UN Climate Change Summit -- Paris 2015 - continued

[Charles Anthony]

Guys,

Please avoid thread drift even if a fellow member provokes thread drift.

Thanks!

[hitops]

If you want to go through their methodology, it is here: http://static.berkeleyearth.org/papers/Methods-GIGS-1-103.pdf

But obviously that would take a long time and I haven't bothered to go through it. Kriging makes sense because it results in the estimates being 'best linear unbiased estimates' and from what I have read BEST tries to deal with differences between land and ocean sensitivity (I don't think Cowtan and Way do this).

Right so the paper states that kriging works if the underlying assumption "the true field contains

normally distributed random variations about a common mean" is true. I think we agree that this assumption is not satisfied in the arctic, as there are no land readings from which to establish a mean in the a region. I do not actually see anything in the paper addressing the arctic. They only talk about the antartic (and other sparsely monitored areas), where the methods may apply because there are in fact some reading stations there.

Incidentally I was surprised to learn that some stations show large variations which they presume to be errors (pg 3 second paragraph).

So I do not think this paper deals with the arctic issue. Actually it does not mention it.

Incidentally Judith Curry is one of the authors. She explains her objection to the Cowton and Way approach here:

http://judithcurry.com/2013/11/13/uncertainty-in-sst-measurements-and-data-sets/

Her conclusion is the same as the position I hold

"The bottom line remains Ed Hawkins’ figure that compares climate model simulations for regions where the surface observations exist. This is the appropriate way to compare climate models to surface observations, and the outstanding issue is that the climate models and observations disagree."

The main point that I cannot get beyond, is really this, that the models fail to predict observations.

It's not so much mitigating vs not mitigating, it's also how much. Trying to meet the 2 C target is probably a net negative to humanity relative to no mitigation. But some mitigation does seem preferable to zero mitigation, at least that is the result of integrated assessment models like DICE and FUND. Is it possible that there might be a net benefit? Yes, that is quite possible given all the uncertainty that exists.

I'm aware there are models, but those models will only work if we have high confidence in the damage functions or relationships that they describe. We do not. If we knew for example how much flooding or drought or species loss or increase yields etc might happen with a few degrees increase, we could plug those into the model. But we really just don't, and we have previous little to go on to try to estimate them. Previous predictions about increases in adverse weather or human costs have been proven wrong (the human situation has actually dramatically improved since the first report). I cannot think of a good reason they would start becoming accurate now, using the same tools.

Edited November 2, 2015 by hitops

[biotk]

as for your continued comments on Hansen, he also recognizes there is little to no gain to be had by trying to front 3rd gen nuclear - hence his more recent emphasis on safety concerns and 4th gen research/trial... your own linked 'Hansen et al' letter speaks directly to that.

Not true the letter speaks to both development and deployment. Nowhere does it say that it referring specifically to 4th gen. Nowhere does it say that it is only talking about deployment after future development.

As Hansen made clear during an interview at the time of the letter:

"The best candidate to avoid that is nuclear power. It's ready now. We need to take advantage of it."

thanks for answering the question; however, your answer, "Yes, I would have us do nothing about emissions reductions until we are serious about actually doing something."... rings hollow in the face of accumulated and accelerating emissions growth. Would you wait, what... 5 years, 10 years, 15... or more? While doing nothing else, waiting for a politicized acceptance of nuclear deployment?

A policy to do nothing is better then a terrible policy to march off in the wrong direction. We have already seen that Sweden and France achieved far, far more in a much shorter period of time despite no policy than any nation has done since with policies directed by anti-science ideologues.

There are plenty of other things that can be done instead of wrong-headed energy policies. They could use a fraction of the money to tackle malaria, HIV/AIDS, education and health care in the third world. All of which would help bring down reproductive rates and result in long term positive effects of emissions. They could use a fraction of the money to educate their own populations against the constant drum of misinformation spread by organizations that profit from creating baseless fear leaving the population in a better position to deal with problems in the future. They could invest a fraction of the money in R+D.

If a company goes to a government and tells them that they must approve a certain policy that benefits the company in question, but is bad public policy while being more expensive for achieving poorer results. I expect the government to say no - and do what it is the public interest. I expect the same thing when environmental groups are demanding bad public policy. If the government is not willing to accept their wrath and implement good, science and evidence-based policies, then I expect them to at least have the decency to not waste time and money on bad, fear-mongering and pseudoscience-based policies - and in the process rewarding the very groups who are responsible for the baseless fear and misinformation within their populations.

[hitops]

A policy to do nothing is better then a terrible policy to march off in the wrong direction. We have already seen that Sweden and France achieved far, far more in a much shorter period of time despite no policy than any nation has done since with policies directed by anti-science ideologues.

There are plenty of other things that can be done instead of wrong-headed energy policies. They could use a fraction of the money to tackle malaria, HIV/AIDS, education and health care in the third world. All of which would help bring down reproductive rates and result in long term positive effects of emissions. They could use a fraction of the money to educate their own populations against the constant drum of misinformation spread by organizations that profit from creating baseless fear leaving the population in a better position to deal with problems in the future. They could invest a fraction of the money in R+D.

Could not have said it better.

[-1=e^ipi]

Right so the paper states that kriging works if the underlying assumption "the true field contains

normally distributed random variations about a common mean" is true. I think we agree that this assumption is not satisfied

Such an assumption is rarely, if ever satisfied. But in order to quantify uncertainty some base assumptions about random variations need to be made. Assumptions of normally distributed random variables are common in statistics for a good reason (see central limit theorem https://en.wikipedia.org/wiki/Central_limit_theorem).The question is how badly violated the assumptions are (and there are ways of testing for violations). If the assumptions are not strongly violated then it isn't a big issue.

I think we agree that this assumption is not satisfied in the arctic, as there are no land readings from which to establish a mean in the a region.

Of course there are weather stations in the arctic. For example, Alert, Nunavut has one.

I'm aware there are models, but those models will only work if we have high confidence in the damage functions or relationships that they describe.

I disagree. The models don't require high confidence, and decisions can be made under uncertainty.

But we really just don't, and we have previous little to go on to try to estimate them.

We have probability distributions.

Previous predictions about increases in adverse weather or human costs have been proven wrong (the human situation has actually dramatically improved since the first report).

To be fair, those estimates were generally not rigorously done and often had biases. Estimates just need to be done more rigorously.

Edited November 2, 2015 by -1=e^ipi

[hitops]

Such an assumption is rarely, if ever satisfied.

But in order to quantify uncertainty some base assumptions about random variations need to be made. Assumptions of normally distributed random variables are common in statistics for a good reason (see central limit theorem https://en.wikipedia.org/wiki/Central_limit_theorem).The question is how badly violated the assumptions are (and there are ways of testing for violations). If the assumptions are not strongly violated then it isn't a big issue.

I agree, but this is just technical splitting of hairs. The point which I'm sure is not lost on you, is that the assumptions are minimally violated in densely sampled areas and in oceans (due to homogeneity), and greatly violated in the poles and to a lesser degree in Africa. Kriging is only proposed as a solution to those latter areas, not as a necessary interpolation to readings worldwide.

Of course there are weather stations in the arctic. For example, Alert, Nunavut has one.

Again splitting hairs, the point is that there are not enough such that more complex statistical calculations are required to try to deal with it, introducing uncertainty. I contend that the methods are inadequate to deal with those uncertainties, including Kriging and BEST. Especially since re-interpretation of the artic temperatures is in fact the entire basis for the proposition that there has been no pause.

This is the same view as shared by Curry, as previously linked.

I disagree. The models don't require high confidence, and decisions can be made under uncertainty.

A decision can be made with no certainty at all. But not a good one.

We have probability distributions.

To be fair, those estimates were generally not rigorously done and often had biases. Estimates just need to be done more rigorously.

All estimates will necessarily have huge biases on this topic, due to lack of data points to put into the estimate. The fact that we can construct a probability distribution does not automatically mean that such a distribution would be helpful.

The fact that we can do math to describe a situation or risk assessment, does not automatically mean that said math better informs us to any meaningful extent.

Edited November 2, 2015 by hitops

[-1=e^ipi]

A decision can be made with no certainty at all. But not a good one.

Decisions have to be made at some point, so you might as well use the best available information.

All estimates will necessarily have huge biases on this topic, due to lack of data points to put into the estimate.

Not necessarily large biases, but certainly large uncertainties.

[hitops]

Decisions have to be made at some point, so you might as well use the best available information.

Doing nothing also qualifies as a decision. There is no reason to assume doing something is always better than doing nothing. It is perfectly rational to not act when uncertainties are large.

Not necessarily large biases, but certainly large uncertainties.

That's what I meant. When confidence intervals are so large they cover all plausible values, they do indeed describe the situation, just not in any useful way.

It would be analogous to you needing a ride, and asking me what I can offer you. I then describe what I have for you as metallic, larger than 1 cubic centimetre, and including the colour black. And then you having to make a decision on whether that would be adequate to your transportation needs.

Edited November 14, 2015 by hitops

[bush_cheney2004]

Canada's brave new Liberal government has already gone lapdog for U.S. policy on binding limits from COP-21 negotiations. Any carbon reduction targets would not be enforceable as minister Mckenna did the usual Canadian deferral to whatever the Americans are/are not doing.

The announcement by Environment Minister Catherine McKenna could irritate host nation France, which wants any deal to be enforceable. That would be politically impossible for the administration of U.S. President Barack Obama, however, since it is clear the Republican-dominated Congress would not ratify any treaty imposing legally binding cuts on the United States.

http://www.cbc.ca/news/politics/cop21-canada-climate-change-deal-1.3341423

[Guest]

Just heard on the beeb that India is going to mine a billion tons of coal every year until 2030. One assume it's not for new buildings.

But, so long as we leave some of those oil sands in the ground we'll be okay.

[Argus]

Just heard on the beeb that India is going to mine a billion tons of coal every year until 2030. One assume it's not for new buildings.

But, so long as we leave some of those oil sands in the ground we'll be okay.

As I have said before, the idea that Canada spending billions every year to lower our CO2 levels is going to have any noticeable impact on world CO2 levels is infantile. We might as well just be spending it on fighting toenail fungus, for all the good it would do the environment (except spending it on toenail fungus would actually accomplish something). India alone plans to increase its CO2 emissions by he ENTIRETY of Canada's emissions every year for the next twenty years.

The only noticeable impact on spending all those billions will be that our health care will have less money, and that progressives will feel good about themselves.

Which is all, in the end, progressives care about.

Edited November 28, 2015 by Argus

[Smallc]

If they actually do that there's a good chance, according to the scientists at the First Minister's meeting, that Canada's temperature will rise, on average, by at least 5C.

[Argus]

If they actually do what? Spend billions on combating toenail fungus?

Edited November 28, 2015 by Argus

[Smallc]

Sorry, India.

[Argus]

Sorry, India.

India has already said expanding their economy is their main priority. And wait till China comes out of the economic doldrums and starts expanding at a higher pace again.

This is a waste of money. The only thing which is going to reduce CO2 emissions is scientific advancements in alternative energy and things like electric cars. Probably as important is the need to figure a way to store large amounts of power for later re-use. Once that's figured out then things like solar and wind power will be much more economically viable, as will electric cars.

[Guest]

India has already said expanding their economy is their main priority.

Justin's gonna have a word with them. That should sort it out.

http://news.nationalpost.com/news/canada/trudeau-to-meet-with-indian-prime-minister-in-hopes-of-convincing-him-to-reduce-countrys-emissions

[Hal 9000]

Justin's gonna have a word with them. That should sort it out.

http://news.nationalpost.com/news/canada/trudeau-to-meet-with-indian-prime-minister-in-hopes-of-convincing-him-to-reduce-countrys-emissions

This is my favourite line in the article:

Trudeau also said Friday at the Commonwealth that Canada will contribute $2.65 billion over the next five years to help developing countries reduce their emissions and adapt to climate change. India is a country that could qualify for some of that funding.

[On Guard for Thee]

This is my favourite line in the article:

Trudeau also said Friday at the Commonwealth that Canada will contribute $2.65 billion over the next five years to help developing countries reduce their emissions and adapt to climate change. India is a country that could qualify for some of that funding.

If you are going to try and rein in CO2 emissions, wouldn't it make a bit of sense to target/help one of the largest emitters?

[Hal 9000]

If you are going to try and rein in CO2 emissions, wouldn't it make a bit of sense to target/help one of the largest emitters?

You and Justin are adorable when you're this naive!

[On Guard for Thee]

You and Justin are adorable when you're this naive!

So I guess you'd just ignore India. Now we are talkin' naive.

[bush_cheney2004]

If Trudeau cares so much about climate change, he can pay for all the assistance to "developing nations".

[On Guard for Thee]

If Trudeau cares so much about climate change, he can pay for all the assistance to "developing nations".

Luckily he's not the only one who cares about the environment. Perhaps that's why Obama said no to Keystone.

[Guest]

So I guess you'd just ignore India. Now we are talkin' naive.

I think if they gave them the entire amount India might cut back on one or two of those 15 billion tons of coal they are going to use before 2030.

[Hal 9000]

I think if they gave them the entire amount India might cut back on one or two of those 15 billion tons of coal they are going to use before 2030.

There are two options for India; they don't take our money and continue to do as they very well please...or, they do take our money and continue to do as they very well please.

[On Guard for Thee]

There are two options for India; they don't take our money and continue to do as they very well please...or, they do take our money and continue to do as they very well please.

Have you forgotten already, Harper is out of power, there is actually a chance at cooperation between us and other nations now.