Friday, 14 October 2011


This article explains that scientists are now predicting the prospect of up to thirty years of cold winters. Being a sceptic I tend to take any weather forecast with a large pinch of salt, but even if it has any truth in it, it should spell the end for the idea that CO2 controls our climate. For a more in-depth look at this this article from the excellent GWPF is worth reading.


Dol said...

Hello. Finished PhD now, up for some good solid climate discussion. Having followed the GWPF link, it then links to a met office study, which is misrepresents. This is something I see repeatedly and don't understand - the argument appears to go like this: 1) we've found that some other factor (in this case an 11-year solar cycle) has an effect on climate, therefore 2) all other factors don't matter, or possibly 2a) that proves how uncertain everything is.

But the met office article GWPF links to says itself: "Compared with the effect of man-made emissions over the last century, solar variations still have a very minor effect on long-term global climate trends, but this study shows they may have a detectable influence on winter climate."

I was listening to an interview with Paul Nurse a couple of days ago: he was discussing his 3rd year undergrad science project, where he was trying to detect the effect of cell division on temperature, I believe. He found a signal he was expecting, an oscillation, and was very happy - but then thought he'd better do the controls. So he checked again without the cells - and found the signal again. Turned out he'd just been detecting the thermostat turning off and on. Since then, he always got his controls in early.

The point: you're citing the sun, and claiming it "spells the end for the idea that CO2 controls our climate." How are you so sure? You've weighed all the different factors that control climate, and concluded CO2 has only a tiny effect?

Dol said...

p.s. another question, if we're looking for some good starting points: do you agree with this other david whitehouse article that there's been no warming since 1998?

Derek Tipp said...

Hello Dan and welcome back after your studying.

On your second post, I accept the data which confirms that there has been no "statistically significant" warming, as agreed by the infamous emailer, Dr Phil Jones of CRU.

What I am saying is that CO2 does not "control" climate. I accept that it does have an effect, but it is one of many, and there are others which can override it - hence the lack of correlation between temperature and CO2.

Dol said...

Thanks for getting back to me. How's your stats? I'm currently OK with working out significance for non-time-series data, but I'm about to embark on working out the stats for just that problem. Is that something you're already familiar with? If not, fancy working through the problem with me?

Similarly for co2: if you have time, I'd like to find out what you think the factors are that control climate and temperature, and what the proportionate control exerted by co2 is. I'm not sure quite where we start the conversation, but if you're up for discussing it more, it'd be great to work through it.

Derek Tipp said...

Well Dan, on your first point, I have to admit that my knowledge of statistics is extremely rusty to say the least. Steve McIntyre is the guy to help with that, on his Climate Audit Blog.

As to the different factors which affect climate, let me refer you to the infamous IPCC report as a starting point. This link takes you to WG1 and if you load up the report (about 7Mb) you can then scroll down to about page 202 you will find a table there which lists about 16 factors. Interestingly in the table they list the level of scientific understanding (losu) for each one and apart from the first four or so, the others are either medium/low or low. So how anyone can claim that we have any real understanding of the extent to which these factors contribute, I just don't know. My argument is that we just don't know enough, and it is not possible to say that CO2 plays a significant role.

Dol said...

You mean this famous graph of IPCC forcing factors? Yes, there's uncertainy, but it's highly quantified uncertainty. What's your problem with uncertainty?

Pretty much the whole scientific endeavour is about quantifying uncertainty. Just because life is uncertain, does that mean you don't take out insurance? If you do (or if e.g. you think companies should) why should that differ to your attitude to climate change?

As I read 2.9.1, LOSU is not used in that document: "The concept of LOSU has been slightly modified based on the IPCC Fourth Assessment Report (AR4) uncertainty guidelines. Error bars now represent the 5 to 95% (90%) confidence range (see Box TS.1). Only ‘well-established’ RFs are quantifi ed. ‘Well established’ implies that there is qualitatively both suffi cient evidence and suffi cient consensus from published results to estimate a central RF estimate and a range." (p.199-200)

Derek Tipp said...

Insurance is no use if the cost of it is so high and the risk is not clear and likely to be low.

I love your term "quantified uncertainty". This sounds like an oxymoron. Do you mean that we can state how uncertain we are? Sounds a bit of a Donald Rumsfeldian statement.

Dol said...

"Insurance is no use if the cost of it is so high and the risk is not clear and likely to be low."

You've kind of made my point for me there - you've attempted to quantify the uncertainty and make a decision based on it. Don't get too hung up on the specific example of insurance - I was only using that to suggest no-one makes binary decisions based on whether something's 100% certain or 100% uncertain.

'Quantifying uncertainty' has been going on since about 1733 when the normal distribution curve was first discovered. Concepts from that were very quickly used to, for example, help quantify measurement error in a precise way.

Some people have recently been arguing that scientific uses of the term 'uncertainty' are confusing the public, and perhaps 'range' should be used instead - I happen to disagree. It's not a complex concept to understand what the most likely future outcome is, that a range around the most likely exist, and that we can quantify that.

But I guess, then, we come back to what you were implying in your original point: you believe, I think, that the costs are going to be too high compared to the benefits of taking action. Is that right?

I'm still troubled by the point about the IPCC's attempt to quantify uncertainty, though. You were arguing there were so many factors, it was all too uncertain. That's a very different argument to saying, the costs will be too high. I think perhaps you're saying - as you did with the insurance - your own conclusion on the uncertainties is that the risks are low. But doesn't that contradict your point about saying it's too uncertain? High uncertainty = high risk: we have less idea what the outcomes will be. Surely you'd want to insure yourself in that situation??

Dol said...

Nice related quote from an interview with Gavin Schmidt:

"Q:I hear you talking a lot about uncertainty, and that’s something a lot of people are paralyzed by: they don’t want us to take these actions because they think everything might be fine on its own. What’s your response to that attitude?"

"Any decision that you’re making now that has to do with the future is uncertain. We make decisions all the time: where to invest money, whether to buy a house – these things aren’t certain, and we still have to make decisions. The issue with climate is that no action is a decision in and of itself. That one is actually laden with far more uncertainty than if we actually try and produce energy more efficiently, try and use more renewables, adjust the way we live so that we have a more sustainable future. The uncertainty comes with what would happen if we don’t make a decision, and I find that to be the dominant uncertainty. But climate change is not unique in having to deal with decision making under uncertainty. All decisions are like that. It’s nothing special about climate change in that there’s uncertainty about what’s going to happen in the future. Any time we decide to do anything, there’s uncertainty about the future, yet we still manage to get out of bed in the morning."

Derek Tipp said...

There is another way to deal with an uncertain future and that is to react to events by adapting. This seems a very sensible option when dealing with climate. If we have c town near a river we can prepare for a flooding incident. If sea levels were to start to rise more steeply than in the recent past we can build defences.

Interestingly the people of San Francisco do not all move away even though there is a strong likelihood of a serious earthquake there in the next century. They have prepared by trying to 'quake-proof' their buildings, but apart from that they simply carry on and ignore the risk. The same is true of many other towns and cities with similar fears.

Even if reducing CO2 was proven to be essential to save us from catastrophic climate change(and it has not), the only way this could be achieved is by a world-wide binding agreement by all major industrial powers. I seriously doubt that such an agreement is possible.

Dol said...

Yes indeed, the old adaptation vs mitigation thing, and weighing costs. So: if climate change loads the dice towards more extreme weather events, you're saying everyone should just adapt? Taking recent US history as an example, that the various extreme weather events they've had (whose probability will increase as co2 goes up) - they should just adapt?

I mean, we will *have* to adapt to what's already in the pipeline. What I wonder, though, is how you and me have a conversation about what we think is a reasonable cost to pay?

What about the fact that those people least able to afford to adapt are the ones most likely to see the impacts?

On one point we're in agreement: I am also mightily skeptical that we stand any chance of building governance structures capable of controlling our co2 output. But I suppose the difference between you and me is that I have to hope we find a way, because just ignoring our increasing co2 output will mean it'll still be increasing as a percentage of the atmosphere up to 2100 and beyond.

I'll have a think about how we can (both as interested amateurs) effectively discuss the cost-benefit thing. Thanks for the dialogue.