Mike, is it OK if I critique this article? This is a very poor attempt at forecast evaluation.

Fred, unlike Mike, I did read the actual Nature Climate Change article.

I will refer you to the following:

"As with IPCC reports since, the 1990 report did not define the
terms of application of the prediction in an observable context4.
Here we take it to refer to the least-squares linear trend of annual
mean temperatures projected over the 19902010 period, with
estimates of the actual global mean trend experienced obtained
from measurements of limited global coverage."

Whether the IPCC authors used a linear projection or one that is linear in its polynomials (non-linear deterministic) is immaterial. They are comparing a least-squares linear trend of actual temperatures with the projections. The standard method of evaluating a forecast is to compare actual realisations with the projections. For instance, their starting point is more than 0.1 degree more than that of the projection. If we start from the actual value in 1990, the last value falls within the non-significance region.

Furthermore, nonstationary time series are confusing for some, but the authors quite rightly point out that the series is what they call "red noise", which means that tomorrows value is equal to today's plus some indeterminate amount. Note that they state:

"climate time series generally resemble red noise processes, yet quantifying the degree of autocorrelation in the complex and nonlinear climate signal is not straightforward"

Since its not straightforward, just ignore it?

The authors also qualify their findings:

"As is always the case in science, we cannot know for certain that
the 1990 prediction was accurate for the right reasons"

which I agree with, and then:

"but, given
the apparent absence of any credible alternative theories and the
robustness of the prediction, this evaluation strongly supports the
contention that the climate is responding to enhanced levels of
GHGs in accordance with historical expectations."

which I do not agree with.

Mike, grow up.

http://icecap.us/images/uploads/OllierPaine-NoIceSheetCollapse-AIGNewsAug.2009.pdf

"The data shows that the 20yr temperature increase is only 70% of the IPCCs".

Peter, there may be a paywall blocking you from reading the article, but the actual temperatures in the diagram of the article begin in 1990 at more than 0.1 degree above where the projections begin. What they are comparing is a best linear fit of actual versus predicted temperatures. So if we start from the actual starting point, then the increase is actually closer to 0.25. Less than half the IPCC forecast and not a statistically significant increase.

Andrew, you should read the article. Maybe if you email Sunanda she will send you a copy?

For an expert evaluation of climate forecasting:

"Validation and forecasting accuracy in models of climate change"

Robert Fildes∗, Nikolaos Kourentzes

International Journal of Forecasting 27 (2011) 968–995

My impression is that the article perpetuates an excessively simplistic view of climate change - namely, that climate change is adequately represented by a roughly constant rate of global average temperature increase.

Andrew Vincent is correct to point out that this is unlikely; for example, here are a few relationships with which I might make a first attempt to project climate:

dQ/dt ~ k( [CO2] - [CO2th](Q) ),
Q = Q(atm) + Q(ocean + cryosphere), and
T = T(atm) = Q(atm)/M(atm)/Cp(atm)

where

Q = earth system heat content
= Q(atm) + Q(ocean + cryosphere)
dQ/dt = net rate of heat accumulation
[CO2] = atmospheric CO2 concentration
[CO2th](Q) = theoretical [CO2] that corresponds with earth system heat content Q.
Q(atm) = T(atm) x Mass(atm) = heat capacity(atm)

Whether Mr Vlachos's criticisms are or are not motivated by a Denialistic streak is irrelevant; what matters is that, in this case, there may be a certain veracity to his argument.

Later generations of projections may be less open to any such criticisms, in which case I would not such criticisms to be made.

Toby James: "the UAH satellite-based measurements from the high point of 2010 to the present show a drop of about 0.28C."

Not statistically significant. But you would never use that term except to regurgitate from a denialist blog.

Toby, you are not seriously trying to extrapolate meaningful implications from less than three years of data, are you? Surely, you are not that desperate.

What I'm suggesting Doug and Chris, is that choosing 1990 to 2010 is cherry picking. 1990 to the present tells a different story. Of course the author referred to the half-way mark (two years ago), presumably because it looked a lot more attractive than the period from 1990 to the present.

Chris, you should get out more - your denialistifacatory comment is so yesterday.

Toby James: "choosing 1990 to 2010 is cherry picking"

So, you think finishing with a double La Nina is NOT cherry-picking?

Oh, the irony. And what a hypocrite.

Look, old bean, no matter which way you slice it, on your very own graph the rate of increase 1990-2000 is markedly greater than the rate of increase 2000-2012 - while in the same period the emission of CO2 has continued to rise.
In 22 years you have got 35% of your warming, thats like scoring a double bogey.
You only have 18 years left, my advice is rush out and buy a great big SUV now and start pumping out cO2, or you mightnt make the cut at all.

Sean, you agree that trend over less than 30 years are too noisy to be useful, right? So comparing trends over two individual decades is meaningless, right? If still in doubt, look at this graph, which includes ocean heat content http://www.skepticalscience.com/graphics/Total_Heat_Content_2011.jpg

Makes your argument look a bit threadbare, old bean.

Not really, that graph does the basic statistical trick known as "suppressing the zero". Temperature anomaly graphs do that also, but at least you known what the zero point is when you look at them. With this graph I have no idea what the total heat content of the oceans is, so I have no idea if that is a big change or a small change.
The change could have been as tenth as big or ten times as big and the graph would still look the same.

"I'm only seeing .2 degreesC since 1990 in current data."

HadCrut3 is out of date and no longer considered accurate enough.

So you haven't heard of the ENSO cycle then?

"So you haven't heard of the ENSO cycle then?"

Thanks for answering the question "If it takes 30 years to measure climate then what exactly are you talking about when you the say the climate of (say) 2011?"

You have such a sharp mind.

David McRae,

You are another of the people who think only those who accept the view preached by the orthodoxy should be heard. What do you think should be done with the others? Exterminated?

Luckily for the world, your group of 'true believers' and alarmists is failing. This chart shows it: http://climatechange.carboncapturereport.org/cgi-bin/topic? Look at the activity time line. Notice the decreasing media interest over the past 4 years. Notice Doha climate chat is not even a blip. No a heart beat.

You attempts to ruin the world with you insane policies has failed - thank God (or more correctly, the US Congress, China and the rational people like Monkton and Nigel Lawson and others). It was the rational thinkers who without funding, got the message out that we were being duped by the massively funded orthodoxy.

What is needed is robust polices, not the ridiculous policies proposed by the orthodoxy scientists and their followers.

There is enormous uncertainty about climate science. Uncertainty about the problem is a given. Uncertainty about the proposed policy is inexcusable.

But that is exactly what we have. Enormous uncertainty about whether the proposed polices will make any difference to the climate. Most people do not believe that carbon tax and renewable energy will make the slightest difference to the climate. Most people also realise it is virtually impossible to get world agreement to penalty schemes like carbon pricing and emissions targets and timetables.

Bad policies:
- Penalty schemes
- Targets and penalties
- Taxes, penalties and restrictive trade policies
- Carbon pricing
- Renewable energy
- Smart grids
- Aid/bribes

Good policies:
- Reward schemes
- 'No Regrets'
- Free trade
- Globalisation
- Wealth creation for whole world
- Adaptation
- Remove impediments on low-cost nuclear
- Developed countries develop low-cost, low-emissions alternatives to fossil fuels
- Produce them commercially and compete to produce them for the world – competition to bring costs down and improve the technologies

Adaptation is the main component – it will provide improved conditions immediately, whereas mitigation may have an effect in 50 years and beyond.

Developing countries need free trade and infrastructure strengthening (i.e. improved governance), not aid. Giving underdeveloped countries aid is equivalent to giving an unemployed person the dole rather than skills development and a job. We give out aid while restricting trade. And we block development of the least cost way to reduce emissions. We have it backwards.

Peter, referring to Judith Curry as an authority is almost legitimate, even she is a fringe dweller, because she has appropriate qualifications and has published in the field - refer the BEST project, that confirmed warming is occurring.

Appealing to Chris Monckton as an authority, however, deflates your credibility to zero. He has no qualification of any relevance and has not published in a reputable journal on any climate-related topic. His only claim on the media spotlight is that he inherited a title from his father: for some reason, some people are awed by the fact that he is a lord of the realm.

Please try to call on credible evidence when making your arguments.

Doug Hutchinson,

You say: "Appealing to Chris Monckton as an authority, however, deflates your credibility to zero."

That is in the opinion of those inside the tent. And there in lies the problem. Those inside the tent have only one narrow perspective - their belief. They suffer from group think. They think only they understand what is important. Unfortunately, they do not understand the real world we live in. They do not have the much broader perspective of economics and policy formulation. Monkton does. Get it? (I suspect you wont understand the point. Most inside the tent don't.)

Peter, thanks for the reply. From the context of your comment, I took the impression that you were relying on Chris Monckton as an authority on the science of climate change, but I see you have qualified that to you mean you meant he is an authority on policy. I have no knowledge of his credentials as a politician, but I am glad we agree that he has no place in a discussion of the science.

Unfortunately, Dave and Doug, there is a scientific field called forecasting that has emerged from statistics. A good answer to the question given in the article could be given by a forecaster, or someone would publishes in that field. Climate scientists cannot invent their own forecasting methods or evaluation methods, and there is no rationale for those methods to be different whether you forecast temperature, price, or the size of tumors. They also cannot hide in their tent, as Peter calls it, and expect that the rest of science does not exist and is not relevant. The authors of the Nature Climate Change papers are not experts in forecasting. The authors of the following paper are, and have published in a leading forecasting journal:

"Validation and forecasting accuracy in models of climate change"

Robert Fildes∗, Nikolaos Kourentzes

International Journal of Forecasting 27 (2011) 968–995

Unfortunately, Spiro, weather has forecasts; climate has projections. Totally different animals, as I'm sure you know.

Doug, are you being comical, or do you not know?

A projection is a mathematical operation, often used in forecasting.

Even Monckton knows that!

Spiro, are you claiming that a climate projection derived from models = a mathematical projection = a forecast, or am I misunderstanding you?

@Spiro says

"Climate scientists cannot invent their own forecasting methods or evaluation methods, and there is no rationale for those methods to be different whether you forecast temperature, price, or the size of tumors"

Are you doing your best impersonation of Tim Curtin? That would have to be the among the most absurd claims made by any poster here since TC claimed that only econometric models explained the climate.

You have not heard of physics? Or GCMs?

Maybe you would like to start here.
https://theconversation.edu.au/explainer-climate-modelling-3162

Thats hilarious Mike! Lets do a forecast (or linear projection?) using Physics. No statistics, no mathematics, just physics.

Astounding! How to argue with that?

@Spiro says

"Lets do a forecast (or linear projection?) using Physics. No statistics, no mathematics, just physics."

Spiro. You are embarrassing yourself. Do yourself a favour and do a science for non-science majors course.

@Mike says "You have not heard of physics? Or GCMs? "

As I said, to make a theory you need Mathematics, to make a forecast you need Statistics. What is it about that you do not understand Mike. Are you an adult? I don't want to be arguing with children here.

Doug, a projection is a method used in forecasting. If you want to estimate what the value of anything is in the future, you are forecasting.

Spiro, now you have argued yourself right round in a circle and you are back to equating a climate projection with a forecast. At least try to be consistent when your are being contrary. I give up.

@Doug

You are quite correct. Spiro is either being deliberately obtuse or he simply does not understand.

The temperature projections from climate models depend on different emission scenarios which is why they are called projections and not forecasts.

http://www.abs.gov.au/websitedbs/a3121120.nsf/home/statistical+language+-+estimate+and+projection

Mike and Doug, you both don't have a clue about stats, but to inform you, the Nature Climate Change article used as a reference the article, by the same author David Frame:

Frame, D. J. et al. Constraining climate forecasts: The role of prior assumptions. Geophys. Res. Lett. 32, L09702 (2005).

But, you two have no purpose in enlightening yourself.

Thanks, Mike. I was starting to get giddy from the circular argument Spiro has been putting up. I am sure he just does not realise there is a life beyond statistics and most of us dwell there.

No worries Doug, if there "is a life beyond statistics" go and find one.

And now we have skreeds of denier cut and paste rubbish .. well done Conversation

The cranks talk of orthodoxy.

Paedophilia is good for kids
Vaccination bad
Fluoride in water a conspiracy to program us
Climate scientists conspire to tax us

It's all the same. These creeps demand, and get, greater representation and respect than scientists. We cannot talk of an AGU lecture pertaining to prediction in a comment thread on an article of prediction without it being sunk in a sea of cut and paste anti-science crankery.

Dave McRae,

Could it ever occur to you that it might be the quankery of the CAGW alarmists that has brought so called 'climate science' (or is it Climate Scientology) down to the level of contempt is it now held in. Climate scientists have been exaggerating, misleading and misinforming the population and pushing Left wing agendas that are repugnant to most people. They will not answer straight questions and instead engage in long rubbish comments to divert from answering the question.

I've asked repeatedly on may threads and web sites for a solid clear explanation and quantification of the impacts of AGW. It is never answered directly with unemotional, quantitative answers, without scary adjectives. I want an engineers and economists type of straight, quantitative answer. The question has been avoided for so long, I know I will never get a straight answer. I also recognise the evidence of catastrophic consequences doesn't exist. So it boils down to a cost-benefit analysis and robust decision analysis methods.

Before you dive in with some simplistic answer, that is not an answer at all, you might like to have a look at:
http://judithcurry.com/2012/12/01/open-thread-weekend-4/#comment-273000
The comments continue down to the first comment by Chief Hydrologist.

Dave Mcrae,

Further to my previous comment, to justify high cost mitigation policies, there needs to be strong evidence of the supposed catastrophic impacts of warming. After 20 years of scaremongering, we still cannot get clear, well quantified, scientific, definition of the impacts.

Uncertainty about the problem is a given; uncertainty about the chosen solution is inexcusable. This is to say, we should be confident that our solutions are going to be effective, and the more expensive the solution the more confident we should be.

http://www.intrade.com/v4/markets/contract/?contractId=707800

LOL @ Spiros.

So far that intrade has three shares.

Since 2010.

Not much of a sample size.

http://www.intrade.com/v4/markets/contract/?contractId=706206

That one is a bit more liquid. A 72% return if it is realised.

Cliff, no need to get all religious now. The article was about climate forecasts.

No, Spiro, the article was not about climate forecasts, it was about climate projections. Please read the title.

Spiro, religious?

As an atheist, that would be close to being the last thing I'd be.

Sorry Cliff, I had just noticed the pronoun "Nature" in your post and made an assumption.

Dave Phillips: "I will also draw your attention to a link I provided further up in this thread regarding Antarctica and Greenlands LACK of ice sheet melting"

That newsletter article says nothing about measurements of ice loss in Greenland. Several years ago (which was cooler in the Arctic than this year), Greenland was losing an average of 195 km3 per year (citation at http://en.wikipedia.org/wiki/Greenland_ice_sheet#The_melting_ice_sheet ).

Saying that Greenland lacks ice sheet melting is just garbage.

Spiro, I suggest you get some literacy, if you think the title reads 'forecasts' instead of 'projections': as we agreed elsewhere, these terms are not synonymous.

Doug, a projection is a type of forecast. Always was and always will be. Read the Nature Climate Change article and then go to a university library and look up James D Hamilton "Time Series Analysis" and you will find a chapter called Forecasting with a subsection Updating a Linear Projection. Then come back and tell me if you do not agree that a projection is not a forecast.

Spiro, you said "tell me if you do not agree that a projection is not a forecast". You are arguing in circles. I am the one saying a climate projection is not the same as a forecast. You are the one saying this article has 'forecast' in its title. Do you have a point with all this convoluted argumentation?

One does not have to be identified in a comment to be aware that the comment is directed at an individual, The hype I refer to is the Al Gore, we are all going to die a horrible death hype, as opposed to articles and publications from both sides of the issue. You say the link i provided you will not read, then you consider the good folk at the University of Western Australia not worthy of 5 minutes of your time, or is it because it may contain information you would prefer not to hear? So the document I refer to is an academic paper therefore not full of hype and hysteria, so often the case with climate change and global warming "activists" I say activists verses scientists. who it is disappointing to learn have themselves succumbed to a bit of hype and hysteria to get their message across. Again I emphasise that how we treat our planet is important, particularly how we are killing our oceans and rivers and lakes, so in that we are in agreement. However in the context of this article the document you will not read is a balancing or counter example of things happening to our environment regardless of our input. I again emphasise that man is his own worst enemy and we will be long gone before the doomsday advocates have us frying under a blood red sun. We will be gone long before that as we will have destroyed the source of life here on earth, the ocean.