View previous topic | View next topic

Global Warming is a Hoax

Page 3 of 25
Goto page Previous  1, 2, 3, 4 ... 23, 24, 25  Next


What do you think?
Global Warming is a hoax
7%
 7%  [ 5 ]
It's real, but is mostly natural
27%
 27%  [ 19 ]
It's real, and is mostly affected by man
57%
 57%  [ 39 ]
Ooh look, a brown dog outside my window...
7%
 7%  [ 5 ]
Total Votes : 68

barbados
648501.  Mon Dec 21, 2009 11:25 am Reply with quote

dr.bob wrote:
barbados wrote:
Here's an interesting thought touted by Dennis Miller recently.

How reliable are the figures quoted to prove that global warming is occuring?


As I understand it, the figures proving that global warming is occurring are pretty damn irrefutable. What's less clear cut is what proportion of the change is caused by human activities.



It was more aimed at in the 1920's the measurement would have been made some mercury in a tube and a pair of eyes, today it would be down to a computer.
Would the reader translate around 28.5 degrees as 28 or 29?

 
PDR
648515.  Mon Dec 21, 2009 11:58 am Reply with quote

dr.bob wrote:
That's a pretty reasonable argument, but sadly the world is a very uneven place.

For the last couple of hundred years, countries in the west have built large, successful economies by burning fossil fuels. Now, just at the time when some newer economies are doing the same thing, a bunch of people are turning 'round and saying "sorry, we've just realised all this burning coal thing is really bad. Instead you'll have to use some other, more expensive energy, which will slow down the development of your economy."

Needless to say, some of these emerging economies think they're getting the shitty end of the stick here, so tend to reply "that's fine, let's save the world. You just give us some of the cash you made by burning coal to make up for the fact that we're not allowed to and we'll agree with your ideas." At which point, the rich western nations start to shuffle their feet and mumble about how it wasn't really all that much money that they made, honest. Especially with the rising cost of school fees and everything.

And so, yet again, everyone fails to reach an agreement.


I have to say that is probably the most succinct summary of the political (rather than the merely technical) problem which I have seen so far. With your permission I may reproduce it in other places Bob (with proper attribution, naturally).

PDR

 
gruff5
648522.  Mon Dec 21, 2009 12:11 pm Reply with quote

barbados wrote:
It was more aimed at in the 1920's the measurement would have been made some mercury in a tube and a pair of eyes, today it would be down to a computer.
Would the reader translate around 28.5 degrees as 28 or 29?

Interesting question. Actually, digital technology with their fancy display of decimal places can give a false impression of accuracy. For example, those automated blood pressure cuffs that my gym uses and surgeries are being forced to use are not as accurate as the old mercury manometers.

So, I would guess that a scientifically-calibrated mercury tube thermometer would be adequately accurate. You're right that the step of human judgement can introduce errors with a mercury tube, though. But this should be taken care of by many, many readings being taken in a year and the average of those readings being used.

 
barbados
648524.  Mon Dec 21, 2009 12:17 pm Reply with quote

That's the thing, with a digital thermometer to a hundred decimal places 28.75 degrees is 28.75 degrees, howerver in a thermometer it can only be about 28 and a half ish, which isn't really that reliable.

don't get me wrong, I'm not one of these naysayers that denies it is happening, I do however think there are two flaws in the theory.
Firstly, the disputed part - are we really responsible for it bearing in mind the temperature of the earth goes up and down by itself.
Secondly we are doing the wrong thing by trying to avoid it - we shold be working out how to cope with it

 
Leith
648558.  Mon Dec 21, 2009 2:09 pm Reply with quote

barbados wrote:
That's the thing, with a digital thermometer to a hundred decimal places 28.75 degrees is 28.75 degrees, howerver in a thermometer it can only be about 28 and a half ish, which isn't really that reliable.

I've been reading a little about thermometer accuracy recently, as it happens. From what I've learned so far it seems that even today it is common for individual readings to be rounded to the nearest degree for use in climate models. As gruff says, The precision and accuracy of model temperatures is achieved through averaging large numbers of readings.

In these instances, the accuracy of the individual readings does not lead to false precision in the derived temperature trends. You can see a demonstration to that effect here:
The Blackboard: False Precision in Earth’s Observed Surface Temperatures?

..and a more detailed treatment of the rounding issues with Excel examples:
The Blackboard: Rounding of individual measurements in an average.

As far as I understand it, what areas of uncertainty there are in climate models come not from measurement inaccuracies, but from physical mechanisms that we don't understand well enough yet (e.g. the impact of clouds and aerosols in atmospheric heat transfers), and from unexplained anomalies in historical temperature reconstructions (e.g. the "Divergence Problem" in certain North American tree ring records).

 
barbados
648564.  Mon Dec 21, 2009 2:42 pm Reply with quote

Here's an example of what I mean.

What is the temperature reading? It's about -32, It could be -32, t might be -34, or -33.

However if you look at this one

It reads 37 degrees, no doubt about it.

You also need to take into account the air pressure in a lquid thermometer.

Assume that the air pressure hasn't had an effect on the liquid thermometer for simplicity what is the difference in the two reading? 69, 70, or 71 degrees. Who can tell catagorically what the difference in temperature is when you change the method of gathering the information, or at the very least a different person interpretting the information as they are reading it.

 
Leith
648636.  Mon Dec 21, 2009 5:48 pm Reply with quote

So it's the transition from one instrument type or measurement policy to another you're thinking about?

If you change the measurement methodology, you could see a step-change in absolute temperature readings if:
a) your methodologies' rounding policies produce different measurement biases
and
b) you haven't calibrated them against each other to compensate for this.

What you wouldn't expect to see is any long term upward or downward trends caused by the change, and in climate models it is the trends that are of interest rather than the absolute temperature at any given station. When you look at the rate of change of temperature, a significant change in method bias should show up as a spike. The spike can be discarded from the trend data and potentially used to detect and correct the anomaly in the absolute temperature data.

 
barbados
648717.  Tue Dec 22, 2009 1:02 am Reply with quote

Again you are trying to make a simple thing complicated with methodologies and cross calibration, when the problem lies with something that wouldn't change, the readers eyes.

The liquid thermometer is, however you want to dress it up, open to interpretation, the picture shown above shows that.. It reads either -32 -33 or -34. That is a three possible three degree shift just by having three different people read it on three seperate occassions, nothing to do with what is actually happpening, more to do with personal interpretation.

 
PDR
648733.  Tue Dec 22, 2009 3:49 am Reply with quote

...and in any series of measurements this might appear as a three degree uncertainty. But even if it did there would be no "trend" within the variation of the readings (or to be strictly accurate - there is an extremely low probability of the random variation looking like a trend). Pick a simple rolling average from this data and these variations (if they exist at all) will targely be nulled out. For more sophisticated techniques we have statistical and mathematical techniques from the humble mean-of-subgroups through to kalman filtering that will extract consistancy from the fuzz in "noisy" data.

So no, I don;t think there's much risk that the issues you've outlined would corrupt the measurements sufficiently to give the apprearence of a trend where no trend exists.

PDR

 
djgordy
648753.  Tue Dec 22, 2009 5:03 am Reply with quote

Did anyone see Tony Robinson's programme "Man on Earth" last night. He was talking about the Maya and described their social system as one in which the people paid taxes to the elite and, in return, the elite promised to provide good weather. I couldn't help noticing that this is exactly the same arrangement the politicians are trying to bring about now!

 
PDR
648756.  Tue Dec 22, 2009 5:17 am Reply with quote

...but without the promise of good weather.

:0)

PDR

 
dr.bob
648771.  Tue Dec 22, 2009 5:54 am Reply with quote

PDR wrote:
I have to say that is probably the most succinct summary of the political (rather than the merely technical) problem which I have seen so far. With your permission I may reproduce it in other places Bob (with proper attribution, naturally).

PDR


Oooh, fame at last. Consider the permission granted.

 
Leith
648772.  Tue Dec 22, 2009 5:56 am Reply with quote

djgordy wrote:
Did anyone see Tony Robinson's programme "Man on Earth" last night. He was talking about the Maya and described their social system as one in which the people paid taxes to the elite and, in return, the elite promised to provide good weather. I couldn't help noticing that this is exactly the same arrangement the politicians are trying to bring about now!

They're not doing so well with the weather right now. Clearly we need to step up the human sacrifices until the traffic chaos is sorted.

 
Leith
648773.  Tue Dec 22, 2009 5:59 am Reply with quote

barbados wrote:
Again you are trying to make a simple thing complicated with methodologies and cross calibration, when the problem lies with something that wouldn't change, the readers eyes.

Sorry, I was moving on to talking about changes in method bias assuming I'd got the point across about averaging the readings. PDR's post covers this, but just to summarize the main points I was trying to make:

For the sort of observation errors you are talking about:
1) Random observation errors are cancelled out when the readings are averaged.
2) Systematic observation errors could affect absolute temperature readings but are highly unlikely to affect the appearance of temperature trends.

 
gruff5
648840.  Tue Dec 22, 2009 7:59 am Reply with quote

barbados wrote:
You also need to take into account the air pressure in a lquid thermometer.

I think Leith & PDR have nicely answered your other points about human judgement of temperature, but on this particular point of air pressure I would say that you don't need to take it into account.

The tube of a liquid thermometer is sealed and so the air content inside is unchanging. Once it is calibrated and the scale inscribed, then it should correctly measure temperature, no matter what the external atmospheric pressure is.


Last edited by gruff5 on Tue Dec 22, 2009 8:05 am; edited 1 time in total

 

Page 3 of 25
Goto page Previous  1, 2, 3, 4 ... 23, 24, 25  Next

All times are GMT - 5 Hours


Display posts from previous:   

Search Search Forums

Powered by phpBB © 2001, 2002 phpBB Group