I tend to agree with Bruno's statement:
". how do climatologists do this, bearing in mind that the results should
take thousands of years to appear? They test their hypothesis in computer
models, which are not quite the same thing as the reality yet."
During grad school I worked at the Atmospheric Sciences Center of the
University of Nevada System under Dr. James W Telford,
http://scholar.google.com/scholar?as_q=
<http://scholar.google.com/scholar?as_q=&as_sauthors=Telford,+James+W>
&as_sauthors=Telford,+James+W
His expertise was cloud microphysics and atmospheric instrumentation;
although most of his papers were theoretical in nature, he also made
contributions to instrumentation. And the reason for his work on
instrumentation goes to the point of Bruno's statement. Dr. Telford's main
complaint about GCMs (global climate models) is that they were way too
simplistic, and did not have enough real-world data about some key elements
at work in the atmosphere with cloud and surface albedo. There are numerous
GCMs, and many are prone to a very wide range of outcomes depending on very
small 'adjustments' in the variables. Just how good the current models are
is definitely a debatable issue. Telford designed, built and then flew his
instrumentation on aircraft thru clouds to get real-world data to help him
validate his theoretical models for cumulus clouds. He always was skeptical
of trying to model things on a global scale. Current science is still
working on understanding enough of what happens in the atmosphere to
generate accurate models. but one is still faced with the fact that Bruno
brought up. that all the models in the world are at best only a guideline
when we don't have enough detailed historical data, AND accurate details of
all the processes at work which affect the atmosphere, AND secondary and
tertiary effects which have not been anticipated, AND accurate data over the
relevant timeframe of hundreds or even thousands of years with which to test
the models. Perhaps scientists will discover ways to tease out some of
those details by creative means, like looking at CO2 levels in ice cores,
but there are still very significant unknowns which make it difficult to
build accurate global models.
Point. I just read a recent paper on this topic and nearly every sentence of
the quoted scientists' had at least one variety of weasel-word (like the
word 'could', or 'might'). As I have said in a previous post today, and a
number of times over the years, a good scientist is VERY careful about the
words they use. and there's a reason for that.
-Mark
From: Bruno Santos [mailto:[email protected]]
Sent: Monday, July 30, 2012 7:46 PM
To: [email protected]
Subject: Re: [Vo]:Koch founded climate skeptic changes sides
I read the literature. That is why I know that what you state is wrong. And
for two reasons:
1 - Insurance companies and epidemiologists have good, reliable data.
Climatologists don't. It is not their fault, is it just impossible to have
good data on the large periods of time required to study planets climates.
You see, just a few decades of reliable data isn't enough when we know that
the climate changes in scale of thousands of years.
2 - When an insurance company gathers all information, they can test their
hypothesis with what happens in the real world and see if it works. That is
how you know that smoking makes a difference in people's lifespans, but
other things don't. Epidemiologists can test on actual living beings how
diseases spread. Now, tell me, how do climatologists do this, bearing in
mind that the results should take thousands of years to appear? They test
their hypothesis in computer models, which are not quite the same thing as
the reality yet.
Large, complex phenomena are easier to study when it is based on the law of
the large numbers. That is precisely what happens with insurance and
epidemiology, but not with climate. There is no reality check in climate.
Those predictions based on large number of (bad) data are tested on
scenarios that are in a computer.
I am an economist, and we have the same problem. We do have good prediction
models, they are quite sophisticated, but not totally reliable. Otherwise,
one would not see economic crisis, economic downturns nor unemployment. And
I am pretty sure that economic data is far more accurate than climate ones.
Economists cannot test hypothesis in a lab. Neither can climatologists.
But that was not even my point. I believe that anthropic global warming is
possible, even probable. I just don't care, because the alternative,
poverty, is far worse.
2012/7/30 Jed Rothwell <[email protected]>
Bruno Santos <[email protected]> wrote:
Do we have a climatologist here? That would help the debate.
It would help if people would first read a credible, expert account of
global warming theory!
However, I have to say that I have my doubts when it comes to predictions by
these experts. You see, we do not have any credible scientific model for
weather prediction that works for periods longer than a week . . .
I do not like to be harsh, but that is a prime example of a mistake made by
an amateur critic who has not read the literature. You completely
misunderstand the technical issue. What you are saying is similar to this
assertion:
"Life insurance companies have actuarial tables predicting how long a person
is likely to live, based on present age, sex, the person's weight, whether
he or she smokes and other factors.
However, a life insurance agent cannot tell me whether I will live another
20 years. I might be run over by a bus tomorrow. I might die of cancer next
year.
Therefore, life insurance is a scam. They pretend they can predict the
future, but they cannot."
Needless to say, that is nonsense. You can predict the remaining lifespan of
a large group of people, even though it is impossible to predict the
lifespan of any given individual. Large scale complex events involving many
elements are sometimes more predictable than individual events with fewer
causes and less complex causes. That is counter-intuitive but it happens
with many natural phenomena, including climate, epidemiology and so on.
- Jed
