One of the main lines of evidence used by the Obama administration to justify its global warming regulations doesn’t exist in the real world, according to a new report by climate researchers.
Guest essay by Michael Bastasch, reprinted with permission
Researchers analyzed temperature observations from satellites, weather balloons, weather stations and buoys and found the so-called “tropical hotspot” relied upon by the EPA to declare carbon dioxide a pollutant “simply does not exist in the real world.”
They found that once El Ninos are taken into account, “there is no ‘record setting’ warming to be concerned about.”
“These analysis results would appear to leave very, very little doubt but that EPA’s claim of a Tropical Hot Spot (THS), caused by rising atmospheric CO2 levels, simply does not exist in the real world,” reads the report by economist James Wallace, climatologist John Christy and meteorologist Joseph D’Aleo.
“Also critically important, even on an all-other-things-equal basis, this analysis failed to find that the steadily rising atmospheric CO2 concentrations have had a statistically significant impact on any of the 13 critically important temperature time series analyzed,” they wrote.
When EPA released its CO2 endangerment finding in 2009, it used three lines of evidence to bolster its argument that greenhouse gases threatened human health through global warming.The crux of EPA’s argument rested on the existence of a “tropical hotspot” where global warming would be most apparent. That is, there should be enhanced warming in the tropical troposphere — the “fingerprint” of global warming.
EPA’s endangerment finding is the legal basis for agency global warming regulations, including the Clean Power Plan (CPP) now being fought over in federal court. CPP aims to cut power plant carbon dioxide emissions 32 percent by 2030 and could cost $41 billion a year, according to independent estimates.
D’Aleo and his colleagues looked at the data and controlled for El Ninos and La Ninas. What they found was that once natural oceanic warming and cooling events are accounted for, there’s no warming trend.
“El Nino is not by any means new,” D’Aleo told The Daily Caller News Foundation. “The El Ninos and La Ninas do not occur at a regular frequency but tend to cluster as we showed in our paper.”
Tropospheric temperatures are mainly measured by satellites and weather balloons, which collect data from the lowest few miles of the atmosphere. Satellites already show only slight warming since 1979, but they are sensitive to El Ninos and La Ninas.
Removing El Ninos and La Ninas from tropospheric temperatures creates “temperature time series each having a flat trend.” Basically, D’Aleo and his colleagues found oceanic warming events are responsible for virtually all the warming since 1977 when El Ninos became more frequent and stronger.
On the flip side, the recent “hiatus” in global warming can be explained by more frequent La Ninas, according to D’Aleo.
“It is an accepted fact that El Ninos bring global warmth and La Ninas cooling,” D’Aleo said. “It is thus not at all surprising that the period from 1947 to 1977 brought cooling, 1977 to 1997 warming and we had a flat trend from 1997 to current.”
With El Ninos and La Ninas adjusted out of the data, only volcanoes are left — base on EPA assumptions — to impact the climate, and D’Aleo’s report acknowledges “it was still possible that the volcanic activity was hiding CO2’s impact.”
Volcanic aerosols can have a cooling effect on global average temperature; the 1991 eruption of Mt. Pinatubo caused a dip in satellite temperature data that lasted for a couple years.
“The temperature data measurements that were analyzed were taken by many different entities using balloons, satellites, buoys and various land based techniques,” reads the report. “Needless to say, if regardless of data source, the results are the same, the analysis findings should be considered highly credible.”
Climate scientists have been debating for years over the existence of the “tropical hotspot.”
Christy, who co-runs the premier satellite temperature dataset at the University of Alabama in Huntsville with climatologist Roy Spencer, has presented evidence that climate models overpredicted warming in the tropical troposphere.
Ross McKitrick, an environmental economist at the University of Guelph in Canada, also ran the numbers and found climate models overestimated warming in the tropical troposphere.
McKitrick provided evidence of a phase shift in 1977 from dominant [La] Ninas to El Ninos — just like D’Aleo, Christy and Williams found.
“Over the 55-years from 1958 to 2012, climate models not only significantly over-predict observed warming in the tropical troposphere, but they represent it in a fundamentally different way than is observed,” McKitrick wrote in a 2014 study.
Originally published at The Daily Caller
From the paper:
The Tropical Hot Spot– CONCLUSION
The analysis above has shown many times over that the THS simply does not exist. Recall from Section IV:
The proper test for the existence of the THS in the real world is very simple. Are the slopes of the three trend lines (upper & lower troposphere and surface) all positive, statistically significant and do they have the proper top down rank order?
And that, quoting from Section XVI above:
Adjusting for just the ENSO impacts via only MEI variables, NOT ONE of the Nine (9) Tropical temperature time series analyzed above were consistent with the EPA’s THS Hypothesis.
That is, adjusting for just the natural ENSO Impacts over their entire history; all tropical temperature data analyzed above have non-statistically significant trend slopes -which invalidates the THS theory.
In short, if on an-other-things-equal basis, CO2 in fact has had a Statistically Significant impact on tropical temperatures, its impact has been offset by other Non ENSOrelated
Natural Variables over the past 55 plus years. In fact, some climate scientists effectively now claim that, while the THS apparently cannot be found in the trend slopes of the relevant empirical temperature data, the CO2-generated warming has to be hiding somewhere yet to be found. This “Missing Heat” subject has been boiling up for some time and this heat has so far
not been found.
Nevertheless, alarmist scientists are still claiming record-setting warming in the Contiguous U.S. and globally caused by rising CO2 levels. If true, this CO2 -caused missing heat has to be warming the planet by a currently unknown mechanism operating somehow outside the tropics. Therefore, this analysis moved on to test this new, never formally claimed before, hypothesis by ENSO adjusting the relevant Temperature data.
Now it is to late to teach Obama to read, but there must be someone in the global political world who can add 2+2 without coming up with 8.
idiots in idiots out !
Think of him as a lawyer – hired advocacy.
Yes as a lawyer he cites the authorities and doesn’t check them. His background will not be in science or history.
So El Ninos and La Ninas tend to cluster? Well, imagine that. Just like warm and cold years cluster, or wet and dry years, and decades and centuries. Mandelbrot called it the Joseph Effect.
Also known as Hurst”s phenomenon
I’ll look out for this news on the BBC and in the Guardian.
Bazinga.
An interesting read. Does anyone have some graphs to show the El Nino/La Nina claim? I had a look, but since 1950 (no idea if the data is good), they look fairly even. Only difference seemed to be more strong El Ninos as opposed to moderate La Ninas. Is that enough for .3? No idea.
Reviews that I have been reading state that the full pdf (top link in article) is packed with graphs – I’m waiting to download at office tomorrow.
BC, correct. Figure 6-2 gives MEI and cumulative MEI. Sourced from ESRL.NOAA.gov.
The clustering of El Nino/La Ninas creates the PDO.
http://climate.ncsu.edu/images/climate/enso/PDO_Phase.gif
Which is not an actual physical process, just the pattern created by the general clustering of these ENSO events.
Absence of evidence is not evidence of absence, or some such baloney.
While true, absence of evidence is enough to bust a theory.
Que!?
Is this some sort of claim that proves zero is not null or visa versa?
You are overlooking the fact that observations fail to prove a hotspot, instead the observations prove the lack of a hotspot.
That is not absence of evidence.
That’s not what they say. They say positivity of trends are not statistically significant. That means roughly that there is a 5% chance (or more) they were not positive, consistent with the null hypothesis. But that still leaves a 95% chance that they were.
“But that still leaves a 95% chance that they were.”
No, that is not correct. Firstly, you should have said: “That means roughly that there is a 5% chance (or more) they were not positive, consistent with the null hypothesis. But that still leaves a 95% chance (or less) that they were.”
But, even without that quibble, what you have stated is not in the spirit of hypothesis testing. There is a reason we typically demand such high levels of significance. It is because we rarely know the actual underlying distribution. If we get a clearly significant signal, then we can reason that, even if our underlying model is imprecise, the SNR is high enough that we are probably on safe ground going forward on the assumption that we have it more or less right. If we can’t get to a overwhelmingly high level of confidence, then all bets are essentially off.
One would reasonably expect that, if the hypothesis of anthropogenic global warming were true beyond any doubt, there would be a significant trend consistent with it that would be easily discernible. That appears not to be the case. Ergo, the hypothesis, in its current form, gains no support, period.
“If we can’t get to a overwhelmingly high level of confidence, then all bets are essentially off.”
But that’s not the claim here. They claim “And, therefore, the THS does not exist.”. And they go on to say, therefore the GCMs are wrong, and so the EPA is FUBAR.
In fact the big doubt here is the measurement methods. The question of homogeneity of radiosondes is very real. And trying to sort out levels in satellites is very uncertain – just see the change in UAH from V5.6 to v6 (still in beta).
The questions regarding the quality of satellite or radiosonde data are chaff. Those concerns pale to a particularly wan shade of white nothingness in comparison to questions regarding the quality of the surface data.
Again, one would reasonably expect that, if the hypothesis of anthropogenic global warming were true beyond any doubt, there would be a significant trend consistent with it that would be easily discernible. It is not. Therefore the range of alternatives to the hypothesis is wide, with the AGW hypothesis given no greater weight than many other candidates to explain 20th century warming. Which means, the odds of it being correct are small.
So, perhaps saying the GCMs are wrong is too strong, but saying they are highly unlikely to be correct is accurate.
Actually, when you have a null hypothesis, absence of evidence IS evidence of absence.
Anthony: “McKitrick provided evidence of a phase shift in 1977 from dominant El Ninas to El Ninos — just like D’Aleo, Christy and Williams found. Shouldn’t that\t read “..La Ninas…”?
Las Ninas to Los Ninos.
Que pasa, Bart?
“To dominant La Niñas” is correct. In English the conceptual word “La Niña” is a compound construct. It is right to speak of “an El Niño” and “the El Niños”, as in “During an El Niño event” (http://www.unocha.org/el-nino).
I’m a native English speaker aged 65, but am fluent in Spanish and have been speaking it among native speakers for 50 years. Even in Spanish, one can speak of “un El Niño”.
For example, this Spanish Wiki entry once uses “del Niño”, but in all other instances says “de El Niño”, which is correct if the two parts, article and noun, together form a single word.
https://es.wikipedia.org/wiki/El_Ni%C3%B1o_(fen%C3%B3meno)
About time this PC rubbish stopped. We use our way of speaking here, and are not arrogant enough to demand Engllish ways in a Spanish context. I know what they would tell us if we tried, and rightly so! Enough now..
So once ENSO is compensated for, there is no tropical hotspot. I am fairly sure that the true believers will argue the compensation is flawed, as the hotspot must be there according to the models, and the models are never wrong.
“…true believers will argue the compensation is flawed…” Cue Nick Stokes in 3, 2, 1…
Time to show the weather the instruments of torture to persuade it to confess to its manifold sins and wickedness.
They will pull from their magic sophistry hat and claim that it’s more correct to interpolate tropospheric temperatures from the surface based temperature data, then claim it’s worse than we thought.
Convection is just working it’s magic. So now, back to the “weather, which didn’t used to be climate, but now IS climate” excuse; the last refuge of scoundrels, who claim every weather related death “proves” their point. Shameful.
Isn’t this what Bob Tisdale suggested about 5 years ago?
Maybe now someone has done some more serious analysis, it may get looked at more seriously.
Lot’s like all these natural “oscillations” are not necessarily net zero effects after all. Maybe we’ll have to stop automatically ‘detrending’ any and all climate variables, attributing the trend to AGW and calling the rest an “oscillation”.
Jo Nova went over this a few years ago too,showing there is no “hotspot” existing.
To my knowledge Mr. Tisdale was around the edges of this, but did not do the same analysis, which goes quite a bit further than what he has done – that I have seen. (Of course, I don’t know everything Mr. Tisdale has done.) I will be very interested to see what he has to say about this, given his prior work.
B.Tisdale has little capacity for analysis beyond fitting linear trends in Excel and I was quite critical of the way he thought he had shown ENSO was the cause. This study takes a much more serious look at that.
My point was just that BT’s posts were the first time I’d heard it explicitly pointed out that ENSO is two separate processes that are asymmetrical, not a necessarily net zero “oscillation”.
For someone with his lack of formal training I think that was quite insightful.
It is this whole ASSUMPTION of linear trend + internal ‘oscillations’ which needs to be thrown out . It has no basis other than preconceived biases which then get imposed of the data and the way it is analysed.
+1
the point is: not all oscillations are exactly the same length.
as musician i look at them as “sound waves” and i found that the IOD, PDO, AMO and ENSO do have their “own frequency and amplitude”. the last 18 years PDO and AMO were opposing each other while ENSO was pretty “near zero” (moderate la nina’s with a few El nino’s) As el nino’s are stronger then la nina’s their effect cancels a back to back la nina.) Yes just like sounds they amplify each other and cancel each other out.
now we had a strong El nino, a very strong PDO and still the AMO just over it’s peak level. That this results in a record warm year is “logic”.
what’s less known is that the main oscillations also had a short combined positive peak in the 30’s followed by a combined negative dip in the early 60’s. The sharp rise in the 70’s combined with first the rise of the PDO and switch o more positive ENSO, followed by the AMO.
even more: PDO and ENSO when they go in a positive combined phase can “overpower” the AMO,
with all that i am also convinced that the southern oceans do have a similar oscillation, hence the antarctic sea ice that broke record extends in 2015.
how the IOD fits in this is also a bit harder to say as i found no real graph of it yet. all i know of PDO ENSO and AMO is that the last 100 years the positive episodes did stay longer “in synch” then the negative episodes. The only strong “triple negative” is from 1958 till 1977
okay before shooting me: i know AMO and PDO are two things you can’t “just add up” as it are not the “in the same unit datasets”. (did read that here as well but can’t find the article from Bob Tisdale)
but i’m pretty sure that if the AMO graph would be converted to the same units as the PDO graph some very natural warming phases will be shown (AMO is a detrended value while PDO is a real visible oscillation value)
Antarctic sea ice broke records only in late 2012, all of 2013, all of 2104, and the first half of 2015. By late 2015, Antarctic sea ice was near-normal, and has stayed right at normal since then.
As I understood, Bob was saying that the warming consisted of ENSO jumps, which somehow made it less real. Insofar as that is true, adjusting to remove ENSO may well diminish the warming.
“This “Missing Heat” subject has been boiling up for some time and this heat has so far
not been found.”
Well, if it helps, I’ve noticed that 9 of the last 10 ribeyes I have ordered over the last 2 months have come back well done.
oops. ordered medium rare
Your ordering model is faulty!
97% of diners say that steaks if cooked for longer become less rare. This has lead to a consensus that all steaks are becoming overcooked due to human influence of more efficient cookers. Models have shown that if the same amount of power input is used as was used in preindustrial times all steaks will now be overcooked, ‘the tipping point has already been reached’. Suggestions that a rare steak can still be cooked if less power is used has been ridiculed by CAGSW (Catastrophic Anthropogenic Global Steak Warming) theorists and dissent has been likened to denying that the moon landings ever happened
“Our kids will never taste a rare steak in their lifetime unless we take action now. It is essential that all cooking is done on wood or coal burning stoves if steaks are to be saved” said a CAGSW spokesperson. Steaks are being destroyed by the use of electric and gas cookers that have been proven by modelling to overcook steaks due to the increase in conversion of fuel into heat
It has been pointed out that the influence of humans on the heat involved in cooking steaks is minimal and even though the increase in efficiency is caused by humans this can be seen to be having an overall beneficial effect by reducing the cost of cooking enabling the purchase of many more steaks
It has been proven polar bears do not eat cooked steaks on the sea ice, therefore, cooking steaks with fire on the sea ice threatens the existence of polar bears in the wild.
This requires more research to establish the long-term trend of barbecue on the vanishing sea ice caused by climate change and sea level rise, send more money.
I wonder if this research may be presented in evidence at CPP trial next week??
Devastating to EPA.
Lets hope so. Very timely. Opens a second line of legal attack on CPP, the first being Tribe’s unconstitutional brief.
Michael Bastasch, please provide a quote or at least an actual, specific place in the EPA’s documentation to support your contention “When EPA released its CO2 endangerment finding in 2009, it used three lines of evidence to bolster its argument that greenhouse gases threatened human health through global warming.The crux of EPA’s argument rested on the existence of a ‘tropical hotspot’ where global warming would be most apparent. That is, there should be enhanced warming in the tropical troposphere — the “fingerprint” of global warming.”
In reality, neither climatologists in general nor the EPA have ever claimed that the tropospheric hotspot is a fingerprint of CO2-caused warming, which is what you strongly imply in your two sentences. Instead, a tropospheric hotspot is a consequence of the lapse rate, and is expected to increase if global warming occurs, regardless of the cause of that warming.
1. It is not Bastach’s assertion, it’s the assertion in the paper that Bastach is writing about.
2. The following is from EPA’s Technical Support Document in support of the Endangerment Finding, p. 47:
The Technical Support Document is available here.
https://www3.epa.gov/climatechange/endangerment/index.html#tsd
Similar statements are found at 74 Fed.Reg. 66,523.
Hopefully that answers your question.
Quinn, Bastasch did not provide that as a quote from the “paper,” nor did he even say he was paraphrasing what the “paper says. So it is clear that Bastasch himself is making that claim, in addition to the paper’s authors making it. Nothing in your citations, Quinn, supports Bastasch’s contention that “The crux of EPA’s argument rested on the existence of a ‘tropical hotspot’ where global warming would be most apparent.” Bastasch and the paper’s authors have erected a strawman.
The paper explicitly refers to the three lines of evidence, see, e.g., p. 12.
The paper does not give the citations to the Endangerment Finding TSD that I provided.
The physical understanding of climate relied upon by EPA requires and predicts the hot spot. So yes, it’s the crux of the matter.
“Hopefully that answers your question.”
It doesn’t. The quote says nothing about a hotspot. What the paper claims is
“The crux of EPA’s argument rested on the existence of a ‘tropical hotspot’ where global warming would be most apparent. That is, there should be enhanced warming in the tropical troposphere — the “fingerprint” of global warming.”
That “crux” is not in your quote.
Nick, is it your position that the hot spot is not a central element of the physical understanding of climate?
“Central element”? No, you could have warming without it. GCM’s predict it; if they turn out to be wrong, then one would have to figure out why. But recent literature says that there is a hotspot. And this paper does not show that there isn’t. It says, on the basis of a statistical model, that there is at least a 5% chance that there isn’t. That still leaves up to 95% that there is.
NS, I ewas the paper. You have your probabilities upside down. Perhaps comes from being down under.
he did –
https://thsresearch.files.wordpress.com/2016/09/wwww-ths-rr-091716.pdf
Bubba, nothing in that linked document supports Bastasch’s contention that the crux of the EPA’s evidence on CO2’s effect is the tropospheric hotspot. Nothing.
TD, you are diesrincorrect in several ways. Read the very general ‘three lines of EPA evidence’ posted in the replies to your comment. 1. General understandings. CO2 is a GHG, yes. But that says nothing about sensitivity or attributiin. Natural variation is overlooked, specifically the period 1920-1945 that is essentially indistinguishable from 1975-2000. The former cannot be caused by delta CO2, which means attribution of the latter to CO2 cannot be proven. 2. By comparison to paleo. That is provably false, just as Mann’s hockey stick and Marcotts mess are. 3. By models. Those models all produce a tropical troposphere hotspot that does not exist in reality, thereby showing the models cannot be relied upon.
This new research directly attacks 1 by showing there is no warming 1975-2000 after correcting for ENSO using NOAA’s official MEI. And it directly attacks 3 by showing there is no observable tropical troposphere hotspot. It does not need to attack 2 because McIntyre already has.
And all that is admissable evidence in a court proceeding attacking the EPA endangerment finding.
Well the Australian climate establishment is ahead on this hot spot belief
‘EMAIL NUMBER 6
Subject: Climate commission report
Date: Tue, 24 May 2011
Hello ,
I have just finished reading your report and I must say I am quite disappointed in the end product, my reasons for being disappointed are simple. I was of the understanding that the sole purpose of the commission was to explain the science to the Australian public, help build and foster a consensus among the people as outlined in the link below. Therefore I was waiting with great anticipation for this report as I hoped it would clarify I few issues I have with AGW. http://www.climatechange.gov.au/minister/greg-combet/2011/media-releases/February/mr20110210.aspx
However after reading your report I do not believe this has happened, in fact not even attempted. One example which springs to mind is your statements regarding the hot spot, you stated on page 16 and I quote,
“An apparent inconsistency between observations with greenhouse theory was the alleged failure to find a so-called “tropical hot spot”, warming in the tropical atmosphere about 10-15 km above the Earth’s surface. In reality, there was no inconsistency between observed and modeled changes in tropical upper troposphere temperatures, allowing for uncertainties in observations and large internal variability in temperature in the region. Furthermore, recent thermal wind calculations have indeed shown greater warming in the region (Allen and Sherwood 2008), confirming that there is no inconsistency and providing another fingerprint of enhanced greenhouse forcing.”
We both know the hot spot is a central plank in the theory of AGW, the hot spot is an accumulation of hot air generated by a positive feedback to increasing greenhouse gases by water vapor. If this hot spot does not exist then the AGW theory is falsified. The problem I have here with this whole sorry saga is the scientific process that has unfolded, for example we have two independent sources of temperature data being satellite and radio sonde. Both these sources of data have been rejected on the grounds that the data is erroneous and have been replaced with GPS data from the very same radio sondes that supplied the thermometer data that was rejected.
Can any scientist seriously condone the actions of Allen, Sherwood and Santer? Can any scientist seriously expect the general public to believe thermometers with the capacity to measure temperature to one decimal place to be more erroneous than a cheap throw away GPS?’
Quadrant Doomed Planet
This is part of our undeniable climate war history in Australia.
So it is hard to defend a statement like
‘In reality, neither climatologists in general nor the EPA have ever claimed that the tropospheric hotspot is a fingerprint of CO2-caused warming,’
The fingerprint hypothesis was all over our news bulletins.
Our Climate C omission reports on it.
The failure in the hotspot hypothesis was one of the reasons I became a ‘denier’, when before I thought the CO2 theory at least plausible.
All you have to do is follow the data.
Mr. Dayton,
The EPA specifically commented on questions about he “fingerprint” in justifying their decision in the endangerment finding. They used the flawed studies such as Allen/Sherwood (2008) using wind date to help the EPA ignore the inconvenient facts of the temperature data. They also used Santer (2008) to try to justify the differences between modeled and observational trends. To argue that climate alarmists and the EPA have not tried to save the “fingerprint” argument as a key element in “proving” AGW is either naive or a prevarication.
From the EPA own website:
https://www3.epa.gov/climatechange/endangerment/comments/volume3.html
Comment (3-7):
Many commenters (e.g., 3215.1, 3330, 3446.1, 3596.2, 4003) indicate an anthropogenic warming is missing in the vertical and horizontal profile of the atmosphere in the tropics. They argue that the distinct human fingerprint of warming concentrated in the tropics between 30 degrees N and 30 degrees S and increasing with altitude to 10 kilometers (km) is the mechanism for amplified warming but that this fingerprint exists only in the models and not in the empirical science. They state that models predict significantly increasing warming with altitude up to 10 km and warming continuing beyond 15 km. Yet, they find except for surface temperatures, observations show limited warming that is statistically significantly less than the warming projected by the models. They indicate observations show a cooling with altitude beyond 13 km while the models still project a warming.
Response (3-7):
EPA is aware of the emerging literature on this issue and the challenges in identifying the anthropogenic fingerprint in the tropics. The TSD???s characterization of this issue is consistent with the assessment literature as well as the most recent studies, which find that when uncertainties in models and observations are properly accounted for, newer observational data sets are in agreement with climate model results.?? ??
In light of this comment, EPA reviewed the assessment reports and newer literature on this topic.?? As one commenter notes, Christy et al. (2007) find discrepancies between surface and tropospheric temperature data in the tropics, and Douglass et al. (2007) report model results that are in disagreement with the observed trends. ??However, Haimberger et al. (2008) analyze weather balloon (radiosonde) records of tropospheric temperature data and find ??????.we note that the temperature trends from RICH???RAOBCORE version 1.4 [a homogenized radiosonde record] are more consistent with trends from recent climate model runs than earlier radiosonde datasets. In the tropical upper troposphere, where the predicted amplification of surface trends is largest, there is no significant discrepancy between trends from RICH???RAOBCORE version 1.4 and the range of temperature trends from climate models. This result directly contradicts the conclusions of a recent paper by Douglass et al. (2007).??? They further note: ???A robust warming maximum of 0.2???0.3K (10 yr)???1 for the 1979???2006 period in the tropical upper troposphere could be found in both homogenized radiosonde datasets.???
Another paper by Allen and Sherwood (2008) reports: ???Climate models and theoretical expectations have predicted that the upper troposphere should be warming faster than the surface. Surprisingly, direct temperature observations from radiosonde and satellite data have often not shown this expected trend. However, non-climatic biases have been found in such measurements. Here we apply the thermal-wind equation to wind measurements from radiosonde data, which seem to be more stable than the temperature data.???Warming patterns are consistent with model predictions except for small discrepancies close to the tropopause.???
Finally, Santer et al. (2008) analyze differences in trends between observed surface and tropospheric temperature records, and also compare the observational trends with the models. They conclude: ???There is no longer a serious and fundamental discrepancy between modeled and observed trends in tropical lapse rates, despite [the Douglass et al., 2007] incorrect claim to the contrary.???
The TSD summarizes this issue and cites the conclusions of the latest major assessments. It states: ??????an important inconsistency may have been identified in the tropics. In the tropics, most observational data sets show more warming at the surface than in the troposphere, while almost all model simulations have larger warming aloft than at the surface (Karl et al., 2006). Karl et al. (2009) claim that when uncertainties in models and observations are properly accounted for, newer observational data sets are in agreement with climate model results.??? EPA concludes that the TSD???s summary of the current state of the science on tropical tropospheric warming as reflected in the underlying assessment literature is accurate.??
Beyond the EPA advocacy of this, any cursory review of the IPCC AP2 report will identify the fingerprint meme and the work of Santer to try to use this as confirming AGW. Your argument here is without merit and is entirely false in its claims.
All of the arguments about Santer 2008 fall away the moment you read Santer’s evidence he presented to the American Physical Society Workshop on Climate Change in Jan 2014. The transcript may be some 500 pages long, but Santer unequivocally confirms that the tropical hotspot is missing. At the same workshop Christy presented four independent studies of the troposphere temperature profile changes over 30 years, and contrasted them to 22 separate model runs. The experimental evidence was for a warming of ~0.1 deg C per decade across most of the tropical troposphere; the models averaged 0.4 deg C per decade in the upper troposphere. Experiments should be able to tell the difference between 0.3 and 1,2 deg C warming – and they did.
Firstly, well done to Christy et al for putting this together. However, whatever happened to the NULL hypothesis? Isn’t the proper statistical approach to this : what are the sufficient conditions for rejecting the hypothesis that there is no THS ?
Correct scientifically, but not legally. EPA endangerment finding is based on their conclusion that the tropical troposphere hot spot exists. To attack that legally required fact finding to declare CO2 a regulanle pollutant, you show it to be factually wrong. As this unassailable research does.
Good point. So the fact that there is no detectable hot-stop means that EPA has no basis to make the regulations they have made.
“Isn’t the proper statistical approach to this : what are the sufficient conditions for rejecting the hypothesis that there is no THS ?”
Indeed so. The paper has that elementary logic wrong. The muddle starts in the first sentence of the abstract:
“These analysis results would appear to leave very, very little doubt but that EPA’s claim of a Tropical Hot Spot (THS), caused by rising atmospheric CO2 levels, simply does not exist in the
real world.”
Firstly there’s the usual problem with sceptic “refutations” that they don’t quote what they are actually refuting. Not once in all those words do we learn what the EPA actually said about the hotspot.
But then, their basic claim is this:
And as a matter of elementary logiuc, that is just wrong. Statistical significance never proves anything. It may fail to reject the null hypothesis. That actually relies on a theoretical uncertainty model, which may be inappropriate. But at best, all you can deduce is that you didn’t have enough data. It’s a failure, not a proof.
According to Jo Nova data from 28 million radiosondes has failed to find it but, like the black swan, it may exist, however:
“In some circumstances it can be safely assumed that if a certain event had occurred, evidence of it could be discovered by qualified investigators. In such circumstances it is perfectly reasonable to take the absence of proof of its occurrence as positive proof of its non-occurrence” — Irving Copi, (American philosopher, logician) Introduction to Logic (1953), p. 95 (Wiki Evidence of absence).
The test of the null hypothesis assumes a null sampling distribution. It assumes that you have noise and only noise. In so doing, you also define a threshold at or beyond which the associated measurement values are unlikely to have occurred if the null hypothesis (and its null sampling distribution) is in fact true. Here, you are chiefly concerned with falsely ‘rejecting the null’.
Put another way, the null hypothesis in this case reflects the possible truth that there is no measurable evidence of a hot spot (subject, of course, to the measuring technique used). The null sampling distribution, given our measurements, is our best guess at what you would expect to observe from your measurements in the long run. in other words, the null hypothesis maps out what you would expect to find if the truth is that there is nothing ‘out there’ but random noise in your samples and measurements.
Evidence in favour of the null hypothesis can in fact mount over the course of repeated tests of the null hypotheses (with different samples and different measurements, of course). In fact, failing to reject the null hypothesis would be the critical factor when deciding which of two competing alternative hypotheses to believe. For tests of the null hypothesis that are designed to test for evidence of the same underlying conceptual hypothesis, repeated ‘failures to reject the null’ would lead the (genuine) scientist to become ever more and more confident that the null is in fact true.
If you are interested in testing for evidence of the alternative hypothesis in a way that allows you to conclude something more than literally simply “well, anything BUT the null hypothesis could be true” (because that is all you can really say when you ‘reject the null’, yes, is quite crude isn’t it), then you are now chiefly concerned with statistical power, the probability of rejecting the null hypothesis when the alternative hypothesis is in fact true. Critically, you’ll now need to specify the values of a host of parameters that together describe the distribution of this alternative ‘possible truth’. The sampling distribution of the alternative hypothesis is our best guess at what you would expect to observe from your measurements in the long run, using a test of the null hypothesis in this world in which your alternative hypothesis is in fact true.
Considering the test in question appears to be based on simple OLS linear regression over a relatively brief sampling window (and my guess would be t-test of the regression coefficient against 0), the underlying statistical parameters and their ‘behaviour’ are very well known. Once the oceanic cyclical trends are removed, the noise around the slope is probably normally distributed no matter where you are across the time series, and the number of measurements in said time series will be well beyond that needed to detect a meaningful (i.e. large positive value) slope (regression coefficient) with a high probability of rejecting the null.
Very important study. Just read it and saved the linked .pdf. Federal Courts defer to agencies like EPA on findings of fact. But this paper shows not only thatnthe endangerment finding based on the tropical hot spot is factually wrong, but also the method the EPA used was logically flawed. Opens up an alternative line of attack to CPP than unconstitutional, and maybe even a bottoms up challenge to SCOTUS Mass. v. EPA.
“…and maybe even a bottoms up challenge to SCOTUS Mass. v. EPA.” I sincerely hope so. What a travesty.
Do the authors of this “study” intend to submit it to a journal with peer review?
You and your lifestyle and level of development are outlawed out of an abundance of caution. Because the political appointees and their consultants said so. I think that merits a full presidential debate and not just one carefully wordsmithed question with the usual canned answer.
“Researchers have published results in Environmental Research Letters confirming strong warming in the upper troposphere, known colloquially as the tropospheric hotspot. The hotspot has been long expected as part of global warming theory and appears in many global climate models.”
“”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.””
http://phys.org/news/2015-05-climate-scientists-elusive-tropospheric-hot.html
http://iopscience.iop.org/article/10.1088/1748-9326/10/5/054007/meta;jsessionid=1CAB18DBB76087611D98B15137CB788E.c2.iopscience.cld.iop.org#footnotes
Can I ask if the authors of the study contradicted by the above, have had theirs peer-reviewed?
Can I ask if the authors have seen the above paper?
And if so have any comments?
Toneb, the second Sherwood paper you cite is pal review garbage. Go read it. Was Sherwood’s second peer reviewed effort to find the mythical hotspot. His first was shredded by his peers. His second is so methodologically bad that nobody has even bothered. And you don’t need an advanced degree in statistics to spot its flaws. Jo Nova did all by herself.
Further to your peer review hurdle, the initial part of the paper contains peer review endorsement from 9? Ph.Ds in the relevant physics, meteorology, and econometrics (regression analysis and resulting statistical significance) disciplines behind the research.
Any fair and rational government would take this study and raise the question of whether they were wrong. They would then proceed to fund four independent studies to check the result out of their multibillion dollar annual budget. We don’t live in that fantasy public policy world though, not even to the slightest extent of questioning the need to consider such possibilities of departing from the current policy push. The policy and advocacy prize of killing coal first followed by all other fossil fuels second comes ahead of truth. Right Gina?
I’m not sure that it’s fair game to remove El Ninos and La Ninas.
The energy for an El Nino has to come from somewhere. Maybe it is the case that global warming causes more El Ninos.
Is this paper the stake through the vampire’s heart? Perhaps not.
ENSO is a natural phenomenon.
AR5, WG1 Chapter 14, Section 14.4, p. 1420
It is natural but it is not magic. El Ninos are associated with warming and that takes energy from somewhere.
CB, all ENSO water warming comes from sunlight, as does all energy input to Earth. The GHE is only a diminution in offsetting cooling. The westerly equatorial tradewinds pile the warmd water up in the western Pacific. When the ‘pile’ gets to ‘high’, the trades reverse and the water sloshes east in an El Nino. My kindergarten version of Bob Tisdale’s recharge oscillator basic explanation.
Correct, CommieBob–El Ninos shift energy accumulation toward the atmosphere at the expense of the oceans, and La Ninas the opposite. There are some small effects on the Earth’s total accumulation of energy, due to changes in clouds. But for the most part the Earth continues to gain or lose energy about the same regardless of ENSO, with ENSO merely changing the ratio of accumulation in oceans versus atmosphere. That fact is clear from looking at the total energy in the total system–atmosphere, surface oceans, land, and oceans all the way down: http://www.skepticalscience.com/graphics.php?g=66. If you don’t trust my word on the validity of the total energy, maybe you will trust Roger Pielke? https://pielkeclimatesci.files.wordpress.com/2009/10/r-334.pdf
I forgot to mention that ENSO affects total energy accumulation not just because of changes in clouds, but because of El Ninos keeping/bringing energy closer to space (i.e., out of the oceans) so it radiates to space more easily. La Ninas the opposite.
The article says that, if we take out the effect of the El Ninos and La Ninas, the trend goes away. That’s fine. It doesn’t mean we can ignore them. They aren’t magic. They are part of the energy balance. The trend is whatever it is … and because of adjustments and uhi we aren’t even totally sure of that.
Something causes the El Ninos. It is probably the same thing that is responsible for the long term trend.
No, not the opposite. La Nina does NOT shift energy in the atmosphere back into the ocean. It accumulates solar energy into the ocean.
That is whole problem when climatologists and IPCC present this as an “oscillation”, it is NOT oscillating, it is varying. It is a variable throughput of energy in and out of the OHC reservoir.
NONE of that energy is swinging back and forth between atmosphere and ocean it is NOT oscillating.
What is their rationale for using *cumulative* TSI and *cumulative* MEI?
when I got to CUM MEI I stopped reading. Thats as bad as cum TSI..
And that seems to be a sure sign that your mind has closed.
Earth accumulates energy, then dissipates energy. It happens every day from dawn to dusk. Similarly, let’s say we had a year of remarkably clean air, with little particulates or SO2, as sure as the sun rises, you can see energy accumulate and put Earth at a higher equilibrium state than the year before when the skies were a bit more opaque, translating to a higher temperature. The reverse has been measured empirically, when Pinatubo went off, skies became instantly opaque, and Earth accumulated less energy. Temperatures dropped.
Anthony, your reply did nothing to address my question of why they used cumulative TSI and MEI rather than instantaneous values. Please answer that question, or ask this author’s article to do so, or as the paper’s authors to do so.
Can anyone explain how El Nino / La Nina explain all of global warming, when the trend of El Nino years is starkly up, the trend of La Nina years is starkly up, the trend of neutral years is starkly up, and all those trend lines’ slopes are nearly identical? http://variable-variability.blogspot.com/2016/09/global-warming.html
Perhaps you should read the linked Christy et. al. paper before commenting. Therein you will find an analysis from 1947 that answers your question rather thoroughly.
No, ristvan, that paper does not explain it. This is simple. It is impossible for those three trend lines each to be increasing with the same slopes, if as Christy & Co. claim El Nino / La Nina explains all of warming. Plotting the three sets of years separately is a rough but completely reliable means of examining to what extent El Nino / La Nina explain global warming. If Christy & Co. were correct, then each of those three lines should be approximately flat. The El Nino line would be higher than the neutral line which would be higher than the La Nina line, but all three lines would be flat. They are not.
TD, if you had read the paper, looked at Figure 6-2 and then studied the following segmented regressions, you would not have made so silly a comment. Read the paper before commenting on a blog post about the paper. Always.
No, ristvan, Figure 6-2 does not at all answer my question. If increasing temperature were due to an increasing number of El Nino years, that would leave the temperature flat across the set of all El Nino years, but instead the trend is starkly up across the set of all El Nino years. It would also leave the trend flat across La Nina years, and flat across neutral years, but instead both those trends are up nearly identically to the trend of the El Nino years.
Before we get to that, I have a question about your question: How is it possible for the trends of El Niño, La Niña and neutral years to all three be starkly up *at the same time* when they are mutually exclusive?
Quinn, the answer is that factors other than ENSO cause the year-to-year increases in temperature–factors that do not depend on whether the year has an El Nino, La Nina, or is neutral. Factors such as…greenhouse gas increases.
You said the trends in 3 mutually exclusive phenomena – El Niño, La Niña and neutral ENSO were all simultaneously “starkly up.” That don’t make no sense. Then you said that things other than ENSO make temperatures go up. But the paper shows there is no upward trend in any of 13 ENSO-adjusted temperature time series, which refutes your statement.
What else you got?
Quinn, responding to your “You said the trends in 3 mutually exclusive phenomena – El Niño, La Niña and neutral ENSO were all simultaneously “starkly up.” That don’t make no sense.”:
Temperature each year is affected by non-ENSO factors and by ENSO. You can control for (i.e., eliminate) the ENSO effects in a rough way by looking at the trend across all and only El Nino years. By definition, El Nino versus La Nina will have no effect on the trend within that set of all and only El Nino years. So any trend within the set of El Nino years must be due to factors other than ENSO. Likewise for the trend across all and only La Nina years And the same for the trend across all and only neutral years. Look at the graph partway down the page I linked to: http://variable-variability.blogspot.com/2016/09/global-warming.html
Your statement “So any trend within the set of El Nino years must be due to factors other than ENSO” you might be on to something. If you mean that any trend in the residual of ENSO-adjusted temperature series must be due to factors other than ENSO – the authors of the paper agree 100%, and say so explicitly. The finding of the paper is that the trend in that residual is flat – there’s no trend left in the residual that could be caused by anything else, including CO2. That is the point of the paper. If you can find an error in the paper it would be good to know.
In 2007 Richard Lindzen said of this tropical tropospheric warming, that is the signature of the greenhouse effect:.
Quote:
Roughly speaking, the (tropospheric) warming in the tropics is about two to three times larger than near the surface regardless of the sensitivity of the particular (climate) model. This is, in fact, the signature (or fingerprint) of greenhouse warming. Stated somewhat differently, if we observe warming in the tropical upper troposphere, then the greenhouse contribution to warming at the surface should be between less than half and one third the warming seen in the upper troposphere.
Endquote.
So this tropospheric warming should be substantial, and redly apparent. If anyone claims 1.5oc of surface warming in the tropics, then the troposphere should warm by 3 to 4.5oc, which should be easily detectable. And yet the DLR radiation record sugggests there has been no tropospheric warming at all.
Ergo – whatever warming the surface has experienced over the last 20 years or so, it was not caused by the greenhouse effect.
Ralph
ralfellis, Lindzen was incorrect in his claim that that the hotspot is a signature of “greenhouse” warming. It is a signature of warming regardless of cause. Basic physics; look in any textbook. I’ll leave it to you to speculate why Lindzen said something that is incorrect.
If that is correct, i.e., if the basic physical understanding of climate that says that is the case, then the absence of a hot spot absence refutes warming from any cause, human or otherwise.
There has been observed warming, but there is no observed hot spot.
So either the observations of warming are wrong or the theory is wrong.
Which is it?
“It is a signature of warming regardless of cause …”.
==============================
It is evidence of water vapour feedback due to warming (regardless of cause) as I understand it.
… Or would be evidence of water vapour feedback.
Tom,
From the evidence presented in IPCC AR4 I think Lindzen is clearly correct that the tropical hotspot is primarily GHG caused. c) is GHG only and f) is total – it is quite clear that c) is by far the predominant cause of the hotspot. See below for the figure from the IPCC report.
http://s19.postimg.org/vsryy4ler/IPCC_AR4_Fig_9_1.jpg
Hoplite: Box c in that figure shows the strongest hotspot only because that particular set of forcings is more strongly positive than the forcings that created the other boxes. For example, if the solar forcing in box a was increased to be the same amount as the GHG forcing in box c, then the hotspot in box a would be just as large as that in box c. Box f’s hotspot is smaller than box c’s hotspot, because box f sums all the positive and negative forcings, so the negative forcing of aerosols offsets some of the other, positive forcings. For more details see the Intermediate tabbed pane and then the Advanced tabbed pane here: http://www.skepticalscience.com/tropospheric-hot-spot.htm
Ok, I’ll grant that’s a subtlety I missed. I understood the warming wasn’t identical for the different forcings not only due to magnitude reasons. However, it doesn’t alter the simple fact that the strongest forcing by far expect was the GHG forcing in c) and that hasn’t happened. Ergo, it is perfectly reasonably to state that that is a far greater problem for CAGW theory than for other modelled heating mechanisms.
How about disproving what the EPA said, not what Lindzen said.
NS, this new study goes a long way toward doing just that. Refutes their lines of evidence one and three rather thoroughly. Doesn’t touch on their line two; no need, as Steve McIntyre already destroyed it.
Rud,
Well, this article says:
“The crux of EPA’s argument rested on the existence of a “tropical hotspot” where global warming would be most apparent.”
And still no quote. But the paper does not disprove the existence of a hotspot. It just claims that positivity of trends are not statistically significant. That doesn’t prove models are wrong; it just means their analysis failed to eliminate the null hypothesis at 95%. That could mean there is a 5% chance (or more) that the models are wrong. That is still an up to 95% chance that they are right. Not a convincing refutation.
NS, I am a true skeptic, on everything. That has garnered 13 US issued patents in four areas so far. I answered your reasonable question in a comment above thread. Based on my status as a licensed lawyer. It is all about the EPA endangement finding. And citing Sherwood 1 or Sherwood 2 in apposition is a guaranteed loser.
We are going to have to rename you ‘logic pretzel’.
“We are going to have to rename you ‘logic pretzel’.”
They already tried that. But it’s a simple and familiar argument. If a prosecution case fails, it doesn’t mean that the defence account was proved. It means that the prosecution failed to reach “beyond reasonable doubt”. Statistical testing quantifies that. It says something is significant at 95% if there is less than a 5% chance that the null hypothesis is consistent with it. If it isn’t significant, that means that there is more than a 5% chance that the null hypothesis is consistent with it. That isn’t a proof of the null hypothesis.
Nick Stokes September 22, 2016 at 3:37 pm
Well, this article says:
“The crux of EPA’s argument rested on the existence of a “tropical hotspot” where global warming would be most apparent.”
And still no quote. But the paper does not disprove the existence of a hotspot.
—————–
Well it also doesn’t disprove the existence Sasquatch. Santa or the Easter Bunny, but from a scientific method standpoint doesn’t mean they exist either.
The burden of proof is on the one proposing the theory, not the other way around.
Show mean one climate model from twenty years ago that successfully and accurately predicted global temperature since then, given the rise in CO2 over that period. Can you?
Okay, Reg. How about a whole bunch of predictions even from way earlier than 20 years ago? See Ray Pierrehumbert’s 2012 AGU lecture: https://youtu.be/RICBu_P8JWI
Tom Dayton September 22, 2016 at 5:11 pm
Okay, Reg. How about a whole bunch of predictions even from way earlier than 20 years ago? See Ray Pierrehumbert’s 2012 AGU lecture: https://youtu.be/RICBu_P8JWI
————
Hey, Tom, feel free to answer my original question.Again I ask, can you?
There have been literally thousands of peer reviewed paper published in the last 35 five years. What percentage of them have been correct? I haven’t found one.
Nick Stokes
Except that the “prosecution” of CO2 was based on the presumption of guilt and the defendant was not allowed to enter a plea nor a defence council.
The test of beyond reasonable doubt was never examined.
Nick
I am pretty certain that the statement in bold is incorrect. Hypothesis testing doesn’t present evidence of the alternate hypothesis being true.
“Hypothesis testing doesn’t present evidence of the alternate hypothesis being true.”
OK, there is an up to 95% chance that the models are not wrong. What hypothesis testing certainly doesn’t do is to allow you to assert that the alternative hypothesis is disproved when the null can’t be rejected at 95%. And that is what they are doing.
Nick,
I have now read the paper and the case they present is very convincing – particularly for the tropical troposphere. The residual slopes for the suitably MEI adjusted temperature series are as flat as a pancake. Stating there is a 95% chance they are wrong in their conclusions doesn’t fit with the case presented in my view.
“Stating there is a 95% chance they are wrong in their conclusions”
Well, I didn’t say that either. It’s disjunctiveness that is their basic fallacy. Their hypothesis testing just tests the null. You may be able to show the result is so inconsistent with it that the null can be ruled out. That gives significance. But if you can’t rule out the null (even if there is only 5% chance), that doesn’t disprove an alternative hypothesis, as they claim. Either could be true. And often, the alternative may be more consistent with the data, but still not succeed at 95%.
Warmistas and Alarmists are proof that even when your head has fallen off, you can still keep on walking and talking, and even writing.
Oh yes you can.
http://www.miketheheadlesschicken.org/
Reblogged this on Climatism and commented:
But the (missing) “Hotspot” exists in junk-in, junk-out computer models, which are apparently evidence enough to justify the destruction of capitalism, and the spending of trillions of taxpayers $£€¥ on useless schemes and scams in a hubristic attempt to control the weather and halt an increment of (modelled) warning that, most probably, would be beneficial to humanity.
The “Missing Hotspot” IMHO is one of the most important (missing) pieces of the global warming aka climate change debate…
No Hotspot = Global Warming theory fail.
https://climatism.wordpress.com/2014/01/28/the-missing-hot-spot/
“…only the atmospheric CO2 levels are deemed anthropogenic, that is, impacted by human activity such as the burning of any fossil fuel. The three other explanatory variables are considered “natural” variables. By natural is meant that each of the variables’ values are not impacted by human activity.”
“GAST = F1(CO2, SA, VA, ENSO)”
Let me see. If temperatures are explained by the MEI then they could be explained by the MEI and not necessarily CO2. Their assumption seems to be that ENSO is not affected by CO2. To put it another way, if the MEI correlates with temperature, CO2 is pushed more to the sidelines.
It doesn’t matter to me that ENSO is chaotic and not too predictable. But is it all natural? If ENSO is impacted by CO2 we’d have:
CO2 > ENSO > GAST
Or is it:
ENSO > GAST
If it is:
CO2 + ENSO > GAST
Then I think the test done needs to be changed some.
ENSO has been around for thousands of years, anthropogenic carbon dioxide, not so much. QED, natural, not impacted by the evil gas.
Here is my email of earlier today set to the authors of the report:
———————————————–
Subject: Comments on “On the Existence of a “Tropical Hot Spot “& the Validity of EPA’s CO2 Endangerment Finding”
I read your research paper with interest, website here. https://thsresearch.wordpress.com/
Unfortunately, the section VIII contains misleading graphs in that the red lines on the three graphs are labeled “trends” whereas they are two horizontal line segments. The term “trend” implies it is a best fit regression line of the data including a slope, rather than a horizontal line through the average of the data.
Figure VIII-1 shows the tropical 200 mb balloon temperature anomaly with horizontal red lines from 1959 to 1976 and from 1977 to 2015. This curve is labeled “Step Trend”, implying that the 2 line segments are regression best fits, but they are not.
Figure VIII-2 shows a red line consisting of two horizontal red lines, each through the average of the data point corresponding to the two line segments, but the curve is labeled “Step Trend”.
I created a graph of the MEI with horizontal lines through the data average and trend lines for the two segments separated by the 1977 climate shift, shown below.
The Annual MEI and the red curve are identical to Figure VIII-2, but I labeled the red curve “Step Average”, indicating the the line is showing only the average values, not trends. The green curves are the Excel calculated best fit lines, indicating declining trends.
Page 23 says “Below the Annual MEI Step Trend rose by 0.66 in 1977.” The actual rise based on the horizontal red lines is 0.645.
Figure VIII-3 is a graph the the tropical 200 mb balloon temperatures, with the title “MEI trend adjusted”. However, this section doesn’t explain how the temperature data was adjusted by the MEI trends, or even show what the MEI trends were. The NOAA MEI data page says “Bimonthly MEI values (in 1/1000 of standard deviations), starting with Dec1949/Jan1950, thru last month. … All values are normalized for each bimonthly season so that the 44 values from 1950 to 1993 have an average of zero and a standard deviation of “1”.”
The MEI data are in unitless values of “standard deviation”, while the balloon data are in units of ºC. The MEI values must be converted to units of ºC to determine the adjustment, but the paper does not explain how this was done. In fact, it was not done as explained below.
This figure has a horizontal red line through the data a value exactly 0 ºC, but the curve is labeled “Adjusted AV3 Trend”, implying that it is the best fit regression line through the data. It is not.
I recreated the figure shown below.
Instead of showing a horizontal red line, I show the linear trend in black. This is not a valid trend as the MEI adjustment in invalid.
The “purple” curve is actually a blue MIE Adjusted temperature curve with a thinner width red curve on top of it. I digitized the MEI adjusted temperature data from figure VIII-3. The red curve labeled “T+step” is the 200 mb temperature data adjusted by adding a constant value of 0.507 ºC from 1950 to 1976, which is the 200 mb temperature step change from the average of 1959-1976 to the average of 1979-2015 temperatures. I also subtracted 0.09 ºC from the temperature date from 1959 to 2015 to make the average value over the total period sum to zero. The red curve is mostly within the blue curve, so you can see that I have reproduced what you have done.
The MEI adjustment is invalid because you did not use the MEI data to make the MEI adjusted temperature data, contrary to the title of the figure. You simply adjusted the 1959-1976 temperature data upward by the amount of the 1976 temperature shift based on the horizontal lines.
The paper says at the bottom of page 25 “Adjusted Tropical 200 mb and 150 mb temperatures do not have a statistically significant trend line slope. -suggesting that CO2, taken together with all other omitted variables, is not the cause of the rise in this Tropical Balloon temperature data.” Readers are mislead to believe that the horizontal line is a trend line, which is not shown. I would think to make this claim the paper should show the trend and its statistical significance. The analysis does not support your conclusion. Dr. Roy Spencer published a paper that shows the effect of ENSO on temperature rise. Accounting for ENSO he determines the equilibrium climate sensitivity (ECS) is 1.3 ºC indicating the the temperature rise can’t be explained only by ENSO, see here. ECS us actually less than 1.3 ºC as Spencer’s analysis does not take into account other causes of climate change, including the millennium scale natural warming from the Little Ice Age, and the 65-year ocean oscillations.
Possible typo:
Page 19 of the full report say “The results are as depicted in the Figure VII-4 and Table VII-1 below.”
However, the figure and table are labeled Figure VI-4 and Table VI-1, respectively. (Section VII has no figure or table.)
Regards,
Ken Gregory
Friends of Science http://www.friendsofscience.org/
Fourteen years of providing independent
climate science information
In response, Joe d’Aleo wrote in part:
Please go back and read the preface that explains what Jim an econometrician with 50 years experience in data analysis actually did. … He used the Cumulative MEI which shows the cumulative effect of El Nino and La Nino (strength and duration). It correctly showed the cooling to 1977, warming to 1997 then stasis.
I replied,
The preface doesn’t not refute anything I wrote. Figure VIII-3 did not use MEI nor cumulative MEI. I accurately reproduced the figure, see below without using MEI information.
50 years experience is not an excuse for publishing misleading graphs.
CO2 is not a source of energy so for it to cause global warming, added CO2 would have to cause an increase in the insulation properties of the atmosphere. One measure of the atmosphere’s insulation properties would be the natural lapse rate. An increase in the lapse rate would signify that the thermal resistance of the atmosphere was increasing and a decrease would signify that the thermal resistance of the atmosphere was decreasing. If CO2 did actually cause global warming then the increase in CO2 over the past 30 years should have caused at least a noticeable increase in the natural lapse rate in the troposphere but that has not happened. To date there is no real evidence that CO2 has any effect on climate.
No scientist has claimed that CO2 is an energy source.
The lapse rate is the negative of the change in temperature per change in altitude, which is currently about 6.5 C/km. An increase in the lapse rate would cause a tropical cool spot in the mid-troposphere, rather than the predicted hot spot. The predicted decrease in the lapse rate causes the hot spot in models, but not in reality. Since greenhouse warming occurs in the mid-troposphere and the surface temperatures responds to that warming via the decreasing lapse rate, the lapse rate change is a negative feedback. That is, if the lapse rate declines from 6.5 to 6.0 C/km, the surface would warm slower than the mid-troposphere. But this is not a measure of the thermal resistance. The value of the greenhouse effect IS the measure of the thermal resistance, which is the difference between the effective radiating temperature as measured from the top of the atmosphere, and the surface temperature. According to CERES satellite data, and the HadCRUT4 data (which is contaminated by urban warming), the transient climate response (TCR) (to double CO2, about 130 years) is 0.74 C +/- 0.5 C. See here:
https://friendsofscience.org/assets/documents/CERES/CERES_Climate_Sensitivity.pdf
Correcting for urban warming, the TCR is 0.3 C [0 – 0.8 C]
However, using the IPCC greenhouse forcing, Steven’s aerosol forcing, HadCRUT4.4 and correcting for millennium scale warming from the Little Ice Age and UHI, the TCR is 0.85 C [likely 0.7 – 1.1 C].
See https://friendsofscience.org/assets/documents/AB_Climate%20Plan_Economic_Impact_Gregory.pdf
It depends on what data you trust the most. In any case, using the high numbers, the net social benefit of carbon dioxide emissions is 17 US$/tonneCO2, that is benefits are much greater than alleged harmful effects.
How temperature changes with altitude is a measure of the insulating effects of the atmosphere. Derived from first principles, the lapse rate is a function of the heat capacity of the atmosphere and the pressure gradient and has nothing to do with the LWIR absorption properties of so called greenhouse gases. The initial calculations of the climate sensitivity of CO2 were to great by a factor of 20 because of the fact that the doubling of CO2 would cause a slight decrease in the dry lapse rate which is in itself a cooling effect, was ignored. There is no radiant greenhouse effect anywhere in the solar system, even on Venus.
For the ‘hot spot’ model forecasts of enhanced CO2 warming to be validated, the temperature trend in the tropical troposphere 9 – 12 km must be shown to warming at double to triple the surface rate (hence ‘hot spot’):
http://www.climate4you.com/images/EquatorSurface300hPa200hPaDecadalTempChange%20BARCHART.gif
Diagram showing observed linear decadal temperature change at surface, 300 hPa and 200 hPa, between 20oN and 20oS, since January 1979. Data source: HadAT and HadCRUT4. Click here to compare with modelled altitudinal temperature change pattern for doubling atmospheric CO2. Last month included in analysis: December 2012. Last diagram update: 4 May 2013 (Climate4you).
And this is why this AGW theory is plain old wrong.
The authors make the following statement:
“Strictly speaking, the ENSO-adjusted temperatures represent the estimated combined impact of CO2 as
well as the two natural variables, solar and volcanic activity. For example, Volcanic Activity could be hiding CO2’s impact. So, for example, if GAST, or any other temperature time series, has a statistically significant, positive linear trend slope and ENSO-Adjusted GAST does not, then the positive trend slope in GAST can be totally explained by the natural ENSO impacts alone.”
However it has been shown that ENSO-adjusted temperatures do show a positive linear trend slope:
http://www.realclimate.org/images/gistemp_mei_correct.jpg
It has also been shown that when adjustments for volcanic activity and solar variation are added positive linear trend slopes still result:
As the authors say in their report:
“Since the Atmospheric CO2 concentration levels are independent of ENSO variable values, removing only the ENSO-related impacts on the temperature time series does not require the specification of a more complicated (i.e., multi-equation) climate model and therefore the use of simultaneous equation parameter estimation techniques – for that matter, neither does removing SA or VA impacts.”
GISStemp is no longer a valid data set,due to numerous adjustments in recent years.
But RSS and UAH are?
Despite their “numerous adjustments”
There is a difference Toneb,but will you care to understand it?
Sunsettommy September 23, 2016 at 6:37 am
GISStemp is no longer a valid data set, due to numerous adjustments in recent years.
Some adjustments are necessary, just as the adjustments were necessary for the satellite measurements, those measurements aren’t even of the same quantity as they were before the adjustments.
In any case the same adjustments can be made on the satellite measurements with the same result.
“There is a difference Toneb,but will you care to understand it?”
No difference at all.
Unless you invoke conspiracy my friend.
What is it about the two sat data records that are more *believable*?
Even though in the case of UAH there is no account of the latest *adjustments* available for peer-review.
Even their Wiki page has not been updated to include V6 beta even though is says updated 16th Sept 2016!
There are multiple data records available for the surface – all saying the same thing. Are they all fraudulent or is just one enough??
Additionally UAH is drifting away from RATPAC sonde data sine 2000 (change of sensor on new sat)….
Just as RSS is STILL even after
changes from V3.3 to v4.0
The surface record has been homegenised for good reason. Those reasons can be found easily, as can the raw data.
Toneb September 23, 2016 at 7:41 am
But RSS and UAH are?
Despite their “numerous adjustments”
Yes, because RSS and UAH don’t need to adjust because of ridiculous, “bucket ship” data.LOL
Is it really that difficult for your to understand? Seriously, is it?”
You can’t adjust GISTemp for El Ninos or La Ninas because there is no ENSO signal remaining in GISTemp. Its been changed to a straight line going up.
Now, it wasn’t always that way. From 2005. And note of course we are talking about “TROPICS” here, the topical hotspot.
http://data.giss.nasa.gov/gistemp/2005/2005_fig2.gif
“Yes, because RSS and UAH don’t need to adjust because of ridiculous, “bucket ship” data.LOL
Is it really that difficult for your to understand? Seriously, is it?””
No on the contrary – I find it very easy to understand.
Your prefference and distorted, myopic view of the two data sources, that is.
On the one hand we have SURFACE data (where we live) and the other hand we have data from a large slice of the atmosphere that includes (in the case of TMT) – the cooling stratosphere.
On the one hand we have data that is measured by a thermometer in thousands of locations that has both raw and homegenised data sets available that show **er all difference in the period since CO2’s +ve forcing overtook aersol’s -ve and actually warm the past.
http://4.bp.blogspot.com/-opy7LoBO__w/VNoo9u5ynhI/AAAAAAAAAg4/_DCE5Rzm9Fw/s700/land%2Bocean%2Braw%2Badj.png
On the other we have a single instrument (both RSS and UAH use the same sat/sensors).
The instrument is moving in orbit and measures brightness. This has to be modelled by a complex algorithm (that in the case of UAH v6 bet is currently secret). The algorithm includes a mdel BTW. The satellite drifts and this has to be estimated. The sensor is calibrated against REAL WORLD radiosonde (except it isn’t v RATPAC and both RSS and UAH are currently too cold).
From RSS:
“A method is chosen based on an optimized second harmonic adjustment to produce a new version of the RSS dataset, version 4.0. The new dataset shows substantially increased global-scale warming relative to the previous version of the dataset, particularly after 1998. The new dataset shows more warming than most other midtropospheric data records constructed from the same set of satellites. It is also shown that the new dataset is consistent with long-term changes in total column water vapor over the tropical oceans, lending support to its long-term accuracy.
Would you be so sanguine if they were too WARM?
Therin lies the reason I find your position very easy to understand my friend.
Toneb claims to be a meteorologist and when asked the name of the law of thermodynamics for solving atmospheric and gas mix temperature, couldn’t do it.
He’s another fake physics and mathematics bumbler who can’t predict which way a thermometer will go, prompted with correct answers ahead of time.
He even thinks the stratosphere is cooling. It isn’t cooling, it hasn’t cooled any for the past decade, at least.
He’s a science darkener, preaching about the GHE that makes him unable to predict or even know the names of the laws he claims to be an expert on.
The stratosphere and it’s alleged cooling: fake. Toneb thought that sh** is real.
http://planetaryvision.blogspot.com/2015/09/stratosphere-not-cooling-as-predicted.html
He even thinks the stratosphere is cooling. It isn’t cooling, it hasn’t cooled any for the past decade, at least.
It certainly is, check out RSS MSU results for the stratospheric layers C10 and above, C12 has a slope of -0.477K/decade for example.
“Other climate models differ in their assessment of future El Niño events. Some suggest the ENSO cycle will become more intense, others say it will weaken, and some find there will be little change. According to Schmidt, “There is a very large variation in ENSO statistics (frequency/magnitude) over time, and so detecting a shift due to climate change is very challenging. Models as a whole are all over the shop, and so it doesn’t fill one with great confidence.””
http://blogs.ei.columbia.edu/2016/02/02/el-nino-and-global-warming-whats-the-connection/
So it is unknown if CO2 changes ENSO.
ENSO is suggested to impact the Greenhouse Effect:
A Hiatus of the Greenhouse Effect
http://www.nature.com/articles/srep33315
Facts will not deter the true believers.
I ran my Nino 3.4 Index, AMO, Volcanoes (AOD Index) and Ln (CO2) model on the average of HadAT, UAH and RSS Tropics temperatures and the monthly model provides an extremely good fit.
The Nino 3.4 (lagged 3 months) coefficient is 0.288 which means the ENSO runs the Tropical troposphere. Think of February 2016 when a coefficient of 0.288 * 2.95C (Nino 3.4 lagged 3 months) = 0.85C of temperature impact by itself.
Here is the model (just regressing all these variables against the temperature in the Tropics troposphere). Why would this be so darn close across the whole 58 years and 690 individual months including ALL of the individual up and downs?
http://s16.postimg.org/iynkqumyd/UAH_RSS_Had_AT_Tropics_Model_Aug16.png
And the left-over Ln (CO2) warming trend is basically NOTHING. 0.029C per decade ??? By the year 2100, we only get a total warming of 0.35C at this rate. There is certainly NOT a tropical hotspot. It is warming less than anywhere else. If anything, there is tropical cooling spot.
http://s15.postimg.org/dw2h9wpff/UAH_RSS_Had_At_Tropics_Warming_Aug16.png
When looking at a chart of sunlight TOA and surface, there’s only one class of gases reducing energy to the surface. Green House Gases.
There’s more to this.
There’s also a large bar, going from the bluest light to the reddest light, where the entire sunlight spectra are simply not reaching the planet
This removal of white class light is also: solely by green house gases, this being primarily water of course which has it’s noted white clouds, snow in them, ice crystals, etc.
There’s a reduction to the planet surface by oxygen which is why daylight skies are blue.
The Green House Gases are refractory in nature. They deflect light to space that never reaches the surface.
This reduction in surface energy density is cooling. There’s no such thing as putting refractory media between a fire and something it’s warming, and the reduction of light to the surface of that object create higher temperature.
In other words there’s no way to make more source-dependent light come out of an object by placing more refractory media between it and the fire illuminating it.
The entire GHG/AGW scam is a gigantic fraud. If any of it were real, ***the people who believe in it wouldn’t be the ones in the early 21st century blamed for trashing science worse than anyone alive has known.***
We all say this, scientists across every field of endeavor: it’s all fake, and all it’s proponents are incompetent kooks; and it proves itself true again and again, day in – day out.
“Ross McKitrick, an environmental economist at the University of Guelph in Canada, also ran the numbers and found climate models overestimated warming in the tropical troposphere.”
Hm. Climate models do automatically consider an ‘increased backradiation’ with increased CO2 at the top of the atmosphere, all other things being equal. If I correctly remember it was something like +3.5°C per doubling of CO2. To this added ‘feedbacks’…
This is not how the physics happen in the atmosphere, the increase in CO2 would mainly only shorten the heat exchange path at the bottom of the atmosphere between the surface and the first 10-20 meters of air. Even if considering a broadening of the waveband through Doppler effect, it is still at the surface.
With this difference in mind it is clear why climate models get automatically a hot spot in the troposphere and why it never happens in real life…
Ooops, sorry where is my mind.. it was 3.5 W/m2 not °C