PDA

View Full Version : Question of pjclarke



DimPrawn
18th May 2011, 13:43
http://kenskingdom.files.wordpress.com/2011/04/co2-vs-t-1979-111.jpg

http://members.shaw.ca/sch25/FOS/GlobalTroposphereTemperaturesAverage.jpg

CO2 is rising as you would expect. Lots of Chinese and Indians becoming westernised for example.

However, global temperature hardly moves.

Where's the hockey stick, where's the positive feedback meaning large rises in global average temperature?

DimPrawn
18th May 2011, 13:57
Did global warming stop in 1998? (http://www.skepticalscience.com/global-warming-stopped-in-1998-intermediate.htm)

I'm not saying it stopped, it never even happened.

BlasterBates
18th May 2011, 14:03
If you look at the current global temperature today:

Daily global average temperature (AQUA cho5 v2) (http://discover.itsc.uah.edu/amsutemps/execute.csh?amsutemps+002)

Compare 2011 to the average of 30 years. Tick the Average box and press redraw.

It's smack on average.

Then add 2008 (slightly below average) and then add 2009 (slightly above average). Then add 2007 (about average).

Only 2010 is high.

BlasterBates
18th May 2011, 15:15
Good to know the temperature at 14,000ft is bang on the average since 2002,when AQUA was launched. Not sure it tells us ground-dwellers much about the longer term impact of GW.

Hey - rather than cherry-pick 5 months cooled by La Nina, why not tell us what the actual 30 year trend in the near-earth satellite data shows?

and if it is bang on average or slightly below average what does that tell us about the trend after 2002?

Bonus point if you get it right.

A clue is to look again at Phil Jones answer on a question he was asked.

Now what was it he said?

EternalOptimist
18th May 2011, 15:59
A clue is to look again at Phil Jones answer on a question he was asked.

Now what was it he said?

I know this -

he said 2 things.
first he would get hung by his mates if he said there had been no warming in the last ten years


second, he said there had been no statistically significant warming in that period

in other words, dont let people try to fool you with fancy graphs - its simply noise


:rolleyes:

pjclarke
18th May 2011, 19:50
No, he said, last February (http://news.bbc.co.uk/1/hi/sci/tech/8511670.stm), that the warming trend since 1995 did not achieve statistical significance. The period was chosen to get the soundbite. The statistically challenged understood it to mean 'no meaningful warming' which of course it does not. It means there is insufficient data in the period to determine if the positive trend achieves significance to the 95% level.

1. When estimating a trend in a time series, the accuracy of the estimated trend increases with the number of data points. The normal threshold for 'statistical significance' is 95%, that is, there is only a 1 in 20 chance that the estimated trend is the result of chance, rather than a genuine trend in the data.

2. In the global temperature series such significance is generally achieved around about the 15-16 year mark.

3. In the BBC Q&A Jones was asked several questions including 'several gathered from climate sceptics'. The choice of period was cherry-picked to be just below the threshold. How do we know this? Because Richard Lindzen had earlier briefed his mate Anthony Watts priming him with this very number


"Look at the attached. There has been no warming since 1997 and no statistically significant warming since 1995. Why bother with the arguments about an El Nino anomaly in 1998? "

3. Watts is experienced in getting out a simple message and appreciates that to a large number of people the nuanced difference between 'no significant warming in 15 years' and '15 years of warming just below the significance level' will be lost. Sure enough (http://www.foxnews.com/scitech/2010/02/15/global-warming-insignificant-years-admits-uks-climate-scientist/).

4. At the time the question was asked, the warming trend was significant at the 90% level, at 95% if the start point was shifted to 1994, or a different temp index was used. It was a massive cherry-pick in other words. In fact the trend since 1995 pretty much carried on the warming (http://woodfortrees.org/graph/hadcrut3vgl/from:1950/plot/hadcrut3vgl/from:1950/trend/plot/hadcrut3vgl/from:1995/trend) from the previous decades, asked for an update recently Jones reported that the trend is now over the 95% significance level.

USeful Summary Round and round we go with Lindzen, Motl and Jones | Deep Climate (http://deepclimate.org/2010/03/02/round-and-round-we-go-with-lindzen-motl-and-jones/)

BlasterBates
19th May 2011, 07:02
http://www.woodfortrees.org/graph/hadcrut3vgl/from:2001.25/to:2011.25/plot/hadcrut3vgl/from:2001.25/to:2011.25/trend

Looks like there's a been a slight negative increase over the last 10 years.

pjclarke
19th May 2011, 19:41
<0.05C in a decade. Basically flat. And if you use an alternate dataset, e.g. NASA GISTEMP, you get a +ve slope. This indicates the trend calculation is not robust.

Because 10 years does not give you enough data to achieve statistical significance.

Try also looking at what the ENSO (http://www.esrl.noaa.gov/psd/enso/mei/) was doing over the same period. Might just have an effect on the short term trend, who knows?

Such multi-year periods of flat or negative growth are predicted by the models and are nothing new.

http://woodfortrees.org/graph/hadcrut3vgl/from:1950/mean:11/plot/hadcrut3vgl/from:1970/to:1979/trend/plot/hadcrut3vgl/from:1987/to:1997/trend/plot/hadcrut3vgl/from:1979/to:1988/trend/plot/hadcrut3vgl/from:2000.5/trend

Question is - do these periods combine to form any sort of a longer-term pattern? Whaddya reckon?

DimPrawn
19th May 2011, 21:34
http://graphjam.files.wordpress.com/2008/03/funny-graphs-crocodile-dundee-knife.gif