Transterrestrial Musings

Comments

Hansen has always been pretty passionate and bold about his stuff. He is probably fed up with the latest "scandal" that was spinned from the US temperature data corrections and all the lies spread. Stuff like what you Rand wrote, mixing up world and US temperature.

"Looks like 1998 wasn't the hotterst year, after all".
and
"Guess the trend wasn't as trendy as they thought."

Even in 2001, when the data was adjusted the last time, it was noted that practically 1934 and 1998 were a tie. The difference was so small that the error bars would make it pretty insignificant.

There are always things like measurement time differences, urban heat island effects etc that change the data and have to be taken into account, and still of course can not be taken into account completely. We live in an imperfect world. This doesn't say measurements are impossible. We just have to have error bars and look at trends etc.

It has to be realized that most of AGW criticism is entirely nonserious. The error finding in itself was a serious thing but the blog reporting was again the usual crap, claiming huge differences and often omitting the fact that it was only US temperature and not the world, just saying "warmest year on record".

Posted by mz at August 18, 2007 08:24 AM

No, the scandal is actually that he didn't really conduct any science at all by not sharing the raw data and methods used.

As far as I know he still isn't. That means he's not fit to work as a scientist and has proven it beyond any doubt.

Further it also shows that whoever used his "work" in their own science simply accepted his conclusions on face value.

The reply makes it fairly obvious that he knows this and is trying to hide it by invoking conspirational and political language.

Anyone prepared to lend him credence at this point is the definition of a sucker.

Posted by Habitat Hermit at August 18, 2007 09:37 AM

No, the scandal is actually that he didn't really conduct any science at all by not sharing the raw data and methods used.

Yep, if you cannot recreate the experiment from the methods and data provided, (his refusal to share his method), then it is not science.

Posted by Dennis Wingo at August 18, 2007 10:12 AM

If one calls one's critics clowns (or jesters, as if that makes a difference) when they insist that you release the raw data and algorithms used to produce your conclusions - rather than simply producing the raw data and algorithms - means that one either has no understanding of the scientific method, or else that one knows that such a release would prove the conclusions false.

Since Hansen has the title "Dr." appended to his name, I assume that he has at least a passing familiarity with the scientific method.

Posted by Ed Minchau at August 18, 2007 10:13 AM

Hansen's main defense for non-disclosure is "The actual FORTRAN we use is too ugly to be exposed to the public."

Posted by Al at August 18, 2007 10:15 AM

mz "that practically 1934 and 1998 were a tie."

I would think that any sane person would wonder, if 1998's high was attributable to AGW, then why was 1934 so hot? And if you say 1934 was attributable to AGW also, what happened to 1935-1997? If AGW, assuming it exists (the "A" part at least), only causes 2-3 overly hot years out of every 60 plus years Algore and "Dr" Hansen are full of it.

Hansen is a kook, and as far as I am concerned anyone who can't see that is also a kook. Or at least as HH put it, a sucker.

Posted by Cecil Trotter at August 18, 2007 10:23 AM

Hansen's main defense for non-disclosure is "The actual FORTRAN we use is too ugly to be exposed to the public."

FORTRAN? That's soo 1970's.

I wonder if he has any understanding of numerical granularity in a FPU.

In this day and age, no one should still be using such an archaic computer language.

Posted by Dennis Wingo at August 18, 2007 11:02 AM

FORTRAN? That's soo 1970's.

I wonder if he has any understanding of numerical granularity in a FPU.

In this day and age, no one should still be using such an archaic computer language.

I hate to break it to you but a lot of code used in the physical sciences is still written in FORTRAN. From the standpoint of computational speed, a more modern language like c++ or java doesn't hold any particular advantage over a purely procedural language like FORTRAN.

Posted by Matthew at August 18, 2007 11:17 AM

"Since Hansen has the title "Dr." appended to his name, I assume that he has at least a passing familiarity with the scientific method."

Well, that's optimistic. Not done much science, have you?

Posted by sjv at August 18, 2007 11:47 AM

If/when there's an AGW disaster, the court jesters will be those associated with the Gorical and his televangelist road show. Long before there was a "consensus" the same folks were screaming bloody doom from the rooftops and greeting every contrary finding with howling insults. It's rather difficult to take that kind of "science" seriously and the poor signal to noise ratio is overwhelmingly caused by the warming fraternity. Hansen is just another clown in that circus.

I find an analogy with the cold war in this. Within a decade after WW2, roughly one third of humanity were living under a communist government. Asia was almost entirely communist. They had the H bomb and promised to bury the west. In the context of a very real danger, creatures like Joe McCarthy crawled out of the woodwork to claim that the crisis was too severe to tolerate the contrarians.

I don't know how this crisis is going to play out, but in the absence of world shaking cataclysms, Hansen will still be seen as a jerk.

Posted by K at August 18, 2007 11:50 AM

"Not done much science, have you?"

Sorry, that comes across much more snarky than I intended. You, at least, do seem to understand. It's just that I've met too many Phds who can ape the scientific method, but don't really understand it. Discouraging.

Posted by sjv at August 18, 2007 11:52 AM

It has to be realized that most of AGW criticism is entirely nonserious.

It doesn't matter. What matters is whether it's accurate or not. In fact...even that doesn't matter. What matters is whether the theory of AGW can stand up to it or not.

Science doesn't say that only criticisms offered in good faith are valid and must be considered seriously. That's religion.

Posted by Carl Pham at August 18, 2007 12:02 PM

I didn't take it as snark, SJV, I know exactly what you mean.

Posted by Ed Minchau at August 18, 2007 12:03 PM

I've met too many Phds who can ape the scientific method, but don't really understand it.

I think part of the blame for this rests with the unholy alliance of university and Federal research money, which is strongly focussed on training graduate students. It's expensive for a department to dump or discourage someone in a PhD program, as long as he can do some useful work washing bottles or whatever.

I once sat on an engineering PhD oral exam as the outside member, and was horrified at the performance. So were the engineering guys. After the poor schmo was sent outside to await his fate and the committee had contemplated their shoes for a few glum minutes, the chairman asked me: 'So what would [my department] do if you had a candidate who shouldn't really be given a PhD?'

I said (being young): 'I suppose we wouldn't give him a PhD.' They all smiled at my naivete. But then one of them said that, given the fact that he didn't hide his light under a bushel, there was zero to no chance he would get a job in a profession close to the university, with someone who knew the department from which he came. So he would probably just disappear and the fact that they'd given him a PhD wouldn't come back to haunt them. They all brightened up at that.

I don't mean to sound cynical about any of the parties involved here. There's not much else the department could have done, and the guy involved was trying hard, it wasn't his fault he just didn't have what it took.

If blame is to be found for this and related examples of people with PhDs who are, nevertheless, utterly unqualified to be doing science, then I would put it on the crack-like addiction of the modern research university on Federal money, and on the tendency of lower grades to buck problems upwards by passing people who shouldn't. The tragedy in that is that if people realize they can't swim when the water is up to their neck, they can learn. If you wait until the water is totally over their head, they just drown.

Posted by Carl Pham at August 18, 2007 01:12 PM

sjv said

"Well, that's optimistic. Not done much science, have you?"

IMHO every scientist and engineer pursuing a Ph.D. should be required to read and write out in their own words what Richard Feynman meant in his speech "Cargo Cult of Science".

Pertinent exerpt:
" But there is one feature I notice that is generally missing in cargo cult science. That is the idea that we all hope you have learned in studying science in school -- we never say explicitly what this is, but just hope that you catch on by all the examples of scientific investigation. It is interesting, therefore, to bring it out now and speak of it explicitly. It's a kind of scientific integrity, a principle of scientific thought that corresponds to a kind of utter honesty -- a kind of leaning over backwards. For example, if you're doing an experiment, you should report everything that you think might make it invalid -- not only what you think is right about it: other causes that could possibly explain your results; and things you thought of that you've eliminated by some other experiment, and how they worked -- to make sure the other fellow can tell they have been eliminated.

Details that could throw doubt on your interpretation must be given, if you know them. You must do the best you can -- if you know anything at all wrong, or possibly wrong -- to explain it. If you make a theory, for example, and advertise it, or put it out, then you must also put down all the facts that disagree with it, as well as those that agree with it. There is also a more subtle problem. When you have put a lot of ideas together to make an elaborate theory, you want to make sure, when explaining what it fits, that those things it fits are not just the things that gave you the idea for the theory; but that the finished theory makes something else come out right, in addition.

In summary, the idea is to give all of the information to help others to judge the value of your contribution; not just the information that leads to judgement in one particular direction or another."

End of exerpt

If a scientist receiving a Ph.D. cannot understand and put into practice these ideas, then that person does not deserve the title Ph.D.

Posted by JAH at August 18, 2007 01:13 PM

Note that one of the "clowns," as Dr. Hansen calls them, is Dr. Robert Jastrow, the founder of the Goddard Institute for Space Studies and its director for 20 years.

Under Dr. Hansen, the Goddard Institute seems to have forsaken most other forms of space studies in favor global warming horse.

http://www.giss.nasa.gov/research/

Posted by Edward Wright at August 18, 2007 01:33 PM

Note that one of Hansen's own papers, featured on his own web site, says that "observed global warming has been caused mainly by non-CO2 GHGs."

Which would be heresy if most people said it.

http://www.giss.nasa.gov/research/features/altscenario/

Posted by Edward Wright at August 18, 2007 01:39 PM

Have you ever written comparative programs in these languages?
FORTRAN is obsolete, incredibly so.

Matlab, Mathematica, and other similar languages are far superior to FORTAN in processing scientific information. The productivity, the debugging, and data presentation are all light years ahead of FORTRAN.

Posted by Dennis Wingo at August 18, 2007 01:44 PM

Cecil, you're mixing global and US temperatures once again.

It's just a sad truth that in politics many other things than science matter. And many of these other inputs are very vocal anti-AGW opinions and lobbiers, perhaps posing as science. Just look at what kind of idiocies are spouted by politicians about the matter. Myths and incorrect soundbites.

I think McIntyre has a 1 degree C hunch from the doubling of CO2 with some iris theory that increasing CO2 will have less effects. I haven't studied it in detail. I'm not sure if I recall correctly, but I think he has some history of posing as a denialist champion in "discrediting" various studies that actually weren't discredited. Seems he found something actual this time.

IPCC has gathered lots of studies that ended up between approximately 1.5 to 4.5 C of warming for CO2 doubling, I think with various different assumptions.

Hansens (et al) temperature correction methods were described in the 2001 paper. I guess this stuff could and should be more open.

Posted by mz at August 18, 2007 02:04 PM

JAH

That comment of yours was brilliant, made my day. I need to look up that document.

Posted by Dennis Wingo at August 18, 2007 02:08 PM

IPCC has gathered lots of studies that ended up between approximately 1.5 to 4.5 C of warming for CO2 doubling, I think with various different assumptions.

I have yet to see anywhere in any paper the exact calculations related to the effect of increased CO2 on absorption of infrared radiation.

Please do not use realclimate.org because they don't do it either.

Posted by Dennis Wingo at August 18, 2007 02:17 PM

Matlab used to be much slower in many ways than fortran though. I don't know how it's nowadays, but that was definitely the case in 2001.

Of course the data handling here is probably quite simple. The advantage with matlab is that it could probably be much shorter code because of the helpful functions.

Many supercomputers have fancy fortran implementations as far as I know. It's probably a leftover from the time when you always had to write the low level code for all simulations by hand. Nowadays it's not always the case. I have a friend who studies ground water dynamics and writes Fortran stuff when modeling.

Posted by mz at August 18, 2007 02:30 PM

Matlab, Mathematica, and other similar languages are far superior to FORTAN in processing scientific information

If you're just messing around with the data to graph foo vs. bar, yes.

But if you're trying to do some serious computation (solve an integral equation, do a Monte Carlo integral, integrate a few zillion equations of motion) then these are just toys.

I'm willing to do most of my work in C, because I don't do seriously intense numerical work. But FORTRAN is still the gold standard when it comes to speed of execution. Only raw assembler is faster.

Posted by Carl Pham at August 18, 2007 02:53 PM

Carl

Speed of execution is only a single metric. Also, if graphing is the only thing that you think that Mathmatica and matlab can do, you are seriously out of date in your information. C and C++ are very good as well. I would like to see you implement a Runge-Kutta equation in assembly.

Carl I have been designing computers since the early 1980's including writing the assembly language test routines for system test, I understand what FORTRAN is. It is incredibly obsolete in that a scientific researcher, in order to do a really good job in using FORTRAN, has to learn FORTRAN in a very competent way. This gets in the way of the science, especailly when these new fourth generation languages might be 50% slower worst case. Big deal with the level of computational power that we have at our fingertips today.

I would make a bet that part of the problem with Hansen's code is that he did not understand it well enough to see that the problem was even there. There are other, even more subtle issues associated with computational granularity that can wildly skew results. It was the running of a mathematical program that found the floating point error in an Intel processor about 10-12 years ago.

Posted by Dennis Wingo at August 18, 2007 05:19 PM

Rand

Have you seen the graph from the pdf from Hansen? You really ought to check it out. The graph on the right (U.S. data) where he shows McIntyre's changes to his data has a vertical scale of 0.5 degrees per division. The graph on the left (Figure 1) with the global data has a vertical scale of 0.2 degrees per vertical division.

Why is that? This is the essence of the crap that comes from him these days. MZ you need to take notice of that.

Posted by Dennis Ray Wingo at August 18, 2007 09:45 PM

"I have yet to see anywhere in any paper the exact calculations related to the effect of increased CO2 on absorption of infrared radiation. "

I did a model a while ago, and that's exactly how mine worked. It modeled CO2's absorbance of infrared. It integrated across the infrared spectrum with the absorptivity of CO2 at each wavelength. Then it integrated vs altitude, adjusting for density. However, I know my model will be innacurate. It doesn't account for re-radiation, re-absorption, or any sort of atmospheric convection. (I'd like to see how you would do turbulent atmospheric scale convection on a desktop computer without inventing tons of fudge factors)

I mainly wanted to see the relative effect of water vapor to CO2 in terms of influence. With an average atmospheric water vapor concentration, I got a 0.0096% increase in absorbed energy for doubling CO2. When I ran it for zero water vapor concentration, the CO2 had an effect: 28% increase in reabsorbed energy.

Basically, CO2 shouldn't be having an effect on the greenhouse properties of any column of air containing water vapor (~1% average) will not be effected in any meaningful way. It is only in arid regions that more reabsorption of energy will take place.

The reason why is that water vapor saturates the spectrum where CO2 has it's main absorption peak at 15 micrometers. CO2 can only really effect the absorption of light at 4 micrometers, where it has another narrow absorption peak. At 14 micrometers, anywhere the air has moisture, the atmosphere is already opaque.

So, only at very high altitudes and in arid regions of the world will CO2 produce an effect.

Posted by anonymous at August 18, 2007 09:47 PM

Of course, the reason why that model is innacurate is that it treats the atmosphere as a single pane of glass. There would be re-radiation and re-absorption at different altitudes, so the upper atmosphere should also play a part in blocking radiation originating just a step below it.

Of course, that ties it into the temperature of the gas at each altitude, which ties it into convection, which makes the model far too complicated to run reliably without doing a full up simulation of the earths weather.

Which means by the time it's done running, my grandkids will have grown beards, and turbulent instabilities would ensure it's innacuracy anyway.

Posted by anonymous at August 18, 2007 09:50 PM

Oops, that was with 0.25% atmospheric water vapor, not 1% - the average over altitude, not jsut on the surface.

Posted by anonymous at August 18, 2007 09:57 PM

There is another aspect to this that I have been running down. There are three modes of absorption modification for CO2 in the presence of more of itself and other gasees.

1. "Natural" broadening

This is the broadening that relates to how CO2 Quantum mechanically absorbs and re-emits radiation. It is a minor variable.

2. Doppler Broadening

Doppler broadening is related to the statistical distribution of the velocity of CO2 molecules in the atmosphere from the surface all the way up to space. It is strictly dependent upon the temperature of the molecules that it runs into and therefore is a feedback and not a forcing mechanism for climate.

3. Collision or Pressure broadening.

Both Collision and Doppler broadening shift what is normally a gaussian absorption/emission line into a Lorentizian line. This is where the "wings" of absorption come from when there is a discussion about CO2 and global warming.

Collision broadening shifts the frequency of the emitted radiation from a CO2 molecule proportional to the square root of the velocity of the collision velocity with other molecules. It is pressure dependent, but the pressure dependency is proportional to the total pressure of all of the atmospheric gasses, not just more CO2. Therefore the Lorentz distribution variation is not only pressure dependent, it is also dependent on the square root of the velocity of molecules that it runs into. So to the first order, all CO2 wavelength broadening influences are temperature dependent.

This is without considering the overlapping effect of the absorption of infrared radiation by water vapor. The education on this subject is quite torturous.

Posted by Dennis Ray Wingo at August 18, 2007 10:45 PM

Dennis, there's still roles for Fortran. You also have considerable overhead with Mathematica and Matlab. For example, I wouldn't use them in an embedded application, that is, in a computer with limited resources attached to something like say a processor in a satellite that does some signal processing.

Personally, I'd use C++, Java, or scripting language (Python or Perl) with calls to Fortran libraries. So the heavy lifting is done by Fortran, but you can use the object oriented features of these other languages to make an understandable program. There's some technical issues, like insuring that data is passed accurately (eg, passing from a Fortran float to a C float and vice versa may involve a little loss of data or introduction of round-off error which is a proble for some applications) and error handling (if an error happens in a Fortran library how does your Java program handle that?).

Doing it that way, the complexity and quaintness of Fortran is hidden from the programmer.

Posted by Karl Hallowell at August 19, 2007 05:33 AM

Yet again, discussion of an important question is diverted down a blind alley.

Fact: CO2 warms the Earth, by direct and indirect paths, some of which have positive feedback loops and some of which have negative ones. (The estimate of Earth's mean temperature without CO2 is about -10 deg C, I believe - at which point the contribution of water vapour would be essentially zero, so it might well go lower than that.

Fact: CO2 and other greenhouse gases such as water vapour and methane are on an upward trend which currently shows no signs of slowing down, never mind stopping.

Fact: More energy dumped into Earth's atmosphere and oceans will make Earth warmer and the weather more violent. (There is a reason why there are no hurricanes in the Antarctic Ocean.)

The next fact is a negative one. We don't know what the net effect of increasing CO2 is going to be in the range likely within the next century or so, because there are many secondary effects (higher temperature leads to more cloud cover which reduces insolation which reduces temperature, for example). We just plain don't know. There are numerous attempts at modelling the processes involved, including a distributed effort on the Internet, but nothing positive has really come out of it.

Fact: The only way to be sure is to do the experiment - which might well have negative effects for humanity to say the least. And, as a matter of fact, we are de facto doing just that. See my third paragraph.

Fact: Get CO2 and other greenhouse levels high enough (or, to be fair, low enough) and the results for the climate will be unacceptable and possibly lethal to the biosphere. To see an extreme example; with the planets in the right position, look into the dawn or dusk sky at the point of light you will see.

With those facts in hand, what should we do? Answer: Stop doing the experiment. Do as much modelling, with as many factors thrown in as possible, as we can - and try to get some solid data, or at least as solid as simulations can get.

And in the meantime, develop solutions to the problem, if there is one, that can be deployed if it becomes necessary. As a side benefit, most of the possible solutions would create more food for humanity (ocean fertilisation with iron), energy independence (any of the space-based solutions, nuclear power or ocean thermal) or both. The second of which would help with another problem, global terrorism, at the same time. Terrorism is actually quite a long way down the list of problems in terms of seriousness, but that can't hurt.

And, lastly, the space-based solutions give us the power to protect the human race from sudden extinction - both by spreading us around and by giving us the power to prevent another Dinosaur Killer.

But, instead of all that, America continues to spend blood and treasure on useless desert. And to use more energy, by a large margin, than necessary to carry on with daily life. As an example, a 1-ton car gets you where you're going just as fast and just as comfortably as a 4-ton Hummer (and uses less resources and energy in the making) while using half the energy or less.

To an outsider (I'm British, by the way) the protests of certain Americans regarding greenhouse emissions sound much like those of a 5-year-old told that, no, he can't take his entire toy collection on holiday with him.

If one likens humanity to a family, then Americans are the spoiled brats.

Posted by Fletcher Christian at August 19, 2007 07:28 AM

Well, we certainly can't argue with Fletcher Christian! After all, he's British -- and not only that, he preceded all his statements with "Fact"! Game over!

Posted by Andrea Harris at August 19, 2007 07:47 AM

"If one likens humanity to a family, then Americans are the spoiled brats."

And I suppose that makes the Brits the once great patriarch who now wallows in his own filth and dementia after a slide into sad entropy far beyond the days of his previous glories?

As far as your 'facts' go, did you read what Anonymous posted above you?

Let me see if I can dumb it(his post)down for you:

What color is your yellow car if I paint it blue? How yellow is it now?

Posted by Mike Puckett at August 19, 2007 10:14 AM

Fact: More energy dumped into Earth's atmosphere and oceans will make Earth warmer and the weather more violent. (There is a reason why there are no hurricanes in the Antarctic Ocean.)

Well yes. Wanna admit that the Sun is dumping more energy into the Earth's atmosphere?

Posted by triticale at August 19, 2007 10:34 AM

Fletcher.

Actually, although _releases_ of methane are on an upward trend, the atmospheric concentration appears to be hyperbolically approaching a maxima. There's something unclear about exactly what's happening to the methane, but the data for the 1900-2000 time period are pretty clear on this.

Then there's the whole CO2 is in a buffered solution with the oceans line. Who knows how much that is relevant.

But the threat of becoming "Venus-like" is just nuts. We've already run the experiment "What if the oceans were completely boiled off, and the concentration of, say, oxygen was all the way down to zero due to greenhouse gases."

Posted by Al at August 19, 2007 10:55 AM

Fletch is correct in his facts so far as presented and advocates a rationally conservative answer, stop doing the experiment, which I agree with but without hysteria and destruction of society in the process.

However, he did not state the important fact that for the entire post Cambrian time, the Earth's average temperature has been several degrees C warmer. Ice at the poles is not the 'normal' climate. What is really happening?

Posted by philw at August 19, 2007 11:09 AM

Oh I agree with the above. However, for use in climate modeling, it is woefully obsolete, leading to errors such as Hansen's.

Posted by Dennis Ray Wingo at August 19, 2007 11:25 AM

Yep, that is a fact, but also not relevant to the discussion. Every competent scientist agrees that today, at the bands where CO2 absorbs, that the absorption is saturated, that is, it CO2 cannot absorb any more radiation, except for that.....

The argument in the AGW world is that an increase in CO2 increases the pressure broadening of the CO2 absorption lines, which results in more absorption, which results in higher temperatures, except for that......

As I stated before (and I am still educating myself on this so I could still be wrong), that doppler broadening is strictly temperature dependent (square root of temp), which means that increased partial pressures of CO2 do nothing to increase absorption. Also, collision or pressure broadening is dependent on the total pressure of all gasses in the atmosphere as well as being dependent upon the square root of the temperature.

Therefore, at concentrations that are realistic in the atmosphere today (between 180 ppm in the depths of the ice age and 500 ppm posited as the max for AGW based upon how much oil we have) CO2 has little effect on increased absorption.

I am still studying but this is what I understand so far. The quantum mechanics of CO2 is very interesting.

Posted by Dennis Ray Wingo at August 19, 2007 11:35 AM

Dennis,

If you're serious about studying CO2, it it worthwhile to note that botanists studying prehistoric eras have made comments about the ice core CO2 measurements. Essentially, they're arguing 'calibration error', since 180ppm is too low for the fossil record's accounting of the amount of plant and animal life at the height of the ice ages.

The trends are supposed to be fine, just remember the plots as more 'relative CO2 concentrations' than 'absolute'.

Posted by Al at August 19, 2007 11:44 AM

Inspired by JAH's reference to Feynman, I wrote an article about "Anthropogenic Global Warming and "Cargo Cult" Science.

http://wingod.newsvine.com/_news/2007/08/18/905075-anthropogenic-global-warming-and-cargo-cult-science

Rand, if it is not ok to do this here please let me know as I don't quite know all the rules regarding cross posting these days.

Posted by Dennis Ray Wingo at August 19, 2007 12:07 PM

I'm not sure if Dennis' ideas are right, that it's claimed that the only effect is that the increased CO2 increases pressure broadening.

Posted by mz at August 19, 2007 02:16 PM

philw:

As far as I can find out, the reason for more ice at the poles (at least some of the time) in recent geological history has been simply the configuration of continental land mass. For example, the reason for ice at the south is the very fact that there is land there; and the reason for ice in the north is that the ocean is essentially landlocked and the ice can accumulate around it, with ocean currents unable to circulate because said land is in the way.

This is what makes ice ages possible. The reason for ice ages in an era when they are possible is another matter.

Posted by Fletcher Christian at August 19, 2007 02:47 PM

Probabaly far too late for this thread but...

Fletcher

There are about four main theories for the advent of ice ages,

Tectonic, (as you mentioned but also including the joining up of south and north America)

Orbital

Solar

and GHGs

and combinations and permutations thereof.

The theories for Interglacials are more prolific and include our position in the Spiral arm of the Galaxy.

Personally I'm more interested in the interglacials, and how they end since that represents the most serious climatic danger to both Civilization and the proliferation and diversity of life on Earth. Remember kids, as far as life is concerned, warm and wet is good, cold and dry is bad. (Oh, and plenty of CO2 helps as well)

Posted by Kevin_B at August 19, 2007 05:00 PM

Speed of execution is only a single metric.

Yes, Dennis, but it's the most important metric if you're doing serious number crunching. All else is secondary. It's not that Fortran is by itself so cool, it's that because it is so close to the metal, and because it's been around so long, Fortran compilers generate the fastest code. Not to mention most of your really good, really fast numerical libraries are implemented in Fortran.

I'm not saying other languages don't make sense in their context. I use Java, Perl, a little shell scripting, C, Mathematica or Macsyma, whatever makes the most sense. Sometimes, especially for a one-off calculation, you want to minimize the time you spend debugging, so a structured high-level language is what you want. But if you're working on the numerical bleeding edge -- and you're not a front-line scientist if you're not -- then you need raw unadulterated speed. Often enough, Fortran is still where that's found.

Also, if graphing is the only thing that you think that Mathmatica and matlab can do, you are seriously out of date in your information.

Dude...don't teach your grandmother to suck eggs. I used Mathematica when it was beta, before it was even in public release. I know its abilities very well.

I would like to see you implement a Runge-Kutta equation in assembly.

Why? It wouldn't be hard, merely tedious. But if I needed to solve 1D differential equations very fast, I would.

You mentioned the difficult of learning Fortran. C'mon. If you can noodle out a new algorithm for multiple time-step MD simulation, or you're into lattice guage theories, then learning a computer programming language is like a mental holiday, difficulty-wise. Especially Fortran, which is exceedingly simple.

Posted by Carl Pham at August 19, 2007 05:16 PM

Fletch, here are the major problems with your argument:

(1) CO2 and temperature are both going up. You've got a correlation. But as any serious statistical scientist will tell you, correlation does not a cause make, and you still have the problem of which way the causation runs. Id est, is the higher CO2 causing the higher temperature? Is the higher temperature causing the higher CO2? Or are both being caused by some third factor?

Yes, I'm aware that higher CO2 can cause higher temp -- in the laboratory, and that it makes logical sense that it would in the wild. If you think this is conclusive, you need to talk to the medicinal chemists. They will tell you 80% of the time they have a drug that does something useful in the laboratory, and no one can see any reason why it would not in the body, too, it just doesn't work. Complex systems are, well, complex.

(2) You are looking at what are (by geological standards) incredibly short-term trends, and extrapolating to the future wildly, backed up by theoretical arguments and arguments based on massively simplified computer simulations. In any other, more established, scientific field, this would be considered as mere speculation, to be backed up by factual data from repeated, careful experiments. But, as you've pointed out, we can't afford to do the experiments, so we are relying very heavily on theoretical arguments and predictions. Frankly, that alarms me, given how very often theory that isn't proved out by experiment goes wrong.

(3) You argue we should stop the experiment. But that's not possible. We're here. If, as you argue, we are already big enough, ecologically speaking, to have major effects on the ecosphere, then anything we do is in the nature of an experiment. There are no choices that allow us to go back to an Edenic state where we have zero influence on the ecosystem, where we truly "stop the experiment."

For example, you think we should drastically reduce CO2 emissions. Now, you know perfectly well that isn't going to happen by rolling the world back to 1820, which could only happen by slaughtering 75% of the people now alive.

So how will it happen? How are we going to continue to support the same number of people, the same level of industry, roughly the same level of energy usage -- but not emit all that CO2? Well, something. Lots more nukes, lots more windmills and dams, tidal power stations, pumping CO2 into old oil wells, massive changes in farming practises, et cetera and so forth.

Every one of which is an ecological experiment in itself. We can't "stop the experiment." All we can do is switch the experiment, so that we're taking some other unknown course into the future. Is that wise? Are any of the practical, effective alternatives to a combustion-based economy better? Or are some of them worse? Are all of them worse? I don't know. No one does, probably. It's certainly worth a lot of study, and it's certainly worth acting cautiously.

But I tire of naive arguments that imagine that we can just "stop" emitting CO2, and don't even ask what that implies. It's as naive and unhelpful as telling a drowning man he can save himself if he merely stops breathing for a while, until he can swim to shore.

Posted by Carl Pham at August 19, 2007 05:36 PM

Here's the original James Hansen letter:

http://www.columbia.edu/~jeh1/realdeal.16aug20074.pdf

Posted by mz at August 19, 2007 06:43 PM

This is now become a religious discussion which ends its utility. I am glad that you are happy with FORTRAN and all that it can do for you.

Posted by Dennis Wingo at August 19, 2007 06:53 PM

I'm not sure if Dennis' ideas are right, that it's claimed that the only effect is that the increased CO2 increases pressure broadening.

I am more than willing to be enlightened on the subject.

Posted by Dennis Wingo at August 19, 2007 06:55 PM

Yo Fletch,
Icehouse Earth is not just a fortuitous config of continents but instead a function of much higher NORMAL temps for this planet throughout many different continental layouts over half a billion years.

http://www.scotese.com/climate.htm

Scroll down and you'll see that usually this planet is much warmer than now. There's supporting data for all the temperatures presented on this site. Why this fact is ignored in climate hysteria is an interesting question.

Posted by philw at August 19, 2007 07:59 PM

Dennis Wingo wrote:

Every competent scientist agrees that today, at the bands where CO2 absorbs, that the absorption is saturated, that is, it CO2 cannot absorb any more radiation, except for that.....

No competent sciensist agress with that, because it isn't true. Moreover, even if it were true, adding CO2 would still cause the surface to become warmer, since it would increase the effective altitude at which the thermal energy is radiated back into space.

http://www.realclimate.org/index.php/archives/2007/06/a-saturated-gassy-argument/
http://www.realclimate.org/index.php/archives/2007/06/a-saturated-gassy-argument-part-ii

Posted by Paul Dietz at August 19, 2007 08:07 PM

I agree, Carl. The missing consideration in Fletcher's chain of logic is what are the relative costs of continuing the "experiment" versus a vast altering of the world's energy infrastructure. You don't just say, "golly, choice A could be really expensive so let's not do that." That's because choice A may be better than your alternate choices, ie, the lesser of evils. But you won't know if you just consider it in isolation.

Posted by Karl Hallowell at August 19, 2007 08:48 PM

Paul

The only one to deal with is part II because they just blather about a bit in part 1.

In part II they say that even 10,000 times as much cO2 would not saturate the atmosphere at the wavelengths absorbed:

Well that is not what the textbooks say.

From: "The Detection and Measurement of Infrared Radiation", Smith, Jones, and Chasmar, Oxford Press

The earth's surface, however, is at a temperature of only 290k and here a completely different set of conditions prevail in that large rgions of the radiated spectrum are completely absorbed by water vapour and carbon dioxide, and to a much smaller extent by other constituents of the atmosphere. In fact, less than 30 percent of the earth's radiation goes directly to outer space, even on a clear night.

Paul, you will note that on realclimate they never actually do any of the calculations related to collision broadening. That is because collision broadening is only in the context of the entire atmosphere, not CO2 itself. Also, if you look closely at the graph they make an error that makes the broadened band look wider than it actually is.

If you really want to figure this out I challenge you to read up on the quantum mechanics of absorption and reemission of radiation by CO2.

Sorry Paul but just quoting realclimate.org does not get it anymore.

Posted by Dennis Wingo at August 19, 2007 09:27 PM

If you really want to figure this out I challenge you to read up on the quantum mechanics of absorption and reemission of radiation by CO2.

I don't see how the QM is especially important here. This is straight IR rotational-vibrational spectroscopy, well understood even in the 40s and 50s. Undergraduates learn it without difficulty.

I suggest the difficulty here is in the statistical mechanics. We would like to make all kinds of simplifying assumptions about the density matrix of the CO2 (and other) absorbers, so that we can use some simple math to calculate what we want. Alas, the real world doesn't necessarily want to cooperate. The atmosphere is not really in thermal, mechanical or chemical equilibrium. How far it deviates -- and how this effects the calculation -- is surely the meat of the problem.

Posted by Carl Pham at August 19, 2007 10:34 PM

I don't see how the QM is especially important here. This is straight IR rotational-vibrational spectroscopy, well understood even in the 40s and 50s. Undergraduates learn it without difficulty.

Maybe this is the source of the problem, that people think that this is the case. spectroscopy cannot be fully understood without its quantum mechanical basis. This is the whole problem with realclimate.org. I may still be wrong on this but from what I am seeing, both the pressure broadening, and doppler broadening have a temperature dependence (square root of the temperature only for doppler and a combination of temperature and total pressure for pressure broadening).

Statistical methods come into play when looking at very large numbers of atoms, which can predict the distribution of wavelengths on the "wings" of emission and absorption lines but there is still a temperature dependency, which puts CO2 into the realm of a feedback, not a pure forcing mechanism as is commonly described.

Posted by Dennis Wingo at August 19, 2007 11:08 PM

Well...I'm not totally sure where you're going, Dennis. The most basic form of pressure broadening is just a version of lifetime broadening. The uncertainty relationship between energy and time means the shorter the lifetime of your excited state, the broader the absorption line. There's a natural lifetime (which comes from coupling to fluctuations in the photon number), and that can be shortened up when you have other nearby molecules colliding with your absorber. Hence, broader lines. Even this is a statistical issue, because as you can see a key question is how many nearby atoms are there and how often do they collide with your absorber.

But beyond that, the question of how much light is absorbed by a large blanket of gas is a question of what is the density of absorbers and amplitude of light at various altitudes and latitudes, and how often do the absorbers interact with other molecules, and how. Because I would guess (I could be wrong here) that even in the stratosphere the density of molecules is high enough that the lifetime of excited states is strongly influenced, if not usually determined, by collisions.

So what I'm saying is, I suspect the tricky part here is figuring out what kind of environment exists up there for CO2 molecules. What's the density, what else is around, how often are the collisions, what happens during those collisions, et cetera. These questions would be fairly easy to answer if the atmosphere were in equilibrium. Maybe it is on the time scale necessary. I'm not sure, it's not my field.

But, yes, surely spectroscopy makes no sense without QM. I mean, classically, there's no such thing as an absorption line. Everything's just a black body, eh?

Posted by Carl Pham at August 20, 2007 12:25 AM

Rabett Run has done some postings on the pressure broadening issue:
http://rabett.blog spot.com/2007/07/how-well-can-we-model-pressure.html
(remove the space)
There are many more about the same subject. I haven't followed it that closely.

Posted by mz at August 20, 2007 01:26 AM

But, yes, surely spectroscopy makes no sense without QM. I mean, classically, there's no such thing as an absorption line. Everything's just a black body, eh?

yer funny. Too bad too many people discussing the issue think that yours is a true statement.

Hence, broader lines. Even this is a statistical issue, because as you can see a key question is how many nearby atoms are there and how often do they collide with your absorber.

Actually the calculations are pretty straightforward (ok it took me six months of searching before I figured out that I had the answer in my own library) for a single molecule and it is interesting that to a fairly close order, you can figure out the frequency deviation (that is what line broadening actually is) as a function of pressure. However, it is also dependent upon temperature, hence the notion that it is a feedback rather than a strict forcing mechanism. Self broadening as a contributor to the total Lorentz broadening is almost non existent and is ignored by most calculations.

Quantum uncertainty does not really come into play at infrared wavelengths to optical wavelengths, it becomes dominant at much shorter wavelenths. However, that being said, the state transitions and energy absorption and emission are strictly governed by the quantum mechanics involved and things like rotational modes are quite trival in comparison with doppler and collison (pressure) frequency shifts.

It has taken me a while to muddle through all of this and I am still working on it. However, I have a pretty good level of confidence that most of what is at realclimate on this issue completely misses the point. The HITRAN data is good but so far I am going behind HITRAN to the orginal sources, which do not support the conclusions drawn by realclimate.org and other advocate sites.

Posted by Dennis Wingo at August 20, 2007 07:48 AM

Rabett Run has done some postings on the pressure broadening issue:

This discussion is very incomplete and the Voigt expression is not used at normal pressures (atmospheric pressure or somewhat lower). I am still learning about the variations here in the different methods as they are thrown out there.

Still gotta track down and cross t's and dot i's but at least this site is working in the right field, as realclimate is off in left field on this one.

Posted by Dennis Wingo at August 20, 2007 07:59 AM

I may still be wrong

You're completely wrong and you won't learn much by spreading fake peacock feathers about HITRANS and quantum mechanics. You have been saying that the atmosphere is already as insulated as possible with carbon dioxide and adding more won't matter. In reality, there is a NASA satellite instrument that very precisely measures IR absorption from each greenhouse gas using a camera with more than 2000 spectral channels.

http://www-airs.jpl.nasa.gov/Mission/WhatIsAIRS/

Here is a map from that satellite that shows just how wrong you are:

http://www-airs.jpl.nasa.gov/Products/CarbonDioxide/

The pages that Paul Deitz linked at
realclimate.org have it exactly right. The "saturation" argument is an 80-year-old physics mistake that confuses underwear with overcoats. Yes, the atmosphere is IR-opaque layer at the ground because of water vapor. But that layer glows with heat too, and it has a dry blanket on top where the CO2 matters. NASA satellites can easily see it.

Now, the absorption models for CO2 are useful both for predicting the future and analyzing AIRS data. However, just because realclimate.org doesn't show every plus sign and decimal point, that doesn't mean that the calculations don't exist. Why would NASA launch a billion-dollar satellite without doing basic grad-school calculations to interpret its data?

Posted by at August 20, 2007 12:13 PM

Yawn

I musta struck a nerve. You need to read Dr. Feynman's 1974 Caltech commencement speech on scientific integrity. I grow weary with all the, as you call it, peakcock feathering about how wonderful this or that model is.

Yep the AIRS data is interesting and I will get to it as soon as I finish reviewing the fundamental physics and the data that stands behind all the current efforts for climate modeling. That is what science is, not taking anything for a given, or anything for granted.

Quantum mechanics is where the study of CO2 begins and without that fundamental understanding, how can you hope to interpret the data? I have already found that there is a temperature dependency for both pressure broadening, and for doppler broadening. If that is the case, which it most certainly is, then CO2, at some level is a feedback and not a pure forcing mechanism. Sorry, you can preen all you want about how wonderful this model or that model is, or the quality of this data set or that data set, and not understand the quantum mechanics is to not understand the entire subject.

I have yet to see, in any paper anywhere, what I have seen in the textbooks regarding how adding CO2 actually effects absorption and reemission using the equations that underly the beer-lambert law and other buzzwords tossed around on this subject.

Yep I am still learning the subject. I have the background in quantum mechanics and physics to see and understand what is going on and then work through it. It may be in the end that I am your most ardent supporter, but that will not happen until the entire chain is complete. So far realclimate.org is full of horse hocky and they commit the same errors in misleading people that James Hansen does. If you look at the graph that they have in their part II, showing the effect of more CO2, they do not use realistic concentration increases, they use a 4x level, while also broadening the response line from the first graph to the second by a micron and a half for no reason whatsoever.

Sorry but that is just crap science and the only way to demolish crap science is with good science and good science cannot be done in a weekend of googling websites.

The absorption models are crap if they do not (show me otherwise), incorporate the temperature dependency of the Lorentz transformed function for the absorption and reemission of CO2 spectral lines. There are two functions combined that give the Lorentz transformed line and only one of them is pressure dependent (the pressure dependent line is also temperature dependent to the square root power).

Let see if you even understand that paragraph and then we can talk.

Posted by Dennis Wingo at August 20, 2007 12:57 PM

Yep the AIRS data is interesting and I will get to it as soon as I finish reviewing the fundamental physics and the data that stands behind all the current efforts for climate modeling.

You don't need to read a physics book to tell whether the sky is blue; all you have to do is go outside and look. Yes, a complete explanation of the color of the sky involves quantum mechanics and it's a fascinating story. But Feynman knew as well as anyone that experiment and common sense should come first. The AIRS satellite instrument can see the color of the atmosphere, in the IR range, from space looking in, which is the direction that matters. It isn't remotely IR-opaque from CO2.

Yes, I know what a Lorentz transformation is. I also know what Rayleigh scattering is, and I even know why the stretching and bending modes of CO2 have quantum numbers. But anyone can see that the sky is blue without knowing Rayleigh scattering; and you don't need to write down the Schrodinger equation to have a basic grasp of the AIRS data either.

Posted by at August 20, 2007 01:45 PM

If you know that then you should know of the temperature dependence of the particular modes of absorption and reemission of infrared radiation of CO2. Temperature dependence means temperature dependence, end of story. Yes the Sky is blue, that is because of oxygen absorption and reemission in the shortwave visible spectrum. Yes you do need to know these things to interpret the data. No one talked about the Schrodinger equation in this context. This is fundamental physics of how radiation is absorbed and reemitted by CO2. This can explain the reason that CO2 is a lagging indicator of climate coming out of the ice ages. I have not seen an explanation from the AGW crowd.

Sorry, to rely on data from the last chapter of a book does little to inform you on what really happened in the story.

Posted by Dennis Wingo at August 20, 2007 02:51 PM

I think you are confusing Lorentz transform with Lorentz profile. A Lorentz transform would only be relevant in the case where the air molecules are moving relativistically. I'm not a climate scientist myself but I'm pretty sure that those working in the field are familiar with all of the QM effects that you discuss. You also seem to think that pressure broadening is some new phenomenon that no one's ever heard of before when in fact you can find a discussion of it in any introductory text on radiative transfer theory.

Posted by Matthew at August 20, 2007 03:56 PM

A Lorentz transform would only be relevant in the case where the air molecules are moving relativistically.

Or the observer, har har.

you can find a discussion of it in any introductory text on radiative transfer theory.

You don't have to go that far. Donald McQuarrie's standard graduate text on statistical mechanics discusses it briefly in Chapter 21 in the context of the (ahem) semiclassical time-correlation approach to spectroscopy. Peter Atkins' undergraduate p-chem text makes mention of it in Chapter 16, on rotation-vibrational spectroscopy, although he calls it "collisional" broadening.

Posted by Carl Pham at August 20, 2007 05:06 PM

I'm pretty sure that those working in the field are familiar with all of the QM effects that you discuss. You also seem to think that pressure broadening is some new phenomenon that no one's ever heard of before when in fact you can find a discussion of it in any introductory text on radiative transfer theory.

Of course not, pressure broadening is one of the things most talked about at realclimate.org. That is the point, that in discussing this, they don't ever delve into what it means.

Thus the problem. Everyone assumes, this that or the other when to do the science you cannot assume anything, question everything, and recreate the entire chain of development.

The textbooks talk about the lorentz transformation from a gaussian to a lorentz function for measuring the divergence from the gaussian of the absorption and emission of radiation and optical or infrared wavebands. This is where the Q and P branches that are talked about in context of global warming come from. They are temperature dependent (doppler broadending) and pressure and temperature (collison or pressure broadening) dependent.

Collisional broadening is the equivalent to pressure broadending in the literature.

Posted by Dennis Wingo at August 20, 2007 08:03 PM

Remember Fleishman and Pons? They claimed to have achieved cold fusion, and when they published they published everything - the experimental setup, initial conditions and so forth - with the intention of having their results duplicated by someone else. Well, nobody could conclusively replicate their results so obviously they made an error somewhere.

Contrast this with James Hansen, who not only refused to publish his raw data and algorithms, but when those were eventually ferreted out and shown to have a huge error, called his critics "jesters".

Fleishman and Pons had scientific integrity. Hansen apparently wouldn't recognize scientific integrity if it kicked him in the nuts.

Posted by Ed Minchau at August 21, 2007 12:08 PM

Global warming is based on something a bit more than 0.1 C GISS temp difference around 2000 in USA.
GISS has corrected the data already.
Yes, they should be more open.
No, this doesn't disprove anything.

Posted by mz at August 22, 2007 08:40 AM

Post a comment