Statistics

Best statistical, scientific talk on global warming

Some careful readers to this blog have pointed out the work of Australian geologist Bob Carter. If you have not yet seen his work, you should. So I want you to drop whatever you are doing and watch his talk below. It is the best statistical and scientific public talk I have yet seen. I’ll write more about this later, after you’ve seen his talk.

Part 1

Part 2

Part 3

Part 4

Categories: Statistics

18 replies »

  1. Dear Administrator (is that you Dr. Briggs?)

    I had this thought while sweeping the snow off my car this morning.

    I spend a lot of time lurking on the ClimateAudit site ( the Sarbanes-Oxley of Climate Science) and they spend a lot of time discussing Dr. Hanson’s GCM approach.

    I don’t know if I can ask this correctly, but here goes.

    If some unknown force or forces are causing the earth’s temperature to raise are the predictions of GCM statistically significant considering we don’t know without a shadow of a doubt what is causing the increase. ie within the bounds of error wouldn’t any hypothesis be able to be modeled (that modeled an increase) have a chance of being “in the ballpark”?

    While I think the Climate Audit site is a fabulous resource, I have a layman’s frustration with it because I have an intuition that they may be spending a lot of time on something that is maybe irrelevant? This may be totally unfair, and if so feel free to admonish me.

    Is this the right way to ask a question of you? Is there a better way of doing this?

    Thanks

    Steve H :]

  2. Steve,

    Yep, Administrator = Me. Your question is excellent.

    The best way to tell if climate models work is if they can predict, for example, temperatures that have not yet happened. If they can, and they can beat a simple forecast of “the average”, then they are said to have skill, and we can believe what they have to say.

    Do climate models have skill? Nobody knows. In part, this is not the fault of climatologists because we don’t have enough independent years measured yet. We have plenty of years worth of data used to fit climate models, and the models can of course be “tuned” to re-produce the input data, in a broad sense (but they cannot do it perfectly). But the value of models can only be proved by making predictions of data we have not yet seen. Because, of course, this is independent evidence that the models work. And this will take several more years of observations to tell.

    What we can say is that the IPCC “scenarios”—a euphemism for forecasts—have not done well and have, so far, tended to over-predict. That is, they do not appear to have skill. Take a look at Roger Pielke’s blog for more on this;

    So, of course, since they do not have skill, people have abandoned them and moved on with other avenues of research, right?

    That’s a brief answer: there’s actually much more to say on the subject (skill is what a forecast has to have to be “statistically significant”). I’ll post more later.

    Briggs

  3. Mr Briggs,

    Perhaps this is a question that you may find worth investigating:

    Temperature trends are autocorrelated which means the future values are not completely random and depend on previous values. This should mean that a purely random time series will have discernible trends that will average out to zero over the the long term but could deviate pretty far in the short term.

    Is it possible to calculate the probability of a X degC/decade trend over Y years given a set of autocorrelation parameters for a time series? If so it might be a useful sanity check to determine whether the recent temperature trends could be explains as a probable outcome of a random process.

  4. Raven,

    Actually, work like this has been done. Many years ago I read a paper comparing temperature trends as limits of something called the “arcsine” law. That law shows what happens in the limit of “random walks”, or cumulative coin flips, if you like.

    In a random walk, sometimes you go up, sometimes down, sometimes up a lot, sometimes down a lot, and so on. If you were to plot the level with “time”, the curve the results can be described by the arcsine function (increasing wiggles up and down) and looks remarkably like the actual temperature time series.

    While that has some interest, it’s far better to try to explain why the temperature meanders as the result of physical processes. Only, that’s really hard, and the models used to do this aren’t that predictive. The “random”—aka unknown–forcings are still just as important as the known forcings.

    Briggs

  5. dear admin.

    the giss data is a collection of points. the points are from various geographic areas. you can output a curve of temperature anomaly by latitude. but i haven’t seen a graph of temperature anomaly per sq mile, nor a summation of global temperature anomaly per square mile. Shouldn’t the temperature data be normalized or corrected to a surface measurement? Otherwise 1 square mile of arctic counts the same as 100 sq miles of equator. Most of the anomaly is in the artic areas during their winter or the antarctic during its winter. there is very little temperature change between -30 and 30 latitudes, or in the artic during summer or the antarctic during the northern hemisphere winter. Doesn’t the lack of normalization to area really really bias those graphs you see of global warming?

  6. What a blot on the world of Science this whole area is; no doubt, when the orthodoxy changes, the same unscientific scientists will again float to the top – a bit like the effortless morphing of leading Russian communists into leading Russian capitalists.

    I guess that the saddest comment of all is that the word “skeptic” is now a pejorative term when applied to a scientist, instead of being a base-line, must-have quality. On such foundations are dark ages built. I wonder how witch-burning will be done in future – perhaps frying on a solar array?

  7. In my estimation, the reason AGW proponents refuse to be sunk by the torpedoes Dr. Briggs outlines is because AGW proponents also believe the earth is entering a cooling period ? and they want to be able to claim credit for it. AGW proponents hope to confuse the public by claiming correlation as causation.

    If the world economies continue business as usual and global cooling occurs, AGW proponents will have missed their window of opportunity to gain hegemonic control over the world?s economies. No global carbon tax scheme can be forced upon industrialized economies if AGW is proven false or insignificant.

    So many disparate groups stand to gain financially from AGW and its momentum is so great that it might be unstoppable. Academics gain through endless government and private grants to study AGW affects; the UN gets control of $billions to redistribute around the world; environmentalists get to destroy capitalism; and the worlds largest financial trading centers and businesses stand to reap untold fortunes in fees while administering carbon trading schemes. Dr. Briggs?s torpedoes are hitting a ship protected by an impenetrable wall of money. Too many powerful groups stand to lose if AGW is proven false or insignificant.

    AGW proponents understand they are in a desperate situation ? up against the clock ? and they must force their CO2 reducing schemes upon the world now, right now, or lose their chance to control.

  8. Louis,

    Good point: there can never be data that shows human activity cannot cause dangerous global warming. But from that fact you cannot infer that humans are causing dangerous global warming.

    Briggs

  9. You blog is really good….I have also a blog on global warming which give information about global warming effects, what is global warming, global warming causes, prevent global warming, global warming hoax, global warming articles, fight global warming, global warming news, stop global warming.

  10. At this point I’ve doine quite a bit of amateur reading of blogs AND of some papers over the last 3 years. I listened to Carter for the first 5 minutes and didn’t see much in terms of references. It seemed like PR, not like the aha of new ideas…or of a carefully layed out examination of what is known. I suspect that he uses the studies which help him the most, within that temp-time discussion.

  11. I really appreciate the skeptical attitude shown in the beginning; unfortunately the presentation degraded into a massive propoganda rather than actual scientific discussion. Many global warming indicators have been ignored. The truth on this subject will only become clear once the temperature does, or does not exceed either the magnitude or the rate recorded within an arbitrarily chosen period of time (whatever a specific scientist choses to prove his/her point). Regardless of that, telling me that there is no warming is ridiculous – I have my own eyes and see the changes over the past 30 years of my life that without a doubt point to warming of the climate in the Northern Europe and North America. Also, there is no question that the rate at which we are releasing fossilized CO2 into the atmosphere is high and increasing. This CO2 was sequestered during times with much, much higher temperatures. Computer models are useless – I agree there, but simple logic shows that one would expect some effect of that. Are there natural variations – you bet there are. Has it been warmer than now at some point in the past – everyone knows so – where are the news? It looks like in his displeasure with some questionable science surrounding many AGW promotion efforts Dr. Briggs misses the point of the discussion: how much we as humans can afford to destabilize the ecosysems on this planet before it shifts rapidly to the point that dramatically reduces it’s habitability? The simple answer is that we don’t know. Neither pro-Global warming, nor anti- GW flks have no idea. I will side with ones that urge caution. As my old mentor used to say, “don’t try to save the Earth, it will be fine. Worry about the people…”

  12. Hi.

    I’m trying to locate statistical process control (SPC) charts of temperature data from various sites around the world. I’m beginning to realize that this data hasn’t been compiled yet.

    My thinking is that the use of CO2 data from ice cores is too indirect a way to confirm global warming. If we were to take a “population” of sites around the world, and plot SPC charts on reliable data, a better case about global warming could be made.

    MIke

Leave a Reply

Your email address will not be published. Required fields are marked *