IPCC (Intergovernmental Panel on Climate Change) doesn't actually do the
science - merely reports it to policy makers.  There are summaries and more
technical information, but it is all written to be used (eventually) by
politicians.  Thus it seems reasonable to me to take a single scientific
data point and attach a lesser "feel good" confidence than one that has been
corroborated 16 times.  Policy makers need to know if they might be dealing
with a wild point in the data or one that is solidly proven by peer review
and duplication - or if the result was arrived at from a multitude of
corroborating angles.  Personally, I applaud the international community for
forging ahead despite our (US) corrupt and dishonest administration.  Talk
about scientific dishonesty!

As for being on or off topic, the OP brought up the 800 year lag between
temperature and the CO2 curve and I presented what I consider to be a
relevant posit for why it doesn't apply to the current data, so you have,
indeed, missed the point again.  As far as I know, no one has stated it the
way that I did, which to me at least, proves the disconnect between natural
CO2 and manmade; i.e., there would need to have been a massive temperature
event 800 years ago to cause the equivalent massive CO2 shift today.  There
wasn't, so there is a different mechanism at work.

-----Original Message-----
From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu]On Behalf
Of Russell McMahon
Sent: Wednesday, October 17, 2007 8:27 AM
To: Microcontroller discussion list - Public.
Subject: Re: [OT]: CO2 peak may lag warming by 800 years ? *.edu source


A second bite from a (slightly) different direction

> The great thing about this new Science speak is that it gives
> scientists a new freedom to say things that they would have been
> stoned for until someone thought of this idea.

> If I've got this wrong, by all means do point out my error.

>>>> Well, yes, I do believe you've got it wrong.  The point is that
>>>> there is
> uncertainty - a lot of it - but there is significant consensus in
> the
> scientific community that there is cause for alarm.

No. That's not the point :-).
It's another point and an entirely good area to be looking at.
But what I was trying to say (and I apparently did it poorly :-( ) is
that they have not only changed the rules but are now acting as if the
new rules were always there.

> ...  and nitrous oxide is +2.30 [+2.07 to +2.53] W m-2, and its rate
> of
> increase during the industrial era is very likely to have been
> unprecedented
> in more than 10,000 years  ...

and it's surrounding material *feel to me* to have been written by a
scientist for scientists. If we lesser mortals come and look at it we
sort of feel that we are on hallowed ground as observers. Maybe.

BUT in the middle of the 'science speak' (radiative forcing and the
rest) we find "very likely".
Now, in the not at all distant past I feel (and maybe this is the Rip
van Winkle effect) that that term would not have been there at all.
The terms in the toolkit were, AFAIR things like " ... a statistically
significant ..." and " ... no statistically significant ..." with a P
value in brackets. I don't recall that in the dark ages they ever used
to say "tending towards significance" or similar but just maybe they
did. My wife deals with scientific reports all day every day (an
editor for a medical reports reports publishing company) and she says
that they now do use a term like 'tending towards significance' or
indications of significance. Whatever.

My point is (if I remember it :-) ) that in the past the passage above
would almost certainly NOT have been written at all as it appears now
BECAUSE he would have had to write " ... there was no statistical
significant indication that ...". As the whole point of what he is
saying is that the data tends to support a premise of a much larger
rate of occurrence of whatever than in kiloyonks. And he could not
have said that.
He would have had to say either that it didn't appear to be so OR that
there was perhaps some indication that the figures might support the
premise.

BUT now they have been given a whole gradation of terms to use which
can fit into the sentences they write and give a graded probability
curve right down to P=0.5 !

In fact !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

They can now write with a straight face eg

> ...  and nitrous oxide is +2.30 [+2.07 to +2.53] W m-2, and its rate
> of
> increase during the industrial era is more likely than not to have
> been unprecedented in more than 10,000 years  ...

and actually MEAN between P=0.5 and P= 0.33

cf

IPCC speak table (courtesy some PICList poster) which the above
rapporteur was (presumably) using.

> Virtually certain > 99% probability of occurrence,
> Extremely likely> 95%,
> Very likely > 90%,
> Likely > 66%,
> More likely than not > 50%,
> Unlikely < 33%,
> Very unlikely < 10%,
> Extremely unlikely < 5%.

This is utter travesty.
P=0.5 is so close to the mean it matters not a whit.
P=0.33 is one standard deviation out.
The Azores butterfly could push any global warming indicator out to 1
SD from norm with a single wingflip.

I haven't yet SEEN any 'scientist' saying "more likely than not' in
such a context BUT it is as legitimate to do that as it is to say
"very likely".

50% < X <=  66% DOES NOT mean "more likely than not" in statistics. It
never has and it never never never will. It means far far far inside
the are of the normal curve where random variation let's this sort of
thing happen as of right and with no essential meaning at all.

And having > 66% (about 1 SD) is worse.

But when Joe/Josephine Sheeple hears a scientist say "it is likely
that ..." he/she THINKS they know what the scientist mean. They don't
know that nowadays when a scientist says "likely" they really mean
"random noise".

ALL of this is lies and moving of markers and unequal weights in the
weighing bag. It's just a matter of scale between "very likely" and
"more likely than not".

Do you think that something is broken.
I think it more likely than not that it is.


        Russell


.








--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist