Not sure whether to put this here or in GQ, but I figure it’ll end up here anyway, so I’ll phrase it as debate.
So, for background I’ll refer you to Popper’s essay on falsifiability. Basically the Popperian standad of science is “the criterion of the scientific status of a theory is its falsifiability, or refutability, or testability.” IOW a theory is only scientific if “the theory is incompatible with certain possible results of observation” Or to put it another way the theory should predict something, and if that prediction fails then the theory is considered wrong. A theory that doesn’t do this isn’t science.
Now there are many other non-Popperian definitions/standards of science. But the Popperian definition is the most widely accepted. It is also the one with the greatest utility when trying to separate psuedoscience such as Astrology form real science, and this is why it is invariably the definition used in the various court cases concerning Intelligent Design. So I’m not looking for debate on the Popperian definition itself. For the purposes of this debate I’m accepting that the Popperian definition is useful and accurate.
So, what possible results of observation are incompatible with the theory of AGW?
Not the fact that the temperature is increasing in line with CO2 emmisions.
The increase in temprature is the very observation that AGW theory was created to explain. For the purposes of this discussion I want to accept as axiomatic that small-c climate change is occurring. Temperatures are rising, they have been rising steadily since since C1860 and more erratically since C1600.
Not the link to CO2 levels. Atmospheric CO2 levels have been rising in a nearly perfect exponential fashion since C1600 and show little to no evidence of recent increases. Moreover historicallly CO2 increases followed temperature increases, just as they have in the present warming event. Cause can’t follow effect. this isn’t to say that CO2 doesn’t cause warming, but the mechanism by which CO2 levels increased started in 1600, well before the invention of the internal steam engine so if a lack of correlation falsifies AGW then it has already been falsified. Now if someone had predicted that temperatures would continue to rise paralleling CO2 increases that would be acceptable of course. But the models don’t; do that. they all work on complex interactions. that’s not a bad thing in itself, but it does mean that the mere fact that CO2 levels increase roughly inline with temperature isn’t a falsifiable prediction because if they don’t the hypothesis isn’t rejected. For example, for the last 11 years temperatures haven’t risen at all despite record atmospheric levels of CO2.
What is being debated is the theory that the rises are anthropogenic. The IPCC notes that all the rises up to 1950 were just as likely to be natural as anthropogenic. So I am working here with a null hypothesis that human activities have made no detectable difference to global temperature. IOW that temperatures had been rising up to 1950 as a result of natural factors and have continued to rise as a result of natural factors.
Now I understand this is a complex field and relationships aren’t simple. But the question isn’t why predictions can’t be made. The question is whether predictions have been made. It’s not science if the theory isn’t incompatible with certain possible results of observation.
In previous threads people have suggested that Hansen’s 1988 models qualify as falsifiable predictions. The problem I have with them is that they predict everything. They cover such a wide range of scenarios. If the temperature trend had continued unchanged then model scenario B would have predicted it. And if temperature had increased scenario A would have predicted it. And if temperatures had levelled off then scenario C would have predicted it. IOW the model predicted any plausible outcome that might have occurred if global temperature changes were purely natural. To paraphrase Popper “I could not think of any climate behavior which could not be interpreted in terms this model”. And as Popper notes this is not scientific. A theory that predicts everything predicts nothing. Hansens models are not incompatible with any possible results of observation that we might get if climate change is entirely natural.
Lest you think I don’t understand how complex this is, I’ll let you in on a bit of my autobiography. I’m an ecologist/plant physiologist. I’ve been employed as a scientist to study climate change, and I’ve been published on the subject in Global Change Biology, the premier journal in my field. I’ve collected data and constructed models. I understand that it’s difficult. But I also know that I have never needed to engage a hypothesis that wasn’t falsifiable. Even though most of my work has been conducted in natural ecosystems where control is impossible I know that you can always construct falsifiable hypotheses if you are engaging in good science.
So I ask those of you who accept AGW and who believe it is good science: what evidence could I show you that would lead you to reject AGW? Obviously a prolonged period of cooling despite an ioncrase in atmospheric CO2 levels won’t do it. In the last 10 years global temperatures have been stable, and for the past five years the temperature trend has actually been negative fig 1. We’ve already had recent papers stating that temperature may not increase over the next decade, but nobody is predicting that they must decrease. IOW this isn’t itself a falsifiable prediction. In fact this makes me even more skeptical of AGW as a theory. It really does predict everything, since it can be used to predict prolonged periods of global cooling as well as periods of global warming.
To pre-empt questions of what evidence I would accept that would lead me to accept AGW. I’ll accept it when someone can show me a peer reviewed paper that 1) Provides replicable evidence for its conclusions 2) Presents a prediction that will lead to AGW being rejected. 3) Analyses observations and concludes, based on a previously acknowledged statistical test, that there is a >95% chance that human activities have produced a measurable effect on global climate. Not a 95% estimate but an actual 95% probability. That’s the evidence that I require to accept everything else as science. AGW isn’t allowed a lowered standard.
To pre-empt comparisons with evolution and other similarly complex topics that have been proved by a body of evidence, I’ll point out that Darwin gave a falsification 100 years ago: if any structure can be shown to exos that couldn’t; derive form previous structures. To which I will add that if anyone can show me a bird fossil in pre-Cambrian sediments or show me that dog DNA is less similar to human DNA than to jellyfish DNA I’ll also consider it falsified. It isn’t any harder to produce falsifiable predictions for theories built on a body of evidence. It’s usually much easier.
So the GQ and the spark for debate is: what evidence could I show you that would lead you to reject AGW? One again I’d like to stress that the question isn’t why falsifiable predictions can’t be made. The question is whether falsifiable predictions have been made. According to Popper it’s not science if the theory isn’t incompatible with certain possible results of observation. So what possible results of observation could I show you that would be incompatible with AGW?