Post by Dr. Ryan Maue (cross posted at my Policlimate)
“Must watch TV: Nye expounds on theory of racism”
Much “debate” has erupted in the liberal mainstream media concerning the effects of global warming on Hurricane Irene. With a few notable exceptions (Henry Fountain awesome), many of the journalists butchered the science and generally constructed disjointed narratives that quoted a variety of favorite experts which left me wondering why they even bother (Politico). Rush Limbaugh provided a compelling alternative explanation for the hurricane hype: “Politics is part of everything. The weather’s been politicized; the climate’s been politicized…Both Obama and the media were hoping for a disaster to revive his presidency and help prove climate change theory…The New York Times is trying to say that this violent hurricane is indeed indicative of global warming. It was a tropical storm when it left New York.”
But Bill Nye takes the “anti-science” crusade to a new level by showing up on Fox Business with my KFI 640 Saturday friend Charles Payne and embarrassing the hell out of himself. Once you watch the video and read the transcript, you will be left in amazement at his utter lack of comprehension of the topic at hand on national television! But, alas, Media Matters thinks Nye owned Payne (h/t to Andrew Revkin to Tweeted this). And CBS News headlines it as a story! Unbelievable!
The left actually thinks Bill Nye is a brilliant ambassador for their brand of global warming alarmism — a legitimate guy that understands the science and can articulate an explanation. However, Nye has no credentials or expertise with respect to global warming and hurricanes, at all. Not one iota.
Video is embedded or to go to CBS News and watch the Fox Business embedded video there. “Heady stuff, but Nye receives my respect for retaining his patience in outlining a life’s worth of work in a six-minute segment.” says Andrew Nusca. He has no idea that what Bill Nye is saying is disjointed and amateurish. Intricacies? Nye got almost everything wrong.
I transcribed my own transcript from the first 3 minutes of this (all I could take). Emphasis — bold and italics are my comments.
Charles Payne: While hurricane Irene brought more than just wind damage and flooding to the east coast, it’s revived a national debate as to whether global warming might be causing an increase in hurricanes and other extreme weather. In fact a recent cover story in Newsweek declared that this kind of wild weather may be quote “the new normal”. Here with insights on this is Bill Nye, otherwise known as the science guy.
Ok Bill, I’m going to come right at you. Um…Hurricane Irene – proof of global warming?
Bill Nye: Oh, I don’t think the word proof is what you are looking for – evidence of, a result of, yeah, yeah. Now here’s what the people will tell you that run these climate models. Now everybody, the word model in this usage is a computer program. A very sophisticated computer program. So you take data from satellites about the thickness of clouds and the extent of cloud-cover over the sea. You take data about the temperature of the sea surface. You take data about the existing weather say in North America or the Gulf of Mexico as this storm moves into it. Then you compute how much rain fell out of it, how much energy must have been put into it to create that much rain. It takes many months to analyze an event like Irene. Now the climate colleagues that I have will not tell you today that Irene was evidence or a result of climate change but check in with them about March next year after they have a few months to collect all of these millions and millions of data from weather services and satellites and compile them and run a climate model and show that Irene was a result of the world having more energy in the Earth’s atmosphere.
(Ryan: First of all, charitably, I think Nye is confusing a real-time operational weather forecast with a climate model. Climate models do not assimilate satellite observations of a given event — and it wouldn’t take months and months to compile the data. I have everything sitting on my server which generates my old FSU weather map page. Check back with them in March — that’s when they’ll have their climate model results back proving Irene was the result of more energy? This is a pretty unconventional way of doing climate or extreme event attribution. Bill Nye follows the “anti-scientific” method: I’ll give you the answer now, and then in 6-months, check back when I have the proof. )
CP: But here’s the thing here bill, ever since Katrina, right, we’ve heard that every year the hurricane season is going to be more devastating and apocalyptic, and the reality is we haven’t seen that. So how can Newsweek say “hey, this is a new normal”? is that irresponsible – is there any science behind that?
(Ryan: this is a great question by Payne. Since global hurricane activity — the number of storms, hurricanes, and Accumulated Cyclone Energy (ACE) is at historical lows — collapsing since Katrina — as I showed in my recent GRL peer-reviewed paper, how on earth can you attribute one hurricane (Irene) to climate change.?)
BN: well there’s a lot more science behind that than just saying it’s not. But, uh, that aside. That’s only 6-years – in geologic time in terms of climate events, is not very long. Furthermore there is a lot of debate about this cool thing or remarkable thing is that the Sea-surface in the Pacific gets warmer, in the Pacific Ocean! Okay, tens of thousands of nautical miles away. As that gets warmer, it will strangely serve to decapitate certain hurricane or cyclonic storms off the coast of Africa – and actually get a few fewer hurricanes.
(Ryan: no kidding Nye, however, you haven’t come up with any science. Nye then launches into a tortured explanation of the El Nino Southern Oscillation warm phase — El Nino where the waters in the tropical Pacific cyclically become anomalously warm. But, it’s not “tens-of-thousands nautical miles away” — that’s more like the distance to the moon. There is actually little consensus in the climate community about the future of El Nino as the planet slowly warms. The CMIP3 models used for the IPCC AR4 report fail to reproduce historical ENSO events or variability, and therefore are useless prediction devices for the future. We already have a pretty good handle on the “teleconnection” effects of El Nino and La Nina on Atlantic hurricane development with research pioneered by Dr. Bill Gray and furthered by Dr. Phil Klotzbach who produces Colorado State’s seasonal hurricane forecasts. 2011 is a neutral-to-building La Nina year, so we should expect weaker vertical shear in the Main Development Region of the tropical Atlantic. It’s bizarre that Nye brought up El Nino which contradicts his original assertion that Irene was evidence of global warming.)
CP: But Bill, that’s not…
BN: This is another thing that’s very hard to show.
CP: But the Pacific Ocean, getting warmer, but that’s not from man.
(Ryan: excellent point again Charles. The tropical Pacific does not have a strong global warming signal over the past 30-years, which is due to the cyclical nature of ENSO on 2-7 year time scales. Our sea-surface temperature (SST) records get worse as you go backwards from the beginning of the satellite era in 1979. Nye has no answer.)
BN: (waving hands): you’re acting that you are dismissing those things like they they are not relevant.
(Ryan: Nye is defeated, and he knows it. After wagging his finger like Judge Judy, he pretty much has spent his arsenal of facts on this issue.)
CP: I’m not dismissing it, but you have so much information, I want to get to all of it. Are you saying though that it’s manmade, though?
BN: Well the world is getting warmer, uh, everybody, the world is getting warmer. I believe the debate is whether humans are causing it…Do we not agree that the world is getting warmer?
(Ryan: The world is getting warmer — so Irene has to be influenced by global warming. Maybe Irene did NOT reach its maximum potential because of global warming — has anyone considered that. Why must ALL of the climate change effects be a certain sign? Why didn’t Irene reach Category 5? Why did it weaken so fast if the SSTs were so warm? This is where the real tropical cyclone researchers will take over from the media hacks, and, yes, they will come with an answer in March. But, they will follow the “scientific” method and not the “I’ll get the proof later” Bill Nye “anti-science” method.)
CP: I have no idea. Someone told me that it’s warmed 1-degree over the past 100-years. I’ll take their word for it.
(Ryan: Charles is right.)
Show continues to talk about racism and shows the Al Gore “racism” clip – but Nye then really goes off into a different realm discussing that. I’m convinced that Fox News booked Nye knowing that he would butcher the science, and force me to write this post.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
No one said it was irrelevant to the task. But consider your position and how it has changed through this thread. You began with how simple it was, and correlation was the only important factor. Swiftly debunked by the diet soda causes obesity example. Then your explanation became, welll, you need some reasoning, and three variables not two, good luck debunking that. Your logic above gets even more complex while you try to maintain that it isn’t easy, but that correlation remains the only factor. At each step you keep adding more steps and requirements to make your oiriginal position true. You’ve now arrived at a point where neither your theory nor your example are true.
Your theory (and I quote)
if C ==> X ==> A&B then C is still a cause of A&B. It really doesn’t matter that X may have other causes unless you need to know ALL of the causes of A&B. That in itself is irrelevant to determining if C is a cause A&B.>>>
WRONG! In your own example, C is NOT a cause of A&B. C is an INDIRECT cause of A&B. ANYTHING that causes X is an INDIRECT cause of A&B. Since C is NOT the cause of A&B in your own example, the balance of your argument similarly falls apart. It isn’t “unless” you need to know “all” the causes, it is that you DO need to know ALL the causes in order to eliminate false positives…which your reasoning above provides an example regarding the importance of. Your last sentence is completely false as a result as well.
Your real world example (and I quote)
Pulling the trigger causes the gun to fire if all else is working properly. It is still true regardless of whether the gun might fire when dropped. >>>
The gun fires when the firing pin strikes the bullet’s primer jacket with sufficient force to ignite the primer. This could be accomplished by:
pulling the trigger
dropping it (assuming it is a cheap piece of poorly designed crap)
hitting the gun’s hammer (assuming it has an exposed hammer) with a rock
If a gun has been fired, was it because the trigger was pulled? It was dropped? Someone hit it with a rock? Answer: You do not know. In fact, you do not even know that the gun was fired at all. Old ammunition becomes unstable. The gun might have been exposed to heat. And so on.
In other words, “unless” you need to know “all” the possible causes is a complete break down in logic. You MUST know all the possible causes in order to design an experiment that produces data of any value at all for quantifying ANY given cause.
Correlation alone is sadly, and completely, ineffective.
davidmhoffer September 2, 2011 at 4:41 pm Correlation alone is sadly, and completely, ineffective.
I’m afraid your diet coke problem only showed how to abuse concepts with inappropriate application. It did nothing to negate what I’ve said. I made a correctio though stating the need for “reason” as well. Perhaps, I should have added “common sense”.
Please show me where I have substantially changed what I’ve been saying.
Other than that, correlation alone is what is used everywhere (in science vs. philosophy or math). Show me a case where cause has been shown through the use of actual data and not done so through correlation. I bet I can show you that wasn’t the case.
Just for a quick example (which I hope you won’t abuse): Suppose one intends to determine if A cause B or vice versa through experiment. One of the ways would be to take control of the states of A and or B. This actually introduces two variables A-controlled and B-controlled and maybe others such as A-controlled-how, etc. The results are evaluated using relationships of those and the original variables. It is their correlations which is used. All the experiment did was supply the information. If pertinent data were lying about then those could have been used instead — a practice quite common in astronomy or epidemiology.
Another example: if you are testing a model, you are looking for correlation with your predictions.
No matter how you look at it, correlation is the final arbiter.
Shall I go on?
Just incidentally (and this may have been a typo in your case) but I think you misunderstood my simple example. For one, I said C is a cause if A&B become uncorrelated given C and NOT when C is removed. It also is a requirement that all three are mutually correlated. It only applies in the case of three variables. It is a rule like x+0=x. Your objections really amount to: “Oh Yeah? What about x+12?” Secondly, I never addressed the absence of C. If A&B become uncorrelated in the absence of C it’s likely that A&C==>B or B&C==>A. More testing needed, of course. Send more money!
My! Such certainty. In my world there is Cause and there is Effect. They don’t come in different flavors and colors. An indirect cause is still a cause. Red cars don’t stop being cars because they are red. In fact, one (cause) is the tail of an arrow and the other (effect) is the head in a directed acyclic graph. The arrow can be replaced with a chain when appropriate. Likewise, a chain can be replaced with an arrow. Insisting on such distinctions is pointless. You will never find “The” cause as there are is always one more thing in the chain. Answering Why is an infinitely recursive process. Where would you draw the line?
In the gun example, it’s not the finger pulling the trigger that fires the gun but is instead the activation of the the firing mechanism. In turn, it’s the firing mechanism hitting the primer. And even, then it’s really the reaction of the primer when it is stuck. Care to bet if I can find more reasons along the way to determining the “Ultimate Cause” of the gun firing? Why distinguish between them? Maybe it depends on what you are trying to do? Until I know, they are all causes to me
You are right. I don’t. I have been talking about cause in the sense of how the flow of causation would be diagrammed. It is a first and necessary step in determining past events if knowledge of those is desired. It is what is being done when examining possible causes of X. The question in climatology today is not so much “DID CO2 cause warming?” but is instead “DOES CO2 cause warming?”. The latter is the form I am addressing but what I’ve said doesn’t change the ways I would go about determining the DID answer as it would still be necessary to know if DOES is true.
I suspect I’ve studied this and thought about it much longer and in more detail than you have. If that is your only true objection, you have entirely missed the point.
This conversation is quickly reaching a dead-end. Don’t you agree?
It’s been fun. Read Judea’s book if you get the chance.
Yup, it is a dead end.
You don’t understand what a false positive is, nor do you understand that drawing conclusions from indirect causes without differentiating them from the direct causes is a complete breakdown in scientific process. I suggest from your answers which are rooted strictly in mathematical constructs, and your appeal to authority (read Judea’s book) that you have done little or no work with actual real world systems.
I suggest you do if you get the chance.
Oh, I know what false positives are.
Logic is a strictly non-“real world” construct. It should be used more.
The graphing of cause is applying logic to causality and even answers some of the questions you have raised but you have to understand what is being said first. One of my fields of study is ontology (in the computer science sense). But then knowing how we know things is not exactly “real world” is it? Cause is one of its subjects of discourse.
Objecting to use because it is a “mathematical construct” is like objecting to using statistics because it, too, is a “mathematical construct” (unless you are a REAL mathematician, that is 🙂 ). The same with calculus. What exactly is the problem with using logic and math to get solutions?
I do use this stuff for real world applications. Every day, too. Even to the point of placing my money with it. Haven’t gone broke yet. Maybe those evil chickens will come back to roost someday, eh? Many of the same concepts are used in some of my robotics works, also. But of course, I’m not all that knowledgeable about the real world as some so what do I know?
your appeal to authority (read Judea’s book)
Tsk! Is being insulting really necessary? I suggested Judea’s book because his explanations might be better for you (and Richard) and he supplies far more detail. Plus I feel I owe him a book push 😉 He also covers things like experimental design which I believe you tizzied about or at least had only questions without answers (I could have answered some them but fear you think you know best). The book is one of the clearest on the subject (i.e., most easily read). But there’s little point in reading it if you already know everything there is to know. Horse to water and all that. Where did you get your ideas BTW? It doesn’t sound like you’ve ever really thought about this stuff in any detail or for very long. Your objections sound like rote responses instilled during some far-off training. It might not hurt to take a step back and ask how you know what you think you know. Then adjust accordingly. I do that every day. Try it, you might like it.
Ciao
DAV:
At September 2, 2011 at 3:25 pm you say to me;
“Think about what you are saying.
If “Absence of correlation disproves causation” then the opposite (i.e., its presence) must imply the possibility of causation.”
No!
And your assertion of what you think “must” be demonstrates your error.
I explain the matter for the last time as follows, and I shall ignore any further posts from you about this issue.
If one effect causes another then the two effects correlate.
Therefore, if two effects do not correlate then one does not cause the other.
But two effects may correlate for any of several reasons; e.g.
chance,
both are affected by something else,
both are caused by something else,
etc.
Therefore, if two effects correlate it implies that e.g.
they have varied with syncronicity by chance,
or
both are affected by something else,
or
both are caused by something else,
or
etc.
Something which implies almost everything really implies nothing.
Correlation implies nothing about causality.
Absence of correlation disproves causality.
You need to think about what I am saying.
Richard
Richard,
The only way its absence could disprove anything is when its presence is otherwise required. Think about it.
DAV;
The “insult” stands, it is an excellent ripost to your pontification from a position of self appointed superior authority. You’ve implied that I said things that I didn’t, and nothing pisses me off more than that. You imply you have answers to questions I raised, but won’t bother to answer them. Almost as bad.
The final retreat of anyone who’s argument is hollow is to a barrage of verbal complexity that establishes nothing but a broad vocabulary. G’day sir. I’m dropping this thread. Richard S Courtney got it right, and you would do well to think about his words as well.
The “insult” stands, it is an excellent ripost[sic]
Sorry to hear insult was intended although I’m surprised you overlooked the tie-in to Hitler. Ah, Well. The shame is yours. When an insult becomes an “excellent ripost[sic]” it says much about how you perceive the strength your argument. You would do well to avoid such ripostes in the future.
As for Richard, if presence implies SOMETHING it is not obviously implying NOTHING. I said it allowed the possibility and he claims that was in total error but so be it.
DAV wrote:
” I said it allowed the possibility and he [i.e. me] claims that was in total error but so be it.”
He is in total error. I summarise the matter for any who have become confused by this discussion.
The fact that nobody has sampled the center of the Moon means it is possible that the center of the Moon is made of green cheese, but this fact does NOT imply the center of the Moon is made of green cheese.
Similarly,
The fact that two parameters correlate means it is possible that they have a causal relationship, but this fact does NOT imply they do have a causal relationship.
However, correlation is a prerequisite of a causal relationship and, therefore, absence of correlation between two parameters proves they do not have a causal relationship.
A similar point of logic which confuses some people is as follows.
Absence of evidence (for something) is not evidence of absence (of the something).
And, no, I am not going to explain or elaborate on that.
Richard
“The fact that two parameters correlate means it is possible that they have a causal relationship, but this fact does NOT imply they do have a causal relationship.
Indeed, that is what I have been saying all along. Go back and read my original post You even quoted me saying If “Absence of correlation disproves causation” then the opposite (i.e., its presence) must imply the possibility of causation.” and the very first word in your response was “No!”
See here:
http://wattsupwiththat.com/2011/08/30/bill-nye-is-the-anti-science-guy-when-it-comes-to-global-warming-and-hurricanes/#comment-735435
“Implying the possibility” is implying SOMETHING (i.e., the possibility). You have repeatedly said the implication is NOTHING. It’s not just word play. Jumping from SOMETHING to NOTHING is muddled reasoning. You have spent quite a bit of bandwidth promoting it. That kind of reasoning gets in the way of understanding.
What I have been saying is that eventually it is correlation (or its absence) that used to make the ultimate determination. I invite you to find a case where that is not so.
To Nye: 2 “tens of thousands of nautical miles” would have you about where you started, having circumnavigated the globe. Duh.
Skaggs: the queasier you “sustainable” double-talkers are made to feel, the better. Your mantra is political code for reversion to imaginary past “balance points” which never existed. Anyone using the word is wholesaling BS.
I have to laugh when I hear people say, “Our computer models prove global warming!” What do they think a computer model is? I’m a software engineer by profession, so I can tell you: It’s a set of rules and formulas, that describe how you THINK the thing you are modeling behaves. It’s a game. If the rules of the game are close enough to the real world, then the game will be “realistic”. But anyone who’s played a computer game can tell you that sometimes the rules are wildly unrealistic, whether to make the game more playable or because real life is just too hard. Like, in many computer games, when you get killed, you promptly come back to life to play some more. Would anyone in his right mind claim that this is PROOF of life after death?
If you can feed data for, say, 1900 to 1950 into your computer model, give it no data after that date, and it successfully predicts 1951 to 1960, then you might have some evidence that your rules are accurate. If you can then take exactly the same set of rules, feed in data for another set of years and successfully predict the years following, and you can do this for any given set of years, then you’ve got something very convincing. But this is exactly what the global warming folks computer models DON’T do. They can’t even successfully predict the past, but we’re supposed to believe they can predict the future.