I am taking bets on a scientific establishment coming out with a statement
that "there 'could' be a rise in global temperatures of 20C in the next x
years", (where x = 50, 100 or 200 etc....... take your pick- it doesn't
really matter) within the next few weeks.

"Pat Slattery" wrote in message
...
I am taking bets on a scientific establishment coming out with a statement
that "there 'could' be a rise in global temperatures of 20C in the next x
years", (where x = 50, 100 or 200 etc....... take your pick- it doesn't
really matter) within the next few weeks.



No, its just swung the other way, bloke just been on five live from some
university in Liverpool that is convinced we are really heading towards an
ice age!

I don't really care which we get, +11°C or -11°C as long as it happens soon!
Just bring it on, lets enjoy it.

Stuart wrote:
"Pat Slattery" wrote in message
...

I am taking bets on a scientific establishment coming out with a statement
that "there 'could' be a rise in global temperatures of 20C in the next x
years", (where x = 50, 100 or 200 etc....... take your pick- it doesn't
really matter) within the next few weeks.




No, its just swung the other way, bloke just been on five live from some
university in Liverpool that is convinced we are really heading towards an
ice age!

I don't really care which we get, +11°C or -11°C as long as it happens soon!
Just bring it on, lets enjoy it.


The temperature ranges from 1.9C to 11.4C, that's a massive difference,
how can a model with that much uncertainty be scientifically useful?

Felly sgrifennodd Karatepe :
The temperature ranges from 1.9C to 11.4C, that's a massive difference,
how can a model with that much uncertainty be scientifically useful?

Firstly, because it shows an increase (whatever the amount). This is useful
information.

Secondly, because any scientific prediction is based on probabilities.
Virtually no prediction is certain. Scientists are used to working with
probabilities. Knowing probabilities is very useful to a scientist.

Go and study some statistics if you don't think this is useful. Pay
particular attention to the section on normal distributions, model error
and probabilities.

Whilst you're at it, see if you can educate certain politicians as well!

As an analogy: suppose a frost is forecast tonight, with a 70% probability.
This means there is a 30% chance there won't be a frost. So, you would argue
that this is useless information? Is it useless for the council gritters
to grit the roads, then?

Adrian

--
Adrian Shaw ais@
Adran Cyfrifiadureg, Prifysgol Cymru, aber.
Aberystwyth, Ceredigion, Cymru ac.
http://users.aber.ac.uk/ais uk

Adrian D. Shaw wrote:
Felly sgrifennodd Karatepe :

The temperature ranges from 1.9C to 11.4C, that's a massive difference,
how can a model with that much uncertainty be scientifically useful?


Firstly, because it shows an increase (whatever the amount). This is useful
information.

That's the only piece of useful information.

Secondly, because any scientific prediction is based on probabilities.
Virtually no prediction is certain. Scientists are used to working with
probabilities. Knowing probabilities is very useful to a scientist.

Go and study some statistics if you don't think this is useful. Pay
particular attention to the section on normal distributions, model error
and probabilities.

Whilst you're at it, see if you can educate certain politicians as well!

As an analogy: suppose a frost is forecast tonight, with a 70% probability.
This means there is a 30% chance there won't be a frost. So, you would argue
that this is useless information? Is it useless for the council gritters
to grit the roads, then?

Adrian

Firstly, because it shows an increase (whatever the amount). This is
useful
information.

Secondly, because any scientific prediction is based on probabilities.
Virtually no prediction is certain. Scientists are used to working with
probabilities. Knowing probabilities is very useful to a scientist.

Do they know the probablity distribution of the huge range of
temperatures quoted? If so, why not tell us? I have not seen any
quoted probabilities, or a time scale, and without these the figures
are quite meaningless.

Tudor Hughes

"Tudor Hughes" wrote in message
oups.com...
Firstly, because it shows an increase (whatever the amount). This is
useful
information.

Secondly, because any scientific prediction is based on probabilities.
Virtually no prediction is certain. Scientists are used to working with
probabilities. Knowing probabilities is very useful to a scientist.

Do they know the probablity distribution of the huge range of
temperatures quoted? If so, why not tell us? I have not seen any
quoted probabilities, or a time scale, and without these the figures
are quite meaningless.

Tudor Hughes

At the end of the news last night the timescale quoted was the end of this
century. Not surprisingly nothing was said about probabilities.

Alan

Felly sgrifennodd Tudor Hughes :
Do they know the probablity distribution of the huge range of
temperatures quoted?

I would have expected any scientific study into global warming to produce
such statistics. I have certainly seen studies with such results published
in the New Scientist.

If so, why not tell us?

Because they think we won't understand?

Admittedly, since different studies come up with different results, the issue
is clouded. But I suppose one could lump them all together and produce a
single probability distribution based on their combined results. That would
of course not be easy as there would be arguments about which to include,
etc.

Adrian
--
Adrian Shaw ais@
Adran Cyfrifiadureg, Prifysgol Cymru, aber.
Aberystwyth, Ceredigion, Cymru ac.
http://users.aber.ac.uk/ais uk

In article .com,
says...
Firstly, because it shows an increase (whatever the amount). This is
useful
information.

Secondly, because any scientific prediction is based on probabilities.
Virtually no prediction is certain. Scientists are used to working with
probabilities. Knowing probabilities is very useful to a scientist.

Do they know the probablity distribution of the huge range of
temperatures quoted? If so, why not tell us? I have not seen any
quoted probabilities, or a time scale, and without these the figures
are quite meaningless.

It's all in the article:

http://www.climateprediction.net/sci...st_results.pdf

Harold
--
Harold Brooks
hebrooks87 hotmail.com

On Thu, 27 Jan 2005 09:00:22 -0600, Harold Brooks wrote:

It's all in the article:

http://www.climateprediction.net/sci...st_results.pdf

Very interesting -- I didn't know this project was running. BOINC was
already installed on my computer for SetiatHome, and I've just added the
Climateprediction project. Now, how much time to allocate to each??


Mike