Global Warming for Dummies

[Spacecadet]

Bigger than it's ever been in spite of CFC's being banned 20 years ago:

http://earthobservatory.nasa.gov/IOTD/view.php?id=7044

Banned != no longer in use

[BlasterBates]

is owned by National Weather Services who place restrictions on it

really?

http://wattsupwiththat.com/2010/03/0...-availability/

[pjclarke]

So in 1988 they proclaimed a few years of global warming was due to CO2.

Who did?

Works one way but not the other, 12 years is significant in the 30 years or so of warming. Is it not?

Scientists tend to use the word 'significant' in its statistical sense, which usually means a 95% confidence. Its unlikely that a 7 year trend in the global temp record has that significance, >15 years is more likely, 30 years almost certainly.

...oh and remember the Satellites show a statistcally significant cooling.

Nonsense. Both show the same long term trend of +0.18C/decade, in line with climate model projections. http://www.ncdc.noaa.gov/sotc/?repor...-month-lowtrop

The 11-year solar cycle is estimated to vary global temperatures by about 0.2C peak-to-trough, though the error bars are large. What IS genuinely interesting is the last few months have seen a historically quiet solar minimum which coincided with the warmest January in the satellite record.

[BlasterBates]

So in 1988 they proclaimed a few years of global warming was due to CO2.


Works one way but not the other, 12 years is significant in the 30 years or so of warming. Is it not?

Scientists tend to use the word 'significant' in its statistical sense, which usually means a 95% confidence. Its unlikely that a 7 year trend in the global temp record has that significance, >15 years is more likely, 30 years almost certainly.

...oh and remember the Satellites show a statistcally significant cooling.

warmest January in the satellite record.

who did?

1. James Hansen to the senate commitee

0.18 per decade

2. Factually incorrect for 1998-2010, the last 12 years have been cooling

warmest January in the satellite record

3. but not the warmest month on record, which was in 1998 (i.e. still cooler than 1998, when the sun was more active), with record cold temps in the Northern Hemisphere, and the solar cycle lasts longer than a month, 13 years.

[pjclarke]

The most powerful El Niino of the C20th was in 1998!

The solar cycle is about 10.66 years on average and this winter saw the end of solar cycle 23 and the start of cycle 24, that is, a minimim in solar activity, which makes the record temperatures all the more remarkable.

+0.18C/decade is the trend you get if you use ALL the satellite data. I wonder why you chose 1998 as the start point of your trend? It is not cooling, and it is not significant.

[BlasterBates]

The solar cycle is about 10.66 years on average and this winter saw the end of solar cycle 23 and the start of cycle 24, that is, a minimim in solar activity, which makes the record temperatures all the more remarkable.

+0.18C/decade is the trend you get if you use ALL the satellite data. I wonder why you chose 1998 as the start point of your trend? It is not cooling, and it is not significant.

I see 1980 0.2 below average, and now 0.2 above average from the trend you display.

that is 0.4 degrees over 3 decades, that is 0.13 per decade not 0.18!

...I quote

UAH v5.2 finds a trend of +0.13°C/decade.[4]

http://en.wikipedia.org/wiki/Satelli...e_measurements

Now subtract 0.2 cooling from 1940 - 1970

...and now what do you get?

0.2 degree warming over 70 years that is 0.028 dgrees warming per decade since 1940.

That's why they fiddled HARD CRUT because the satellite doesn't say anything scary.

..and the cooling cycle has only just begun, in 20 years we may be back to 0.

[sasguru]

Roy Spencer and Richard Lindzen were the 2 who answered no to Q2, I speculate. Spencer is free to post whatever he likes on his blog, of course, though it rarely stands up to scrutiny.

His publications that have survived peer-review and made it into the literature tell a rather different story. There is also a question mark over his mathematical ability - he posted a regression analysis showing 'remarkable correlation', when his technique would have shown the same correlation between any two variables plucked at random.

Ho Hum. Regarding the CO2 correlation with global temp, suppose for a moment that there was in fact, no increasing greenhouse influence on the temperature. Would you expect it to trace a perfectly horizontal straight line? Of course not, there is noise in the signal - the solar cycle and natural variability and other factors cause the trace to zigzag. So why would anyone expect a perfectly straight positively-sloped line in the presence of an external forcing? The reality is that over the shorter term the 'zigzags' can temporarily swamp the greenhouse signal, but take a longer view and the 'noise' averages out, leaving a rising curve correlated well with the increase in greenhouse gas concentrations.

Shocking - they cover that in Stats 101.
Amazing how many so-called scientists make this elementary school-boy error - you notice they are usually the older ones who were trained when stats was much less manadatory on science courses than it is now
Given that the universe is inherently probabilistic rather than deterministic, a discovery that has bypassed most on CUK, no scientist worth his salt should be ignorant of statistical fundamentals.

[pjclarke]

The satellites were never designed for climatic monitoring (there was one that was so designed but George Bush cancelled it, that's another story). In a sentence the brightness temperature of the lower troposphere has to be converted to an actual temp. There are several adjustments and choices to be made and two main agencies UAH and RSS use slightly different techniques against the raw data, and there is another re-analysis using adjustments described in Fu et al used by the US NOAA. It is this last process that gives 0.18C / decade.

Strange as it may seem, there are more robust methods of computing a linear trend than (cherry) picking two data points and joining them up with a ruler ....

[BlasterBates]

The satellites were never designed for climatic monitoring (there was one that was so designed but George Bush cancelled it, that's another story). In a sentence the brightness temperature of the lower troposphere has to be converted to an actual temp. There are several adjustments and choices to be made and two main agencies UAH and RSS use slightly different techniques against the raw data, and there is another re-analysis using adjustments described in Fu et al used by the US NOAA. It is this last process that gives 0.18C / decade.

Strange as it may seem, there are more robust methods of computing a linear trend than (cherry) picking two data points and joining them up with a ruler ....

The adjustments have been refuted by Christy and Spencer in 2003.

These adjustments are therfore not included in the official data set over which Christy and Spencer officially preside.

The best that can be argued is that the UAH dataset has been challenged, and the challenge has so far failed.

The official UAH dataset -

UAH v5.2 finds a trend of +0.13°C/decade.[4]

[threaded]

Indeed. You have evidence of faked data, I presume, that's a fairly serious allegation.

Are you denying the data is faked? Statistical tests have shown it to be quite dodgy, and I've already given publicly available examples of some of the tricks the AGW loons have used to fake data. There are many others examples that are waiting to crawl out of the darknet ;-)

So, what bizarre 'evidence' are you going to proffer that the whole shebang isn't a complete fabrication to further some crazy political ideology, line pockets, and is nothing to do with science at all really?