Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Prof. Hal Lewis resigns from APS


Hanson makes the argument in his "Grandchildren" book that we aren't relying on computer models when we say we know CO2 will raise the temperature. He says (I have not found the scientific paper) there were natural experiments in the geological past that show warming driven by CO2. One example he talks about (besides the feedbacks that pushed climate out of ice ages) was the Paleocene-Eocene thermal maximum. Hanson seems to think that a large injection of carbon into the atmosphere (the cause of which is poorly understood) resulted in a sharp 6C global temperature increase.


----- Original Message -----
From: "brian whatcott"<>

On 10/12/2010 6:28 AM, LaMontagne, Bob wrote:
You seem to take it as an insult that I would say that "the serious
investigators have not found an effect that rises out of the noise" and
then you go on to say how the results are not reproducable "100% of the
time". I would characterize that as not rising out of the noise. /snip/
Bob at PC

Interesting fusion of the concepts of
significance and repeatability coming through here?

Of course one could counter with how significant is a result if it is not
repeatable? However, this is probably the nature of beast when we try to
talk about global climate.

Something that concerns me and the degree to which we are depending on
computer modeling to deal with the topic, is that certainly these are all
dynamic models. That is, the models themselves are constantly being updated
and fed new data. That's how it should be, but when somebody counters with
a question about why this or that has or hasn't happened when predicted by
the models and the answer comes back 'the current conditions are consistent
with the model' (or something along those lines) is this 'consistent' with
the old model or with the updated model that has recently been fed the
particular data or conditions that might be in question? Ultimately then
the question becomes just how predictive the models really are. How much
should we rely on today's models say for 10 years hence. Turned around, how
accurate were the predictions of 10 years ago. Of course this gets
complicated by advances in the science and the computers such that we might
well expect today's models to be better than those of 10 years ago.
Still--models are just that, and climate models are extremely complex with
way many (too many?) variables!


Richard W. Tarara
Professor of Physics
Saint Mary's College
Notre Dame, IN
Free Physics Software

'Before you open your mouth, just remember,
the empty wagon rattles the loudest.'
-- my dad

kyle forinash 812-941-2039