3 ms·

This may come down to our individual philosophies, but I'm far more interested in what would convince you that a model is wrong. It's a Popperian sentiment, whi

by guscost 11y ago

This may come down to our individual philosophies, but I'm far more interested in what would convince you that a model is wrong. It's a Popperian sentiment, which may be considered quaint these days, but I rarely hear any kind of answer. Anyway, I don't think we are talking about even close to the same meaning of "complex", but one possible situation that would help convince me is if we had a "control" planet, and an "experimental" planet, with nearly identical climates and biodiversity and so on, and then we raised the CO2 level of the experimental planet, keeping in mind that all kinds of other inputs to the non-linear system should remain the same, like land use, fishing, other industrial activities, etc. If the experimental planet diverged significantly from the control planet, it wouldn't be solid 100% irrefutable proof but it would be far better evidence than anything we have today.

drewrv 11y ago
>one possible situation that would help convince me is if we had a "control" planet So on this one issue it's impossible for you to be convinced.
- guscost 11y ago
  Not at all, there could be plenty of other evidence, I just can't imagine what it would be. Other than some kind of control planet, or maybe a complete overturning of nonlinear dynamics and/or chaos theory maybe. Of course, if the ocean was engulfing whole communities faster than people could run uphill, or people were choking on CO2, I'd probably be a whole lot less rational about it and say "let's throw things at the wall to see what sticks." But so far, and after reading what I figure are among the strongest arguments out there, it still looks like your typical doomsday story dressed up as science.
  - nitrogen 11y ago
    Okay, ignore warming if you like. We still need to stop producing so much CO2 because it's acidifying the oceans and destroying ecosystems we depend on for food.
  - lotharbot 11y ago
    the thing about nonlinear dynamics/chaos is that it still allows for some degree of control and prediction -- not in the sense of being able to predict one specific trajectory, but in the sense of being able to make statistical predictions about likely regions for the trajectory to pass through, and even to be able to apply small controls to change a trajectory drastically and potentially keep it out of entire regions. See, for example, http://onlinelibrary.wiley.com/doi/10.1046/j.1461-0248.2001.00223.x/abstract http://onlinelibrary.wiley.com/doi/10.1046/j.1461-0248.2001.... (yet another tribolium paper. The favorite insect of nonlinear dynamics and chaos books everywhere.) I don't think we need to overturn anything mathematically to be able to talk sensibly about climate models, predictions, complex forcing mechanisms, and control. We just have to stop putting the cart before the horse (or the shady politician before the scientist) and look at what can be done to stabilize the ecosystem -- not necessarily worrying about catastrophic predictions, but simply looking at some of the measurable negative effects (like acidifying oceans, brought up in a sibling comment) and determining how to control the system in order to slow or even reverse those effects.
  - jberryman 11y ago
    > I'd probably be a whole lot less rational... I think you imagine that you are being very rational, because you've studied some climate science and believe you're intelligent enough to form your own judgement, but you're wrong. It's extremely irrational to dismiss the consensus opinion of climate science researchers based on your own (epistomologically limited, non-professional-climate-scientist) assessment of their research. > it still looks like your typical doomsday story dressed up as science. And say the likelihood of this doomsday scenario being correct is 10% by your assessment: do you think a rational risk analysis would conclude that we should do nothing and go about business as usual? Do you also enjoy playing russian roulette when your not skimming climate science articles?
jfoutz 11y ago
No, i'm with you. Falsification is a huge deal. As far as "complex" goes, i'm thinking of the old strange attractor. I don't know if it'll rain on sept 19 2016, but i think it's safe to say there will be 3 inches of rain next september. If we refine our understanding and language, there's a nice probability density function for quantity of rain over the month of sept 2016. Not much probability around 0, not much beyond 6, but a big lump right there at 3. I outright reject the "no model is possible" argument. If nothing else, it's that way because god wills it. Some folks argue weather is related to frequency of gay sex. I'm confident the above paper outperforms that model. The nice thing about the above model is it is testable. we know historical variation around the prediction. We can make a prediction based on increasing CO2, then increase it, and see how our prediction plays out. (edit and we did, and the model was pretty good, but earth actually got warmer faster than the paper predicted http://www.realclimate.org/index.php/archives/2012/04/evaluating-a-1981-temperature-projection/ http://www.realclimate.org/index.php/archives/2012/04/evalua...) So, anyway, weather works "somehow" and we can make forecasts. Although it's never been tested, i'm pretty sure Pat Robertson is wrong and it's not gay sex driven. So, i still dont' get why it's not settled. The best models we have are refinements of the above paper. Is there some other model that gives better answerers?
- guscost 11y ago
  I'll agree that your favorite model is lot more likely to work than the "because gay sex" model, because your favorite model presumably has causal relationships based on measurable reality. And I like the idea that atmospheric composition could be a strange attractor in the climate system. But then you'd have to throw the simple forcing model for CO2 out the window, and come up with a hypothesis for why there is a strange attractor somehow causing CO2 and temperature to exhibit a linear relationship. Wouldn't your favorite model have to be rewritten at that point? And then you'd need an explanation for how the water vapor feedback forcing (a critical part of most alarmist models) fits into those dynamics. And why a similar albedo forcing or plant cover forcing or fracking-earthquake forcing or Chinese economic forcing does not. I don't think you will find it simple to describe at anything but the most abstract level. But it is an interesting idea.
  - 11y ago

by guscost 11y ago

drewrv 11y ago

>one possible situation that would help convince me is if we had a "control" planet So on this one issue it's impossible for you to be convinced.

guscost 11y ago
Not at all, there could be plenty of other evidence, I just can't imagine what it would be. Other than some kind of control planet, or maybe a complete overturning of nonlinear dynamics and/or chaos theory maybe. Of course, if the ocean was engulfing whole communities faster than people could run uphill, or people were choking on CO2, I'd probably be a whole lot less rational about it and say "let's throw things at the wall to see what sticks." But so far, and after reading what I figure are among the strongest arguments out there, it still looks like your typical doomsday story dressed up as science.
- nitrogen 11y ago
  Okay, ignore warming if you like. We still need to stop producing so much CO2 because it's acidifying the oceans and destroying ecosystems we depend on for food.
- lotharbot 11y ago
  the thing about nonlinear dynamics/chaos is that it still allows for some degree of control and prediction -- not in the sense of being able to predict one specific trajectory, but in the sense of being able to make statistical predictions about likely regions for the trajectory to pass through, and even to be able to apply small controls to change a trajectory drastically and potentially keep it out of entire regions. See, for example, http://onlinelibrary.wiley.com/doi/10.1046/j.1461-0248.2001.00223.x/abstract http://onlinelibrary.wiley.com/doi/10.1046/j.1461-0248.2001.... (yet another tribolium paper. The favorite insect of nonlinear dynamics and chaos books everywhere.) I don't think we need to overturn anything mathematically to be able to talk sensibly about climate models, predictions, complex forcing mechanisms, and control. We just have to stop putting the cart before the horse (or the shady politician before the scientist) and look at what can be done to stabilize the ecosystem -- not necessarily worrying about catastrophic predictions, but simply looking at some of the measurable negative effects (like acidifying oceans, brought up in a sibling comment) and determining how to control the system in order to slow or even reverse those effects.
- jberryman 11y ago
  > I'd probably be a whole lot less rational... I think you imagine that you are being very rational, because you've studied some climate science and believe you're intelligent enough to form your own judgement, but you're wrong. It's extremely irrational to dismiss the consensus opinion of climate science researchers based on your own (epistomologically limited, non-professional-climate-scientist) assessment of their research. > it still looks like your typical doomsday story dressed up as science. And say the likelihood of this doomsday scenario being correct is 10% by your assessment: do you think a rational risk analysis would conclude that we should do nothing and go about business as usual? Do you also enjoy playing russian roulette when your not skimming climate science articles?

jfoutz 11y ago

No, i'm with you. Falsification is a huge deal. As far as "complex" goes, i'm thinking of the old strange attractor. I don't know if it'll rain on sept 19 2016, but i think it's safe to say there will be 3 inches of rain next september. If we refine our understanding and language, there's a nice probability density function for quantity of rain over the month of sept 2016. Not much probability around 0, not much beyond 6, but a big lump right there at 3. I outright reject the "no model is possible" argument. If nothing else, it's that way because god wills it. Some folks argue weather is related to frequency of gay sex. I'm confident the above paper outperforms that model. The nice thing about the above model is it is testable. we know historical variation around the prediction. We can make a prediction based on increasing CO2, then increase it, and see how our prediction plays out. (edit and we did, and the model was pretty good, but earth actually got warmer faster than the paper predicted http://www.realclimate.org/index.php/archives/2012/04/evaluating-a-1981-temperature-projection/ http://www.realclimate.org/index.php/archives/2012/04/evalua...) So, anyway, weather works "somehow" and we can make forecasts. Although it's never been tested, i'm pretty sure Pat Robertson is wrong and it's not gay sex driven. So, i still dont' get why it's not settled. The best models we have are refinements of the above paper. Is there some other model that gives better answerers?

guscost 11y ago
I'll agree that your favorite model is lot more likely to work than the "because gay sex" model, because your favorite model presumably has causal relationships based on measurable reality. And I like the idea that atmospheric composition could be a strange attractor in the climate system. But then you'd have to throw the simple forcing model for CO2 out the window, and come up with a hypothesis for why there is a strange attractor somehow causing CO2 and temperature to exhibit a linear relationship. Wouldn't your favorite model have to be rewritten at that point? And then you'd need an explanation for how the water vapor feedback forcing (a critical part of most alarmist models) fits into those dynamics. And why a similar albedo forcing or plant cover forcing or fracking-earthquake forcing or Chinese economic forcing does not. I don't think you will find it simple to describe at anything but the most abstract level. But it is an interesting idea.
- 11y ago