New global water vapor findings contradict second draft of IPCC Assessment Report 5 (AR5)
Guest post by Forrest M. Mims III
I was an “expert reviewer” for the first and second order drafts of the 2013 Intergovernmental Report on Climate Change (IPCC) Assessment Report 5 (AR5). The names and reviews of all the reviewers will be posted online when the final report is released. Meanwhile, reviewers are required to not publish the draft report. However, the entire second draft report was leaked on December 13, 2012, without IPCC permission and has subsequently received wide publicity.
My review mainly concerns the role of water vapor, a key component of global climate models. A special concern is that a new paper on a major global water vapor study (NVAP-M) needs to be cited in the final draft of AR5.
This study shows no up or down trend in global water vapor, a finding of major significance that differs with studies cited in AR5. Climate modelers assume that water vapor, the principle greenhouse gas, will increase with carbon dioxide, but the NVAP-M study shows this has not occurred. Carbon dioxide has continued to increase, but global water vapor has not. Today (December 14, 2012) I asked a prominent climate scientist if I should release my review early in view of the release of the entire second draft report.
He suggested that I do so, and links to the official IPCC spreadsheet version and a Word version of my review are now posted near the top of my homepage at www.forrestmims.org.
The official IPCC spreadsheet version of my review is here. A Word version is here.
A PDF version (prepared by Anthony from the Word version) is here: Mims_IPCC_AR5_SOD_Review
A relevant passage from the AR5 review by Mimms (added by Anthony):
The obvious concern to this reviewer, who has measured total column water vapor for 22.5 years, is the absence of any mention of the 2012 NVAP-M paper. This paper concludes,
“Therefore, at this time, we can neither prove nor disprove a robust trend in the global water vapor data.”
Non-specialist readers must be made aware of this finding and that it is at odds with some earlier papers. Many cited papers in AR5 have yet to be published, but the first NVAP-M paper was published earlier this year (after the FOD reviews) and is definitely worthy of citation: Thomas H. Vonder Haar, Janice L. Bytheway and John M. Forsythe. Weather and climate analyses using improved global water vapor observations. GEOPHYSICAL RESEARCH LETTERS, VOL. 39, L15802, 6 PP., 2012. doi:10.1029/2012GL052094.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
MiCro says: “My data shows the same flat temps for the last 16 years or so, and I used all of the actual measured data (well 114 Million records out of ~120 million), no cherry picking involved.”
The cherry picking in this case wasn’t the stations used, but the period selected. The period of 16 years was optimized not to show any trend.
Mister Böehm, maybe it is superfluous to say so, but my one counter example is logically sufficient to disproof your rather strong and absolute statement that “the alarmist crowd has not got anything right yet”.
vvenema,
I looked at your blog, which you claim is your authority. That is somewhat circular, no?
You report on the unreliable Surface Stations network, but I would prefer to listen to Anthony Watts, who has done considerable work on the USHCN network. Anthony shows that USHCN data is useless. When you input worthless data, what do you expect to see in the output? Your ‘study’ is a perfect example of GIGO.
vvenema:
Concerning the recent 16-year period of no discernible warming at 95% confidence, at December 18, 2012 at 12:02 pm you assert
Bollocks!
There is NO such “cherry picking”.
“The period selected” starts from now and considers how far back in time one has to assess before discerning global warming at 95% confidence.
This is important because in 2008 the US National Oceanographic and Atmospheric Administration (NOAA) stated in its State of the Climate Report for 2008 (page 23)
Please note the strength of that statement: it says the models’ simulations RULE OUT a zero trend of 15 years or more, but that has happened.
In other words, the 16-year period of no discernible global warming (at 95% confidence) demonstrates the climate models are wrong.
Your excuses are unpardonable.
Richard
vvenema says:
December 18, 2012 at 11:59 am
“This study was blind, so that these “alarmists” could not cheat. The results showed that homogenization improved the temperature data as promised. ”
If your data has Min/Max daily temp records, I’d be more than will to examine that data.
“The cherry picking in this case wasn’t the stations used, but the period selected. The period of 16 years was optimized not to show any trend.”
But as you can see, I did no such picking, actual measured data has the trend flat for the last 16 (or so ) years.
Why this is critical is that models, being coded (through the CS algorithms) to force an increase of temps with increasing co2, can not and do not show the actual measured flat temps.
Böehm, I do not like authority that much, I prefer good arguments and clear formulations. My blog helps me to have to type a little less when commenting. I understand that you would prefer to listen to Anthony Watts.
richardscourtney, I feel the statement of NOAA is much too strong and would say that climate models are not that well suited to study natural variability and are bound to underestimate it because not all relevant processes are taken into account.
But NOAA are save by luck. The 15 year temperature trend they mention is significantly positive, just as the 17 year temperature trend. It is only the 16 year temperature trend that is marginally not significant. And if you take into account that David Rose searched for such a case, you have to take multiple testing into account. (Multiple testing: If you search for relationships between x and 20 other variables, one of these 20 relationships will be statistically “significant” at the typical 5% level by accident.)
MiCro, yes the validation study includes minimum (Tn) and maximum (Tx) temperature, as well as mean temperature (Tm). You can download the homogeneous, inhomogeneous and homogenized data from my homepage. There you can also find a link to the open access article with results and a report with all the detail on how the data was generated. Have fun with the analysis. It is monthly data, however, and without corresponding humidity, thus it would not help you to study the relationship between the DTR and humidity.
Climate models are not forced to respond to an increase in CO2 with a temperature increase. They compute a simplified 1-dimensional version of the radiative transfer equation, an equation that is also used in weather prediction models, in astronomy and to interpret laboratory experiments with electromagnetic radiation. If you searching for an error in the climate models, radiative transfer would be the last place I would look. You have more chances of finding problems if you look at feedbacks, such as humidity.
Given its importance, I feel that humidity is studied much to less. Thus I am looking forward to any results you will show in future, at least if you perform them in a reliable way. If you can proof that one of the standard steps leads to wrong results, that would be very interesting and a perfect reason to use a better alternative. If not, please use the best tools around. Many of them can be freely used, for example here is a list with open source homogenization algorithms.
[…humidity is studied “much to less” ? Mod]
humidity is studied much too less.
vvenema
you seem to be saying that, no, you cannot accomodate your theory to observations.
vvenema says:
“I would also love to get out of that rut. If I could do so with good arguments, I would get a Nobel price. Can you offer any help?”
=========================================================================
I would, if you promise to share your Nobel prize with me. Try the saturation theory. If CO2 is at saturation, then the effect of added increments is now negligible. This is an opportunity going begging. What you should do is compose about ten pounds of documents and charts, roll it into a big roll, and smite climate scientists with it. I promise that you will be awarded a Nobel Prize and freedom medals galore from a greatful world.
vvenema says:
December 18, 2012 at 1:38 pm
But NOAA are save by luck. The 15 year temperature trend they mention is significantly positive, just as the 17 year temperature trend. It is only the 16 year temperature trend that is marginally not significant.
It depends on your data set. For example RSS has a negative slope for the last 15 years and 16 years. And while the slope from the last 17 years to the last 23 years is positive, it is NOT significant at the two sigma level. Here are the numbers.
For RSS: +0.130 +/-0.136 C/decade at the two sigma level from 1990
For RSS: +0.135 +/-0.147 C/decade at the two sigma level from 1991
For RSS: +0.142 +/-0.159 C/decade at the two sigma level from 1992
For RSS: +0.107 +/-0.166 C/decade at the two sigma level from 1993
For RSS: +0.069 +/-0.174 C/decade at the two sigma level from 1994
For RSS: +0.043 +/-0.190 C/decade at the two sigma level from 1995
For RSS: +0.036 +/-0.210 C/decade at the two sigma level from 1996
For RSS: -0.003 +/-0.229 C/decade at the two sigma level from 1997
For RSS: -0.045 +/-0.250 C/decade at the two sigma level from 1998
Werner Brozek,
Thank you for correcting vv enema. Unfortunately, he will not get out of his rut because he keeps making incorrect statements.
Correction: vvenema. Sorry.
@Victor Venema: You wrote: “That is the main difference with people who think that the sun is responsible for the recent temperature rise, they might have a correlation, but they do not have a working mechanism.”
I believe there are working mechanisms – but I will counter in this way.
The IPCC do not have a “working” mechanism or model.
They only had correlation, which they immensely relied upon. Their entire theory is predicated on the “past” belief that CO2 was rising and this causes the temperatures to rise. I believe that many of IPCC’s members no longer believe CO2 drives climate such that it will lead to catastrophe. That means I think those who do not speak out are dishonest.
Werner Brozek says: “It depends on your data set. For example RSS has a negative slope for the last 15 years and 16 years. And while the slope from the last 17 years to the last 23 years is positive, it is NOT significant at the two sigma level.”
I only mentioned the significance of the 15 year trend, to illustrate that the 16-year period selected by David Rose was cherry picked. Sorry, if I do not worry that there is no significant positive trend in a short dataset of doubtfull quality. The long term trend in good quality data is clearly positive. No one has explained to me why I should see the worst trend estimate as more important.
By the way, do the authors of the RSS dataset say something about the uncertainty of its trends? I just read a somewhat older paper in which the trend in the RSS dataset was said to be double of the trend in the uah dataset. As these differences resolved yet, or should we still see this difference as a measure of the uncertainty of satellite estimates of surface temperatu
re trends?
D Böehm says: “Thank you for correcting vvenema. Unfortunately, he will not get out of his rut because he keeps making incorrect statements.”
Dear Mister Böehm, I just caught you making an incorrect statement. If you do not give an argument why my argument was wrong, I will presume that you acknowledge that you made an incorrect statement.
Especially in that case, it would be good form to explictly say which statement of mine you see as incorrect, so that we can discuss this topic.
May I ask, are you from Austria and can you read German. Your (distant) family member Reinhard Böhm, who unfortunately recently died, has written a beautiful book on climate change: “Heiße Luft – Reizwort Klimawandel Fakten – Ängste – Geschäfte”. You may like it, he is far from an alarmist and maybe you do trust family.
Mario Lento says: “I believe there are working mechanisms – but I will counter in this way.”
One would be sufficient.
Mario Lento says: “The IPCC do not have a “working” mechanism or model. They only had correlation, which they immensely relied upon. Their entire theory is predicated on the “past” belief that CO2 was rising and this causes the temperatures to rise. I believe that many of IPCC’s members no longer believe CO2 drives climate such that it will lead to catastrophe. That means I think those who do not speak out are dishonest.”
Do I understand you right, that you are even doubting that the CO2 concentration is increasing?
Our understand of the climate is not based on correlation. It is based on our physical understanding of the physical processes in the climate system, which are partially implemented in global climate models. If the results of the global climate models would not fit (correlate) with reality that would be a problem, but you should not forget the physical reasoning.
What do you call a catastrophe? Estimating what the consequences will be is very difficult as you need to know how people will respond to climate change. That is no longer natural science and consequently more a matter of believe. Feel free to spread as much doubt about this as you can. What worries me about this blog is that the basic facts are doubted without good arguments and that posts that misinform such as this guest post do not create an outcry. Don’t you want to well informed?
Victor Venema:
I write to offer some sincere and genuine advice.
You are floundering and it is obvious to all that you are floundering.
You need to learn the first rule of holes; i.e. when in a hole then stop digging.
Richard
Victor Venema says:
December 19, 2012 at 2:31 am
By the way, do the authors of the RSS dataset say something about the uncertainty of its trends? I just read a somewhat older paper in which the trend in the RSS dataset was said to be double of the trend in the uah dataset. As these differences resolved yet, or should we still see this difference as a measure of the uncertainty of satellite estimates of surface temperature trends?
As for the uncertainty in RSS, I just relied on the numbers from
http://www.skepticalscience.com/trend.php
The long term trends for both RSS and UAH are virtually the same, but there are significant differences lately. UAH is working on a version 6 but it has not been completed yet. One of the satellites seems to be getting worse all the time. The four slope lines below show what I mean.
http://www.woodfortrees.org/plot/rss/from:1979/plot/rss/from:1979/trend/plot/uah/from:1979/plot/uah/from:1979/trend/plot/rss/from:1997/trend/plot/uah/from:1997/trend
Thank you Werner Brozek, looks like the problem of the strong trend differences in the satellite records has been solved. The trend of RSS and UAH is very similar; they could still both be wrong; one is always most sceptical about the things you know least about, in my case satellite data.
The uncertainty in the trend in the Skeptical Science tool is estimated purely statistically, according to the paper they link to. If there were a bias in the trend, such a statistical estimate would not notice. Thus the true uncertainty is probably larger.
The trend from 1979 to now of RSS is: 0.133 +/- 0.073 °C per decade at the 2 sigma level. I have no idea what you should take as uncertainty due to remaining inhomogeneities in the satellite record, but that may be significant. Interesting is that you reported:
For RSS: +0.130 +/-0.136 C/decade at the two sigma level from 1990
While the trend from 1979 to 1990 is basically flat. This shows how important having a sufficiently long period is to get significant results.
A really nice tool. It also shows for GISTEMP that between 1940 and 1980, a 40 year period, there was no trend. That puts a puny 16-year period in perspective. In both cases it is cherry picking the right period, naturally.
Dear Richard S. Courtney, given that the situation is probably symmetrical and we both are doubting the honesty and mental capacities of the others, it might be more productive to discus very specific questions as general advice.
That is also how science progresses. Although scientists can be weird, unconventional and sceptical people, if you split up a problem in sufficiently small sub-problems, in the end you end up with problems which are sufficiently small and clear that every rational being can agree upon the answer.
vvenema:
Thankyou for the comment in your post at December 19, 2012 at 10:26 am which says
I agree, and I appreciate your taking my comment to you with the genuine sincerity with which it was offered.
I see you have started to discuss the MSU data with Werner Brozek along the lines you suggest of “sufficiently small sub-problems”. Werner has made himself expert in use of the on-line tools for assessing linear temperature trends. Indeed, I have adopted the practice of quoting him when citing such trends because I trust his expertise. However, I doubt that linear trends are appropriate for assessing global and hemispheric temperature time series. Despite my doubts, I accept that those who present the time series say they like linear trends so your interaction with Werner may prove productive.
Unfortunately, I am to leave on one of my absences tomorrow so will be isolated from the internet again until some time in the New Year. You are new to WUWT so are probably unaware that I often go on these absences: please be assured that it is not a measure to avoid interaction with you and I would welcome such interaction upon my return.
Richard
vvenema says:
December 19, 2012 at 10:26 am
It also shows for GISTEMP that between 1940 and 1980, a 40 year period, there was no trend. That puts a puny 16-year period in perspective. In both cases it is cherry picking the right period, naturally.
The above shows that CO2 could not have been the driver it was claimed to be. It is generally accepted that CO2 really went up after around 1945. However as for the “puny 16-year period”, you will have to take that up with NOAA. They believe 15 years of no warming is highly significant.
[snip. How does it feel being persona non grata here? Troll elsewhere. — mod.]
Werner Brozek says:
Except that what they said differs from your 15 years claim in two very important aspects:
(1) They talked about the temperature trend after correcting for ENSO in a way described in a paper they referenced.
(2) They said that a trend of zero for 15 years would be outside of the 95% significance cone of the modeling results. However, the trend hasn’t been zero for the last 15 years. Claiming the trend is zero vs claiming that the 95% significance cone on the actual trend includes a trend of zero are two very different things.
joeldshore says:
December 19, 2012 at 5:13 pm
(1) They talked about the temperature trend after correcting for ENSO in a way described in a paper they referenced.
(2) However, the trend hasn’t been zero for the last 15 years.
I did not go deeper into the report, but if your point 1 is true, and I am not saying it isn’t, then CO2 is clearly NOT the huge driver they thought it was. After all, there was an El Nino in 2010, and at least 4 data sets still have 1998 as the hottest year.
As for your point 2, it depends on the data set.
Data sets with a o slope for at least 15 years:
1. HadCrut3: since April 1997 or 15 years, 7 months (goes to October)
2. Sea surface temperatures: since March 1997 or 15 years, 8 months (goes to October)
3. RSS: since January 1997 or 15 years, 11 months (goes to November)
See the graph below to show it all.
http://www.woodfortrees.org/plot/hadcrut3gl/from:1997.25/trend/plot/rss/from:1997.0/trend/plot/hadsst2gl/from:1997.1/trend/plot/hadcrut3gl/from:1997.25/plot/rss/from:1997.0/plot/hadsst2gl/from:1997.1
However in view of the significance of the 16 years lately, I would like to elaborate on RSS. The slope for 15 years and 11 months from January 1997 on RSS is -4.1 x 10^-4. But the slope for 16 years and 0 months from December 1996 is +1.3 x 10^-4. So since the magnitude of the negative slope since January 1997 is 3 times than the magnitude of the positive slope since December 1996, I believe I can say that since a quarter of the way through December 1996, in other words from December 8, 1996 to December 7, 2012, the slope is 0. This is 16 years. Therefore RSS is 192/204 or 94% of the way to Santer’s 17 years.
Werner Brozek, CO2 is indeed not a huge driver of the climate. The sun is much more important: simply compare the large size of the day cycle and the annual cycle with the small, about 1 degree, change in the long term climate. And as MiCro has shown also the variability of the daily cycle depends much more on humidity, as much stronger greenhouse gas as CO2. The difference is that CO2 (Methan and Distickstoffmonoxid, etc.) are increasing, whereas we cannot directly change humidity as it would rain out and the long term changes in insolation are too small.
Furthermore, these greenhouse gasses increase the amount of energy available at the surface. Only a small part of this energy goes into increasing the temperature, the largest part of this additional energy is absorbed by the ocean.
For all of these reasons it is important to look at longer periods. There is strong natural variability in the climate system.
Werner Brozek: I don’t think your results mean much. You are carefully cherrypicking just the right interval, just the right data set to maximize the effect of the Super El Nino of 1998 and get the result that you want (i.e., the lowest slope possible). By changing things even a little bit, say, adding one more year to the start of the data, one changes the result significantly. Presumably, this sensitivity would be less true in a data set corrected for El Nino, which is why they talk about the need to correct for it.
As for your statement, “I did not go deeper into the report, but if your point 1 is true, and I am not saying it isn’t, then CO2 is clearly NOT the huge driver they thought it was,” as Victor has explained, you are creating a strawman argument. It is not so much that CO2 is such a huge driver but rather that it is one thing that is changing by a huge fractional amount and is responsible for a slow but steady trend over time. Things like El Nino can have huge effects but they are fluctuations…They don’t give a significant contribution to the trend over the long term (although they can significantly affect trends measured over the short term).
joeldshore says:
December 20, 2012 at 7:25 am
By changing things even a little bit, say, adding one more year to the start of the data, one changes the result significantly.
It is not so much that CO2 is such a huge driver but rather that it is one thing that is changing by a huge fractional amount and is responsible for a slow but steady trend over time.
Victor Venema says:
December 20, 2012 at 2:10 am
Only a small part of this energy goes into increasing the temperature
First of all, with regards to the statement by Victor and the second statement by Joel, I agree, but would add that the influence of CO2 is so small that it is NOT worth spending billions of dollars on things like carbon capture to reduce it.
Now as for the first statement by Joel above, changing RSS from 16 years to 15 years does have a huge affect, but not as you expected! See below. For 16 years, from December 1996, the slope is almost flat with a value of 0.000129266 per year. But from the 15 years from December 1997, the slope is -0.00464267 per year.
http://www.woodfortrees.org/plot/rss/from:1996.9/plot/rss/from:1996.9/trend/plot/rss/from:1997.9/trend