar changes, land use, volcanic eruptions; some models included some of these, others left them all out. It’s little wonder they arrive at vastly different answers and little wonder many don’t come close to actual observations.

By comparing an ensemble “model average” to observations MMH 2010 and Nova are effectively saying Usain Bolt cannot possible exist.

2010 Oct. Is the Western Climate Establishment Corrupt?

Nova reiterates the topics above as part of her “Big Government, Corruption, Cover-up, Conspiracy” series. Nothing new to see here.

2010 Nov. Dessler

From Dessler’s paper …

The five reanalyses analyzed here … unanimously agree that specific humidity generally increases in response to short‐term climate variations (e.g., El Niño). In response to decadal climate fluctuations, the NCEP/NCAR reanalysis is unique in showing decreases in tropical mid and upper tropospheric specific humidity as the climate warms. All of the other reanalyses show that decadal warming is accompanied by increases in mid and upper tropospheric specific humidity.

We conclude from this that it is doubtful that these negative long‐term specific humidity trends in the NCEP/NCAR reanalysis are realistic for several reasons.

• First, the newer reanalyses include improvements specifically designed to increase the fidelity of long‐term trends in their parameters, so the positive trends found there should be more reliable than in the older reanalyses.
• Second, all of the reanalyses except the NCEP/NCAR assimilate satellite radiances rather than being solely dependent on radiosonde humidity measurements to constrain upper tropospheric humidity.
• Third, the NCEP/NCAR reanalysis exhibits a large bias in tropical upper tropospheric specific humidity.
• And finally, we point out that there exists no theoretical support for having a positive short‐term water vapor feedback and a negative long‐term one.

Nova writes …

Essentially this graph reflects the different results from satellites versus weather balloons. If more researchers were paid to find holes in the theory of man-made global warming, there would be more reanalysis of weather balloon data, and more squiggly lines out near the Paltridge one. If anything this graph shows how funding affects the choice of data set for reanalysis.

Yet every dataset uses radiosonde data.

• NCEP/NCAR – radiosonde
• ERA40 – uses radiosonde, satellites, aircraft, ocean buoys and other surface platforms.
• JRA – uses radiosonde (NCEP, NCAR, ECMWF), atmospheric motion vector (AMV) from geostationary satellites, brightness temperature (TOVS, ATOVS). precipitable water (SSM/I).
• MERRA – radiosonde & satellite.
• ECMWF – As for ERA40 (including radiosonde) plus additional changes for Altimeter wave-heights, Winds and clear-sky radiances, Ozone profiles & Radio occultation.

Nova’s (and Paltridge) claim that the newer analysis only shows the warming trend because they exclude radiosonde data, is baseless given that every analysis in Dessler’s paper used radiosonde data along with other sources.

Had Nova and Paltridge spent a little time investigating, rather than simply assuming, they would not have made such a basic error. Nova resorted to calling scientists non-skeptical money chasers rather than taking a few minutes to examine their reports.

2010 Nov. Thorne

Thorne 2010 discusses the history of troposphere measurements and how they compared against the models …

This review documents the evolution over the last four decades of understanding of tropospheric temperature trends and their likely causes. Particular focus is given to the difficulty of producing homogenized datasets, with which to derive trends, from both radiosonde and satellite observing systems, because of the many systematic changes over time. The value of multiple independent analyses is demonstrated. Paralleling developments in observational datasets, increased computer power and improved understanding of climate forcing mechanisms have led to refined estimates of temperature trends from a wide range of climate models and a better understanding of internal variability. It is concluded that there is no reasonable evidence of a fundamental disagreement between tropospheric temperature trends from models and observations when uncertainties in both are treated comprehensively.

As reported by NOAA …

“There is an old saying that a person with one watch always knows what time it is, but with two watches one is never sure,” said Thomas Peterson, lead scientist at NOAA’s National Climatic Data Center. “The controversy started with the production of the first upper-air temperature ‘watch’ in 1990, and it was only later when multiple additional ‘watches’ were made by different ‘manufacturers’ that we learned that they were each a few minutes off. Although researchers all agree the temperature is increasing, they disagree how much.”

Nova’s still using her wind-up analog watch in the digital world – she says she knows exactly what time it is and is happilly ignorant of other “times” that disagree with her.

Instead of rebutting Thorne, Nova repeats the use of one single set of radiosonde data and then says … “Fred Singer is about to release another paper (or three) also refuting Santer.” – we are still waiting. But even if it does, that still leaves a mountain of other evidence to refute.

2011 March. David Evans – ROCKET SCIENTIST!

In this post David Evans lists a number of debunked myths or unsubstantiated claims, but I must admit, I was a little sidetracked when looking at David’ credentials I discovered that he likes to consider himself a “Rocket Scientist”. Apparently (in his opinion) anyone that completes a PhD is entitled to do so.

Enough LOLs; on the topic of the hotspot, David repeats the old balloon myth and disregards all recent science.

2011 Oct. This is 90% certainty

Nova exaggerates the work of Fu, Manabe, Johanson (2011) who says

While satellite MSU/AMSU observations generally support GCM results with tropical deep-layer tropospheric warming faster than surface, it is evident that the AR4 GCMs exaggerate the increase in static stability between tropical middle and upper troposphere during the last three decades.

These are not the words of scientists that think the models are fatally flawed, but they do question some of the models properties.

Seidel, Free & Wang, (2012) counter FMJ …

We repeat that analysis of temperature trends, vertical difference trends, and trend ratios using five radiosonde datasets. Some, but not all, comparisons support the notion that vertical amplification in models exceeds that observed. However, larger ranges of radiosonde trends compared with those for MSU, and the sensitivity of results to the upper-tropospheric level analyzed, make it difficult to conclude unambiguously that models are inconsistent with radiosonde observations. The larger ranges are due to the availability of more radiosonde datasets with different approaches for adjusting measurement biases. Together these two studies highlight challenges of using imperfect observations of tropical tropospheric temperature over a few decades to assess climate model performance.

2012 Apr. So is the hotspot a “fingerprint” or signature?

Both, but not a unique signature/fingerprint.

Joanne Nova finally realises her confusion. The hotspot is a sign of surface warming, no matter the cause. It is NOT unique to greenhouse gases. By contrast the stratospheric cooling is unique.

Whilst backpedalling, Nova highlights the many times the term “fingerprint” is used by the IPCC in their report. The IPCC, however, don’t declare the hotspot as uniquely associated to GHG the way Nova does.

I can’t blame Nova for not knowing that the hotspot was not unique to GHGs, I didn’t know either, but then I’m not a climate scientist, unlike the folks at Realclimate. But then I don’t go around claiming to know better than the experts in the fashion Nova does.

2012 May. Models get the core assumptions

Nova wrongly assumes the hot spot is an assumption. It’s not. The hot spot is a theoretical prediction of what will happen when the surface warms. It is not “core” to the theory of AGW as Nova would have you believe and as explained above, there is evidence of its existence.

Nova says “If the IPCC models are right about the feedbacks, we would see a hot spot 10km above the tropics”

If they are right about the feedbacks, we’ll see a lot more than just the hotspot. The water vapour feedback will occur throughout the atmosphere, not just above the tropics. There are many other feedbacks completely unrelated to the hotspot.

Again Nova presents her image showing model projections over 40 years vs 20 years of incorrect radiosonde data and declares the hotspot doesn’t exist.

2012 Dec. Fasullo and Trenberth

Aside from spelling Fasullo incorrectly, Nova also shows reluctance to accept the new findings. I’ve covered this before, listing John Christy’s mistakes, but will add a little more for fun. As Jo points out, this is not about the hotspot (why then does she tag it as such?), but it’s about relative humidity.

The models that most accurately depict the change in relative humidity RH (a substitute for clouds) are the ones that have higher climate sensitivity. Nova complains that RH is not the same as clouds, and I think we all agree, but in the absence of good cloud data, RH is the next best thing and the relationship between RH and clouds was examined.

Nova, in her “Merchant of doubt” role, is keen to cast doubt about what we do know. Real, skeptical scientists understand that all measurements contain uncertainty; it’s what they deal with on a daily basis. Nova wants you to discard any data that’s not perfect – but only if it disagrees with her ideology; as we know, she’s quite willing to accept raw radiosonde data with its known flaws.

Nova also takes offence at the way the graph is plotted, using an X-axis, that to her reckoning should be reversed (this has NO impact on results); real scientists are well accustomed to viewing data plotted in all kinds of ways. Nova also complains about the use of stars to represent model values. Sheesh!

In substitute for arguments or evidence, Nova repeats her old balloon data story covered earlier.

2013 Feb. Yet another paper …

Hotspot Found!

Nova is so desperate to prove the models are wrong and this time presents Po-Chedley S. and Fu Q. (2012). In her haste Nova didn’t notice that both the satellite measurements and the models show amplified upper tropospherical warming!

The following image shows the amplification of T24 to TLT for models & observations. T24 represents the upper-middle troposphere and TLT is the lower troposphere, divide T24 by TLT to get the “amplification” values – a ratio describing how much hotter the upper troposphere is compared to the lower troposphere. With values above 1.0, both the models and observations show the amplification exists, on both interannual and decadal timescales.

Figure 4. Decadal versus interannual amplification of T24 to TLT from both AMIP and coupled GCM simulations and MSU observations in the tropics (20S–20N) between 1981 and 2005.

On the interannual timescale (the X-axis) many of the models are extremely close to observations, sometimes overestimating, sometimes underestimating. On decadal timescales, they are not as close, almost always overestimating. Unlike Nova, the authors are open to possibilities …

The apparent model-observational difference for tropical upper tropospheric warming represents an important problem, but it is not clear whether the difference is a result of common biases in GCMs, biases in observational datasets, or both.

Unlike Nova, the authors do not claim the theory of AGW has collapsed.

Observations Disagree with Observations

When Nova claims “the models are out”, bear in mind that even the observations don’t agree with each other. In the above graph both RSS and UAH come from the same set of satellite data, but different teams use different methods to arrive at global temperature values for various atmospheric heights. The data has uncertainty. Nova wants you to believe that only the models are uncertain.

2013 Apr. AR5 Hidey Games

Nova’s own hotspot image makes a reappearance. Thanks to Nice One for pointing out two other problems. Aside from comparing different timeframes, it also has a different temperature colour range and it’s showing “GHG Forcings vs Observation” when it should be showing “All Focings vs Observations”. And Nova calls the IPCC deceptive!

Nova accuses the leaked draft AR5 report of “graph-trickery”.

The document shows Nova’s favourite hotspot data in ‘smallified’ fashion (her word, not mine).  The format is not new, Thorne 2008, Sherwood 2008, Dessler 2010, and is useful for comparing multiple datasets in the one graph because they are overlaid. The comparison between model expectations and observations can still be seen.

Yet Nova claims the IPCC did this on purpose to hide the model vs observation discrepancy. If so, why then does the IPCC also state in plain English the following as the opening line for the Upper Tropospheric temperature Trends section

Most climate model simulations show a larger warming in the tropical troposphere than is found in observational datasets (e.g., (McKitrick et al., 2010) (Santer et al., 2012)).

Three Radiosonde Datasets

On the IPCC graph there are actually three sets of radiosonde data, HadAT2, RAOBCORE 1.5, and RICH-obs 1.5 ensemble. Of these three sets Nova only highlights one on her graph, and no surprise, it’s HadAT2, the one showing the largest difference to model expectations. The RAOBCORE and RICH datasets both showed agreement with the models.

Cherry picking is not good scientific practice, but does Nova have a good reason for selecting just the one set and ignoring the others? When asked this, Nova justifies her position saying “RAOBCORE and RICH are both reanalysis sets.”, but leaves out any reason as to why the reanalysis should not be considered.

The Hadley Centre (who produce the HadAT2 dataset) specifically state …

It is important to note that significant uncertainty exists in radiosonde datasets reflecting the large number of choices available to researchers in their construction and the many heterogeneities in the data. To this end we strongly recommend that users consider, in addition to HadAT, the use of one or more of the following products to ensure their research results are robust.

– Radiosonde Atmospheric Temperature Products for Assessing Climate (RATPAC) – RAdiosonde OBservation COrrection using REanalyses (RAOBCORE) and Radiosonde Innovation Composite Homogenization (RICH) – IUK (Iterative Universal Kriging) Radiosonde Analysis Project

The RAOBCORE and RICH datasets both showed better agreement with the models. Joanne Nova’s exclusive use of the HadAT2 data ensures a non-robust conclusion.

Conclusion

• The hotspot is not critical to the theory of AGW.
• The hotspot is a signature of any surface warming.
• It shows up consistently in short term analysis.
• There is evidence of it in more recent long term analysis.
• Nova expects you to ignore all datasets except one set of older balloon data.

Our cherry picking Nova accepts only the older raw weather balloon data, despite the known issues, and ignores all recent data and science that provides evidence for the hotspot and tropospheric warming in accordance with model projections.

The planet is still warming and we know that by looking at multiple lines of evidence. We also know it is due to mankind’s activities. Climate sensitivity studies using various methods indicate that the warming will continue and without intervention, we will cause a radical change to our natural environment.

Scientific understanding is obtained from having multiple studies all telling us roughly the same message. To ignore the mulitple lines of evidence and single out one outlier of data is called cherry picking – quite unscientific. When the future of our planet depends on it, it’s downright dangerous.

Tags: Cherry Pick, Hot Spot, Radiosonde, Troposphere