October 2014

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31  
Blog powered by Typepad

« The Hayekian Case for Legalizing Same-Sex Marriage | Main | On Nation Building »

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d83451eb0069e2014e88a58903970d

Listed below are links to weblogs that reference Why The Severe Weather?:

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

These are just butterfly effects. There's no reason to believe the underlying distribution has changed. Weather is crazy.

I realize there has been more going on than tornadoes, but I thought I would share that there is nothing abnormal going in in that category of destruction.

http://www.ncdc.noaa.gov/img/climate/research/tornado/tornadotrend.jpg

http://www.ncdc.noaa.gov/oa/climate/severeweather/tornadoes.html

From the latter link:

"One of the main difficulties with tornado records is that a tornado, or evidence of a tornado must have been observed. Unlike rainfall or temperature, which may be measured by a fixed instrument, tornadoes are ephemeral and very unpredictable. If a tornado occurs in a place with few or no people, it is not likely to be documented. Unfortunately, much of what we know as tornado alley was very sparsely populated until the 20th century, and so it is possible that many significant tornadoes may never have made it into the historical record."


Why the severe weather? Didn't you hear? The world's coming to an end on Oct. 21st.

But seriously, as happyjuggler show in the NOAA link regarding tornados, the weather has not gotten severer. But, as you say, there are more forms of communication. There is more technology available to the ordinary person to capture images of tornados and severe weather. Also, as the population in the US increases, there is more opportunity for people to be killed and property to be damaged.

And I think the news media has an incentive to scare people into thinking things are worse than ever. You may have noticed that bad news sells extremely well (just look at you local news with the leading stories of murders, fires, and robbery). And I wonder if people have a psychological need to think they are living in extraordinary times, where maybe not entirely believing that the world is coming to an end next week, but something is very wrong with the planet and it is at least getting worse and worse off (e.g. change that light bulb and YOU can save the planet).


Here is Richard Whately (1831) on change in media-coverage misleading us to think that things are changing:

"It should be observed also, that in large towns, and in populous districts intersected by roads which furnish a rapid conveyance of intelligence from place to place, and where newspapers are in common use, much more in proportion is known of every enormity that is perpetrated, than in remote country-districts, thinly peopled, where there is less facility of mutual communication, and where the natural appetite for news is compelled to limit itself to the gossip of the nearest hamlet. Much apparent increase of crime (I will not undertake to say how much) consists, I am convinced, in the increase of newspapers. For crimes, especially (be it observed) such as are the most remote from the experience of each individual, and therefore strike him as something strange, always furnish interesting articles of intelligence. I have no doubt that a single murder in Great Britain has often furnished matter for discourse to more than twenty times as many persons as any twenty such murders would in Turkey. We should remember, that there are not more particles of dust in the sunbeam than in any other part of the room; though we see them more where the light is stronger."

That is from Introductory Lecturers on Poltical Economy, which contains many remarkable passages. Available at the Online Library of Liberty.

Months ago I did a research on the casualties of natural disasters. I read a paper which analyzed data on the number of "disasters", defined as events which killed at least ten people, interested one hundred, and were either officially proclaimed as disasters or compelled government officials to call for international aid.

There was a steady increase in the number of disasters in the last decades.

However, when the authors considered the first quartile, the first ventile and the first hundredile (???) of the distribution, there was no trend at all.

There are no more large disasters nowadays than in the past, but reporting of small disasters have improved throughout the globe, giving the impression that the occurrence of disasters have increased over time.

Nice topic! If you were to pick this up as a research project, I think you'd also need to consider the statistical properties of natural time series. The great Nilologist H. E. Hurst ("Abu Nil") showed that many (most?) natural time series exhibit long memory. This means that they have long waves of ups and downs. (Autocorrelations die slowly in non-ergodic time series.) So it might well be that we're getting, say, more tornados in North America even though there is no particular observable cause for the trend. Under that hypothesis, the uptrend might continue for a century or suddenly reverse for no particular reason. Such issues complement the epistemic points made by Dan, Pietro, and others.

(BTW: I've never seen this logic applied to global warming. It seems an obvious point, so I suppose it's just that I don't follow the scientific debates in that area.)

cafehayek has some data on the subject, mostly deaths. One would expect deaths to increase if weather had not increased in severity just because of the population increase.

It would be interesting to see population waited averages of deaths.

Roger, I have seen a few people apply that logic to AGW, but they tend to get shouted down. If you do a frequency analysis of sun spots, one of the dominant frequencies is 250 years. That correlates well with the medieval warming period and the little ice age.

M: Hm! do you have a cite on the nonergodicity point??

Wow. That's a great topic for economists.

To address the question of whether or not weather patterns have changed in recent times, we first need to explore the question about whether the rules of weather have changed.

Yes, the rules have changed in many ways.

Human beings engage in agriculture, build cities, dams, build roads, plus add aerosols and greenhouse gases to the atmosphere. All of these things perturb sensible and latent fluxes of heat and moisture. These perturbations teleconnect through the system, and change weather patterns.

Therefore, human beings have certainly while unintentionally changed the rules of weather, and therefore have changed the patterns of weather.

But... the frequency of severe tornadoes in tornado alley seems to better correlate with ENSO (El Nino-Southern Oscillation). We are currently in a La Nina cycle with seems to have some correlation to increased outbreaks of very severe supercell thunderstorms across the subject region.

Do these concepts sound strangely familiar to economists?

Is there someone commenting here who formerly went by the name "fundamentalist"?

I thought so. On the previous thread, I felt a disturbance in the force that I have not sensed in a long time.

At least have the courtesy to keep the same name.

All those tornadoes are due to there being so many blowhards on this site, myself included, :-).

I'm surprised that no one has mentioned the fact that weather created disasters give politicians great opportunities for photo ops casting themselves as the saviors. FEMA is an ever expanding and expensive tool in the hands of the ever more pragmatic politicians of DC; "never let a crisis go to waste" is their mantra. And let us not forget about moral hazard. John Stossel had a great show about his beach property that the government rebuilt a few times after is was washed away. Even if the force winds of storms are no more damaging to property than in the past, they seem more interesting to me now (and maybe are more newsworthy generally) because as a taxpayer I'm more and more on the hook as a result of political self interest driving ever more expansive and expensive disaster relief.

Roger, not off hand. It's been a while since I read the stuff.

I think one will find that there are natural cycles to these things, as many have already pointed out. Global temperature correlates with sun spot activity, storms and temperature variations coincide with El Nino/La Nina events, etc. It should be easy to find the research out there covering all of this.

From an economist's point of view, what may be of more interest is effected populations. Are more people affected by hurricanes now than before because of people living along our coasts than have in the past, due to government insurance? Are more people affected by floods now because more people are living in flood plains because of government flood insurance? Are more people living in the more severe-weather-prone areas, like the deep South due to the presence of air conditioning, and the subsequent emergence of more jobs? What about the percentages of people affected? Has that changed?

Troy, good points! And to follow that line, how has government flood control projects had unintentional consequences?

The Mississippi is going to flood and there ain't nothing we can do about it. So when the state builds levies to protect some people, all they really accomplish is to move the flooding to areas that hadn't flooded before, such as tributaries.

In geophysics, "rules" are also known as boundary values. Boundary values govern the patterns of emergence in weather. Changes in boundary values do not diminish weather and its sensitivity to initial conditions, but these changes do move the statistics of weather or weather patterns otherwise known in the aggregate as climate.

In economics, rules likewise govern the patterns of emergence in complex social phenomena.

The study of rules in social orders and rules in climate are very much related, and to have full context for contemporary discussions about human-made climate change, I would argue that one needs a good appreciation for the closely inter-related nature of emergence in complex social orders, and emergence in climate. They are very much two aspects of a particular field of study which in this subject, may be tightly coupled.

I am curious to get feedback as to whether others think I'm on the right track in the above statement.

I suspect that this aspect of the climate issue has been missed by many people, but I'm not sure if it's me who has simply missed hearing others spell out this issue in these terms. The science is *very solid* that the biproduct of human society has changed the climate of the atmosphere in diverse ways, but I think the aspect that has been missed is that emergent human social order and emergence in climate are part of the same field of study, namely rules and the evolution of complex patterns.

IMHO, this recognition may lead us to a completely different and more effective way to analyze the climate issue from a classical liberal viewpoint.

There are similarities, such as the presence of strange attractors, and the emergence of order (hurricanes and tornadoes), but climate is a chaotic system while the economy is a critical system.

Troy,

Thanks for replying. However, you are missing something important.

Climate (a weather pattern defined by some subjective metric) is the statistical aggregation of all the weather systems over a period of time, but weather *patterns* (not distinct weather systems like cyclones) are forced by changes in boundary values (rules).

Therefore the order that emerges is not chaotic in the same way that weather is. There appears to be some statistical stability.

Yes, the climate system is very non-linear and therefore presents a daunting challenge in the prediction of it's future evolution, but climate is not necessarily chaotic in the popular conception of the word.

Weather systems of any kind show a type of emergence that is governed by *intitial conditions*, but this is a somewhat different problem than a mixed initial/boundary value problem that is delt with in the study of climate.

It is almost certainly the case that the evolving social order which has been governed by "rules of behavior" have contributed to changes in the weather patterns. There is a mountain of scientific research which supports the conclusion of human interference in the weather patterns, and I can provide reems of journal articles if someone doubts this assertion.

So this leads me to the conclusion that the rules of social patterns and the rules governing weather patterns feedback on one another in very curious ways. In some way we might make the jump in saying that the rules of social order definately effect the evolution of the climate system, but these rules can never be fully planned. Since the evolution of the social order is unpredictable no matter how much design is made to the rule structure, so will also be the evolution of the climate system. This may be a firm conclusion no matter how much evidence collects that humans effect the climate system.

In the policy realm, what is being proposed by policy makers regarding climate change is a human- designed set of rule changes governing the social order which is ultimately expected to return the climate to some "ideal" egalitarian norm.

Now the lights should be going off to all good economists!

"Since the evolution of the social order is unpredictable"

Perhaps I should not have made this strong of leap. Some aspects of social order may indeed be predictable, especially over short time periods, by prescribed changes in rules. However the full evolution of the system cannot be predicted no matter how much a conscious and articulated set of rules are prescribed.

Tom: "And I wonder if people have a psychological need to think they are living in extraordinary times, where maybe not entirely believing that the world is coming to an end next week..."

HL Menchen wrote that politicians spend most of their time trying to frighten people because they know that few people pay attention to them except in a crisis. They'll invent a crisis is one isn't at hand.

I think that philosophy extends to media as well, though Menchen might not admit it. Nothing sells papers, or advertising, like crises. CNN lives by them.

Nothing draws donations to non-profits like crises. And nothing pries loose federal funding for research like the potential end of the world.

So is it any wonder that everyone in need of other people's money sees nothing but crises everywhere?

When I say "chaotic" I mean the chaos of chaos theory, not mere randomness. That is what it meant by a "chaotic system". In such a system, order can emerge periodically, as I pointed out. But they are ephemeral. No weather pattern lasts more than a week or two (in the case of the most organized of weather patterns -- hurricanes); cities, on the other hand, can last centuries or even millenia. In the latter case, that is evidence of a long-running economic pattern, to say the least. Firms can last weeks, years, decades, or even a century or more. This is evidence of being a critical system rather than a chaotic system, where order emerges but "momentarily". There are similarities, as I said, but the differences are just as important. Both exhibit butterfly effects, but the consequences are often quite different. An economy is closer to a cell or an ecosystem than it is to weather.

Troy,

You still are not understanding the significance of what I am saying above.

The problem may be that I have tried to communicate a geophysical concept in economic terms, so that some not extremely familiar with climate might understand the significance of my thesis. I can see that possibly I need to define some terms more clearly so that we can better communicate.

I will have more to say when I have a few minutes.

I do understand it. I am trying to help you understand why the metaphor is not quite as apt as you think. I have been studying chaotic and complex systems for most of my academic life. There is a difference between complex, self-organizing critical systems and chaotic systems. The climate/weather is an example of the later. Cells, ecosystems, and economies are examples of the former. There are overlapping elements, such as strange attractors, but chaotic and critical systems act quite differently.

Chaotic dynamics are a sub-category of complex dynamics.

Troy, No, after carefully reading your responses, I don't think you understand it. Our definitions of chaos are identical, so this is not the problem.

let's try to clarify a few definitions so that the conversation can continue in a fruitful way. definitions; my hope is that others will take the time to read and chime in on this.

1) Weather-The state of dynamic fluid motion of the atmosphere and/or ocean at a particular time; weather is an initial values problem and is highly sensitive to intial conditions (therefore weather is chaotic). Weather includes short-lived eddies or vortices in tubulent fluid or gas flow such as tropical cyclones.

2)Weather pattern- a pattern of specific types of weather events occurring across a selected area or region measured over a selected time span ranging from very short to very long. Weather patterns can be very short(days or weeks), seasonal (lasting weeks or months)(an example would be a winter jet stream pattern that seemingly becomes stuck for large parts of a season), or long (lasting many years to decades). A long-term drought results from a prolonged multi-year or decadal weather pattern. Multi-decadal patterns of ocean oscillations such as the Pacific Decadal Oscillation may be an example of very long-lived ocean/land weather patterns. Weather patterns are generally considered initial value problems in climate prediction. Changes in weather patterns may be either random (short time scale) or externally forced (long time scale).

3) climate-the statistics of weather and/or weather patterns, usually measured over a relatively long period of time (usually >30 years). Climate, since it deals with a relatively viscous ocean system of long memory has elements of both initial and boundary value problems, but is mainly considered a *boundary values problem*. It does *not* appear that climate is chaotic as measured across multi-decadal periods of time. *Please read this closely*; The statistics of climate are generally well behaved. It is much less sensitive to variations intial conditions, AND THEREFORE IS NOT CONSIDERED BY MOST CLIMATE RESEARCHERS TO BE CHAOTIC.

4) Climate change: changing statistical functions of a given climate metric. Climate change over multi-decadal time periods is commonly thought to occur mainly through changes in external forcing (boundary values).

Before I go any further, I want to make sure we are clear on definitions. If you will check off on these, then we can proceed with my hypothesis.

One more clarification related to my previous comments:

Climate change can be thought of in terms of *long-term changes in weather patterns*. An example would be the change in the frequency of drought across a particular region. Or maybe the multi-decadal change in the diurnal temperature range.

Another example; when Pete asked whether storm frequency or intensity has changed over a period of 50 years, this is a climate problem, not a weather problem. The question boils down to whether there has been a shift in the long-term weather pattern. Following the recent literature on climate, these questions are usually addressed with experiments using general circulation models or couple atmosphere ocean models (AOGCMs), and perturbing intial and boundary conditions and watching the statistics of the model weather evolve.

"Weather patterns are generally considered initial value problems in climate prediction.'

SHOULD READ short term or seasonal weather patterns

I heard a guy from the national weather service today say that weather is not more severe today than in the past but that the population has grown and created more urban areas for bad weather to hit.

In other words, people are confusing the effects of increased population with increased severity of weather.

I understand all of this. But my point stands. Let me now clarify.

Climate affects weather. The resultant dynamics of climate is thus chaotic.

The economy affects human action. The resultant dynamics of the economy is not chaotic. Human action is critical -- on the edge of order and chaos.

More than that, does weather give rise to climate? As you observed, climate is affected by a number of "external" elements -- such as earth rotation, orbit, and tilt, ocean currents, variations in sun spot activity, etc. All of these contribute to climate, which in turn affects weather. Climate may be critical, but the dynamics it gives rise to is not.

However, this is not what happens with the economy. Humans, who are made up of cells with critical dynamics, and think using brains with critical dynamics, act in ways that give rise to processes with critical dynamics. These processes with critical dynamics in turn result in human actions with critical dynamics. The feedback between the emergent process and the underlying elements is an important aspect of the economy. We do not see this in the interaction between weather and climate, for example.

Thus, the dynamics of the two are different.

Dan Sutter has done a lot of research on that topic. His resume lists at least 20 published papers related to tornadoes, hurricanes, etc. He also has a paper on "What Drives Television Weather Coverage?" forthcoming in Environmental Hazards. So I would argue that he's the person to ask all these questions.

You have it wrong. Think about this more closely, and I think you will get it.

You say: "Climate affects weather". This is 180 degrees backwards. Climate by definition is a statistical property of weather and/or weather patterns over a long period. Changes in weather patterns shift the probability distribution functions for a particular climate metric, and this is climate. It is a logical muddle to say that climate changes weather. It does not make any sense by definition.

Over a long enough period, the ensemble mean of weather patterns will form a climate. As best we can tell, climate over multi-decadal periods is not chaotic (thank goodness for life on earth), and the long term statistics only move when acted on by an external forcing (a change in the rules driving weather patterns).

We must get this correct, or we are stalled out in this discussion.

Is it the fact that it snows in the poles that makes it cold, or does the fact that it's cold at the poles that make it snow? Is the fact that deserts occur in a band along the 30th parallels have to do with weather? No, it has to do with the sun hitting the equator and the rotation of the earth, both of which affect air flow. Which affects weather. There are places where weather does affect climate, rather than the other way around, as my examples above show. The American southeast is the exception to the 30th parallel rule because of the presence of the Gulf of Mexico, which pumps moisture into the SE. Mountains can also block rain and thus create weather-created deserts. But these seem to be the exceptions, not the rule.

The Amazonian rain forest pumps so much water into the air that it makes a significant contribution to the rain that then falls on it. It ends up being a feedback loop.

The spin of the earth affects climate. The presence, shapes, and sizes of oceans/seas and continents affect local climate and, thus, the weather. Consider the fact that the great lakes create lake-effect snow -- it is snow because of the climate, whereas it would be only rain if the lakes were further south. However, the lakes are where they are because they were created by glaciers, which means they were created by a climate change.

It is an error to say that climate is the statistical average of weather. It may be the common definition, but if it is, then that definition is in error, unless we are talking about a handful of local climates. From what I understand about meteorology, it is absolutely clear to me that it is climate which creates the particular weather one sees. To say otherwise is to say that the snow is what makes the poles cold, rather than the other way around.

So long!


For some reason, this weblog eats my comments. They post for several hours, then dissappear into the ether.

One more time; Troy, you do not understand what you are talking about. I am willing to teach you some basics about the subject, but you first must be willing to shut up and listen before spouting off.

The conversation will only progress if you accept that I am the teacher, and you are the student in the area of climate and geology. When it comes to Austrian economics, i am not the expert, but in this area i am.

I am a senior scientist employed by a giant international company who you would recognize. This is one of my particular research interests which I am paid quite a large sum of money to pursue.

I know a little more than you about this subject matter and would be happy to share, but I am also equally happy to go away if you want.

K Sralla,

I would agree with you rather than Troy regarding the relationship between climate and weather. However, you have made a rather large error: your claim that the climate system is not chaotic. It is perfectly possible for both weather and climate to be chaotic.

So, sure, it is true that climate has longer periods of staying within certain bounds. However, the 1963 model of Lorenz that demonstrated the possibility of chaos was one of general atmospheric circulation patterns that could be climatic rather than merely meteorological. Thus the sort of shift that could be involved could be one of jumping from a zonal to a meridional flow pattern as exemplified by jumping from one "wing" to the other of a Lorenz attractor.

Thus the late Reid Bryson, who served in Saipan in WW II with Lorenz, demonstrated that the time period during which basic temperature would change and the climate system would change during movements in and out of ice ages was very short by geological time standards, less than a century. Bryson used to argue that this could be explained by Lorenz's model. I can provide references, if you like, but much of this is based on personal conversations with Bryson as well as having studied and published on the Lorenz model, although I do not work for some big unnamed company. I am just an academic.

Yes, It's late and I became a little impatient with Troy.

When we discuss whether or not climate is chotic, many folks often get this question confused with the queston of whether components of the Earth System (dynamic coupled system containing the ocean, atmosphere, and solid earth) display chaotic behavior. The answer to this question is certainly yes. The Earth System is a highly non-linear dynamical system, and these types of systems involving fluids in turbulent flow are nearly always chaotic.

**But**, when it comes to the question of whether or not the *climate* of the atmosphere (measured over multi-decadal periods of time such as 100-200 years) is chotic, the answer is almost certainly NO.

Why are most climate scientists confident this is the case? It is because experiments performed with fully coupled general circulation models do not show such behavior. When these models are spun up and run into the future along these timescales, the unforced model climate (things like mean global surface temperature) is statistically stable. Climate does not randomly drift wildly from one state to the next. When those models are forced with added greenhouse gases and aerosols, the statistics of the resulting changes in weather patterns remain well-behaved.

This point is no longer controversial, so we can beat a dead horse on this, or we can move on to something else.

I'm tired, and will say more tommorrow.

BUT MASSACHUSETTS NEVER HAD TORNADOS THAT I KNOW OF,I GREW -UP IN NEW.HAMPSHIRE..NEW ENGLAND OCEAN..KNOWN FOR HURRICANES,NOT TORNADOS.

K.S.,

Oh, you are talking about simulations of particular large-scale models. Well, this does not convince. I do not know how well acquainted you are with the Lorenz attractor, but it stays within certain zones for very extended periods of time before flipping rather suddenly into quite different ones. If these large-scale models do not contain within them such formulations then they will not exhibit such dynamics. But, the geological record is indeed such that we have seen such fairly sudden shifts.

I will grant that this does not necesarily prove that the climatic system is chaotic, but that a bunch of simulations of some particular set of large-scale models that supposedly represent global climate does not show this also does not prove that the system is not chaotic.

Barkley, Nothing that I have stated is controversial among the broad community anymore. You can stick your head in the sand, or possibly you need to dive into the literature and confirm what i have stated, but you are showing that you are not familiar with the modern literature. As far as geological time, please re-read the careful qualifications that I made in my last comment, and you will see that I covered myself.

I am a geologist. What I have explained is where the modern science of climate and geology (two closely related fields over long enough periods of time) is located.

Yes Barkley, my level of familiarity with chaos theory is well, a tad bit more advanced than Gleick's book.

As is mine. By all means, tell me me where I went wrong in my explanation. I agree with you that climate is stable over long periods. There can of course be sudden shifts, but that is covered by critical systems, which are also bifurcating systems. Where I disagree is that climate should be understood as the statistical average of weather. Higher CO2 levels do not directly affect the weather. They affect climate, which affects the weather. This seems as obvious and clear as day to me as anything could be.

If I am wrong, explain why I am wrong. But don't give me that you are the expert, so I ought to just sit back and listen like a good little moron. I do know enough about this stuff to ask questions and make statements. That is how discussions leading to understanding are supposed to take place. If you either can't or won't address my specific claims and arguments, then that says more about you than me. Let's have a discussion, but don't be giving me this "How dare you question me!" routine.

Look up the definition of climate. Google is a good place to start.

Are we talking about global climate or local climates? Global climate affects weather, weather affects local climate. It is unfortunate we have one word for two concepts. I have been consistently talking about global climate. Your definition only makes sense if you are talking about local climates. Even so, there are climates, such as deserts along the 30th parallel, that are absolutely not explainable by weather. It is explainable by the way the sun hits the earth, the rotation of the earth, etc. The main exception to this rule is explained by the location of large gulf, which itself affects the weather. So which are you talking about? Global climate, which affects weather, or local climate, which is affected by weather and the global climate?

K. Sralla,

So, are you actually Bryan Sralla, the petroleum geologist who is president of Hewitt Mineral Corporation? If not, I would suggest you avoid trying to play one-upmanship games. That I conversed with top climatologists from nearly 40 years ago who are now dead does not mean that I am not doing so with the current batch and have my "head in the sand" and need to read the "modern literature."

Your claim about some consensus that global climate is not chaotic is simply bogus. There is no such consensus. There have been lots of studies of more local climates (a point raised by Troy), with competing results. Just to note two cases, Chang et al from Physica D in 96 found the El Nino system to be chaotic, while Shukla in Science in 98 found tropical rain patterns not to be so, even as Shukla declares that "It is generally accepted that the atmosphere is a chaotic system." He claims his results to hold "for certain zones." Hmmm.

What is certainly true is that the models used by the IPCC to forecast climate change over the next century do not generate chaotic dynamics, a point I already noted, if less precisely, but that is not proof that either the system is not chaotic or that there is a consensus among climatologists that it is not chaotic. Case is wide open, frankly, and you should be a bit more careful about making ex cathedra pronouncements as if you are some great authority (unless you are Bryan Salla, and even then, you should be a bit more careful about what you say).

Barkley,

I have sent you my e-mail contact, and will be very pleased to continue this discussion offline.

I loath to do this since I agree with these guys http://www.realclimate.org/index.php/archives/2005/11/chaos-and-climate/ on very little (especially on policy), but on this issue, they are on the side of most climate scientists (especially ones under about age 60).

Several years ago I thought they were wrong on the issue of chaotic climate. After several years of study on this, I now concede that they are likely to be at least partially correct.

There is some excellent back and forth between the authors and Roger Pielke Sr. (who I respect a great deal, and is right on many things), but on this particular aspect of global climate metrics, he is not on the winning side.

My suspision is that even Roger has softened on this since 2005, and now sees certain global climate metrics (such as ocean heat content changes as measured over multi-decadal periods) as being more stable than he previously suspected. I could be wrong on this, so I hope I am not misrepresenting his view.

This will be my last comment on this issue, and I will leave the others with the last word.

That's what I said. Weather is chaotic, climate is critical.

K. Sralla, you have never addressed any of my actual claims, only vaguely accused me of not knowing what I'm talking about. That's not an argument.

In any case, whether I am right or you are, and average weather = climate (where's the sun, oceans, dust particles, earth orbit, etc. in weather? but I digress . . . ), then that means chaotic weather dynamics give rise to critical climate dynamics. As I pointed out, with the economy, critical cells give rise to critical brains which give rise to critical social interactions, including, in humans, the critical economy. The underlying dynamics is different in the economy than it is in the climate, and thus the climate is not an apt model for the economy. An ecosystem, on the other hand, is.

My last comment on this will be to agree that as one goes to a higher level, lower level chaotic dynamics can get washed out due to the law of large numbers. I happen to have sympathy with the position of Pielke, Sr., although the matter of whether or not the global climate system is chaotic is not readily determinable. I agree with Sralla that for the practical near term forecasting purposes related to decisions made by either private or public authorities, they should assume reasonable bounds of climatic fluctuation for the near future (open to debate, needless to say), although, of course, chaos does imply bounds so that within those bounds there may still be chaotic dynamics, the question of "fat tails" being another matter, (see my in press book).

Beyond that point I would assert that the larger question of the global dynamics here remain up in the air, although we may be dealing with time horizons reduced to near zero present values by our conventional methods for such evaluations, with the historical record suggesting the possibility for very serious shifts of fluctuation patterns from geological time to geological time, as has appeared to have happened, although without us being able to determine whether or not those catastrophic shifts were due to exogenous or endogenous causes.

Well, I said that was my last post on this, but I think it might be worth clarifying something further here, although this has been said above, if in scattered form.

Whether or not the global climate is chaotic or not id not important for longer range projections, which have their problems no matter what. Chaotic dynamics are locally unstable, but are globally bounded, with Sralla having emphasized above the importance of what he called "boundary values." This latter point is not always known by many, but is crucial here. So, in effect what a future projection involves is projecting forward those likely boundaries, along with the implied average, which is what usually gets discussed in public fora. This can be done, even if within those bounds there may be sensitive dependence on initial conditions and local instability, which is why indeed there are such serious limits to short=run weather forecasting.

Regarding the issue initially posed by Pete in his post, the matter of chaos is also essentially irrelevant to the issue of are we seeing more extreme weather and if so why or why not? One can see an increase in those bounds or variance or fat tails with or without there being chaotic dynamics per se. As it is, it is my understanding that it is very much up in the air whether or not the apparent longer term trends are consistent with such increases in extreme outbreaks, although they clearly could be.

The comments to this entry are closed.