Bill Gray was credited by Chris Landsea in his WMO and WTC presentations as the person who discovered the link between El Nino and Atlantic hurricanes. He recently suggesting a connection between Atlantic hurricanes and fluctuations in the world ocean conveyor. Judith Curry told the WSJ that Gray has “brain fossilization” and “nobody except a few groupies wants to hear what he has to say.” AT CA, she said:
I am not going to critique Gray’s paper, it is beyond rational critcism, i will save technical comments for such an unlikely event as any of this actually ever gets published. Bill Gray is not a player in the scientific debate, his ideas reflected in the paper referred to at RC are so flawed that they are unpublishable.
Gavin Schmidt had a hissy fit at realclimate – Gray and Muddy Thinking
Fast forward a few months later. The new big thing in Atlantic hurricanes is their connection with something called the Atlantic Meridional Mode (Vimont and Kossin 2007; Kossin and Vimont 2007). I’ll present some maps below comparing the Atlantic Meridional Mode to Bill Gray’s diagrams. You decide whether Judith Curry and Gavin Schmidt should be spitting and having hissy fits at Bill Gray or not.
Gray’s Presentations
Landsea’s recent presentations cordially acknowledge Bill Gray. See slide 33 of his WMO presentation. See Slide 25 of his WTC presentation. Perhaps these are the “groupies” that Judith Curry is referrring to.
In Landsea’s WMO presentation, after a slide (#73) showing a standard diagram of the world ocean conveyor system, there is the following diagram (slide 74) illustrating two modes for Atlantic hurricanes. On the left is one mode showing an active hurricane season (multiple arrows) associated with warm SSTs in the west Atlantic and cool SSTs in the south Atlantic, a strong meridional ocean current, low pressure offshore Africa and wet west Africa; on the right is the opposed mode showing an inactive hurricane season, cold SSTs in the west Atlantic and warm SSTs in the south Atlantic, weaker ocean currents, high pressure offshore Africa and dry west Africa.
Figure 1. Slide 74 from Landsea’s WMO Presentation, showing two opposing modes.
Now here’s a related diagram from Gray’s controversial presentation previously linked to at CA here, in which you see many of the same features. The top panel shows a strong hurricane season, low pressure offshore Africa and wet west Africa – corresponding to the left panel above, while the bottom panel shows the opposing mode: fewer hurricanes, high pressure offshore Africa and dry west Africa, this time placing the Atlantic circulation modes in the context of global circulation. I have no opinion as to whether these modes are valid or invalid; I am merely observing that the modes presented by Bill Gray are the same as those presented by Chris Landsea.
Figure 2. Diagram from Gray 2006 presentation.
The American Meridional Mode
The Atlantic Meridional Mode is discussed in a hurricane context by Vimont and Kossin, referring to Chiang and Vimont 2004, who show the following illustration (Atlantic on the right is the one of interest here) for the Atlantic Meridional Mode (their MCA Mode 1). MCA mode 1 has cool SSTs in the south Atlantic (compare to cool south Atlantic SSTs in Gray’s strong hurricane mode); low pressure offshore west Africa (again compare to Gray’s strong hurricane mode). While Gray does not refer to Chiang and Vimont nor conversely, it seems to me that the Chiang and Vimont MCA Mode 1 corresponds rather nicely to the cross-equator mode shown in the Gray and Landsea diagrams.
Chiang and Vimont 2004 FIG. 1. Spatial properties of the leading MCA mode 1 in the (left) Pacific, (right) Atlantic. (a), (b) Regression maps of the MCA leading mode SST normalized expansion coefficients on SST and 10-m wind vectors. Wind vectors are plotted where the geometric sum of their correlation coefficients exceeds 0.27 (the 95% confidence level). (c), (d) Same as (a), (b) but for precipitation (mm day21). In general, shaded regions in all panels exceed the 95% confidence level.
Kossin and Vimont
Kossin and Vimont 2007 and Vimont and Kossin 2007 are two recent articles, which associate the Atlantic Meridional Mode with changing hurricane activity. They describe the AMM as follows:
The AMM represents the leading mode of basin-wide coupled ocean-atmosphere interaction between SST and low-level winds [Chiang and Vimont, 2004] and while its amplitude is maximized in boreal spring — via a positive feedback between surface winds, evaporation, 85 and SST [Xie and Philander, 1994; Chiang et al., 1997] — there is a strong signal in the Atlantic during hurricane season, arguably due to simple persistence [Hu and Huang, 2006].
…Unlike the AMO, the AMM emerges as a true dynamical mode of variability intrinsic to the tropical coupled ocean-atmosphere system [Chiang et al., 1997; Xie, 1999], Pronounced coupled ocean-atmosphere variability in the Tropical Atlantic is generated by fluctuations in the Atlantic Meridional Mode (AMM) [Servain et al., 1999; Xie and Carton, 2004; Chiang and Vimont, 2004].
… The AMM is characterized by a meridional SST gradient near the location of the climatological inter-tropical convergence zone (ITCZ); boundary layer winds that flow toward the anomalously warmer water and veer to the right (left) in the northern (southern) hemisphere, in accord with the Coriolis force; and a meridional displacement of the ITCZ toward the warmer hemisphere.
They associate changes in the AMM with changes in hurricane activity in the Atlantic:
It is shown that a large part of the variability of overall “hurricane activity” — which depends on the number of storms in a season, their duration, and their intensity — can be explained by systematic shifts in the cyclogenesis regions. These shifts are strongly correlated with the AMM (Atlantic Meridional Mode) on interannual as well as multidecadal time-scales.
…Vimont and Kossin [2007; hereafter VK07] showed that the relationship between SST and hurricane activity could be viewed as part of a larger relationship between hurricane activity and a dynamical mode of Atlantic variability referred to here as the Atlantic Meridional Mode (AMM), but also known historically as the “gradient”, “interhemispheric”, or “Atlantic Dipole” mode [see the review by Xie and Carton, 2004].
… [The AMM] correlates significantly with hurricane activity on interannual, as well as decadal to multidecadal time scales. The interannual relationships between the AMM and a broad spectrum of local climatic factors that control hurricane activity offer a more complete and physically reconcilable link between climate and hurricanes than SST alone.Although AMM variance maximizes in boreal spring, it also exhibits variability during the Atlantic hurricane season. Variations in the AMM have been shown to be related to principal variations in hurricane tracks over the North Atlantic [Xie et al., 2005].
Their analysis was recently praised by Curry associates, Holland and Webster, as follows:
Kossin and Vitmer (2007) have shown that there is a close relationship between positive phases of the AMM and the eastward extension of the genesis region;
Of course, Holland and Webster also state that:
Mann and Emanuel (2006) have shown that the AMM is potentially increasing because of Greenhouse Warming;
Now there’s no mention in Mann and Emanuel 2006 of the Atlantic Meridional Mode whatever – but hey, this is climate science.
Let’s re-visit Gavin Schmidt’s hissy fit against Gray. Gavin:
The THC is undoubtedly important to climate, because it transports heat from one place to another. However it cannot do magical things. It cannot created energy out of thin air (or thick water), nor can it make energy mysteriously disappear.
… The fact is that neither of Gray’s story lines about the THC is sufficiently well formulated to allow any clear-cut test. Nonetheless, insofar as it can be understood at all, some aspects of Gray’s new story line about the THC are demonstrably wrong.
To the extent that the AMM is a plausible quantification of the modes identified by Gray (and they seem to be describing the same or similar phenomena on one scale), it seems inconsistent to lionize the AMM as an explanation, while not acknowledging that the AMM appears as a component of Gray’s reviled diagram.
References:
Chiang, J. C. H., and D. J. Vimont: Analogous Pacific and Atlantic meridional modes of tropical atmosphere-ocean variability. Journal of Climate, 17, 4143-4158 (1 Nov 2004) abstract / pdf http://www.atmos.berkeley.edu/%7Ejchiang/Workpage/Paper/CV_2004.pdf Kossin, J. P., and D. J. Vimont, 2007: A more general framework for understanding Atlantic hurricane variability and trends. Bull. Amer. Meteor. Soc., accepted pending minor revision.Vimont, D. J., and J. P. Kossin, 2007: The Atlantic meridional mode and hurricane activity. Geophys. Res. Lett., in press. PDF file (1.5 Mb) http://www.ssec.wisc.edu/~kossin/articles/Vimont_Kossin_2007.pdf
Kossin, J. P., K. R. Knapp, D. J. Vimont, R. J. Murnane, and B. A. Harper, 2007: A globally consistent reanalysis of hurricane variability and trends. Geophys. Res. Lett., 34, L04815, doi:10.1029/2006GL028836. PDF file (377 Kb)
http://www.ssec.wisc.edu/~kossin/pubs.html
Climate Audit 
227 Comments
In 1993, Gray proposed three studies for funding: “Further Development of Quantitative Seasonal Prediction of Atlantic Basin Hurricane Activity and Sahel Rainfall,” “Development of Quantitative Seasonal Prediction Schemes for Atlantic Basin Hurricane Activity and African Sahel Rainfall” and “A Global Synthesis of Climate Trends Associated with Atlantic Multi-decadel Variability.” They were all rejected by NOAA Atlantic Climate Change Program manager David Goodrich because “hurricane prediction research is not presently a central focus within ACCP,” “we do not expect that this proposal would be successful,” and “the results would be qualitative and necessarily based on a small number of realizations.” When Gray submitted a proposal to examine “Extended Range Prediction of ENSO,” Kenneth A. Mooney, program officer of NOAA Office of Global Programs rejected him with a rote note.
In September, 2006, Rong Zhang and Thomas L. Delworth of GFDL study “Impact of Atlantic multidecadel oscillations on India/Sahel rainfall and Atlantic hurricanes” in Geophysical Research Letters, which found: “multidecadal fluctuations of India summer rainfall, Sahel summer rainfall, and Atlantic Hurricane activity have been observed during the 20th century. Understanding their mechanism(s) will have enormous social and economic implications. We first use statistical analyses to show that these climate phenomena are coherently linked. Next, we use the GFDL CM2.1 climate model to show that the multidecadal variability in the Atlantic ocean can cause the observed multidecadal variations of India summer rainfall, Sahel summer rainfall and Atlantic Hurricane activity (as inferred from vertical wind shear changes). These results suggest that to interpret recent climate change we cannot ignore the important role of Atlantic multidecadal variability.”
Go figure.
Your Vimont & Kossin 2007 links don’t work for me. Could they be in a directory not open to the general public?
Perhaps, in the end, Dr Grey will be vindicated. There is the possibility of a new teleconnection between the AMM and
Inida summer rainfall variations. It at least looks good enough to at least throw a few dollars to investigate. What could
tie both of these together is the world ocean conveyor. The implications could be huge; we would have direct linkage between ENSO, the AMO, and the PDO -at the heart of all of these teleconnections is the Pacific Ocean’s ability to manage
incoming solar radiation. In light of Lyman’s and Willis’s recent discovery of unprecedented ocean cooling, I would think scientists would be all over this one.
There seem to be a number of “catch-alls” in the AGW debate, such as:
(1) if climate isn’t perfectly average at every place at every time (more or less rain, more or less snow, warmer or colder weather), it’s due to AGW (I recently read an article about an Arctic global warming expedition that was cancelled due in part to “colder than expected weather,” which was considered consistent with global warming because it was “unexpected”)
(2) if there’s a natural explanation for a deleterious weather or climate event/period, then
a) AGW can/will/is increasing the magnitude of the natural event/phase
b) AGW can/will/is increasing the frequency of the natural event/phase, or
c) things will really get bad when AGW is added to the mix in a significant amount
The above quote seems to fit 2a or 2b perfectly.
It’s amazing that, according to AGW folks, there is absolutely no positive impact to a warmer world. Everything is bad.
I’d already tried to discuss Sahel rainfall on RC before. The droughts over there are apparently 100% due to AGW according to the RC folks.
I eagerly await David Smith’s review of this posting. 🙂
Michael Jankowski:
No. Really. Ya think?
I just love the way the “Team” dismisses contrarian science as “way too much wrong to address” ie. “I can’t see anything wrong with the science so I’ll attack the person.” The whole approach of the entire group is to defer any real critique into some future time frame and hope the issues become disparate and forgotten. I find the comment regarding the need for a “clear-cut test” interesting. Would that also apply to the hockey stick? Sad stuff for a scientific community.
Post #1
Seems to be an example of the “real” driving force of separation between the “real Climate Scientist” and all those “skeptics”… which is, who will garner the Government Funds for Research!!
Apparently, if you’re not one of the clique, then you don’t get funded. If you are, then you do. This stuff just makes my blood boil!!
Re#4 – You can’t be anti-establishment and say some of it’s good.
Darwin, re (#1)-
The next question is whether or not Rong Zhang and Thomas L. Delworth got a one-off study funded – or is resistance to the SRP Gray proposed earlier now getting renewed and continuing support?
IN any event, thank you for sharing some affirmative details about the ACW crowding out effect Gray complained about loudly in recent years.
So, how does Gray’s theories apply to the current SST anonalies.
A few strange things are going on. La Nina conditions appear to be strengthening and there is a very cold pool of water on the north side of the Gulf Stream. Have been watching this graph for quite awhile and I have never seen a cold pool formation like this one.
Sorry, chart didn’t show up. It is a little large so here the link to the latest SST anonalies.
There are several interesting twists in all of this. First, here are some comments and conjecture on the history and the personalities, which I’ll do in two parts. Here’s the first part:
The idea that Atlantic storms undergo multidecadal oscillations has been around for decades. In the 1980s and early 1990s Gray (and others) warned that, while the Atlanitc was quiet at the time, the day would come when the cycle would swing and North America would again get pounded by frequent big hurricanes.
Sure enough, it happened (1995). Today we’re maybe one-third to one-half through the active part of a 30-year cycle, according to Gray’s view.
Gray took things one step further and offered conjecture on the cause(s) of the Atlantic’s cyclical behavior, even though the ocean circulation portion is outside his expertise (and outside anyone’s expertise in the 1980s). That’s admirable in my book, part of healthy science, not something to be condemned or ridiculed.
Gray focused on changes in the thermohaline cycle (THC) as the cause of changes in the Atlantic and made several Powerpoints and easy-to-read papers outlining his general hypothesis. I see them as outlines, meant to provoke thought in younger people and written in easy language for people outside meteorology (he’s a senior teacher, after all).
Along comes the increase in hurricanes in the 1990s and 2000s. The AGW folks see hurricanes as something that resonates and “has traction” with the American public and needs to become part of their apocalyptic message. But Gray, the most-visible of the specialists, publicly disagrees with the GW/hurricane hypothesis.
Objective: neutralize Gray.
Gradually, more papers are being published that expand our knowledge of climatological temperature history beyond “Hockey Stick” simplifications. Scientists are also exploring solar irradiance, cosmic rays, ENSO,AMM, AMO, PDO, multi-century glacial records, bovine flatulence,natural carbon sequestering, CO2 variations, urban heat island effects and other areas, I’m sure. The cat’s out of the bag – once published, these papers won’t be buried, in no small part due to CA. And there’s nothing more appealing to scientists than to present a tempermental “consensus” that reveals major chinks in its confidence through ad hominem attacks against and academic expulsion of outlier science and scientists.
But something else is happening – public exploration of the conflicts of interest for the AGW cataclysmologists. I hear about a $5 billion a year AGW industry, but how about the untold amount of money diverted to AGW through provision of public forums, academic facilities, etc.? I have a feeling that this issue is going to be explored, also.
I think that with time, and more temperature records, the percentage of global climate change attributable to anthropogenic causes is going to shrink, while the societal costs of trying to reduce CO2 will stabilize or climb. In effect, the benefits will fade but the costs will increase. The world effort will shift from aggressive prevention policies to aggressive adaptation policies.
Re: #14, part 2
Bill Gray was a problem. Hurricanes catch the American public’s imagination much better than does permafrost. Hurricanes cause visible, dramatic damage, with victims in tears and cities in shambles. It would make the perfect poster child for AGW, if natural hurricane cycles could be eliminated and everything attributed to CO2. But Bill Gray, a very visible figure, was bucking the AGW message. What to do?
Imagine Gray asking for funding from a government where Al Gore is Vice-President and the agencies contain managers like James Hansen. Is there a possibility for mischief, of trying to isolate and starve opposing views? See post #1 above – it makes one wonder.
Gray, well into his sixties when hurricanes increased in the mid-1990s and approaching retirement, became more vocal. He also became more shoot-from-the-hip in issues outside his expertise, making himself an easier target for anyone gunning for him.
When Emanuel and Webster published and publicized their papers in 2005, Gray came unglued. The Hurricane Wars were underway. Gray made enough poorly-worded, and in some cases plain wrong, statements about matters outside his expertise that he was especially vulnerable. So, the kicking began and people like gavin did so with relish. Everything associated with Gray got kicked, including Landsea and including the thermohaline hypothesis.
Things went pretty well for the Emanuel/Webster/Curry side for a while. They got lots of publicity and visibility, in a word fame. Egos expanded.
But, in 2006 the flaws in the Webster/Emanuel work began to show. Klotzbach and Kossin showed that the data does not support claims of a trend towards more and stronger hurricanes, outside of possibly the Atlantic. Michaels pecked away at the hypothesized physical mechanism for stronger storms. Pielke and ClimateAudit chipped away at the claims that the Atlantic was showing a trend, and pretty well showed that whatever trend might exist is confined to a historically poorly-measured region (eastern Atlantic). Enough analysis of SST and storms was done to show that any connection between higher SST and greater storms was either indirect or “classical” (I won’t go into that here).
But, egos don’t allow admission that the 2005 papers were not-ready-for-prime-time. So, some of the players (looks like Webster/Holland) appear to be regrouping to the eastern Atlantic, and hypothesizing that it is indeed more active and that is due to AGW. It’s a small territory they are defending versus their global claims of 2005 and in a way that is sad.
So, Holland/Webster need some kind of physical mechanism to link AGW to eastern Atlantic hurricanes. I’ve been wondering what it will be. It looks like it involves the AMM, linked in some way to AGW. And my guess is that they’ll say that the AMM, historically a natural cycle, is now “broken and stuck in high gear” due to AGW. And that the “stuck” AMM will gin up superstrong, low-latitude hurricanes which will routinely bring apocalyptic conditions to the US.
They probably figured that the broad-sweeping potential intensity model would suffice and that they’d not need to retreat to using a natural mode. But, circumstances changed and they appear to need something natural. But, look for them to say that the natural mode is now broken-in-high-gear by AGW. They are not embracing natural modes, they’re using them.
Next: how does Gray’s model stand up?
Re: #14
The crux of Gavin Schmidt’s hissy fit on Gray I think centers on this statement that he made about Gray’s theories below (from Steve M’s link to RC) and it no doubt involves Gray’s claims about the THC and its being used for explaining the 20th century global warming.
It would appear that some in the AGW crowd can digest less than convincing evidence for the AGW causes of hurricanes and cyclical climate occurrences and their concurring theories while dealing with opposing theories as heretical.
David Smith, I believe you have made the point before that Gray’s theories about cyclical climate conditions such as the THC are considered much stronger by the scientific community, in general, than extending that theory to explain 20th century global warming.
Hello, Ken. I think that Gray is well-respected for his tropical cyclone and tropical climatology work. The farther Gray moves away from those specialties, the less-seriously he is taken.
Some of Gray’s conjectures on global warming are disjointed, missing important details, hard to follow and just wrong. However, I think his overall global warming theme (that ocean circulation varies on multidecadal timescales, and affects climate, and accounts for some of the 20’th century temperature variation) is a robust hypothesis.
And, anything he says with regards to the tropics, especially the Atlantic, should be weighed quite seriously. More on that later.
Too busy to spend much time in the blogosphere, but i try to check in on the weekends.
I think that the Vimont and Kossin paper is very interesting and a useful contribution. Vimont and Kossin point out that gradients in SST, and the location of these gradients, also influences hurricanes. This is not a new revelation, but their analysis definitely contributes to our understanding of what is going on. Vimont and Kossin does not in any way “refute” Emanuel and Webster et al. Rather it enriches our understanding of the different processes that influence North Atlantic hurricanes. There are all sorts of internal oscillations in the climate system associated with individual ocean basins and coupling with the atmosphere. These oscillations interact in complex ways with each other and with the external forcing, and cumulatively influence hurricanes. Vimont and Kossin do not support Gray’s analysis, other than the fact that both include some reference to multidecal oscillations in the Atlantic.
FYI, I stand by my original statements regarding Bill Gray’s paper. Gray seems pretty much to have “neutralized” himself (I haven’t seem him on the list of speakers at any of the big “summits” on this issue that various insurance groups and other organizations have been holding). The only reason he got so much attention early on from scientists actively publishing on this topic was in defense against his very public ad hominem and appeal to motive attacks (it was not the content of his scientific arguments such as outlined in his AMS paper, which lacked both substance and credibility, as per gavin’s blog).
The “punch line” on this story hasn’t changed much in the last year or so; if you warm the sea surface temperatures by a degree or two, on average hurricane intensity will increase. Scientists are debating the magnitude of the intensity increase, not whether intensity would increase. It would be astonishing for average hurricane intensity NOT to increase with warming SSTs by a degree or two, based upon everything we know about how hurricanes work. It remains to be seen whether we can sort out the hurricane intensity data to address this issue more definitively from the observations. Data in the Indian Ocean and the South Pacific lack credibility, but our understanding of data in the NPAC and NATL is increasing and there have no huge changes here. High resolution coupled climate model simulations are also providing considerable insight into this subject.
So scientific research is marching on. The public’s understanding of this issue is maturing and the media is improving in terms of its treatment of this issue. There is more discussion of assessment reports rather than individual papers, and there is discussion on adaptation to the high risk we are facing in the North Atlantic. Catastrophe modeling groups (e.g. AIR, RMS) have inferred that there is a risk that we should be worried about and addressing. They are influencing the situation more than anyone else at the moment, in my opinion.
Re #19, Judith Curry
The insurance industry is one of the potential big winners from AGW alarmism. The last place I noticed a hockey stick was in a recent report on the Lloyds of London site, saying that global warming means they have to raise insurance rates.
What sort of remuneration do you get for speaking at these ‘summits’ ?
Yup. Even the monolith of the BBC is beginning to admit the existence of skeptic scientific opinion. With a bit of luck, they will shortly be displaying the typical bureaucrat’s panic on realising that they have done something for which they could be criticised by the boss – at which point they will overcompensate by flipping to a massively skeptical position. Which will be fun.
Meanwhile, the large segment of the pubic who aren’t eco-cultists, but believed that AGW was real science, are beginning to realise that they have been conned. It is very noticeable that the comments sections even at web sites like the BBC, the Guardian and the Independent are beginning to get filled up with skeptical comments.
Some of the people who have been claiming to speak for science are going to have a very interesting time in the near future…
As always, it’s nice to see Judith Curry’s participation here, as her time permits.
Anyone have any thoughts to where I might go to better understand Dr. Curry’s comment regarding warming of SST leading to stronger hurricane intensity. I am still thinking that with higher global temperature comes higher air temperature as well as SST. My current layman’s understanding is that hurricanes are a mechanism by which heat is transferred from the warm water to the atmosphere. Therefore, even though SST might get warmer, if the atmosphere also warms, wouldn’t the power necessarily released in the dissipation also be attenuated?
In any event, I concur. Pleased to see your comments Dr. Curry. Understanding them is my problem, and a problem I welcome and enjoy.
I will summarize my understanding of Gray’s view of the Atlantic modes:
1. The active mode is characterized by weaker-than-normal trade winds. There is less evaporation, so tropical SST warm. Also, the weaker winds weaken various parts of the North Atlantic Gyre, which reduces the southward flow of cool water in the eastern Atlantic (resulting in further SST warming) and reduce the northward flowing tropical water neat the US coast (resulting in cooler-than-normal temperatures in the west central Atlantic.
The warmer tropical SST affect the location of the ITCZ, moving it northward. This affects African rainfall patterns and puts seedlings into a warmer place where they also benefit from a greater Coriolis effect.
Gray also connects this to an increased THC, which brings warmer water northward into the subpolar Atlantic, and that warmer water may be the cause of the weakening of the trade winds.
2. The inactive mode has stronger trade winds, greater tropical evaporation, greater transport of cool water southward in the eastern Atlantic, cooler SST in the tropics and warmer SST in the western Atlantic (due to faster transport of tropical water northward). The ITCZ moves southward into regions with less Coriolis effect, adversely affecting storm seedlings.
There is less northward flow of warm water into the subpolar Atlantic, so sea ice increases.
El Nino frequency increases, which also adversely affects hurricanes.
My thoughts are:
* Apparently Gray connects the strength of the North Atlantic gyre, the strength of the THC and the strength/frequency of El Ninos. I think these are actually independent of each other and that Gray is wrong about the connections. I think it’s possible to have a strong gyre and a weak THC, while El Nino (ENSO) does its own thing.
* I think he’s right about the SST / wind strength / ITCZ relationships.
* He is, of course, right about the El Nino impact on Atlantic hurricane frequency
Both Gray’s model and Chiang/Vimont’s AMM involve wind-driven changes in SST patterns and shifts in the path the seedlings travel, and these changes occur on both short-term and decadal timescales.
These contrast with Emanuel’s potential intensity hypothesis, which offers little more than a broad thermodynamic argument and hasn’t held up well against papers like Michael’s.
Emanuel and Webster et al are not mentioned in this post. The point of this post has NOTHING to do with Emanuel or Webster et al, but with the similarities/differences of the AMM and the South Atlantic-West Atlantic mode of the LAndsea-Gray diagram. Any problems with Emanuel and Webster et al lie elsewhere.
Well, Vimont and Kossin do not simply “include some reference” to the AMM – it’s about a supposed connection between AMM and hurricanes. I’m just exploring here, so indulge me: explain to me why Vimont and Kossin’s AMM isn’t a quantitative implementation of the South Atlantic-West Atlantic mode in the Landsea-Gray diagram. (No references to Emanuel or Webster et al are required or desired for this.)
I’m not interested in discussing whether climate scientists like Bill Gray or not or whether you invite him to him to your conferences or not. That’s irrelevant to the point of this post. I agree that it would be surprising if hurricane metrics were not associated with SST, but that doesn’t mean that the statistical analyses presented in Emanuel; Webster et al or Mann and Emanuel are valid.
I’d also be interested in whether you share Webster’s opinion that I and this site have treated either you or Webster with “much scorn and derision from some superior position that “climate audit” blog feels it has earned”, terms used by Webster about me and this site in an email to several prominent climate scientists. I feel that you’ve been treated cordially here and that, if you shared Webster’s view, you should have expressed this to me either on the blog or by email long ago.
#23. David, from time to time, the potential impact of north-south ITCZ movements on proxies has been discussed. IT seems to be pretty relevant for centennial changes.
I think that anything associated with trans-equator Atlantic is potentially very interesting. One of the things that I find most interesting about climate geometry are the asymmetries between the NH and the SH: the warmth in the Arctic relative to the Antarctic (place relatively warm Svalbard in the Antarctic); the northward flow of the “conveyor” towards the equator in the Atlantic; the NH bias of the ITCZ. Hurricanes are also asymmetric towards the NH. The earth seems to have a number of tricks for getting energy poleward. Webster’s argued that hurricanes have an important role in the overall energy balance.
It makes a lot of sense to study symmetry-breaking crossing the equator in the Atlantic.
Re: #19
From my viewpoint, as a layman attempting to follow the recent discussions of hurricane frequencies and intensities, I would say that much has changed in the recent past and much of that comes from the analysis of past hurricane and TC data which have taken the edge off some of the earlier correlations and conclusions published by Webster et al. and Emanuel. Certainly it appears that some of these recent analyses cannot continue to be ignored and the NATL basin used in isolation to make scientific points.
I would be interested in references and any details from HR coupled climate models as the models I have read about predict significantly less hurricane intensity effects from increasing SST than some of the published analyses have shown.
Science, I believe in this case, has moved on by taking a deep breath and step back to find a better direction to move on to.
Re #26 Here are the original press releases
Georgia Tech
and
MIT
where phrases like “excellent basis”, “astonishing”, and “significantly more powerful and destructive” were used in these public communications.
The question is, would the authors make these same powerful public statements today?
The debate has been about the extent of changes in storm intensity and frequency, and that debate has shifted dramatically since the 2005 press releases.
RC has never had much credibility in my book, but they really lost it over the past year. After trashing Bill Gray for his failure to provide a complete description of how THC mechanism works, they defended their AGW argument with a similar and vaguely described mechanism after the Lyman et al paper was released. It seems that heat transfers through cyclical global circulations are legitimate explanations of nature if they support the AGW theory, but incredibly stupid if they contradict the AGW theory!
In the end, I believe Bill Gray will be vindicated for helping to pioneer the concept of ocean cycles and climate change on multi-decadal time scales.
David Smith & David Magnam,
Thanks for context and insight on the debate. I concur with your observations and conclusions!
That should be ‘David Smith and Jerry Magnan’. Sorry, Jerry!
I need to learn to read more carefully Jim. The first time I read your post, I saw “cynical global circulations”.
Must be the company I keep.
Re #22 Barclay, I can offer the “classical” view of a SST/intensity connection and give a reference to Emanuel’s potential intensity hypothesis.
The classical view is that warmer seas would have larger regions to sustain development of storms, so that the duration of storms would tend to increase. And, the longer storms exist, the greater their opportunity to intensify.
Also, the warmth would tend to extend deeper into the ocean, whioh is important to sustaining strong storms.
And, there’s higher specific humidity near the ocean, which would more-quickly dilute any dry air.
Quantification of these is far beyond me, especially in a world where the troposphere also warms and perhaps the atmosphere becomes more stable. But, I would expect some effect despite any offsetting tropospheric warming.
Emanuel’s hypotheses are given in his webpage here . I think his position would be that the 15% increase in specific humidity per degree C puts a lot of extra water vapor (latent heat) into the atmosphere, which would increase storm intensity even though the storm’s temperature difference is unchanged.
I have no reason to disagree with Emanuel’s hypothesis, but the effect is only about a 5% increase in windspeed per 1 degree C increase in SST. We’ve seen about a 0.5C increase in tropical SST, so the to-date effect is maybe +3%, which is far too small for historical (and even current) measurement methods to detect.
It’s nice to see how this thread both constructively critiques yet also builds upon the concepts of Gray. That’s all they were meant to be. Thought starters. And now, we have all benefitted due to them. Onward to yet greater understanding! 🙂
I’ve been playing with SST and other data, so as to replicate some of Emanuel’s plots. Several oddities are below.
First, I double-smoothed Emanuel’s SST data ( unsmoothed data given here ). What’s odd is the oscillatory nature of the SST plot, with the oscillations having 10-year frequencies. The peaks are 1958, 1968, 1980, 1989 and 1998.
Then I double-smoothed sea-level pressure for the tropical western Atlantic ( unsmoothed data given here ) . It, too, shows oscillations, though not as clearly as SST.
Finally, a glance at a hurricane activity time series (given in Figure 1 here ) shows oscillations of about ten-year frequencies.
I’m not a solar hypothesizer and offer neither explanation nor conjecture. I simply find these oscillations, plus the one shown by Holgate, to be odd. Perhaps these are illusions caused by smoothing.
Emanuel uses summer SST in part of the tropical Atlantic to show a SST/hurricane activity correlation for 1970-2006. The idea is that warm SST causes increased hurricane activity (count, duration, intensity) via “potential intensity” growth. My replication of Emanuel’s unsmoothed data (note: i use ACE rather than PDI) gives an r-square of about 0.46 on the unsmoothed data.
If I look at 1980-2006, which presumably has better storm intensity data due to improved satellites, I get an r-square of 0.50 for Emanuel’s approach.
My conjecture is that other factors, such as atmospheric pressure and circulation patterns, also play a role, a role at least as important as SST. My proxy for Atlantic circulation patterns is the summer atmospheric pressure in the tropical western Atlantic (unsmoothed data linked in #33). For 1970-2006, I get an r-square of 0.40 while for 1980-2006 I get an r-square of 0.56
The physical model would be that low western Atlantic pressures are associated with a more-poleward seedling path, less upper-air subsidence and weaker trade winds (= some warming of SST). Perhaps the mid-tropical Atlantic SST pattern that Emanuel shows is more like a “fingerprint” of shifts in atmospheric patterns, rather than a direct cause of stronger storms. Conjecture, but fun.
Test
Downloaded the xls document, opened it in Excel 2004 without any problems, and up popped the graph. AFAIK, posting the xls document won’t cause it to run on the web. I suggest that you simply take a screenshot of it and post the resulting graphic and not the spreadsheet.
Re #36 Thanks. Getting closer! I’ll try to get the graphic tomorrow and do a writeup then.
One of the things on the chart (which I cannot explain) is the large extent to which key Atlantic sea surface temperature, atmospheric pressure and hurricane activity vary on 10-year cycles. It has increased my curiosity about a solar-cycle / climate link.
I’ve tried the screenshot and it works, except that the resolution is poor. So, for the moment I offer simply a link to a spreadsheet . Click on the image and an Excel spreadsheet should appear. It contains one chart.
This post contains some background on the content of the chart, and a second post will offer comments on the significant points.
Chart #1 contains time series of four Atlantic Basin variables. The variables are:
1. “ACE”, which is a measure of Atlantic storm windspeed and duration over a season
2. “SLP”, which is the average sea level pressure in the western Caribbean during August thru October of a year
3. “SST, Tropics”, which is the sea surface temperature of the deep tropical Atlantic (the box used by Emanuel) for August thru October
4. “SST, Spain”, which is the sea surface temperature of a mid-latitude ocean region near Spain for August thru October
The data is in the spreadsheet, with sources listed there.
The plots have been double-smoothed, for the purpose of removing higher-frequency variation, like ENSO, and for visual presentation.
The y-axis is simply an indexing, so that I could get the data close together on one chart.
The x-axis is time.
Two of the curves are of SST. One (the yellow line, for the tropics) curve has been well-publicized for a while – it is the SST box which Emanuel uses. It shows a decline until about 1975, after which it rises. It oscillates on a frequency of about 10 years.
The other SST (light blue) is outside the tropics, near Spain. It, too, shows a similar decline as in the Tropics, even though it is a considerable distance away. It, too, oscillates on a frequency of about 10 years.
The third curve (dark blue) is ACE, which measures Atlantic hurricane activity. It showed a general decline until the mid-1990s, when a sharp upswing occurred. The period from 1980 to the present shows oscillations. The period prior to 1980, when intensity measurements were more problematic, does not show clear oscillation.
The fourth curve is the sea level pressure (“the barometer reading”) in the western Caribbean. It rises until the mid-1990s, when it displays a sharp, sudden drop. This curve, too, oscillates.
I’ll offer some comments on several interesting aspects of the chart in the next post.
Re: #38
I can only suggest increasing the axes and legend font sizes and lightening the plot area color.
Re #38 Several comments on the chart in #38:
1. There is correlation (r=0.78) between SST in the deep tropics (Emanuel’s box) and ACE, which is consistent with Emanuel’s SST/potential intensity hypothesis. It is also consistent with Gray’s AMO hypothesis.
Note, though, that there is also correlation (r=0.72) between western Caribbean sea level pressure and ACE, and between Spanish SST (r=0.72) and ACE. These two (SLP and Spanish SST) play roles in Gray’s AMO hypothesis. See here and here for detailed discussion.
Emanuel’s hypothesis is silent on SLP (as best as I can tell). Spanish SST, which is outside the tropics, plays no direct role in his hypothesis.
2. ACE measurement improved as satellite coverage improved. Since 1980 (= modern satellite data), SST in the deep tropics (Emanuel’s box) correlates (r=0.93) with ACE but, something less publicized, is that Caribbean SLP (part of Gray’s hypothesis) also correlates slightly better (r=0.95) with ACE.
3. The western Caribbean SLP pattern is intriguing. It shows a slow, general rise (=consistent with less storm activity) from the 1950s to 1995, then a sharp decline (= sudden resumption of the active phase of the AMO, with many storms). It makes me wonder if the AMO is characterized by a sudden onset of the active phase, followed by a slow multi-decadal decay.
We don’t have good ACE data from pre-1950 (my opinion) so we don’t know how the last active-phase of the AMO played out. I do wonder, though, if it started with a bang (greatest activity in the first decade or so) and then slowly petered out into an inactive phase. If true, that would have implications for us.
4. The oscillations in all four curves, with periods of around 10 years, look uncannily like something related to solar cycles. Not perfect, though. I have no idea of how there could be a connection, so for me it’s a head-scratcher, like Holgate’s sea level rate-of-rise that seems to match solar cycles. I guess, like Shakespeare said, there are things in heaven and earth we’ve haven’t (yet) dreamt.
5. Note how, as SLP rises on a decadal timeframe, the ACE line moves closer to the SST lines (this is a qualitative observation) then, as SLP drops in 1995, ACE “runs away” from the SST lines. This suggest to me that part of the higher ACE since 1995 is due to the changes in atmospheric conditions signified by the SLP drop and not just SST. This is consistent with the Gray’s AMO hypothesis.
Anyway, I ramble too much. All comments, especially those related to better ways to examine this data, are appreciated. As always, I do not mind admitting errors or changing my conjectures inm this exercise: my skin is thick.
David, good work and interesting questions. However, you can’t rely on R^2 figures from smoothed data. The R^2 of the unsmoothed ACE/SST, while significant (p=0.002), is only 0.24. That’s the true number, the correlations between smoothed curves are not meaningful.
w.
willis, thank you for the feedback. You’re in the South Pacific (Tahiti?) I believe and I hope you were unaffected by the tsunami.
For the raw data:
ACE / Emanuel SST r is 0.50 since 1950 and 0.71 since 1980 (when ACE accuracy improved thanks to better satellite coverage and resolution)
ACE / Caribbean SLP r is -0.63 since 1950 and -0.75 since 1980.
I’d say that Gray’s SLP observation performs a bit better than Emanuel’s SST.
A climate model attempted to replicate past Atlantic hurricane seasons, as reported here . It did so with near-perfect hindsight of atmospheric and sea conditions during the season and model “nudging”, which I don’t understand but do wonder if it used to be spelled with an f instead of an n.
The bottom line was that the model did a reasonably good job of replicating seasons, which (to me) indicates that it may be useful as a meteorological tool, sort of a sandbox for playing with air flows and humidities and sea patterns. Whether it can actually predict future activity is a different matter.
I noticed that it created 21 storms last year (2006) whereas the actual storm count was 10.
Re: #43
It appears, based on quick glance at this report and especially Fig. 8, that both models consistently over-estimate both storms and hurricanes. So, maybe they need another nudge factor.
An important paper on Atlantic hurricane climatology, located here , was published this week. I’ll summarize some key points soon.
Vimont and Kossin are sharp dudes.
The paper by Vimont and Kossin is here . First, some background:
* The key questions in the hurricane world are whether hurricane/typhoon activity is increasing worldwide and, if so, why.
* Kossin and Klotzbach have shown that there is no upward trend in hurricane/typhoon activity in the modern satellite era (last 25 years) despite rising ocean temperatures , except in the Atlantic. Atlantic activity switched from dormant to highly active suddenly, in 1995.
* So, the questions become why is the Atlantic different and is it related to SST/AGW?
* Emanuel hypothesizes that hurricane/typhoon activity increases as SST increases, for thermodynamic reasons. It’s a simple hypothesis but is grounded in basic thermo considerations. (The data outside the Atlantic argues against a SST/activity link, but this summary is limited to the Atlantic, where the question is open.) Increased SST is due to AGW forcing, per Emanuel.
* Gray and others hypothesize that Atlantic activity is affected by something called the Atlantic Multidecadal Oscillation (AMO). The AMO is defined by Knaff and others as being characterized by certain atmospheric conditions (moisture, wind shear, vorticity, surface wind velocity). Tropical SST is mostly an effect of the AMO rather than a cause (though there are feedback loops). The ultimate cause of the AMO is considered unknown, though changes in the thermohaline circulation are thought to play a role.
* Enter Vimont/Kossin, who attempt to reconcile the situation. They include another concept called the Atlantic Meridonal Mode (AMM). The AMM is a tropical sea surface temperature pattern that affects the location of the ITCZ. It also affects several other weather phenomena (vorticity, shear, etc) which have clear links to hurricane activity and appear to work in concert in the Atlantic, unlike other basins where the effects conflict with one another or are distorted by ENSO effects.
* V/K found that year-to-year changes in the AMM are more correlated with yearly hurricane activity (r=0.64, unsmoothed) than are SST changes (r=0.45, unsmoothed).
* V/K present evidence (latent heat flux patterns) which suggest that ocean/atmosphere phenomena (basically lighter winds) generate, at least to some extent, the SST anomalies associated with AMM. This is contrary to the standard AGW model, which attributes all SST increases to heat trapped by CO2.
* The AMO “excites” or nudges the AMM into its active, hurricane-supporting state on multi-decadal timeframes. Interestingly, there is evidence that other things, like prehistoric ice sheet variations (Chang and Bitz, 2005) and trade wind variations can also excite the AMM.
* The ultimate “cause” of the AMO remains unknown and is beyond the scope of their paper. V/K leave open the possibility that ocean circulation, or AGW, or something else, could be involved.
All in all, the V/K paper offers considerable support to the concepts of classical tropical meteorologists. It provides an improved physical framework for understanding how Atlantic climate affects hurricane activity (how all the pieces of the machine fit together), which is something that has been needed. Elevated SST in certain Atlantic regions play a role, but it is a supporting role rather than center stage.
This does not mean that the battle of classical vs. AGW is over. I expect that we’ll see papers which argue that the AMO/AMM is, or will be, permanently stuck in a high-activity phase due to AGW. Even so, this retrenchment will be quite a change from the warnings offered in 2005 about a developing AGW-driven global cyclonic calamity.
Thanks David. I really appreciate your summaries.
#46,
Their premise is flawed. Hurricanes, and tropical cyclones in general, are generated by a stochastic process with Poisson statistics. There has simply been an upward fluctation in the storm counts due to randomness and nothing more.
I continue to marvel at charts like solar irradiance in comparison with Atlantic sea surface temperatures and pressure in chart 1 of this spreadsheet .
Irradiance rises by a rather modest 1 w/m2 and tropical Atlantic SST rises by 0.3C-0.5C, which I think is considerably greater than what would be expected.
Obviously, the data covers far too short of a time to be meaningful, but I do wonder if there’s a connection.
David Smith (#49),
Does there appear to be a lag between the solar irradiance and the sea surface temperatures?
Jerry
David Smith (49)
are there any sources for cloud cover over these oceans? A slight decrease in cloud cover (low cloud) could explain the rises of SST with only a small increase in TSI
Re #50, #51
Jerry, the data is too coarse and smoothed to detect any lag between the two. This is a qualitative eyeball exercise, I’m afraid. If there is a direct connection, then a lag seems reasonable.
Gaudenz, cloud cover data may exist from satellites but I know of no source for that. The periods of higher activity (higher irradiance) may see greater cloud cover in the deep Atlantic tropics from the ITCZ shifting northward but I’d expect Spanish cloud cover to be unaffected.
The four geographically-spread measures (deep tropical Atlantic SST, SST near Spain, air pressure near Cuba and hurricane activity) all share the AMO/AMM phenomena. Perhaps higher solar irradiance manifests itself in some SST pattern which then drives the AMM. Alternatively, irradiance affects the stratosphere and the stratosphere affects the location and strength of tropospheric weather, so maybe there’s a connection there.
If someone can figure out plausible physical models for an irradiance/Atlantic climate connection then that may give a clue as to what drives the AMO/AMM.
Chris Landsea has an interesting post about new hurricane research over at Prometheus.
David Smith (#52),
If the TSI has an impact on the tropical ocean temperature, then because the variation over a solar cycle is considered to be small, it takes a period of time before that change in TSI would have an impact on the SST or weather. I can quantify the lag in mathematical terms (also see my ITCZ thread under modeling).
Jerry
If an increase in irradiance were to somehow cause an increase in northeasterlies from the Sahara toward the equatorial Atlantic, I wonder if it might result in fewer clouds there, resulting in a SST increase?
Steve (#55),
This certainly sounds plausible. Even if there were some clouds over the Atlantic, I think you might still expect some increase in SST’s over the period of a fourth of a full solar cycle ( ~ 5 years) because it would be a cumulative effect. Even though the solar variation is small, the coefficient in the atmospheric dynamical equations multipy the impact of the amount by a substantial number near equatorial regions.
Jerry
Re #56 A surprise is that tropical Atlantic SST leads irradiance by about two years, which makes no sense. It should be the other way around.
RE #53, Thanks, Joe, for finding the Landsea comment and the full wind shear article.
I noted this comment from a Prometheus reader:
The more things change, the more they stay the same. To quote gavin, “sigh”.
David Smith (#57),
It would certainly be simpler if the TSI led the SST. How confident are you in that result, i.e. what is the quality of the two sets of measurements and the brevity of the data?
Jerry
Re #59 Not at all confident: the data duration and quality is too poor. This falls into the “intriguing but so what” category.
I’ve been looking at SST data for the full tropics, and there does tend to be about a 0.3C SST response to the peak of a solar cycle. I wonder if that is tied to the Holgate chart that shows sea level rate of rises which cycle over a roughly ten-year period.
David (#60),
It seems to me that this is worth pursuing if the SST data is reasonably reliable. Does it come from satellite measurements and if so what
is the accuracy of those measurements?
Jerry
RE: #58 – It really irks me that when she participates here, she attempts to seem “neutral” or “agnostic” but elsewhere is a clear member of the AGW lobby.
Steve (#62),
I warned Steve M. about such possible duplicity on her part long ago.
Jerry
Joe B mentioned a new paper on what state-of-the-art GCMs say about future hurricane activity in the Atlantic.
The key factors evaluated were wind shear, lower-atmosphere humidity, and (indirectly) sea surface temperature (SST). Hurricanes are strongest when wind shear is low, humidity is high and SST are high.
What the models forecast is that tropical Atlantic wind shear increases by significant amounts, especially in a region around Cuba where many major hurricanes travel. This would tend to weaken hurricanes. Also, an important variation of wind shear increases in the region where storms form, which would tend to hurt formation.
Humidity in the Atlantic decreases slightly, but the impact on storms would be minimal.
Potential intensity (which reflects SST) increases in the far eastern Atlantic and near North America, due to higher SST. Interestingly, potential intensity drops in the key Caribbean and central Atlantic regions, due to the upper atmosphere warming more than the SST underneath it.
This is good news (to the extent that one has confidence in GCMs) for those in the Atlantic region. The tendency would be towards weaker storms, especially in the warmest regions of the Atlantic basin.
It’s a mixed bag in the rest of the world. It looks like somewhat stronger at-sea storms in the western Pacific, but that tendency weakens closer to Asia. It also appears that there would be a higher storm count in the western Pacific.
The eastern Pacific gets both fewer and weaker storms. Australia appears mostly unchanged.
In the big picture, we have
** the recent (last 25 years) data saying that global hurricanes have not gotten worse (exception is the Atlantic since 1995),
** the GCMs saying that future storms probably won’t get worse (probably weaker in the Atlantic) and
** the Vimont/Kossin paper arguing that natural atmospheric oscillations (AMM/AMO) play a major role in explaining the post-1995 increase in Atlantic storms.
I hope we can soon put the hurricane gloom-and-doom hypotheses out to pasture while, at the same time, using better judgment about building communities on barrier islands and below sea level. And, move on to other more important issues.
RE: #64 – Barrier island such as the ones along the south shore of Long Island …. LOL! Seriously, I’ve heard New Yorkers are gnashing their teeth about huge increases in costs of (and even outright cancelations of) home insurance policies.
Re #65 My friend in Perdido Beach, Florida used to pay $300 a year for condo insurance. He now pays over $10,000 a year.
That will do wonders to change peoples’ behavior and decisions regarding where they live and how they build.
Jeff Masters has a good blog on the outlook for 2007 hurricanes. It’s well-written.
He admits (and I agree) that forecasts made in April are nearly worthless, but that doesn’t stop him (or me). Masters is down for 15 storms in 2007, joining the tight cluster of wizards bunched around the number 16.
Re: #58
David Smith, you have been more favorably inclined to see benefits of JC’s appearances at this blog than perhaps I have been, but on careful reading of the comment from a Prometheus reader I could think that JC is not saying anything she has not said here. If her hurricane proclamation is limited to the NATL and after 1970, that is nothing new. She claims that the Webster papers, in which she was involved, started at 1970 because they had decided that was the period of “good” data. If she made claims further back in time (I would want to check that one out personally) then she is playing games by attempting to have it both ways. Her “irrefutable” claims on more intense hurricanes or TSs had to do with the thermodynamics and computer models, but I am not at all sure how that relates to more frequent hurricanes.
If she related all this without reference to what is happening hurricane-wise outside the NATL in the remainder of the hurricane world and without noting that computer model results predict an increase in wind speeds that would be nearly or entirely undetectable then I would say that she had her AGW advocacy hat on that day — hardly something unexpected coming from the scientists/advocates with which I an familiar. Finally, however, I must say that if she failed to carefully “couch” all of her statements, I would be somewhat surprised. Can we get a link to what she actually said?
Re #68 Ken, I’ll e-mail Anders Valland and ask if he can provide the full article (translated is best!).
Judith Curry, like most of us (including me), occasionally plays to an audience. It happens and I have no problem with that. It adds some life and spice into topics which would otherwise be quite dry.
At the same time, though, when the evidence against one’s hypothesis mounts, one should temper one’s position and, in this instance, take a wait and see position. I’m not sure she’s doing that, based on the excerpt.
My views of Curry and Webster are positive. They are good people who inadvertently used bad data (they should have known better, of course). It happens. Now it’s time to acknowledge the limitations and move on.
(Are their views of us CA heretics positive? Doubtful, but that matters not. I think Steve’s and other’s points have stood up quite well, and that’s what matters.)
By the way, Dr Curry’s acid test was not the audience in Norway. Rather, it will come this fall, or next, when a hurricane hits Tampa and the BBC calls her for a quote, hoping for something eye-catching.
RE #61 Jerry, I’ll do some digging on this next week.
This evening I noticed a paper indicating that some aspects of the cycle (neutrino flux, proton events, cosmic ray flux, etc ) peak 2 or so years before and after the sunspot peak. I wonder if some types of solar emission, not just irradiance, play bigger roles in climate and they are what appear to show up in the deep tropical Atlantic SST record.
I also took a more-detailed look at tropical SST during solar cycles and find that the north Atlantic comes closest to a correlation. Other tropical regions show weaker, or no, correlation. There appears to be something special about the north Atlantic and solar cycles, and I wonder if it works through a stratosphere/AMM/AMO connection of some sort.
Solar/climate conjecture never interested me, until now. I doubt there’s anything substantive to this but it’s worthwhile just for the learning exercise.
Re #70:
Douglas Hoyt (and Schatten) has a different way of calculating solar influences on climate, where the peak influence is during the upgoing flank of the cycle, but he may explain that better himself here.
There is empirical evidence for an influence of solar activity on cloud cover. To begin with: an inverse correlation between cloud cover and TSI, see Kristjansson ea., figure 1. Although they disagree with the GCR-cloud connection (but see the reaction of Svensmark), whatever the mechanism is, the correlation between (mainly low) cloud cover is quite clear, with about +/- 1.5 % change in cloud cover vs. +/- 0.5 W/m2 TOA solar irradiance.
From Kristjansson ea.:
and:
As the change in % cloud cover is mainly in the subtropics, this will have its largest effect on SST in the subtropics too, leading to an increase of ocean heat content which is at highest in the subtropics. See Fig. 2 in Levitus ea., 2005.
Besides the short-term 11/22 year cycles, there is some long-term trend, which influences cloud cover in the (sub) tropics, as was derived from satellites: see the work of Wielicki ea. and Chen ea. Warning: the radiation balance (but not the cloud cover) data are outdated, as a recent correction of the satellite data changed the negative trend in radiation balance (TOA) into a positive one for less clouds over a 15-years span.
Why the NH is more reactive to solar changes may be a matter of land/ocean ratio and the presence of mountain ranges. The latter changes the wind patterns from circular West-East (as is the case for e.g. the near-polar SH) to SW-NE in the North Atlantic. Together with the change in jet stream position (see e.g. the story by NASA), this causes changes in cloud/rain patterns over the US, Portugal and Italy.
For more in-depth comment on how solar/ozone induced changes in the stratosphere affect the troposphere, see Baldwin and Dunkerton, for short-term (op to decade) variability. Something similar in Hameed and Lee (costs some money for the full article). Longer-term connections are more scarcely described and often compares with the HS as temperature proxy, which clearly underestimates the real climate variability of the past millennium…
Thanks, Ferdinand, references linked together are exactly what I hoped for. This is a new area for me. I’ll post some data next week, once I return from a trip.
David Smith (#70),
I look forward to your findings. It amazes me that someone has not looked into this possibility earlier.
Jerry
Ferdinand Engelbeen (#71),
Thank you for suggesting references for this study. I am not sure where this will all lead, but it is an area that is worth investigating, especially given the correlations between solar cycles and SST.
The other day I read that the Farmers Almanac (I do not know how accurate it is, but I do know that many farmers swear by it 🙂 uses a formula based on solar cycles and other data (although the exact formula is a secret).
Jerry
Re #72,
Jerry, several people have searched for a connection between solar cycles and climate. While the connection is rather clear for the 11-year (+/- 2 yrs) cycle, there is already discussion about the effect of the 22-year cycle (the sun switches it’s magnetic field every cycle, with exceptions, thus the complete cycle is 22 years). On longer term (like MWP-LIA-current) the link also is less straight-forward, and heavily depends on the method used to reconstruct past solar activity (sunspot number vs. cycle length, cosmic ray produced material deposits or a combination thereof) and the temperature proxy (MBH vs. Moberg) used. The temperature proxy amplitude is very important in defining the influence of solar vs. other forcings. With the MBH proxy, the amplitude of solar variations translates to a 0.1 oC variations in temperature (maximum 0.1 oC volcanic in a 0.2 oC variation between MWP-LIA-current, the difference = solar), while with Moberg the influence of solar is 0.7 oC, which is a factor 7! This has a huge impact on what to expect for the future, as the influence of 2xCO2 is at the high side with a low solar influence and low with a high solar influence, in order to fit the instrumental record of the past century (as far as reliable…).
I had some discussion with Raypierre on RC about the “efficacy” of the different forcings. According to him, most (all?) models, use similar effects for similar forcings, with only small differences. See e.g. the work of Hansen ea., where solar has an effect (“efficacy”) of 90%, compared to CO2, for the same change in forcing. This is very unlikely, as solar changes have their maximum effect in the tropics and in the stratosphere, while most GHGs are more or less evenly distributed over the globe and have their highest effect in the lower troposphere. Because of the propagation from stratosphere to the troposphere and its influence on cloud formation (empirically proven for solar, uncertain modelled for GHGs – even the sign is not known!), the effect of solar is probably much higher than for GHGs for the same forcing. That some models underestimate solar (at least factor 2) is proven for the HadCM3 model in testruns used to see what the effect of increased solar (and volcanic) was on the outcome of the model. See Stott ea.. To be noted that the testrun results are within the constraints of the model, which includes a fixed influence of human induced aerosols (which probably is highly overblown). Without these constraints, the attribution toward solar even might have been increased…
#75. The “efficacy” issue is a big one. I don’t understand why solar – with its short wavelength – should have lower efficacy than GHG forcing. The opposite seems more plausible to me. The Hansen arguments seem like arm-waving to me, but I haven’t waded through this interesting topic in detail. I’d like to do so at some point.
Ferdinand,
I am well aware of solar cycles and the change of the magnetic pole approximately every 11 years. I published a manuscript with Tom Holzer on multiconstituent plasmas that applies to both the upper ionosphere of the earth and the outer atmosphere of the sun. There, as in a number of other areas of science, approximations were made in the unmodified magnetohydrodynamical system that led to a system of equations that was ill posed. Once those assumptions were removed, quite different results were obtained.
It is rather disturbing that the measurements of the TSI at the TOA have just been corrected in one of the references that you gave David. How can we determine what the sun is doing with these types of errors in the measurements? Given that the sun is the main source of energy for the earth, it should be the first forcing that was investigated for changes and yet it appears to be the least measured in an accurate manner according to the reference you have quoted.
Numerical climate models are completely unreliable and cannot be trusted to provide reliable information, especially with regard to variations in the solar radiation (see considerable discussion on this topic on Exponential Growth in Physical Systems). This will have to be done thru mathematical analysis and data. That is why I am interested in helping David solve this problem if it is possible given the short duration of the solar satellite data and the question of its accuracy using the now well understood mathematical knowledge of the slowly evolving solution in time for the lower atmosphere.
Jerry
Re #68
Anders Valland was kind enough to provide both the original news article and a translation into English. This is in reference to Judith Curry’s speech in Norway:
The comment about hurricanes forcing US politicians to address climate change may explain why the proponents of a hurricane/AGW connection, like Curry, continue to vocalize that view even though their scientific ground is shaky. They see this issue as having political traction.
Thanks, David Smith for Valland’s tranlation in post #78. Whether this is the media rendition of what she said or that it is a more or less a complete summary of what she said, the message that comes through is obviously very misleading. Unfortunately rather typical of the sad state of affairs of climate science/advocacy.
I guess what surprises me is how much and often Curry is quoted in the media. From what I have learned of her efforts in her field I would not put up there with the leading scientists in the field. She must be saying the things the media wants to hear.
World Climate Report has an easy-to-read summary of the recent Vecchi/Soden paper
link
Re #61 Jerry, I’m playing with some data, which is available in the Excel spreadsheet linked below. I have no conclusions or pet hypotheses, just curiosity. Opinions and comments are appreciated.
The chart labeled “Chart 2” shows data from 1950-2006 which has been double-smoothed (1-2-3-2-1 smoothing). I have no basis for that smoothing, other than that I suspect there is decadal-scale cyclical behavior which I don’t know how to filter, so I used something simple.
The three variables are annual sunspot number, sea surface temperature (SST) in the deep tropical Atlantic (the “MDR”), and the geopotential height of the middle stratosphere (70mb) over northern North America.
Sunspot number is my proxy for total solar irradiance, as I have not yet found a good set of up-to-date irradiance values.
What is the significance of stratospheric geopotential height over a particular region? Well, what it reflects is an aspect of stratospheric circulation, and changes in it can affect stratospheric wind speed and direction over distant regions of the Northern Hemisphere (think “teleconnections”).
The SST chosen is correlated with the well-discussed Atlantic AMM/AMO and hurricane activity.
All three variables on chart 2 show cyclical behavior of of about decadal length. As solar activity rises, tropical Atlantic SST rises and seems to peak a couple of years before solar activity peaks. Also, as solar activity rises, stratospheric behavior over northern North America changes, with height peaking at about the same time as solar activity peaks.
Not shown on the chart (but present on the data sheet) are Caribbean sea level pressure and SST near Spain, which also seem to cycle decadally.
I have some ideas on how these may tie together, but right now I’m still exploring, and searching for irradiance data, especially any which might break it down into components. Also, the thought crosses my mind that the smoothing may create the illusion of correlation.
Spreadsheet
David (#81),
I think that Ferdinand provided a satellite reference that is at least worth perusing. I would also look at the NASA site for other satellites that measure TSI or its components. As you are well aware, we are not positive that sunspots are the best measure but certainly a possibility
for a solar proxy (especially given the Maunder minimum).
One of the problems with using any digital filter is that when one nears the end points, the filter must be altered. As long as one stays away from those points near the end that are affected by this modification, the filter response will behave as expected on the interior points. For grins, you might also find two points that have roughly the same value (should be some if the data are approximately periodic) and perform a complex FFT on the series in time. This will at least ensure that the data are continuous, but maybe not differentiable (more about this later). The Fourier coefficients might be useful and it is easy to apply any filter in Fourier space.
I can provide a good FFT program and the info how to use it if you do not have one available.
If the SST is measured by satellite, there could also be problems
with that data, e.g. under cloudy areas where the conversion involves
a questionable algorithm? It might be necessary to eliminate those readings
if that is the case?
As we proceed, we can check each of these issues to ensure that the data
are appropriate and sufficiently accurate. Then any conclusions that we draw will be convincing and can be validated by others. I look forward
to our collaboration as this is an issue that has bothered me for some time. And we certainly value any suggestions from others that would like
to help. In this regard, Ferdinand has already been helpful.
Jerry
Dear Ferdinand Engelbeen
thanks for your always interesting thougths. In your thread Nr 75 you write, that solar changes have their maximum effect in the tropics and in the stratosphere. What effect would we see in the stratosphere? As a layman I would expect to see warming, but satellite show cooling. Am I right or wrong?
Can you answer this (maybe dumb) question for me or point me to some literature about this.
Thanks
David (#81),
Another thought. I have seen the convolution theorem used to considerable advantage, i.e. a filter is applied in physical space by using integrals
(or an approximation thereof) in physical space. I can provide
a reference if you are not familiar with the technique.
Jerry
It’s nearly showtime. Season officially begins in a few days.
The following report will be posted weekly to an applicable thread:
http://www.nhc.noaa.gov/text/refresh/MIATWDAT+shtml/251717.shtml?
zzzzzzzzzzzzzzzzzzzzz ……. 😉
Re: #48
Just because the individual data points follow a Poisson distribution is not sufficient to prove that lambda is constant over the whole time series. That is a very insensitive test of the hypothesis that lambda is constant. In fact, both a cusum chart:
http://www.climateaudit.org/?p=1206#comment-92271
and the individual C chart in 2005 reject the null hypothesis that lambda is constant and equal to about 6 with better than 99% confidence. Beginning in 1995, lambda appears to have increased to about 8.
Re #83,
Gaudenz,
As a part of the UV (mainly UV-C and UV-B) are absorbed by oxygen and ozone, this leads to increased stratospheric air temperatures at the maximum of the solar cycle. The peak value is some 0.8 degr.C higher than the minimum value. This also causes a temperature gradient from the equator to the poles, leading to shifts in jet stream position, wind/rain patterns, cloud cover,… The same is true for ozone concentrations (maximum +3% in the higher stratosphere at the equator, as direct result of UV-B/C absorption).
See Hood and Soukharev
The above points are about cycles in temperature and ozone. The long-term trends are negative: for ozone, due to ozone depletion by CFC’s and for temperature due to less ozone (which acts as a greenhouse gas in the stratosphere) and more heat absorption in the lower atmosphere, due to more CO2 and other greenhouse gases. The latter mechanism is not (yet) clear to me, I have to read more about that…
Re #81/82,
David and Gerald, Scafetta has the TSI data in Fig.1 of Scafetta and West, but I received the TSI reconstruction data from Douglas Hoyt (method by Hoyt and Schatten) from 1700 to +/- 1980, expanded with the ACRIM satellite data by Scafetta up to 2005 (yearly averages). The data (.txt) file is here.
Be aware (there we go again) that there is discussion about the minimum of cycle 22, due to difference in interpretation of the satellite data (different satellites, each with their own bias…).
David and Ferdinand (#89),
It seems to me that we should break these into two distinct sets:
one of just satellite data and the other with the reconstruction
further back in time (and possibly more controversial)? I have printed out the first reference (Scafetta and West) and will read it in detail so I can discuss the results more competently.
Jerry
David (#81),
I printed out a copy of Scafetta and West suggested by Ferdinand (it is only four pages long).
I recommend that you do too and then we can discuss its strengths and weaknesses and how to proceed.
Jerry
Re #77,
Jerry, do you have a reference to your work on sun/earth plasma’s? I suppose that there is a connection between the two, especially during (extreme) outbursts of the sun, which may influence weather patterns too (besides radio and grid problems)…
Re #91 Yes, and I’m reading the other papers, too, before proceeding. Fascinating and perplexing content.
I checked other tropical regions around the world to see if they have the appearance of a solar cycle / SST relationship. The western Pacific has has a weak appearance which weakens over time. The other regions have little or no appearance. Whatever the relationship is between SST and solar cycles, it is strongest in the tropical Atlantic.
Re #87
-An increase of expected hurricane count from 6 to 8, if true, would not be alarmingly large.
-Not sure how this cusum algorithm works, but on the surface it seems the estimates of lambda are not independent over time. That would make the inference of a trend a little more tenuous.
Re: #94
I don’t think it’s alarmingly large either, but I think it is statistically significant. The satellite temperature data for the Arctic appears to have changed slope in 1995 as well. Related??? Don’t know.
Cusum is a control chart algorithm that is very suited to quickly detecting small changes in a process. The particular implementation I used is designed for processes that follow Poisson statistics. All you can say for sure from the over the upper control limit values in 2005 and 2006 on the cusum and 2005 on the invidual charts is that there has been a increase in lambda from the value of 6 that is outside the range of random variation. The data since 1995 is consistent with a value of 8. The confidence limit on this estimate is large, just not large enough to include 6.
Increasing lambda from 6 to 8 , however, significantly raises the probability of having years with large numbers of storms. For example, if lambda is 6 the probability of having more than 14 hurricanes is 0.14%. Increase lambda to 8 and the probability of more than 14 hurricanes increases more than an order of magnitude to 1.72%
Re #95 The mid-1990s was the time when the sea northeast of Newfoundland, a key downwelling region of the thermohaline circulation, began strong warming. I interpret such sudden, strong warming as an indication that the thermohaline circulation increased. See the SST time series here .
Another key thermohaline downwelling region is north of Iceland. Its SST time series is here . It, too, shows strong warming beginning in the mid-1990s.
It is as if the thermohaline circulation strengthened in the 1990s, but the strengthening was concurrent with, or lagged, the shift in the AMO/AMM.
I haven’t connected all the dots but I firmly believe that the dots can be connected. Someday, someone will figure it out and we’ll say, “of course!”.
(Also of interest is that neither SST time series shows high-latitude warming prior to the mid-1990s, which is different from what I would expect from increased CO2.)
Re #97 Is that not around the time that the Newfoundland cod stocks collapsed? Overfishing often gets the blame. Audit the fish stock models!
Re #95
1. What is the calculated confidence limit?
2. How is it calculated?
3. Are you sure the estimates calculated before and after 1995 are independent?
4. How sensitive is the analysis to the cherry-picking of the pivot year (1995)?
Ferdinand (#92),
Do a google search on Browning Holzer. The title is
A comparison of the reduced and approximate systems for the time dependent computation of the polar wind and multiconstituent stellar winds
I don’t think it will be much help for the current study, but it is a good way to obtain a feeling for plasma physics issues.
Jerry
David (#93),
When reading the manuscript, I assume that you saw a number of assumptions that could (and should be) investigated. But I was pleased with the points that were raised and discussed by the authors. A good intro to the subject?
Jerry
Re: #99
I’m an industrial analytical chemist, not a statistician. My SPC training was on the job, so I’m not exactly an expert. It looked to me that SPC techniques could be used on the hurricane data, so I did what I could.
1 and 2. I don’t have a calculated lower confidence limit. I say it must be greater than 6 with high confidence because the control charts based on the assumption that lambda is 6 have exceeded the upper control limits.
3 and 4. Each point on a cusum chart contains information from all points since the last time the zero line was crossed. The zero point for the trend that eventually exceeded the upper control limit was 1994. Therefore 1995 as the pivot year in the cusum chart is not cherry-picked, it emerges from the assumptions used to create the chart: lambda = 6, lambda+ = 8, lambda- = 4 and ARL greater than 400. k+ is then 7 and k- is 5. More details in this comment:
http://www.climateaudit.org/?p=1206#comment-92535
On the Arctic temperatures, 1995 was eyeballed as the date of slope change. I plan to do an exponentially weighted moving average control chart of the monthly UAH MSU Arctic lower troposphere temperature sometime soon to see if that shed any more light.
David (# 97),
The SST data are evidently based on NCEP reanalysis data. In the atmosphere, the reanalysis data is based on both an atmospheric model and observational data (weighted average). If the model is in error (especially near the equator) the data can be tainted badly by the model. If the observational data is sparse (e.g. near the equator) the reanalysis can be badly skewed by the model. Are you familiar with the SST reanalysis method, i.e. how is it determined? This issue is crucial to the accuracy and reliability of that data if we decide to use it.
Is there a reliable source of SST data from satellites? When working with Sylvie Gravel, it became clear that the satellite temperature data did not match reality unless there was also a radiosonde sounding at or near the measurement. Is that still the case?
Is it possible to obtain ITCZ location from a satellite (should be the case)? How does it vary with the solar cycle?
Jerry
103,
“it became clear that the satellite temperature data did not match reality unless there was also a radiosonde sounding at or near the measurement.”
I’m trying to figure out how radiosondes manage to get satellite data back in sink with reality.
More teleconnections?
Re: #95
I agree that a cusum calculation can show a statistically significant change in the mean of the Poisson distribution. I think one can make a good case for cyclically varying mean for a Poisson distribution for hurricanes in the NATL. It makes even a better case if one assumes that the early time periods undercounted hurricanes. It is more difficult to explain the constancy over time of land falling hurricanes by using other arguments –although I believe that Webster and Curry will look for better ones.
Re: #105
I should have reminded you all of the post where Willis E showed (graphically) the dramatically improved fit with a Poisson distribution for a detrended NATL hurricane time series and the cyclical component removed. I was able to show this with a chi square goodness of fit test. I can look up the specific posts if anyone is interested.
RE #103 Jerry, it looks like SST measurement is still a work in progress.
Satellite measurements are of ocean skin temperature and can be affected by things like water vapor, cloud cover and near-surface windspeed. Also, satellite technology has changed over time, and I believe that the preferred measurement is now microwave-based instead of IR-based. My understanding is that these satellite estimates are then adjusted in some manner, including possible calibration against buoy and ship measurements.
Problems with ship measurements have been well-discussed at CA. Large scale buoy coverage is a rather recent event and there are still gaps in coverage. Satellites have limitations and weaknesses and cover but three decades. How all this is assembled into global data that is homogeneous over time and space, and useful for climatology, is beyond me.
I’m still looking for an all-satellite, apples-to-apples dataset but my hope is fading.
Mark W (#104),
The inversion of satellite info from channels to obtain temperature is a very tricky business, especially when clouds are present. As I am not an expert in that transformation, I can only give you my opinion.
In the data assimilation program that involves data from multiple sources, possibly the temperature info from the radiosondes provided some kind of anchoring info for the satellite data (wind data is the best source of updating information and the wind info alone provided as accurate a forecast as anyting else – see our periodic updating manuscript). When satellite data was used alone, the results were a disaster. Only when used in conjunction with radiosonde temperature info did the forecast with satellite data start to behave better.
This is my best recollection, but I can get the copy of our manuscript to refresh my memory if you would like. Sylvie ran the Canadian assimilation program and model as black boxes. I had her turn on and off
different forcings (the most crucial in the first 36 hours was the lower boundary parameterization that could be replaced by a much simpler scheme to obtain the same result). Then we tried different data combinations to see which one contributed the most help. Everything was as expected from theory.
Jerry
Re #107/108,
David and Jerry,
SST is only one parameter in heat transfer between sun and earth’s climate. A better indication should be the changes in heat content of the oceans, as these present over 90% of the heat balance of the earth (see Levitus ea. 2005, data at NOAA). Heat content is not influenced by short or long term surface events like ENSO. But even there we have problems in the past (sparce deep sea data for the SH) and present (see the problems with the 2003-2005 “cooling”).
While searching for more heat balance articles, I have found a few interesting ones:
http://www.asp.ucar.edu/colloquium/2001/curry01.pdf (Judy Curry)
http://www.sciencemag.org/cgi/content/full/309/5731/41 (AMO and climate)
“In the descriptions of the measurement process, so essentially simple, one can notice a significant reticence in many courses of mechanics and physics which have become classic. It was my task to establish more determinacy in the problem and, along with that, to show what a great arbitrariness is present in establishing a measurement” (Friedmann 1965, p.16).
Gulev et al in the Journal of climate identify some substantial problems with the measurement sampling for historical ocean-air flux interactions.
Estimation of the Impact of Sampling Errors in the VOS Observations on Air’€”Sea Fluxes. Part I: Uncertainties in Climate Means
Gulev et al JOURNAL OF CLIMATE VOLUME 20 January 2007
ABSTRACT
Sampling uncertainties in the voluntary observing ship (VOS)-based global ocean’€”atmosphere flux fields were estimated using the NCEP’€”NCAR reanalysis and ECMWF 40-yr Re-Analysis (ERA-40) as well as seasonal forecasts without data assimilation. Air’€”sea fluxes were computed from 6-hourly reanalyzed individual variables using state-of-the-art bulk formulas. Individual variables and computed fluxes were subsampled to simulate VOS-like sampling density. Random simulation of the number of VOS observations and simulation of the number of observations with contemporaneous sampling allowed for estimation of random and total sampling uncertainties respectively. Although reanalyses are dependent on VOS, constituting an important part of data assimilation input, it is assumed that the reanalysis fields adequately reproduce synoptic variability at the sea surface. Sampling errors were quantified by comparison of the regularly sampled (i.e., 6 hourly) and subsampled monthly fields of surface variables and fluxes. In poorly sampled regions random sampling errors amount to 2.5°’€”3°C for air temperature, 3 m s 1 for the wind speed, 2’€”2.5 g kg 1 for specific humidity, and 15%’€”20% of the total cloud cover. The highest random sampling errors in surface fluxes were found for the sensible and latent heat flux and range from 30 to 80 Wm 2. Total sampling errors in poorly sampled areas may be higher than random ones by 60%. In poorly sampled subpolar latitudes of the Northern Hemisphere and throughout much of the Southern Ocean the total sampling uncertainty in the net heat flux can amount to 80’€”100 W m 2. The highest values of the uncertainties associated with the interpolation/ extrapolation into unsampled grid boxes are found in subpolar latitudes of both hemispheres for the turbulent fluxes, where they can be comparable with the sampling errors. Simple dependencies of the sampling errors on the number of samples and the magnitude of synoptic variability were derived. Sampling errors estimated from different reanalyses and from seasonal forecasts yield qualitatively comparable spatial patterns, in which the actual values of uncertainties are controlled by the magnitudes of synoptic variability. Finally, estimates of sampling uncertainties are compared with the other errors in air’€”sea fluxes and the reliability of the estimates obtained is discussed.
David (#107),
I too have been concerned by the problems you mentioned. In the end, we will either see that the solar cycle variation influence is completely unknown (which is a result in itself) or find a way to determine its influence (best result). What about ITCZ movement. That can be seen in cloud imagery from satellite. Is there a drift north or south during solar cycles? (Steve M. suggested that this might occur over longer periods
of time.)
Jerry
As we also noted there is degree of arbitrariness in the measurement process for SST reconstructions. Another is this.
Journal of Atmospheric and Oceanic Technology
Article: pp. 476’€”486 Toward Estimating Climatic Trends in SST. Part II: Random Errors
Elizabeth C. Kent and Peter G. Challenor
ABSTRACT
Random observational errors for sea surface temperature (SST) are estimated using merchant ship reports from the International Comprehensive Ocean’€”Atmosphere Data Set (ICOADS) for the period of 1970’€”97. A statistical technique, semivariogram analysis, is used to isolate the variance resulting from the observational error from that resulting from the spatial variability in a dataset of the differences of paired SST reports. The method is largely successful, although there is some evidence that in high-variability regions the separation of random and spatial error is not complete, which may have led to an overestimate of the random observational error in these regions. The error estimates are robust to changes in the details of the regression method used to estimate the spatial variability.
The resulting error estimates are shown to vary with region, time, the quality control applied, the method of measurement, the recruiting country, and the source of the data. SST data measured using buckets typically contain smaller random errors than those measured using an engine-intake thermometer. Errors are larger in the 1970s, probably because of problems with data transmission in the early days of the Global Telecommunications System. The best estimate of the global average random error in ICOADS ship SST for the period of 1970’€”97 is 1.2°C if the estimates are weighted by ocean area and 1.3°C if the estimates are weighted by the number of observations.
Ferdinand (#109),
I was quite frustrated when working in oceanography by the lack of
observational data.
If you think there is a sparsity of atmospheric data there is next to nothing in the ocean. 🙂
My guess is that we will need to find a different indicator than SST and I have made one suggestion to David. But we will keep our minds open until we
have exhausted all possibilities.
Jerry
maksimovich (#110 and #112),
Thank you for returning after my tirade. 🙂 I look forward to any suggestions you have on this matter. I know this is not an easy task,
but the issue is quite intriguing and worth a shot with all of the resources we can muster.
Jerry
Re 114
Thanks,it is a busy time due to IHY and the October presentations.To provide a scientific perspective we are approaching the issues from observing the limitations of “sets or systems in isolation” to one of a series of interactive overlapping sets (of which there are many)that provide an understanding of the interaction of all dynamics of the system to be able to describe the mechanisms, its functions ,and effects .
There is over confidence in the measurement process and it is important to understand the limitations of the instruments including satellites to provide accurate datasets.
This is being addressed from a physics perspective.
We have some limitations due to language for papers etc.But I would recommend Monin Climate as a physics problem where a number of equations and solutions are provided.
To provide an example we can see that sst can change by 1-2.5c simply by the population and taxa of the biota in the location sampled as opposed to similar latitudinal samples due to increased absorbtion (an albedo gradient)
of infrared spectra and phytoplankton pigmentation.
In my reading about solar / climate connections I came across this article on how the troposphere is affected by the solar cycle. The authors attempt to isolate the solar signal from others and see how the troposphere varied during the cycles, using reanalysis data. Interesting reading. Certain aspects of the troposphere change while other aspects are unchanged, sort of like “fingerprints” of changes in solar activity.
This raises the question of whether the important 1976 climate shift bears this solar fingerprint. If it does, if the troposphere changed on a decadal scale in ways consistent with how it varies over a solar cycle, then it suggests that the 1970s climate shift (which changed the globe from net cooling to net warming) may have had some solar aspect. That’s what I’m currently exploring.
(This whole solar / climate thing is like one of those Egyptian pyramid mazes: something important seems to be present but there are also many dead ends, and no guarantees.)
Re: #105, #106
I don’t care whether the variation in lambda is sinusoidal or a square wave or some other function. I agree that hurricane numbers follow Poisson statistics. This is not and should not be news to anyone. My analysis is based on this. However, I strongly disagree with the statement of Paul Linsay in #48 in this thread and elsewhere that the recent variation in NATL hurricane numbers is within the range of random variability with a fixed lambda. It isn’t. The hurricane count for the twelve years from 1983 to 1994 was 57. There were 98 hurricanes from 1995 to 2006. This is all in the modern era so observation bias is unlikely. If the storm numbers came from the same population with a lambda of 6.46 or 77.5 hurricanes/twelve years, the probability of fewer than 58 or more than 97 storms in twelve years are both less than one percent. If you ignore observation bias post WWII, the average number of storms was 5.6/year from 1946 to 1994 or 67 hurricanes/12 years and the probability of more than 97 hurricanes occurring in any twelve year period from random variation is 0.02%.
maksimovich (#115),
I fully agree that the observational data (especially from satellites)
has not been carefully scrutinized.
Jerry
David (#116),
I have printed out the manuscript and will read it tonight.
One reason I reccommended visual imagery of the ITCZ (in addition to Steve M.’s idea) is that it should not involve questionable inversion processes?
Jerry
Need to post early this week, due to brutal workload:
http://www.nhc.noaa.gov/text/refresh/MIATWDAT+shtml/301802.shtml?
Chrp, chrp, chrp, chrp ….. ( …. 9 BABY! …. )
David (#116),
I did a quick perusal of the manuscript. It seems to connect the vertical component of velocity (omega) to solar cycles. This is what might be expected from the balance between the vertical velocity and total heating
if there is a connection between the solar cycles and storms in the areas where the balance holds. I will read it in more detail tonight . If it depends on reanalysis in the atmosphere where the reanalysis is reliable,
then this might be a first clue?
Jerry
Re: #117
DeWitt P, I believe that Paul Linsay originally found a Poisson fit for the hurricane data from 1945-2006 and then again for the period 1851-2006, but the goodness of that fit was shown to allow for a varying lambda (RichardT loked at this) during those periods and the fact that the fits were better when the data was separated into these two periods. Willis E showed graphically (and I was able to show it with a goodness of fit calculation) that extracting a cyclical component (and detrending)improved the fit. Also over the longer periods of time the landfalling hurricanes have had a constant frequency as noted by RPJr. That leads me to view the total observed hurricanes as occuring in cycles with a Poisson distribution and with a changing ability to detect and/or define them. David Smith and Steve M have also made observations/analyses of the spatial relationships of hurricane detection to indicate that changes have occurred into the modern times.
That’s how I view the data. How do you view it?
Jerry, I found some data on the location of the ITCZ across Africa, but it only extends back to about 1988. I’ve seen a reference to data back to 1979 but haven’t located it. Still digging.
ITCZ data in the linked file
Re: #122
Ken F., I think the data clearly shows statistically significant periods of high and low activity in NATL hurricanes. The activity for the last twelve years has been high, especially compared with the previous twelve years. Lumping large chunks of the data into histograms doesn’t really tell us anything new. The surprising thing would have been if annual hurricane counts didn’t follow Poisson statistics. In emission spectrophotometry both UV/Vis and X-ray, one counts photons. It is impossible to predict the arrival time of the individual photons, but the emission intensity from a sample with known concentration can be predicted and measured with high accuracy and precision on a calibrated instrument, provided enough counts are accumulated. These observations are also counting statistics or ‘shot noise’ limited. Just because we don’t have a quantitative (or maybe even qualitative) theory, like we do for emission spectrophotometry, to calculate variation in hurricane annual frequency doesn’t mean we should ignore the strong evidence that there is such variation.
Re: #122 again
Here is the relevant quote from #48:
The attitude I see here, and I may be misinterpteting, is that because annual hurricane counts obey Poisson statistics, they are therefore completely random and unpredictable in any way. Therefore any study of variation can be safely ignored. This is just wrong. A process variable can follow Poisson statistics and be otherwise completely determined, see example above. I believe it’s also wrong about recent variations being statistically insignificant, but I hope I have already made my point about that.
I’m still searching for the European forecast for the 2007 hurricane season, without success. If anyone sees it, please post. I believe it goes by the initials EUROSIP. Thanks
Re: #125
Dewitt P, I made the point to RichardT — when he indicated that if hurricanes followed a Poisson distribution people like Bill Gray would not be able to successfully predict their annual frequencies– that the conditions that produce hurricanes can occur and combine in the form of a Poisson distribution. If one can predict those conditions and their combination one could predict hurricanes. My problem with Gray’s predictions is that they become much more accurate the closer they are made to the event and in fact my calculations showed that they are not statistically significant when made on long term basis.
I have a much clearer picture of what I think are the critical factors in producing hurricanes from reading and participating in these threads. I do not agree with what I might suspect are all other peoples’ conclusions coming out of these discussions, but I have learned from many of the inputs, including yours on the cusum statistic. Maybe if you had used a statistic like maximum likelihood to reach the same conclusion that the lesser known and QC chart used cusum delivered, others here would have been more impressed with the exercise.
I probably have too selfish an approach to participating at this blog, but my primary purpose is to learn and add to my own knowledge base and only very secondarily concerned with changing minds to my view of things.
Re: 127
Ken F., I use the tools I have and maximum likelihood isn’t one of them.
Predicting the number of hurricanes each year months in advance is unlikely to ever be very accurate. The numbers for any one year are just too small. I credit Gray with a successful prediction that the regime would eventually shift to higher intensity even if he wasn’t able to predict when the shift would occur. As my approach to participation is similar to yours, I don’t see a problem with it. I’m here mainly to learn too.
David (#123),
After reading the manuscript in more detail, I guess it is a question of how much faith one has in the reanalysis data in different locations and in the removal of volcanic and El Nino info by the multivariate regression process? At least others also believ there might be a connection.
You are being very industrious in finding useful data. I hope Steve M. is following the discussions here.
Jerry
Re #127
Maximum likelihood is not a statistic. It’s a method of estimation used in computing statistics.
A variable can follow a poisson distribution where the mean of the distribution changes both periodically and as a function of some forcing paramter. It’s still a random poisson process. It’s just not fixed.
Re: #130
Like I said Dewitt P, if you want to get your point across here with those blog participants with extensive backgrounds in statistics you need to use terms like maximum liklihood — but you had better use them correctly.
My point was that maximum liklihood estimation can be used in a more general sense to show a changing mean much as cusum is used and that its use is probably better known to most statisticians than cusum is.
Calling maximum liklihood a statistic is flat out wrong and shows a lack of detailed understanding of the process.
I think that your statement follows what Dewiit P and I have both stated here, but I agree with Dewitt that I am not so sure that there is a general agreement here with this observation or for that matter a number of the side issues that have been pointed to, but not discussed in much detail.
As promised, here (I hope) is the Exponentially Weighted Moving Average chart (lambda = 0.08) of the MSU NoPol lower troposphere monthly data from December 1978 to March 2007:
For comparison, here is a cusum chart of NATL hurricane numbers from 1979 to 2006. The target was the mean of the data set, 6.4. The upper out-of-control target was 8.0 and the lower was 4.8. Control limits were calculated with an ARL of 400 equivalent to a 99.7% confidence limit:
The out-of-control cusum indicates that lambda since 1995 is greater than or equal to 8. I find it hard to believe that the temperature shift and average hurricane number shift is coincidental. Correlation is not causation. But I think it does lend support to Gray’s contention that shifts in ocean currents are a (the?) primary driver of changes in hurricane activity.
RE: #132 – If this were my manufacturing process, I’d be highly suspicious that something fundamental shifted or changed in 1993.
One thing that changed circa 1995 was sea surface temperature (SST) in the far North Atlantic, where cold water sinks as part of the thermohaline circulation.
The SST time series for the six warm months of the year is here . Temperature takes off like the proverbial rocket around 1995. Thermohaline circulation strengthened? Maybe Gray’s hypothesis, at least part of it, is in the ballpark after all?
Re: #132
DeWitt P, why look at such a short period of time when it has been shown that the hurricane activity can be cyclical and on decadal time scales? Did you have a chance to read Willis E’s post on removing the cyclical component of the hurricane activity over time and then looking at the fit for a Poisson distribution with a constant lambda?
As I recall the parameters that Gray uses in his models to predict hurricanes are many and SST is but one.
The data we are talking about is confined to the NATL and hurricane frequency for the world, in general, does not follow these trends with SST.
Re #134
There’s that warm water collapsing cod stocks in the mid-1990s again. Where’s Mark A. York when you need him?
Re: #135
Ken, I restricted the cusum chart to the same time scale as the MSU satellite data. Here’s the chart for 1851-2006. For this chart I used target 5.3, upper out-of-control target 8.0 and lower ooc target 3.6.
You can see something that may look like periodic behavior, but the accuracy of the data before the satellite era makes drawing conclusions questionable. As far as detrending hurricane numbers for an improved fit with the Poisson distribution, I don’t have a problem with the stochastic nature of individual hurricane generation. I also don’t have a problem with radio-isotope dating even though radioactive decay also follows Poisson statistics.
http://www.nhc.noaa.gov/text/refresh/MIATWDAT+shtml/071746.shtml?
zzzzzzzzzzzzz ………
Funny how the MSM are trying to “spin” the cold core intense low off of the SE USA as a sort of “earliest ever hurricane.” interesting aside, it is pretty wierd to see what we on the W. Coast would immediately recognize as a classic Gulf of Alaska storm in that location.
I have summarized some TS and hurricane data below to better show the case I have been attempting to make about the cyclical nature of tropical cyclones in the NATL for the period from 1851-2006. I am suggesting that the named storms would be more difficult to detect and particularly so for those that remain far out at sea compared to hurricanes and finally that land fall hurricanes would be the easiest to detect. As detection processes have improved throughout the 1851-2006 period one would expect to see the detection of named storms improve the most, hurricanes next and finally land fall the least.
The three graphs below show that the improvement reasoned above would fit the actual case if one assumes that the tropical storms have been more or less of constant frequency (on compensating for the cyclical nature of these storms) and that the ratios of named storms to hurricanes to land fall hurricanes has remained essentially the same over the period noted.
Steve M, David Smith and Bob Koss have presented evidence at this blog that indicates that named storm detections have become more efficient over time.
The cyclical pattern of the three classifications of storms seems to follow rather well over time. Willis E showed that by detrending and removing a sinusoidal cyclical component from the NATL named storm time series 1851-2006 that the fit to a Poisson distribution improved dramatically. I calculated chi square goodness of fit probabilities for the raw named storm time series and then again for the series detrended and with the cyclical component removed. The raw data had a p = 0.088, while the modified data had a p = 0.86. The lambda for the Poisson was 8.7 (named storms per year).
Willis E’s post on detrending and removing the cyclical component for the named storm time series (1851-2006 for the NATL) is given in the link below:
http://www.climateaudit.org/?p=1022#comment-77004
Named NATL storm and hurricane data came from:
http://www.aoml.noaa.gov/hrd/tcfaq/E11.html
US land fall hurricane data came from:
http://www.aoml.noaa.gov/hrd/hurdat/ushurrlist18512005-gt.txt
This paper offers a very interesting summary of some of the current thinking on the thermohaline circulation and Atlantic temperatures, and how the atmosphere couples to them. I’ll work on a summary.
Re: #141
I notice the shape of the smoothed dipole index curve in Figure 2 looks to be a very good match to the MSU Arctic temperature anomaly I posted in #132 in this thread. If that correlation continues, we should see cooling in the Arctic in the next decade.
Regarding hurricane activity, this paper in Geophys. Res. Letters is interesting:
http://www.agu.org/pubs/crossref/2007/2006GL028884.shtml
Perhaps you have seen it already.
What happened to all the posts from May 8th to the beginning of the 14th?
RE #143 gb, that’s an interesting abstract. Lightning/thunderstorm activity is probably correlated with the robustness of the seedlings leaving Africa. In general, the healthier the seedlings, the more likely they are to survive the sometimes-hostile conditions of their Atlantic journey.
It’d be nice to see a plot of lightning actvity (strikes per day inside some African ITCZ grid, for instance) and see if that correlates with tropical storm activity a week or two later.
RE #141 Figure 2, the dipole index, broadly tracks Atlantic interdecadal hurricane activity.
The other thing about the strengthening of the thermohaline circulation in the 1990s is that it brought additional heat into the far northern latitudes, likely raising global temperatures. The satellite plots show a step-change in global temperature circa 2000, but that plot is complicated by the strong 1997 El Nino and the subsequent strong La Nina. Maybe the actual warming was four or five years earlier.
At times I wonder if the warming we’ve experienced these last 30 years was, to a large extent, (1) the PDO switch to warm phase in the mid-1970s foolowed bt (2) the thermohaline cycles switching to a warm active mode in the 1990s. Increased greenhouse gases and solar changes could indeed have played roles in this scenario, but may have not been the dominant factors behind the warming.
A warmer would argue that the PDO and TH circulation etc. are not independent from AGW, but instead are major pathways through which heat flows strengthen under AGW. Wouldn’t they? Surely serious climatologists have reasoned through all this?
RE #147 bender, the oceans would have a greater heat content in a warmer world but there’s no clear projection of whether their oscillatory nature would change. I believe the GCMs tend to predict a weaker TH flow and are silent about the PDO.
My conjecture is that these two oscillations went into their warm phases, one in the 1970s and the second in the 1990s, adding warmth to the atmosphere. Projecting forward, they should enter their cool phases in coming decades. If there’s a paper which explains that these oscillations will malfunction in an AGW world, and be stuck in their warm phases, I’ve missed it. The only malfunction I’ve heard about is a weakening of the thermohaline circulation, as in “Day After Tomorrow”.
As mentioned, there could (and I think is) a background warming from GHGs and possibly solar, which add to the oscillatory warming.
Re # 147.
A warmer climate can cause more precipitation and ice melt which implies more fresh water input at higher latitudes. This could potentially weaken the TH. There are a number of papers about this topic.
David, why in a warmer world should TH be weaker? Regardless of the reason, if the sinking water in the North Atlantic should reduce its speed, now or then also the surface, northward current will have to slow down. So, surface water will have extra time to become denser (colder and/or saltier) and quickly restore the cycle or, as you say, to maintain its oscillatory nature. Am I wrong?
RE: #148 – In an oscillatory electrical circuit, when you increase the voltage (and hence, current and power) you certainly do not change the frequency. What does change is that there is greater immunity to noise, and possibly changes in the amount of overshoot and ringing during transitions. In a circuit, lower voltage/current/power tends to be more problematic than the stress induced by higher voltage, within reason. Of course at some point, you’ll exceed components’ voltage ratings and damage will ensue. But I don’t think the climate system is anywhere near that point, but in fact, bumping along the noise floor. Take that to heart, and be very, very worried, for reasons quite different than what the AGW fanatics spew.
RE 3150 Paolo, I think that most of the slowing-THC scenarios involve fresh water diluting the seawater and lowering its density, thus slowing the sinking. The fresh water comes from ice melt (sea ice and Greenland).
The paper, which I’ll get to this weekend, has some fascinating statements about time lags in the THC. Bill Gray may be closer to the truth than is generally believed.
Australia just experienced its eighth consecutive below-average typhoon season and the eighth-weakest (using ACE) season in the last 45 years.
link, see 11 May report
David, out-of-season Tropical Cyclone Pierre has formed this morning in the Solomon Sea and is headed west towards Nth Queensland – they will need to have another look at their seasonal statistics for the Australian region!
RE #154 Carl, I guess that’s the nature of Nature: when we try to pin something down, it wiggles!
Pierre will bring the region to 10 storms, which is dead-on average. The activity index (ACE), which measures the number of storms, duration and intensity, will remain well below average.
Comments on thermohaline circulation (THC) changes in recent decades are at this CA auditblog site . It’s mainly a summary of Latif et al (2005), which I found to be a good paper.
I’m experimenting with this linked auditblog format, which is courtesy of CA and which I find to be user-friendly and forgiving.
The Hong Kong forecast for the western Pacific is given here . Their prediction is for slightly below-normal activity, based on several broad indicators.
The eastern Pacific will almost surely be below-normal if the La Nina develops as forecast.
So, here’s the 2007 scorecard and forecast:
South Pacific – normal (actual)
South Indian – normal (actual)
Western Pacific – near/below normal
Eastern Pacific – below normal
Atlantic – above normal
Earth as a whole: normal
Apocalypse: postponed
The US National Hurricane Center issues its Atlantic forecast tomorrow (May 22). I think they’ll call for 15 to 18 named storms, slightly above the post-1994 average.
A recent PowerPoint presentation by Bill Gray on hurricanes and global warming is here . It’s a redundant presentation, saying the same thing in several ways, but I think he’s just trying to get a point across to a broad audience.
Dr Gray is a hoot – he calls it as he sees it.
Steve M, you might note slide 19, in which he points out the absence of East Atlantic storms in 1933 vs recent times, similar to analysis you did last year. I’m confident his point was that it’s due to detection differences over the decades, not a real absence of storms in 1933.
World Climate Report has a nice summary of a Chinese study on tropical cyclone frequency affecting part of that nation ( link ).
The results are consistent with an earlier Chinese study of a broader area of China. Basically, the studies show a decline in storm frequency, not an increase, despite higher SST.
Have fun folks, this one’s legendary:
http://www.realclimate.org/index.php/archives/2007/06/storm-world/
I think Mike’s in line reponse in comment 41 says it all.
re: #160, 161
Yes, I needed a booster shot on why I am here rather than RC. Let’s see if I got the essence right.
1. Chris Mooney has a great new book supporting AGW and more / stronger hurricanes
2. Thought his last book was a partisan attack, he’s become more nuanced now and the bias isn’t as obvious in his writing.
3. The kind and munificant Michael Mann actually allowed one possibly negative remark so that he could smack the poster down and call him names.
4. While everything the poster said can be reasonably argued, he overlooked that the models say that hurricanes will become stronger, more frequent and that’s all that counts, of course.
5. Poster from 3 recants in sackcloth and ashes since Mike has spoken.
How’s that for summarizing the thread?
RE 162 That’s about it.
My sense is they want to promote a vision of consensus in the postings.
A while back you could actually argue and learn something. Now, its the amen
corner.
Just placed into the RC post queue:
This year’s season is off to an interesting start. On the one hand, in the Pacific, conditions are ENSO neutral bordering on La Nina. However, a slight complication is that we appear to be either at the start of or on the cusp of a negative PDO.
Meanwhile, in the Atlantic, the classic El Nino type of config continues – the jet stream is going straight across from just off the SE coast of the US toward Iberia and the UK. No wavyness there. Also, Sahara and Sahel dust have all but shut down the Cape Verde segment of the storm factory.
The UK Met forecast appears to be well put.
by Steve Sadlov
Nyuk, nyuk, nyuk ….
#164,
I hadn’t seen anything recently about the sahel dust. My understanding from earlier this year is that the region received above average precipitation. My perspective was that this would lead to a reduction in the sahel dust over teh Atantic this year. Perhaps that was wrong. I’m not a scientist, but I did stay at a Holiday Inn Express last night.
RE: #165 – It’s a wet – dry climate, most rain falls in the late summer and early fall. So by now, there’s been months of relatively dry weather. I’d imagine that the amount of dust is mostly a function of wind velocity as opposed to the amount of rain last rainy season.
Re #160 to #163 There’s little new in Mann’s current Real Climate article. However, I did find interesting his use of the words “frenzy” and “theatrics” to describe the 2005 media and political behavior, as if he disapproved of the hype that surrounded the claims of Webster, Emanuel, Mann, Schmidt(sigh), Curry and the others. Yes, it was a media circus, but the circus was fueled and stoked by the press releases and interviews given by Webster, Emanuel, Mann et al. They greatly overstated their case and should not blame that on the press.
Here are a couple of interesting sea surface temperature maps. They are global but just focus on the Atlantic
June 21, 2005
June 21, 2007
What’s noticeable is that, in 2005, there was a small cool pocket surrounded by a “ring of fire” of warm SST (yes, the ring isn’t closed, but “horseshoe of fire” sounds strange). That is due to pressure and wind pattern anomalies (see AMO). The patterns produced a very warm tropical Atlantic, which in 2005 preceeded a very active hurricane season.
In 2007 the pattern appears to be reversing, with a small warm pocket in the center surrounded by increasing anomalously cool waters, including the tropical Atlantic. This, again, is related to wind and pressure patterns. The pattern is reminescent of pre-1995 SST, when the Atlantic was much less active.
Does this mean that we’re mving back to a less-active Atlantic? I strongly doubt it, and see this as just a bobble in a longer-term active pattern. But, I wonder if the active-AMO phase is strongest immediately after a shift occurs (1995-2005) and then we increasingly see bobbles which add up to a slow decline towards inactivity.
Footnote: There is a lot of anomalously cool subsurface water in the key tropical Atlantic region this year which may well upwell. If it does, then the one-two punch of a cool wind pattern and cool upwelling may give us a relatively cool tropical Atlantic. We shall see.
Footnote: The wind and pressure patterns which produce tropical Atlantic cooling also affect wind shear. That (wind shear) impact is the dominant factor in determining whether a season is active or not, not SST.
RE 162
Pardon my late arrival, folks.
Dave Dardinger writes:
“1. Chris Mooney has a great new book supporting AGW and more / stronger hurricanes
2. Thought his last book was a partisan attack, he’s become more nuanced now and the bias isn’t as obvious in his writing.”
I won’t comment on my new book’s quality, biases, or lack thereof. This I leave for reviewers to judge–and for folks like you, if you care to pick up the book.
However, statement 1 is false. I don’t support “AGW and more/stronger hurricanes.” I wrote a narrative account of the ongoing debate over this subject, and the conclusions I reached were more about the nature of such debates than the validity of the science per se. In fact, in the ongoing debate I don’t pronounce that anyone is “right.” I personally find the current evidence on hurricane intensification (such as it is) to be quite worrisome, but I fully admit that evidence is contested (and go into considerable detail about why). As an index to the book, I publish the WMO consensus statement from late last year pronouncing precisely this–the data is contested.
Also–and again remaining consistent with the WMO statement–I would never argue that we’re going to have “more” total tropical cyclones due to greenhouse warming when currently some models show *less.* It could very well be the case that we get less total tropical cyclones but that they’re more intense on average.
If any of you folks are interested in what I actually say I encourage you to read the book. ClimateAudit gets a brief cameo.
Best,
Chris Mooney
RE: #168 – A goodly portion of the Atlantic as well as far Western Europe and NW Africa appear to be in a pattern that is what one would expect in an El Nino. It’s as if someone took the eastern half of the lower image in Fig 2 in Steve McIntyre’s initial post and grafted it onto the western half of the upper image.
re: #169
Please note I was summarizing the thread on RC with a bit of editorial commenting thrown in. I didn’t read your first book, but given both the title and RC approval, I assume it’s from some position to the left of US political center. And your statement above,
does nothing to dissipate that feeling.
It’s possible I’ll glance through your 2nd book at a book store, but am unlikely to buy it or read it completely unless it’s a lot different than I suspect. Now admittedly not everyone here is on the right (specifically Steve McIntyre isn’t). But those who are, are used to seeing bias in writing from those on the left. And that includes writing about science.
For people who like neat internet sites, here’s one that shows hurricane computer model outputs ( link ). It lists seven key models and their 5-day predictions.
Simply click on “submit” and then on the next page click “FWD” to start the map loop.
Three models (mm5fsu, gfdl and hwrf) don’t update until there’s a storm or suspicious area around, so their maps are currently stale. So, look at the other four “live” models.
High pressure (good weather) is reddish while stormy weather tends to be greenish or bluish. Hurricanes are small concentric circles. You can watch the highs and lows progress across the Atlantic. The Azores/Bermuda high is the big reddish thing in the mid-Atlantic, which weakens and strengthens but never goes away this time of year.
Dave,
Thanks for the reply. You don’t have to read anything you don’t want to. But to me, your reply also shows “bias”…you’re literally judging a book by its cover. That’s exactly what I *didn’t* do in Storm World–instead, I went out and met and talked with the chief scientists in the debate on both sides….
cm
RE: #172 – I chose GFS, simply because its one used a lot here on the W. Coast. Both the Pacific and Bermuda/Azores High end up pretty far to the south after 168 hours. The NWS is prog’ing rain pretty far south for this time of year here on the W. Coast, late in the week. Similarly, reports I’ve gotten from SW Europe are unseasonably wet weather. So long as these highs stay south, the truly tropical weather gets pinched down to the ITCZ proper and in between the two highs (e.g. the Gulf and Mexican landmass) will continue to get mostly midlatitude weather.
Re #173 Hello Chris, I have a question unrelated to future hurricanes and which draws on your New Orleans background: how well do you think the communities that lie east and south of New Orleans will recover?
I find their situation and future to be intriguing, because it’s a case of a tough-as-nails village and small-town people, including many Southeast Asian immigrants, meeting absolute devastation.
Ultimately, I bet there’s a book or two to be written about their struggles and (my bet) triumph.
Re 3174 Steve S, one surprise for me is that the Atlantic ITCZ activity so far this year has been quite weak. It’s still too early to know if that will hold true into the peak season.
RE: #176 – Atlantic synoptics are still saying El Nino, in spite of ENSO reality. Is PDO mucking things up? We know very little about what a negative PDO is like, most of us were either kids or at most, young(er) adults during the mid point of the last negative PDO. Satellites really did not come into their own until the last 10 years of it and even at that, for most of that time, were very primitive. Plus, no one even knew what a PDO was back then, so no one knew to even monitor it and its impacts. ENSO was barely understood. We really don’t know much about what to expect from this new scenario.
To Steve and David, La Nina conditions weakened considerably in the last 2 weeks. The El Nino/La Nina area is only 1.5C cooler than average right now. So, the La Nina predictions will have to wait for further data coming in.
Huge cool pool of water in the northern pacific, huge warm conditions in the southern pacific. I don’t know what this implies for the ENSO or PDO or hurricanes.
I know you didn’t ask me David, but I think the smaller towns and villages outside of NO will probably slowly work their way back. For a lot of those places, the people didn’t have two nickels to rub together to begin with, so they didn’t lose much more than that. With a little assistance from outside sources and some scratching and clawing on their part, they will be back to squeaking out their existence within a year or two. The people there are used to doing what needs to be done for themselves, and for that reason i think they will make it back where NO probably won’t.
As for NO, I work for a contractor that is doing alot of the recovery work there (luckily I work in a different group). But I am close enough to hear plenty of stories about what a train wreck it all is. I just don’t think most people realize the level of political and organizational ineptitude that NO is facing. Repairing and rebuilding NO would be a monumental task even for a state and local government that got along well and knew how to properly organize and operate a city with modern infrastructure. The local government in NO is beyond worthless. They are unable to do anything other than harm the rebuilding effort at this point. If the city government were to close up shop and leave town, it would probably be an improvement. The parish governments aren’t much better, but at least most of their efforts are aimed at their own personal enrichment and aren’t actively involved in making the situation worse than it already is. The state government would just as soon forget that NO existed and spend all the federal money building up Baton Rouge. And the only part of the original NO population that has come back seems to be the ones intent on stealing stuff and killing people. The amount of money being wasted trying to clean up and rebuild that city is staggering…and almost all of it is going right back out of town and into somebody else’s pocket. In the end, if NO is going to be rebuilt, there has to be somebody besides thieves and murderers who want to live there…and who can find work there…and they will have to be the ones who take the effort of rebuilding the city on their backs and get it done. New Orleans has never bred much of that sort of attitude in its citizens, and for that reason, I doubt the city will ever return to what it was. Most of the NO residents that came to Houston seem content to stay put as long as the government keeps paying the rent without asking them to get a job. In that respect, their lives haven’t changed much since they were living in NO.
David, I am not sure if you were living in Houston in 1979, but I was just a kid back then, and so far this year reminds me ALOT of 1979. It seemed like it just rained all the time beginning in about April, and the rainfall events kept getting more and more significant in terms of quantity as the summer went on. I know Claudette was the “big one” in terms of floods, but I can remember going out with my family at least 3 other times that summer in a canoe to bring in meals to people living nearby whose homes were flooded. If you were around (and perhaps more aware of meteorological patterns back then), how does our current pattern compare? I don’t remember 1979 as being a big year from a tropical storm standpoint (could be wrong), but I do remember it as being a very rainy and somewhat cool year for all of Southeast Texas. Given the cool start to the year we have had so far, perhaps a repeat is in the cards for us?
Re #179, #180 Bill, I was in the Texas Hill Country this weekend and have never seen it so wet and beautifully green as it is this June. It’s an unusual, and welcome, pattern. My guess is that, globally, we’re into a pattern change, and how that plays out locally is anybody’s guess.
You speak the truth about New Orleans. I have other, even harser comments, based on my many years of living nearby, but those are not for the internet. I’ll be in NO in July and will bring back some photos and stories of the attempt at recovery, on personal levels.
Chris Mooney (#169) opines “I personally find the current evidence on hurricane intensification (such as it is) to be quite worrisome….”
Having just questioned NCAR’s Greg Hollander – one of the hurricane specialists in Chris’s new book – in a public forum last week, I’m not.
His conclusions are triple caveated, to the dismay of ACW proponents, said Hollander. He simplistically defended his open-mindedness, because science needs “diversity of opinion” ‘€” except when it comes to Senator James Imhofe, Ruch Limbaugh, and Exxon/Mobil. (Double standards in Boulder are considered “sophisticated.”) Regarding Bacon’s Novum Organon, “it’s all there,” he said with breathless lack of critical insight into how good honest science is defeated by institutionalist funding and sociological pressures of conformity – which is what Bacon endorsed. Yet Hollander himself is very very very self-critical, he said.
When I challenged him on his “worrisome” inferences about the future because of the lack of predictive skill in the GCMs, all he could say is that “no one he knows” [sic] expects lower earth temperatures, and the latest models claim to resolve cyclones of about 20km. That’s like saying “experts predict crime will be up over the next ten years,” but we’re still working on who will be doing it and against whom!
Nevertheless, I’m going to buy “Storm World,” mostly because I’m a sucker for insider details. I doubt it can do better than his previous diatribe, but I hope it will.
re: #173, 182
The thing is we’ve had a number of the heavy hitters in hurricanes weigh in here and the impression I took away was that there’s no real evidence of hurricane intensification. It’s likely the observations which have indicated the intensification are artefacts of either historical cycles or of random fluxuations. The improved observations from increased trade and satellites mean that there isn’t really much change to be found once it’s allowed for. All that’s left is the modeling; as seems to be the case in most things AGW.
Re: #183
I would agree with you, DD, and point to the threads here at CA which have provided analyses that I am certain will not appear in a book produced for popular consumption. If one were paying attention to those threads, I cannot see how they would not come away with a comprehensive view of the subject.
If Chris Mooney had joined the technical discussions here at CA, perhaps we would know better how knowledgable he is on the subject of hurricanes and been more inclined to read (buy) his book(s). We already know that two people, even two scientists who are expert on the subject, can look at the analyses and come up with different inclinations on global warming and climate adversities — with much depending on their AGW advocacy views as it relates to policy. When somebody such as Mooney writes a book about these subjects one tends to get too much of the policy view and not enough of the understanding of the underlying scientific and statistical matters. Of course, I am prepared to stand corrected if Mooney wants to engage here on the technical matters as contained in his book.
I plan to read Chris Mooney’s book for its perspective on the players (Emanuel, Gray, Webster, Landsea, Curry, Holland, Klotzbach, etc) and how this whole topic grew into the hurricane wars. It’ll be particularly interesting to see how Mooney analyzes the early promotional and publicity aspects of the issue and whether he blames the media for hyping the topic.
It’s been a rough two years for the proponents (Emanuel, Webster, Curry, Holland, etc). Not only have the reanalyzed storm histories been unsupportive of the proponents but the GCMs have also delivered a few blows (see the Vecchi and Soden paper and thread here ).
But, my expectation is that they’ll be back, this time on the Atlantic alone, with an AMO-is-stuck-in-high-gear-and-North America-is-doomed hypothesis.
Re: #185
I would have expected nothing less from a man with your interests in weather, climate and particularly hurricanes. An articulate review of this book, as you have done here with articles on the topic would be much appreciated and of course save me the expense of time and money ‘€” until such time you recommend it or give a thumbs down.
While the NATL TS activity appears to be somewhat subdued for the present they have had two unusual occurrences of cyclones in the Arabian Sea. Maybe that is next fertile ground for WHCC and Emanuel to connect warming SST and tropical cyclones ‘€” you know like moving on from the North Atlantic.
RE #185 For a while it looked like the second Arabian Sea storm would enter the Gulf of Oman. That would be highly unusual but meaningless – simply a chance event. But, had it occurred, we’d see headlines and by next year a peer-reviewed paper, “Oman Typhoons Linked to Global Warming; Cliffs of Dover Next” would appear in Nature. I’m glad the storm instead veered left and broke apart in Pakistan.
Re #187 It should be marked “#186”, sorry.
Steve S, I think the tropical Atlantic season gets cranked up in the next two weeks. That’s about 3 weeks sooner than normal.
RE: #188 – We’ll see. I don’t recall seeing a more or less north to south branch of a polar jet sitting for so long, over the East Coast, so late in the year. There is a cold front prog’ed to eventually reach FL mid next week. I agree that we are now seeing Easterly Waves and one will cause showers in FL over the next few days. But with that jet, I don’t know. A lot would have to change over the next two weeks for the season to actually get going. Then there is the dust issue, still almost reaching the Lesser Antilles.
re #183
I agree with your brief summary of what’s happened here on CA, Dave: “there is no real evidence of hurricane intensification.” But my two-fold point is rather seriously much larger and importantly wider in its implications and pertinence. These issues expose how science has devolved from the Enlightenment ideals it used to uphold into something more tribal, worryingly politicized, and epistemologically polluting.
First, Greg Hollander is an important figure within the ACW community to understand because he was a student of Bill Gray’s, up the road, at Colorado State University. He’s seen as an ACW skeptic who has gone over to the other side. Thus, understanding the answer to the question “why?” gains importance.
It seems he has adapted to the very pro-environmentalist milieu of Boulder, and this has affected the inferences he has drawn from his own research. It has also circumscribed the range of experts he considers relevant to the problem under study, eg, Hollander excludes solar scientists from the answering the ACW problem. In other words, sociology has altered or “bent” his scientific epistemology.
Second, traditional positivist epistemology considers knowledge to be about the rules governing “Justified True Belief.” But the late great philosopher of science Karl Popper held that no knowledge can ever be justified; it is always provisional, subject to new discovery and empirical testing. Therefore, the best we can be certain of is falsehood, ie, that certain explanations are falsified, given certain empirical facts. Popper’s philosophy is about the practical application of the logic of modus tollens to scientific problems, ie, A implies B, if and only if ~B implies ~A. He realized the logical asymmetry that prevails: knowledge cannot be “proved” true, but falsehood can. Therefore, only falsifications are true or false. All other claims are merely provisional.
Third, according to Popper, not only is naïve inductivism unsound, it rewards and idneed thrives on confirmation bias. For example, take the new UN Secretary General’s charge that ACW has caused the genocidal problem in Darfur! But of course.
The battle between traditional climatologists such as Bill Gray – not to mention Roy Spencer and John Christy at UAH – versus the New Guard of ACW insurgents who now dominate publicity is fundamentally about what constitutes good science.
To Gray, climatology is a humble, descriptive, and empirical science where the data and facts are the best guide to understanding the dynamic, complex coupled system that is climate. By contrast, the New Guard that flows from James Hansen’s pronouncements in 1988, believes that we can’t wait for the data to come in: the risk of altering the climate from anthropogenically added CO2 requires new methods that supercede patiently waiting for the data to confirm suspicions of danger. They apply models based on hypothesized impacts of added CO2, based on the physics of fluid dynamics. Thus, theory trumps induction.
Therefore, because of the company he keeps and the sources that fund him, Hollander is torn between two conflicting ideals of how to do good climatology. ACW Believers want the evidence to be more powerful than his own modest research conclusions go. Yet now, or recently, even the traditionally trained humble meteorologist he is wants to draw inferences that support alarm and worry.
This cleavage among scientists was one of the chief lessons of last years climax over M&M and the Hockey Stick with the NAS and Wegman reports. Wherever one looks in ACW debates, the issue breaks down to commitments to, or lack of understanding of, how to properly conduct science, ie apply the philosophy of science to doing science.
Recognizing this, however, allows us to expose the patterned nature of the “understanding” claimed, as well as the pervasive patterned nature of the misunderstanding, the vehement charges and countercharges repeatedly made by rival sides. Judith Curry says Gray suffers “brain fossilization” while William Connelley says Nils-Axel Morner is “a band of one.” Where is any self-critical ethos? Gray counters with charges that the politically-driven New Guard have crowded out the funding of his own more important scientific research program on hurricane prediction. Morner claims the IPCC is deceptive and fraudulent about its claims on sea-level rise. Contrary to Hollander’s declared values, the company he now keeps rejects grappling with the well-founded criticism of so-called “skeptics.”
From time to time, people on CA have claimed that ACW theory has been falsified. But the New Guard marches on as though this never mattered, as the Hockey team has since last year, ignoring that their many year’s commitment to the proposition that “the 1990s was the warmest decade in a thousand years” is known to be false. Therefore, any baseline support for ACW alarm is much harder to shout.. And so critics today see ACW proponents as more hysterical than ever before!
If this debate ‘€” and despite the stridently disingenuous claims otherwise, it is a real debate – is to be clarified, better understood and (put optimistically) transcended, then popular books like “Storm World” should be seized upon as opportunities to expose the persistent meta-context honest observers see operating. In short, the repetitiveness of claims and counter-claims proffered by scientists should be recognized.
Which brings us back to the salient example of NCAR meteorologist Greg Hollander. One of philosopher Mark Notturno’s most remarkable Popperian insights is that inductivism leads to the institutionalism (ie, joining ACW proponents), which is the path Hollander exemplifies. Logical arguments do not establish truth, but merely test the truth of a statement. Unfortunately, committed and trained inductivists like Gray and Hollander do not realize that derapage, or sliding away, tends to result from naïve inductivism. Either one self-consciously resists this slide like Bill Gray does, or a good scientist like Hollander succumbs.
Notturno writes: “the problem with inductive arguments is that they cannot justify their conclusions [eg, “ACW is a worrisome threat to us”]….Since the conclusions of an inductive argument never follows logically from its premises, an inductivist must take a little leap in order to accept it as true. The [slender pro-ACW] belief that [Hollander, for example, accepts as true] is unjustified. But he feels safety in numbers [at government-funded NCAR and Boulder generally], and more justified in accepting it if others whom he respects will at first [like Webster, Curry, Emmanuel] accept it too.” And so Hollander rubbishes Bush and ACW-critics on the “right” in general as all like-minded and uncritical, as he did last week in Boulder.
This last interpolated point explains why Hollander correctly disavows that any one hurricane can “prove” ACW threatens us, like 2005’s Katrina hitting New Orleans, while also condemning NOAA for posting explicit disavowel of any such connection on their web pages! Hollander implied ‘€” probably correctly but dismayingly to ACW-critics ‘€” that surely some meteorologist nonetheless holds this dubious belief! This virtually amounts to doublethink, the simultaneous holding of two contradictory beliefs at the same time! If I’ve ever seen n esteemed scientist do this before, I don’t know when it was.
Notturno concludes: “This is why institutionalism and inductivism are more closely related than one might think. This is why institutionalists are so opposed to criticism. And it is why the idea that inductive arguments justify their conclusions [cf, Al Gore’s AIT] are so pervasive [cf, the ACW proponents that cite AIT].” (Page 88/) Thus, an otherwise self-critical scientist Greg Hollander accepts that AIT is scientifically valid, while calling The Great Global Waming Swindle’s science false.
What is one to do? I could go on in detail about rejecting the doctrine of justificationism in the proper practice of science, but let me just cut to the chase. The best scientific example to follow is Steve McIntyre, who hosts and leads this web-site. Without declaring himself to be on either side of the ACW debate, he has been a shining example of scientific inquiry done with dignity and honesty, patiently examining the data, inviting criticism, following up on leads, while continually accepting that a range of informed and expert viewpoints exist. This is refreshing and far more productive than the conformist institutionalism seen at RealClimate.org.
If there is any fault to place before Steve, it was his naivete last summer that ACW advocates would accept falsifcation of the Hockey Stick as counting against the strength of ACW conclusions. (And I’m rather sure Steve would concur.) Yet by explaining why these divisions systematically arise and sociologically recur, we can anticipate and alay the rise of misplaced optimism in the future.
While both Popper and Thomas Kuhn believed that science was a social institution, they diverged over its character and identity. Kuhn rejected progress as a governing ideal. “And the institutions of science have now devolved into The Scientific Institution,’ a tribal institution [like the IPCC] that not only wield political power, but whose effect is to suppress the freedom of thought.” While Popper believed that science was devised to facilitate the critical examination of competing ideas, Kuhn believed it was to “forge solidarity of belief.” (Page 89.) This division underlies what we see and discuss here.
There continues to be a great deal to learn from Steve’s fine example. I think that most of us here are still learning from him. And that’s why we come back again and again: to argue and to learn, and to test the limits of what we actually know. More often in the hope that Popper was right than Kuhn.
Lastly, let me express surprise that David Orrell’s chapter in The Future of Everything on prediction in climate science has not yet provoked discussion at CA. His work on falsifying models should. This is precisely the topic that commands our attention, and not only for reasons given above. As Eduardo Zorita asks about a recent post at Climatefeedback blog by Kenneth Trenberth entitled “Predictions of climate,”
“if the IPCC climate simulations do not represent real predictions for the future [as Trenberth says], how can the theory be falsified?” Indeed. And so how can sound science prevail?
REFERENCES
Jarvie, Ian C. The Republic of Science: The Emergence of Popper’s Social View of
Science. (2001)
Notturno, Mark A. Science and the Open Society: The Future of Karl Popper’s
Philosophy. (2000) See especially “Introduction” and Chapter 2, “Science and the Institution.”
Orrell, David. The Future of Everything: The Science of Prediction. (2007)
re: #190 T J
Thanks for the long message. I enjoy philosophy of science and wish I had time to respond at length to your message.
I think there’s something I should be getting from this partial sentence, but I’m confused about who’s the honest observers; you and I? Chris Mooney? Someone else? And I’d be happy to discuss what I see as the clear meta-context except that Steve Mc explictly requires we don’t do so here. (And I see his point and go along with it.)
Tell me about it! I just forced myself to read the whole thread on RC concerning CO2 band saturation. I can imagine what Raypierre would have to say to your message, assuming it were allowed to be posted. “You’re mostly wrong and obviously don’t know anything. Try sticking to what we here know to be ‘real science’ if you ever want to be allowed to post here again!” Of course he had his own head handed to him in post 47 but it probably was lost on most people. And there are a couple of glaring errors in the thread which I don’t know whether or not are worth trying to bring up or not. It would require more tact than I think I’m capable of showing.
Re Chris Mooney’s book, a “must read”. Mooney frames the entire book in the context of the very topic that TJ Olsen posts about, the battle between “empiricists” and “theorists”, a battle that was set up on the issue of hurricanes over century ago. MIxing that with personalities and the policy relevance of this issue makes for a very compelling read. The book is surprisingly technical, with plenty of meat for people interested in the actual science. Mooney has provided some important insights into this.
And one other thing. Mooney has definitely been reading climateaudit, it is even mentioned in the book
Re: #190
First off, T J Olson, let me be clear in saying that I have no interest whatsoever in knowing what motivates the public statements of a Greg Holland or a Judith Curry or a Bill Gray ‘€” you get the picture. Otherwise, thanks for your rendition of the philosophical underpinnings of the climate debate.
In my view it is instructive and essential to separate the climate debate into that related to the science and that to policy and to do so down to the level of the individual whose discussions/work is being evaluated. The science can be evaluated based on some rather objective criteria while policy statements, as they use and interpret that scientific evidence, are much more difficult to evaluate. Policy relates to many factors not scientific and while these factors are important to take into consideration they do not lend themselves immediately to an objective evaluation.
A rather black and white example I would give would be 2 scientists who would appear to be interpreting the scientific results much differently when you hear how differently they would apply the current knowledge to climate policy. Scientist 1 has determined, outside his field of expertise, that government programs are required to meet the goals of climate and environmental control and sees them working with a net benefit regardless of any unintended consequences. This scientist sees government mitigation of the potential adverse effects of GW as something that is a good thing, regardless of the eventual need for them, i.e. how well the predictions of climate change turn out. Scientist 2 has, by a similar fashion, determined that the unintended consequences of government mitigation are most likely to have a net negative effect over essentially having it doing nothing and even in the face of the worst case scenario predictions. Scientist 2 has considerably more confidence in humans abilities to adapt on the own than Scientist 1.
Now if we are given the current situation where one cannot place confidence limits on the projections of future climate, and particularly so when it comes to netting out beneficial versus adverse results, in my mind Scientist 1 and Scientist 2 are going to look at these uncertainties much differently — and primarily based on considerations outside their field of expertise.
I think that thread of uncertainty and the various reactions to it come through in reading the FAR reports and definitely in reading the rejected reviewers’ comments. Although the preparatory procedures of the FAR call for a traceable account of how the authors determined the levels of confidence and certainty that they attach to their findings published in the final report, one sees no accounting for these determinations that should be the most transparent and important part of the report. I have requested these documents, but have received no answer to this point in time.
#190 — TJ, that was merely a brilliant essay. Thanks for the pointer to Notturno. I didn’t know about his work.
And you’re right about Steve M. He has been relentlessly scientific in the best Popperian sense of that word. You’re also right that Team Mann revealed its anti-science hand when they launched RC in response to M&M03, rather than correcting their mistakes.
Judith, save this comment for the book cover. In the meantime, I will await my reviewer’s (David Smith) evaluations and, of course, an appearance by Chris and/or you here at CA to engage in discussion about the latest developments in the science of hurricanes.
I’m appreciative of T J Olson’s comments in #190 most especially because he frames the debate in terms of objective knowledge vs everything else. The AGW crowd, including Mann&co. at RC, have polluted science with subjective conclusion-mongering. That is, they already “know” the answer by reference to their inner certainty, and have interpreted their empirical results in terms of that certainty. But their certainty rests on the inner jump in logic, from induction to conclusion. It’s identical to a leap of faith. The intuited conclusion takes precedence, not only over the data but also over any other hypothesis, even a better one. Hence the debate proceeds by polemical discredit rather than by the normal re-evaluations of objective science.
Scientific dispassion has been derided. But its real value is terribly apparent when contrasted against the degrading polemics of the AGW debate, and the macerated and tendentious methodologies that so disgrace AGW so-called science.
Nils Roll-Hansen has written “The Lysenko Effect” in which he shows that this episode included many scientists who actively supported Lysenko. They probably referenced the same intiutive feeling of inner certainty as motivates the AGW faithful. When joined with politics, science becomes muddy and polluted, and an instrument of stultifying oppression. The elements of persecution that resulted in jail and suicide in the SU already show up here, as straws in the wind, with pejoratives like “denialist,” rhetorical accusations of mass murder, and attempts at professional decertifications of skeptics.
#194,
I have no problem with two scientists having a different take on how to deal with the uncertainty in the science.
What I do object to, most strenuously, is for either scientist to try and hide that same uncertainty, in order to better influence policy makers.
I just read the excerpt from Chapter 1 on Mooney’s web site. I have to admit that the positive review by RC makes me “skeptical” and the cautious pro-AGW spin to the first chapter reinforces my skepticism – though it is really well written. So I am off to the bookstore to read Chapter 2 and then to buy it if the argument appears sufficiently objective and balanced as Dr Fred on Amazon suggests.
I just finished “Storm World” and whether you agree with AGW or not it is a good look at the history of hurricane science and at the present state of it. It is very even-handed and goes into many of the data quality details which make the question of “is there a detectable AGW signal in hurricane activity?” so contentious.
Chris Mooney has written a book (“Storm World”) which covers the ongoing debate over hurricanes and global warming. I recommend it.
This is not a science book but rather history-of-science book. Mooney does a remarkably good job of explaining some of the technical aspects but doesn’t go off the deep end. His focus is on the battle of views and personalities.
He also does a good job of removing much of his personal viewpoint on the matter, which is no easy task. Mooney authored a book named “The (US) Republican War on Science”, which is a nod towards his worldview. There are a few slips, like innuendo about Exxon and skeptics (where’s my free gas?), but overall his objectivity is an achievement.
The theme of the book is the battle between two approaches in storm meteorology (theorist vs empiricist) and how that battle plays out through the personalities involved. As Mooney sees it, on the “warmer” side are the theoretically-grounded while on the “contrarian” side are the data-driven empiricists. There’s a touch of an undertone in the book, in my opinion, that data-driven people aren’t firmly grounded in theory (sort of like a collection of Forrest Gumps) while theorists are purified by proximity to computer code, so be forewarned and ignore it. Developed disciplines lack the strong dichotomy Mooney describes and benefit from properly-used observations in developing theory. Facts are friends.
The first portions of the book cover the historical attempts to understand cyclones and how the battle of ideas about cyclones played out through the personalities involved. It’s well-written section. The current battles enter about midway, beginning with sketches of various personalities, especially Bill Gray.
Mooney tends to paint the players in neutral colors, which makes some appear to be more dispassionate and objective than perhaps they deserve. As Steve McIntyre sometimes mentions, some of the players are not unlike mining promoters. Read the book, reflect on some of the exchanges at CA and draw your own conclusions.
I think this is a good summertime read, maybe for the beach or cottage. Few people (including me) will agree with all that is said, but that’s often the mark of a good book.
David Smith thanks for the book review. Since it would appear that you have recommended it as a good summer read and I tend to be a frugal guy with time and money, even in retirement, I must now put out some qualifiers for you to judge before I impetuously purchase the book or even wait for it to arrive at my library.
Let us assume that I have a reasonably good understanding of the difference between empirical and theoretical approaches to acquiring knowledge and would not understand how they would be definitively separated in doing science regardless. Also assume that I am not particularly attracted to a work simply because it is well written with the exception of fiction and poetry. Assume also that I am fairly familiar with the personalities involved in the climate discussions and think that too many times they overwhelm the more substantive and objective parts of the debate. Assume also that I assume Mooney has not added significantly to any new insights into the questions of storm frequencies.
After reviewing my assumptions, would you continue to recommend the book for my personal reading? If, yes, could you please briefly explain why? The only consideration that I am seriously thinking about would be a historical perspective that is unique to Mooney’s book.
Hello, Ken. In your position I think you’d get a good accounting of the history of hurricane research up to about 1990 and, of more interest, an accounting of the blow-by-blow events of 2005 and 2006, with many details which aren’t widely known. Is that worth $25 (higher in Canada)? Hard to say. Personally, I would spend the $25 (higher in Canada), as what I read in the later chapters helped explain the shifts I saw in the various Webster/Holland/Curry Powerpoints and commentary (many discussed here).
I just heard Chris Mooney being interviewed about his book on the radio show Quirks and Quarks. He comes across as being very even handed and well informed.
#206
Ralph! That would be the June 23 podcast of QQ
Looks like 2 cats…Not Cat. 2…Joking apart
Mooney is from New Orleans and his mother
lost her house there in Katrina 2005 and moved to
Arizona…
Al Gore wasn mentioned! I can see
why…
Well you posters really have left me no choice in this matter. I am off to my nearest book store with a handy 20% discount ticket in hand to buy St. Christopher’s latest book. I had a tooth implant this PM and will read the Storm while recovering. I am hoping that the recovering and reading will both be relatively painless.
Re #208 Buy a good red wine while you’re out, just in case…
I am almost half way through Storm World and from my perspective as a skeptic I cannot agree with previous posters here that Mooney has presented an unbiased view of the global warming and hurricane frequency and intensity debate.
To this point in the book Mooney has presented material that I could obtain in the mainstream media and the IPCC. It unfortunately for me has been the same old same old. He essentially sees the skeptics as less than pure in motivation and probably more infulencial than they actually are (fossil fuel interests and backing and all the other non-scientific influences are trotted out) while seemingly having no similar precautions when it comes to the AGW and “bigger and better” future hurricane advocates.
He has used RealClimate for his scientific blog perspective exclusively to this point in the book.
I find it most curious the contrast between Gray and Emanuel and the artificial separation of the climate debate into data driven and theory based camps that Mooney proffers in this work. The theory based people like Emanuel also do the careful data collection and observations while it would appear from Mooney’s reporting that Gray, and by implication more data driven scientists in the climate field, is driven more or less exclusively by data. So far it has been the old literary gimmick to personify differences by introducing a couple or more personalities to pique readers’ interest with, of course, the author deciding on which personalities to contrast. The data driven scientists to this point in the book tend to be the more skeptical. What I would have preferred would have been an extraction of information and analyses from the various players in the field by Mooney without the political implications and his view of their personalities.
I can certainly see why Judith Curry supports Mooney’s renditions of the climate wars as they are very similar to what she has offered here and elsewhere. Now for Mooney to even the score to net a neutral stance on these matters for me, the remainder of the book will have to explore the use of data by the Emanuel and WHCC crowd and point to the critiques of it. He would also have to explain and critique the process of determining a consensus by the IPCC. He no doubt will need to broach the issue of the transparency or lack thereof of the AR4 in revealing the methods by which they arrived at their levels of confidence and likelihoods. He will get into the very essence of measuring uncertainty.
Unfortunately such insights have been lacking to this point in the book and I see no reason to expect a literary u-turn, but read on I must.
Just noted:
Brouhaha at the National Hurricane Center
I finished Storm World and I can only say that I really learned nothing new (well, the jacket pix of Mooney showed that he could be a brother to my cardiologist) and received another rendition of the view of the climate debates from a member of the consensus. In the second half of the book Mooney’s POV of politics and climate becomes more apparent and pronounced as he uses Gray as a straw man and proxy for the skeptics that he gradually grinds into a blithering idiot at the same time he proclaims a certain admiration for the man and all the while he slowly builds Emanuel (representing Mooney’s POV) into a heroic figure. He covers the Kossin revelations that put uncertainties into the WHCC and Emanuel conclusions, but chooses to say little. He puts forth Landsea’s arguments critiquing WHCC and Emanuel papers on hurricane frequency and intensity increases, but without the detail he uses in putting forth Emanuel’s and WHCC’s original arguments.
Mooney spends time on the Bush administration’s attempts to control what government climate scientists say at or which of them attends press conferences. While I have little doubt that this occurred, the administration was unsuccessful in the end and the public relations backfired on them. I somehow doubt at another time and place that Mooney would do a similar expose or write about the lack of transparency in the IPCC and AR4. He seems, in fact, to be totally naive about the political nature of the IPCC in the FAR, TAR and SAR reporting, while being very sensitive to it in the Bush administration.
Finally, I must note here that Mooney even mentions the Curry visit to CA and her attempts to involve her students in an evaluation of the blogosphere and CA specifically. Since I thought that attempt was a waste of time I would categorize Mooney’s mention of it in same way. He does seem to quote Curry rather liberally in the last part of the book and talks about her getting more involved with blogs but for what purpose or point I do not know.
Thank you Ken. You report about what I would have expected fo find in the book. Some people think not overtly calling skeptics names is all that needs to be done to be “even handed”. Bias is generally much more covert / indirect than that.
re: #212
“There are so many wondrous appearances in nature, for which science and philosophy cannot, even now, account, that it is not surprising that, when natural laws were still less understood, men should have attributed to supernatural agency every appearance which they could not otherwise explain. The merest tyro now understands various phenomena which the wisest of old could not fathom. There are so many wondrous appearances in nature, for which science and philosophy cannot, even now, account, that it is not surprising that, when natural laws were still less understood, men should have attributed to supernatural agency every appearance which they could not otherwise explain. The merest tyro now understands various phenomena which the wisest of old could not fathom.”
Checking back for further reactions to Storm World. I heard from a reliable source that Bill Gray liked the book, but this is second hand information, it would certainly be interesting to hear Gray’s opinion of the book. In my opinion Bill Gray was portrayed accurately and generously. WIth regards to the portrayal of Kerry Emanuel, whether or not you like his recent work on hurricanes and global warming, he ranks as one of the most highly cited researchers in geoscience (as per webofscience), he recently received the Rossby Award from the American Meteorological Society (their highest research award), and was just elected to the National Academy of Sciences (pretty much the highest honor available to a U.S. scientist outside the Nobel Prize). There are relatively few atmospheric scientists in the National Academy, and it is certainly not a hotbed of “warmers”, with Lindzen being one of the few atmospheric scientists in the National Academy. So if Mooney portrays Emanuel as a big shot scientist, I would have to say it was pretty defensible to do this. A main part of Mooney’s thesis is that the reputations of the scientists involved and their personalities were a major part of the conflagration that occurred in 2005/2006 and how this played out particularly in the public debate on this issue (including what goes on in the blogosphere).
I am not unaware of the players and the importance of their roles in this debate which makes it all the more curious to me why Mooney would spend so much time on Bill Gray. Bill Gray’s political comments, aging characteristics and personality are well known, but in the hurricane debates there are more important players who are more substantially critiquing the works of WHCC and Emanuel. Gray’s dominant skeptic’s role in Storm World was particularly surprising in the light of the science and non-science worlds’ criticism of Gray’s participation in these debates without contributing to the peer reviewed literature. Emanuel could get a Nobel Prize yet what I would continue to want to see is a hard hitting critique of his work and in this case his use of storm data in an attempt to confirm his theories on increased hurricane intensities/durations. Maybe the separation between the data and theory driven climate scientists, that Mooney attempts so valiantly to maintain throughout the book with inferences that this would appear to hinder Gray’s understanding of the new climate model regime with AGW and by inference and innuendo his allegedly more data driven students like Landsea, should also have been applied to theorist Emanuel’s handling of the storm data.
Judith, you or any number of AGW advocates could have written this book and imparted the same message ‘€” a message that I have heard many, many times. I understand where you and Mooney are coming from but I did not agree with the conclusions or the methods of presenting the issues. Like so many times occurs, but can be avoided, in the blogosphere, the book Storm World allows the personalities (and an inappropriate one in Gray’s case) of overwhelming any decent analysis of the issues at hand.
After my read, I am more inclined to refer to a leading player in the book as St. Kerry and remove the sainthood I endowed on the author. Even so and as I recall a Saint is not infallible. I guess from the author’s view imparted to me on your role in these debates I could also have sainted you, Judy. I, however, did have second thoughts when Mooney quoted you using swear words.
If I had written this book, it would have been very different (but probably not “better” in terms of selling books or effectively communicating with the public). I definitely would not have included bill gray as the major character. I do not agree with everything in the book, but i do think the book is important and well done.
Kenneth Fritsch said
“I am not unaware of the players and the importance of their roles in this debate which makes it all the more curious to me why Mooney would spend so much time on Bill Gray.”
This reminds me of the story of an ancient Chinese general who was asked if there was one factor in battle that he could choose to assure his victory, what would that factor be? The weather? The battle ground? The time of day? The number of troops on each side?
He replied that to assure victory all he needed to choose was his opponent.
Gray is ornery and prone to hyperbole. That makes him easy pickings even if his arguments are valid.
Interesting that you should say that because Mooney in his concluding chapter writes about the personalities involved in the hurricane debates overwhelming the issues and not allowing some kind of consensus policy (even given the scientific contentions involved) coming out of the dialogue. What did Mooney contribute to the debate then by writing a book that very much puts the personalities back into play? Does that make Mooney a tool of the book publishing industry? There is much that Mooney could have contributed to the debate by concentrating on all aspects of the scientific arguments being put forth and letting the personalities involved become secondary. He mentions Pat Michaels in passing as receiving funds from the fossil-fuel industry without so much as a nod to his cogent paper on threshold SSTs and high-powered hurricanes.
I am of the opinion that his personality portrayals added little reader interest to the book and they certainly have not nor will be a contributing factor on how the debates will work or have worked themselves out. I would also have to disagree with Mooney’s contention that scientists need to better hone their skills in translating their scientific findings into climate policy. Since I do not see that sufficient answers have been found and settled for all time by climate scientists, suggesting that they become part time policy advocates is particularly disconcerting to me. It also breeds the potential appearance for conflict of interest that Mooney sees so readily in the fossil-fuel industries. Mooney seems to miss entirely the politicization of the IPCC or at least fails to pass judgment on it.
Judith, I suspect that for one who fails to capitalize “i” in her postings, your writing of this book would have failed to make the numerous references to Judith Curry that Mooney did in his rendition.
I definitely would have had patrick michaels and his paper be a bigger player in this. pat michaels is an interesting case in the world of “skeptics”, and there was one important idea in his hurricane paper. Apart from Landsea and Chan, Michaels is the main skeptic on this issue actually publishing papers. I also think Phil Klotzbach is an interesting part of this story.
i found it interesting that i played a different role in mooney’s book than gray, landsea, emanuel, webster, holland. I was not described personally nor were my scientific points highlighted, but rather my assessments on the sociology and politics of the situation got a fair amount of ink.
but i think mooney is on the money in terms of how all this gets translated into policy. scientists need to sit down and do an assessment (preferably under the auspices of the NAS/NRC or some equivalent international body), send it out for peer review, and then make carefully crafted statements about the range of what the science says might have been happening and what might happen in the future, and how likely all this is given the uncertainties. President Bush’s Climate Change Science Program (CCSP) is in the process of completing the sythesis and assessment report on climate change and extreme weather events (including hurricanes). the lead agency organizing this assessment is NOAA (ugh), but the NRC is coordinating the peer review. It will be interesting to see what this report has to say. This will probably be the first real assessment of the hurricane/climate change issue: the WMO statement wasn’t a real assessment (no external peer review beyond people who attended one particular meeting) and the IPCC really didn’t dig that deeply into the hurricane issue. it will be interesting to see what this assessment report actually says.
Scientists involved in the hurricane issue are increasingly being roped into being involved in policy related to hurricanes. next week is a busy one inside the beltway, dealing with issues surrounding the observing system. There is a congressional briefing on the issue jul 10 (both house and senate) and there is a senate commerce committee hearing on the intersecting issues of noaa, hurricane observing systems, and relevant satellite systems. Scientists working in this environment need to do something other than present their own scientific arguments; they need to present both sides and the uncertainties, in a context that is relevant to the decisions at hand. This is part of mooney’s point: most individual scientists don’t have much experience in doing this or even particularly understand the situation they might find them in and may not know how to be effective in such situations.
That was my point. Besides the fact that there are so many other players in the hurricanes debates that have questioned the WHCC and Emanuel conclusions who have and are writing papers about the issues and making statements on the blogosphere makes, in my mind, Mooney’s intentions rather obvious. My alternative view would be that Mooney was not as informed as he should have been about the alternative (to WHCC and Emanuel) interpretations of the storm data. If he were reading exclusively at Real Climate that might be a legitimate excuse.
Which is in effect saying that scientists must became better politicians and that is what is very disconcerting to me. As one with a libertarian view of politicians, I see them making political statements and doing political posturing during these hearings that has little to do with the questions at hand and in this case the science at hand.
But given that we do not live in my idealist libertarian world, I would venture to propose that climate scientists have little or no business or expertise in directly suggesting policy direction or for that matter proposing the levels of uncertainties or likelihoods associated with their proposals. Lets put the measurement of uncertainty and likelihood to an uninvolved group of statisticians. Also I suggest that instead of attempting to obtain a false consensus from scientists that a majority and minority report be issued by the scientists. In fact what would work best, in this less than ideal world, would be similar to the opinions handed down by the US Supreme Court whereby we have the majority opinion and dissenting opinion and then versions of these by any of the participating justices (scientists). Best they be made in writing instead of in direct testimony to reduce the grandstanding by politicians (and less frequently witnesses) that so frequently render these events next to useless.
I somehow think that Chris Mooney is much too much into the status quo to make proposals such as this or even to contemplate them.
Scientists involved in the policy process are not making specific policy recommendations (with a few exceptions like Hansen). They are informing policy makers of the state of scientific research, and the relevance of this research for policy decisions. So scientists do not need to become politicians, but rather need to become more effective at informing policy makers of the relevant aspects of the science and its uncertainties. The alternative is for policy makers to make decisions based upon the representation of science in the media, or trying to sort out the primary scientific literature. Next wed, I will post a link to the briefing I am giving on Jul 10 as an example of trying to inform policy makers in a way that is relevant to the issues they are considering, without being prescriptive about policy. Last April I gave testimony to the House Special Committee on Energy and Global Warming; I am still getting questions from the Committee as they continue to deliberate the hurricane issue (clarifying: no there is no “consensus” on the hurricane/global warming issue, no, reducing CO2 emissions is not going to help the hurricane issue in the short term). So all this is not just about posturing by the scientists and the legislators, legislation does get passed on these issues and one can only hope that our best current assessment of the science is seriously considered. So am i becoming a politician? No. However I am actively engaging in the policy process in an effort to inform policy makers about the hurricane risks, and this involves becoming knowledgeable about the political process. This is not what all scientists should be doing or would want to do. But I personally think it is important.
Judith Curry, In a posting on the “IPCC AR4: No skill in scientific forecasting” thread, I’ve cited the joint statement made by 17 national scientific academies in “Science” on 18 May 2001 (p. 1261). In this statement, the science academies of Australia, Belgium, Brazil, Canada, the Caribbean, China, France, Germany, India, Indonesia, Ireland, Italy, Malaysia, New Zealand, Sweden, Turkey and the UK said, inter alia, that ‘The ratification of this protocol [Kyoto] represents a small but essential first step toward stabilizing atmospheric concentrations of greenhouse gases.’
As these academies were unquestionably making a specific policy recommendation (that governments should ratify the Kyoto Protocol), it is clear that the exceptions to the proposition that scientists do not make such recommendations are not confined to a few individual scientists. I agree that it is possible for scientists to confine themselves to informing policy makers of the relevant aspects of the science and that many do so. However, there are many others who are not only actively engaged in the policy process but who have effectively become politicians.
Judy-
Working with Democratic staff of the Special Committee in support of their political and policy agenda is overt political advocacy pure and simple — even your April testimony concluded by using the hurricane issue to advocate for actions on adaptation and energy policy:
There is nothing wrong with scientists acting as advocates, and of course one need not advocate specific legislation to be an advocate. You can for example be used (willingly or unwillingly) by lending your authority to one side or another of a political debate, as you have with the Special Committee, which is pursuing very specific legislation.
Problems occur when scientists pretend to be value-free or just presenting science. The hiding of political agendas behind science is an important factor in politicized science.
I discuss this in the case of hurricanes here:
Pielke, Jr., R.A., 2006. Self-Segregation of Scientists by Political Predispositions. Bridges, Vol. 11, September.
Click to access resource-2473-2006.11.pdf
And more generally here:
I commend your participation in the political process, but please don’t pretend to play the role of “speaking truth to power.”
re: #223 ” … not making specific policy recommendations (with a few exceptions like Hansen). ”
But Hansen has a weighting factor of like 100 compared to almost all other scientists and so it gives the appearance of being all about “speaking truth to power” and not about the science.
And more and more as time goes on, many ‘scientists’ in that community appear to be jumping on the PR bandwagon and letting the fundamental principles of the scientific method get trampled under their feet as they rush to get on the bandwagon.
This web site http://www.michaelellenbogen.com provides credible and original content in many different subjects under tips. The site has information that’s useful to the public and stands out for its content, you will definitely revisit to see new tips. I have a copy of the book, which I highly recommend ‘€” “The Insider’s Guide to Saving Money”. The book not only tells you how to save money, but many of the suggestions help in the fight of Global Warming.
I have little to add to what RPJ reveals about the effects of scientific predispositions on policy testimonies. I think his admonition to “get and stay real” with this issue and others discussed here has been helpful in seeing more realistically the bigger picture.
It is my opinion, that the confidence that most climate scientists see in predicting the adverse conditions resulting from potential AGW effects has much to do with their POV on the effectiveness of government mitigation. I think most statistically aware participants in these discussions are aware that the placing of confidence levels and likelihoods on these consequences is much a matter of subjective disposition and therefore the risk of being wrong comes down to, in the end, much a matter of the judgment one has of the effects of mitigation.
I think that Judith has it backwards when she claims that scientists must learn to relate to politicians on their terms. The most refreshing hearings I have had the pleasure to witness have been those where the witness set the mood and agenda of the debate in forceful and straight forward manner and as a result minimized the showboating by the politicians. By the way, a true test of an unbiased testimony, for me anyway, and one that I all too seldom see is when an apparent political supporter of a POV attempts to put words and conclusions into their witnesses’ mouths and the witness forcefully points out that there are countervailing POVs that could also prove correct.
RE: #211 – I sent them a nasty letter wherein I essentially accused them of cooking the books vis a vis the named stormed count by striving to “make tropicial” – i.e. the two “named” storms thus far both of which had their ultimate upstream origins as Aleutian Lows (!).
2 Trackbacks
[…] A lively discussion of the book is going on over at the website Climate Audit, which tends to be critical of much […]
[…] natural cause. Such a perspective supports the views of Joe D’Aleo (see); Bill Gray (see); and Roy Spencer (see). [also see]. [Added June 2 2009: Joe D’Aleo alerted me to another paper […]