Monday, April 28, 2008

Has global warming stopped?

This has been done to death all over the blogosphere (eg here, here and here), so I can hardly bring myself to post about it. However, I do have one minor contribution to make.

First, the basics. Most of you will have noticed that various sceptics have been crawling out from under stones with variations on "global warming stopped in 1998". Of course, the temperature in 1998 was exceptionally hot, due to the strongest El Nino on record. And according to HadCRUT, we haven't quite exceeded that temperature yet (although according to the NASA/GISS analysis, 2005 just pipped it).

Here is the HadCRUT surface temperature record, with 30 consecutive 10-year straight line fits overplotted in blue. Note how the last of these, 1998-2007, still has a positive trend despite starting right at that cherry-picked high point. However, there have been 3 different 10y intervals with lower trends even during this period, with the more recent one covering 1987-1996. That does include the 1992 Pinatubo eruption, but even correcting for that would still leave it no higher than the 1998-2007 trend. Anyway, global warming can hardly be be described as having stopped since 1998 when the trend starting in that year remains positive.

One of the sceptics who has waffled on about global warming having stopped is is a person called David Whitehouse, who I must admit I'd not heard of prior to that linked article. Mark Lynas has written a perfectly adequate rebuttal, so we don't need to dwell on the argument any longer. But thanks to "More or Less" (today at 16:30 on Radio 4, I just discovered) I may have a 100UKP bet with David that there will be a new record temperature by 2011 at the latest. As I said in the interview (whether or not it ends up on the cutting room floor), I don't think this is a sure win for me, but I do think that the odds are clearly in my favour. Even writing off 2008 (since it's started a bit cold), that leaves 3 more years, and if each one has only an independent 30% chance of beating the old record, then that gives me a 0.73= 35% chance of losing and therefore 65% chance of winning. That's without any careful analysis of whether an El Nino is likely to turn up (surely the next decent-sized event will result in a new record), just from eyeballing the graph. The Smith et al Science paper puts the odds better than that, I think.

18 comments:

Silver Fox said...

Well, I'd vote for a little cooling just to get some relief - but I think the likelihood of me winning a bet like that would be kind of low. Yours sounds about right.

crandles said...

I think I would prefer your side of the bet but maybe only just.

Where does your 30% come from?

Since the record in 1981 I counted 5 records in 24 or 26 years. That is more like 20% than 30%. We are in a La Nina episode so maybe that makes an El Nino the next likely phase. OTOH we have had 3 El Nino episodes since 2001 so maybe that indicates we a due a long period of La Nina and we there have been a couple of periods where La Nina has dominated for 3 years.

As you said the Smith et al decadal forecast does help. (However your post seems to indicate the probabilities derived are not much better than you could arrive at with a statistical analysis. This raises the possibility that you are using it when it suits you and dismissing it when it doesn't.)

My opinion doesn't count for much but are the odds that "clearly" in your favour? Perhaps the odds are clearly in your favour but it appears to me that you simply haven't backed up that 30% assessment well enough to show me that it is "clear".

James Annan said...

Chris,

I didn't plot (or explain) all my various investigations, but if you try taking the trend over a longer time (which makes it more reliable and less dependent on endpoints than the 10y trends I showed) it is clear that the underlying trend line is already close to the record, and maybe even exceeding it by 2011. There is also, of course, model output (not just Smith but all the IPCC stuff) showing that a linear trend + noise is a pretty good approximation to the climate system under this forcing and over these time scales.

Chuck said...

But we had snow down to 900m last weekend, so global warming *must* be over.
;)

Rick said...

Why don't you take your graph back into the 1930s? Four of the ten warmest years on record occurred during that decade. If you want credibility, you can't pick a comparatively cool decade as your starting point.

James Annan said...

Rick,

Don't be silly.

Corbett Kroehler said...

Measured as a global average, the month of March, 2008 was the HOTTEST month of March EVER. Global warming continues to accelerate and its correlation to global cooling is very thin. They are seperate phenomena which are not mutually exclusive. For a very simple explanation of the current state of global warming, watch Mr. Gore's update to his Oscar-winning film. It's free and only takes 30 minutes. The URL is HotAsVenus.com

Corbett Kroehler said...
This comment has been removed by the author.
James Annan said...

That would be an interesting factoid were it true, but I see 7 warmer Marches in the last 20 years...

Corbett Kroehler said...

Remember, we're talking about a global average. Besides, the government of the United States has an incentive to report GOOD news.

http://www.sciencedaily.com/releases/2008/04/080418112341.htm

Hank Roberts said...

I've tried to find whatever NOAA source the ScienceDaily people were relying on, and haven't yet.

NOAA links to NCDC
http://www.ncdc.noaa.gov/oa/about/whatsnew.html

The latest there is this:

"Climate of 2008: February in Historical Perspective
March 13, 2008

The combined global land and ocean surface temperature was 0.68°F (0.38deg;C) above the 1901-2000 long-term mean for February, 15th warmest on record."

I've tried bothering ScienceDaily to provide a cite to their source and even (gasp) to consider checking their sources before ... oh, who am I kidding ....

Hank Roberts said...

I've been trying to track down ScienceDaily's source. So far no luck. Anyone know where they got it?

I'm curious if it's SD's website, or NOAA's press release, that makes the claim reported.

NOAA links to NCDC, and NCDC's "news" has nothing about this yet.

mistermucky said...

Surely the really important thing to establish is not weather the climate is warming, but whether it is man made CO2 that causes the change. We know that climate never stands still, but is CO2 a factor or not? We are told over and over that this has been proved, even though it most certainly has not.
The carbon lobby constantly point to the thirty odd years when CO2 rose, and climate warmed, from the sixties to the late nineties. But this is all the evidence we have. All the rest is unproved theory.
But against this slim evidence there stand the period of cooling, up to the mid sixties, while CO2 climbed, and the last 10 years, where rising CO2 caused no rise.
So the link between CO2 and climate is not proven at all, indeed the evidence against the theory is almost as strong as the evidence for. It's ludicrous to call it proved beyond further debate.

James Annan said...

I do feel honoured to be the recipient of such drive-by postings of random drivel...

mistermucky said...

Just non specific abuse, without any detailed rebuttal, gives a strong clue to your debating skills. A child of five could manage that.

James Annan said...

Why don't you come back when you've read the recent IPCC report...

mistermucky said...

Because I've now got a good idea of the standard of debate here.
Not worth my time. Good luck with it.

James Annan said...

It may come as a surprise to you, but I'm not interested in wasting my time "debating" with a random anonymous troll in a corner of my blog archive.