2011-06-30 09:25:27Maue 2011: Recent historically low global tropical cyclone activity
Rob Honeycutt


Since studentnigel brought up this paper in the 2010-2011: Earth's most extreme weather since 1816? comments section, I've been looking into it.

Maue is comparing a standard measure of tropical accumulated cyclone energy (ACE) from 1972 to the present.  One thing that catches my eye is that ACE is, according to others, an inaccurate measure of the destructive force of storms.  The figure is based merely on a measure of surface wind speed.  According to another paper, Powell and Reinholt 2009: New Scales for the Destructive Potential of Tropical Cyclones, a more accurate measure would be Integrated Kinetic Energy (IKE) as a method for better expressing the actual energy potential within tropical cyclones.

This Dr Maue caught my attention especially because, by the looks of his web page he's been featured on WUWT a couple of times and he even has a quote about himself from Rush Limbaugh.  

It looks to me, even though Maue 2011 is apparently a peer reviewed paper, it is severely underestimating the actual energy contained within tropical storms, thus not an appropriate measure relative to climate change.  According to Dr Masters reposted article on SkS the number of tropical cyclones is falling, as anticipated by global warming theory, and the intensity of storms is rising, also consistent with global warming.

2011-07-01 04:59:52
Julian Brimelow


Thanks for highlighting this. Some thoughts on the Maue paper-- and I am trying to be objective here. 

I note some interesting things, and some interesting omissions in the paper.  For example, Maue et al state that:

"There is no significant linear trend in the frequency of global TCs (of at least tropical storm strength), which in
agreement with previous analysis [e.g. Webster et al., 2005; Wang et al., 2010]."

Really?  Webster et al. (2005, Science) analyzed satellite data from the past 35 years and found a “large increase in the number and proportion of hurricanes reaching category 4 and 5. The largest increase occurred in the N. Pacific, Indian and Southwest Pacific Oceans.”  So Maue is, IMHO, not accurately presenting Webster et al's findings.

Wang et al. (2010) did conclude that "the global total number of storm days shows no trend and only an unexpected large amplitude fluctuation driven by El Niño-Southern Oscillation and PDO. The rising temperature of about 0.5°C in the tropics so far has not yet affected the global tropical storm days."

But that metric says nothing about their intensity or duration of TCs.  Further, if Emanuel and Landsea are correct, we should see a decline in the numbe rof tropical storm days.

Maue et al. also make no mention of Emanuel's (2005) finding that the PDI (Power disspitaion Index) has nearly doubled in the past 30 years.

Maue also ignore the findings by Hoyos et al. (2006), namely:

"We used a methodology based on information theory, isolating the trend from the shorter-term natural modes of variability. The results show that the trend of increasing numbers of category 4 and 5 hurricanes for the period 1970–2004 is directly linked to the trend in sea-surface temperature; other aspects of the tropical environment, although they influence shorter-term variations in hurricane intensity, do not contribute substantially to the observed global trend."

Interestingly co-authors on that paper were Webster and Curry (yes Judith Curry)-- so we have 'skeptics' at odds with each other here.

They do cite Emanuel (2005), but do not note that  the PDI (Power Dissipation Index) is a better indicator of tropical cyclone threat than storm frequency or intensity, or that  that the PDI has nearly doubled in the past 30 years.

I'm not at all surprised that this pape ris doing the rouond sin the denialosphere.  I am surprised that Maue appears ot be going the way of Bill gray and shacking up with the WUWT crowd.  IMHO, no scientist of repute would align themselves so closely and so frequently with a group like WUWT.  So I find Ryan's behaviour strange.  He also seems to be falling into a trap that I have notices some scientists make-- their discipline/research focus  is say hurricanes, but the data do not appear to be consistent with the projections in the IPCC reports, and somehow that then calls into question all the science surrounding AGW.

Maue has really latched onto the ACE, but doing so may be unwise as Powell and Reinholt suggest (note too that Maue et al. do not cite that paper or discuss the limitations of ACE), but perhaps those particular data are telling him what he wants to hear?  What I would like to know is how woudl the ACE index respond in a world with fewer, but more intense hurricanes?  Has anyone looked into that?

I suspect that there will be some fallout over this  what appear to me to be a rather poor paper, in particular from Landsea and Emanuel.  In the meantime, it is sadly serving very nicely as fodder for the "skeptics" and those in denial about AGW.

2011-07-01 05:06:03
Julian Brimelow

I'd appreciate people's thoughts on my post.  Thanks :)

2011-07-01 05:13:08
Dana Nuccitelli

I don't know enough about hurricanes or ACE to comment on the content of the paper.  However, I would love to see a response to it.

For those unaware of his history, Maue has been posting on WUWT for years now.  He first drew 'skeptic' attention when he was in grad school at Florida State for his ACE work.  Deniers loved the headline that ACE was at its lowest levels in the past few decades, claiming it disproved the prediction that hurricanes will grow stronger as a result of global warming.

Virtually anytime a denier talks about hurricanes, he references Maue (who is now a postdoc, I believe).  So it would be very valuable to get a good critique of his paper here, from somebody who's familiar with the subject.  As I said, Maue posts on WUWT pretty frequently.  He's basically the hurricane equivalent of Roy Spencer.

2011-07-01 05:19:25
Julian Brimelow

Thanks Dana.  Yup, I'm afraid his work is rather tainted b/c of his chosen affiliations.  But it is worth countering.  ThingsBreak has an excellent summary of the science of TCs and how TCs might be affected by AGW. Maybe he would post here?

2011-07-01 05:26:25
Rob Honeycutt


Alby...  Thanks for all that.  I'm going to try to write up a post on this one.  I shot off an email to Mark Powell and he said he thought it was appropriate to question the limitations of ACE to measure total energy essentially because it relies on a single and often subjective measure.

I get the sense that Maue is really out there.  If you go to WUWT and read his responses to comments he is very much of the WUWT ilk.  It looks like this is Maue's second published paper.  He's a young guy.  What a way to ruin a career before it even gets started.

2011-07-01 05:29:24
Rob Honeycutt


I'm also fine if someone else with more background on this wants to take it on.  I'll do my best to tackle it but it may take a little time.  Tomorrow I'm off to China for 2 weeks with my family.

2011-07-01 05:49:57


Whoever is going to handle this should consult Kerry Emanual's work or Emanual himself on how he would judge the use ACE for comaprisons.  He compares storm strength with Power Dissipation Index and Storm Maximum Power Dissipation Index.  These correlate well with SST in the Atlantic.  This is AFAIK the best data we have and it is enough to give he, and other experts, confidence in the models going forward.

2011-07-01 06:08:04


There is something odd here as Emanual says of PDI in 2005 Nature:

This index is similar to the ‘accumulatedcyclone energy’ (ACE) index19, deļ¬ned as the sum of the squares ofthe maximum wind speed over the period containing hurricaneforce winds.

So why the different conclusions on trends?



2011-07-01 06:57:02
Rob Honeycutt


grypo...  There is this comment from Emanuel in that Nature paper that suggests the relationship between wind speed, which ACE is based on, and the diameter of a cyclone has little, if any, relationship:

"The quantity PD has the units of energy and reflects the total power dissipated by a storm over its life. Unfortunately, the area integral in equation (1) is difficult to evaluate using historical data sets, which seldom report storm dimensions. On the other hand, detailed studies show that radial profiles of wind speed are generally geometrically similar16 whereas the peak wind speeds exhibit little if any correlation with measures of storm dimensions17."

ACE, as I understand it, is solely based on peak sustained wind speed during a 6 hour period.  That could introduce some sizable errors if the sheer dimensions of storms are much larger.

2011-07-01 07:14:02
Julian Brimelow

No worries Rob, enjoy your trip!

This is not my forte either, but I know enought o smell a rat, and this paper has just enough dog-whistling to raise flags for me.  The exclusions/omissions that I highlighted are also worrisome, the reviewers dropped the ball there.

I do not think there is a perfect index, but my gut is that PDI is better, although I think the way Emanuel uses it does not take storm size (i.e., wind radii) into account.

Whoever writes this up should not frame it as an attempt to dismiss/ignore the findings, but rather use it as an example of how people can mislead themselves and others, becasue that is what I think what Maue is doing here (if he is awre of that is another question, but him posting at WUWT does not work in his favour).

I agree with Grypo, it would be beneficial to ave the input of Emanuel on this.  Anyone have his ear?

Grypo asked "So why the different conclusions on trends?"  a very good question that needs to be answeredIMHO.  And it is not only the ACE and PDI that have different trends-- see the findings by Webster et al. and Hoyos et al..

2011-07-01 07:16:25
Rob Honeycutt


I'm not sure I have this right but it looks to me like the power dissipation index (PDI) uses Vˆ3 whereas accumulated cyclone energy (ACE) uses Vˆ2.  And I think PDI also takes radius into consideration where ACE does not.  

My sense here is that Maue is trying to hide half the deck of cards.  Whew!  Lot's of great metaphors built into that one!

2011-07-01 07:29:32
Rob Honeycutt


Here is another paper, Villarini 2011, that states the main difference between PDI and ACE is that PDI is cubed and ACE is squared.

2011-07-01 07:44:18
Rob Honeycutt


Another interesting link.

So, both PDI and ACE do not include the size of a storm.  This seems to be the upshot of the IKE method.  What I'm also gathering is the IKE method is much more complex to calculate but "a group at CSU (John Knaff)" (per Mark Powell) is doing work to address the challenges.

2011-07-01 07:52:29
Julian Brimelow


@7;29 am, that is correct. That is the main diff. as I understand it.  Thanks for the other links Rob.  IIRC, Emanuel designed the pdi to include size/area, but to simply matters reduced it to v^3 accumulated/integrated over time.

2011-07-01 08:22:02


I believe his conclusions are a little like Pat Micheals sea level paper from a few months ago.  It is well known that PDI and  ACE are correlated with SST temps.  Both of these have fallen since the 2005 el nino, so yeah, the "lull" is perfectly in sync with prediction and Emauel's updated charts.  The divergence just seems to the wording surrounding oscillations.  I have no idea where anyone gets the idea that either ACE or PDI are not in an upward trend.  I think Maue is attempting to say this is within natural variation, where other scientists are seeing variation outside the noise.  The important conclusion that never comes out of Maue is that the warmer the SST, the higher PDI or ACE.  It doesn't matter what the cause, El Nino, PDO, or greenhouse forcing.  Instead he says:

While many overall global and basin ACE and frequency record lows have been set, the glaring anomaly in 2010 was  the extremely active NATL basin, which contributed more than 30% of the annual global ACE output. In the context of overall NH ACE variability, even with the doubling of the NATL to NH ACE proportion since 1995, the ENP+NATL proportion has remained  constant since 1979 with the WNP coincidentally reacting to the observed fluctuations of ENSO and PDO [Chan, 2008].  

Kerry Emanual's Page

2006 paper

A measure of total power dissipation by tropical cyclones (the power dissipation index, or PDI) has been shown by Emanuel [2005] to be well correlated with MDR ASO SST over the past half century, during which tropical cyclone wind measurements are most reliable. Although wind estimates prior to the 1940s are problematic, detection of the existence of tropical cyclones is less so, because without aircraft and satellites to warn them off, ships often encountered storms at sea, at least peripherally. A reasonably reliable record of annual North Atlantic tropical cyclone counts is thus available back into the late nineteenth century [Jarvinen et al., 2005]. This record, like the PDI index, shows a strong, long-term relationship with tropical Atlantic ASO SST (Figure 2). The linear correlation between the decadally smoothed series, r = 0.73 (p < 0.001 for decadally smoothed data, and a onetailed hypothesis test), indicates that the overall trend and more than half of the total decadal variance in annual tropical cyclone counts can be resolved by SST variations (r = 0.61; p < 0.001 is obtained if the bivariate statistical model for T(t) is used in place of T(t) itself). R(t), which must include any AMO contribution, explains an insignificant four percent (Figure 2) of the decadal tropical cyclone variance (r = 0.20, p > 0.1 for a one-sided test). In other words, the SST variability underlying increased Atlantic tropical cyclone activity appears unrelated to the AMO.

 It might be argued that other factors potentially associated with the AMO (e.g. changes in vertical wind shear in the tropical North Atlantic) could be responsible for the observed tropical cyclone changes [e.g., Goldenberg et al., 2001]. This possibility was rejected after examining the residual time series that results from removing the statistical fit of the bivariate model for T(t) from the annual tropical cyclone series. The residual shows no evidence of a multidecadal spectral peak (see additional material). Thus, it can be inferred that any factors unrelated to SST that might influence tropical cyclone activity also do not exhibit any detectable multidecadal cyclicity

I find it hard to believe that reviewers wouldn't resolve this issue before publication in GRL (unless I'm missisng something., it apperars these conclusions on oscillations and SST warming signal contradict each other).

2011-07-01 22:43:59


So after looking into the recent research, here is the latest. From Knutson's page:

  • It is premature to conclude that human activity--and particularly greenhouse warming--has already had a detectable impact on Atlantic hurricane activity. However, human activity may have already caused substantial changes that are either below the 'detection threshhold' or are not properly modeled yet (e.g., aerosol effects).
  • Anthropogenic warming over the next century will likely cause hurricanes globally to be more intense (by a few percent on average) and have substantially higher rainfall rates than present-day hurricanes.
  • Anthropogenic warming over the next century more likely than not will lead to substantial fractional increases in the numbers of very intense hurricanes in some basins, despite a likely decrease (or little change) in the global numbers of tropical storms.

Assesment report from last year Emanual, Knutson, Held, et al.

These improvements have encouraged us to raise our confidence levels concerning several aspects of cyclone-activity projections. These include our assessment that tropical cyclone frequency is likely to either decrease or remain essentially the same. Despite this lack of an increase in total storm count, we project that a future increase in the globally averaged frequency of the strongest tropical cyclones is more likely than not — a higher confidence level than possible at our previous assessment6.

Importantly, although some statistical methods project very large increases of about 300% by the late twenty-first century in aggregate Atlantic hurricane activity (power dissipation), such dramatic projected increases are not supported by existing downscaling models or by alternative statistical methods9. Moreover, despite some suggestive observational studies, we cannot at this time conclusively identify anthropogenic signals in past tropical cyclone data. A substantial human influence on future tropical cyclone activity cannot be ruled out, however, and could arise from several mechanisms (including oceanic warming, sea-level rise and circulation changes). In the absence of a detectable change, we are dependent on a combination of observational, theoretical and modelling studies to assess future climate changes in tropical cyclone activity. These studies are growing progressively more credible, but still have many limitations, as discussed in this review

The problem, as I see it, with Maue's paper is the silly focus on the last few years of ACE number since the 2005-06 El Nino.  It's good that he's investigating different basins, but I'm unsure what his paper has new to offer.  I'm wondering whether the paper online is actually the final version after reviews.  The takeaway for me is that the new paper tells us nothing new that the assesment team from 2005 isn't already aware of.  Although I'd like to see an assesment from an expert on the paper before writing anything specific that I don't have the knowledge to defend.

2011-07-02 04:11:35
Rob Honeycutt


Reading through these papers I'm getting the sense there is still a lot of uncertainty involved is measuring cyclonic energy, which is echoed in grypo's post above.  

I think Maue was essentially sneaking his paper in as a kind of Trojan horse.  He doesn't really offer anything that interesting in the paper.  But what he does manage to do is get his charts of ACE data into the peer reviewed literature, and that's all the WUWT crowd wants.  He's throwing a little red meat to the denier set to prop up his own cred with his crowd.  (Why does Curry not hound the WUWT crowd about their tribal antics???)

It seems that both ACE and PDI are, at best, just general estimations of cyclonic energy.  Dr Powell's work on IKE would probably generate more accurate estimations but it sounds like that that research is very new.  I can't figure out how to pull any data out of the tool they have set up on their website.

It would be a very interesting research project for someone to try to do essentially what Maue has done with ACE but using IKE data.  It may be unlikely there is enough data there yet to do it, though.

2011-10-08 08:09:10Honeycutt, Albatross
Peter Miesler

I enjoy the privilege of being a member of this board and you folks tolerating me.
I peek in often and have a few times used information from these posts in my own blogging efforts.
I try to be respectful of giving credit and links to sources, but now and then it seems to be better no to go there.  Also, I try not to trumpet my being permitted in this chat room.

The reason I’m bringing this up is I just finished posting something where I grabbed text from a couple posts above and just wanted to let you know.  If I’ve over stepped bounds or if you have some etiquette suggestions I’m all ears, I want to keep in good standing with you folks.