Re: [asa] Creation Care

On 1/24/07, Bill Hamilton <williamehamiltonjr@yahoo.com> wrote:

> Don makes some good points. I've been involved with engineering projects
> for
> the past 35 years, and seldom have I seen an engineering project
> correspond
> exactly to the frequently extensive modeling that was done in the design
> phases. When we could presumably control all aspects of the design, we
> still
> had to tweak and modify to account for circumstances that loomed larger in
> practice than in the models.

Like you I have 25 years of experience in engineering modelling. In my case,
it is semiconductor device simulation. One thing I have seen in either Don's
or your post is how modelling has changed recently. 25 or 35 years ago we
made numerous simpifying assumptions just because it would take FOREVER to
simulate. Now we have machines that are 5-10 order of magnitude faster than
when we started our careers. This means we can remove the heuristics -- a
fancy scientific word for guess -- decrease our mesh size and time steps and
get really good results. Why do I have confidence about this? Well, we are
communicating through devices that have been extensively simulated where the
measured results match extremely well with the modelled behavior. As time
went on we modelled more and more effects, e.g. resistance, crosstalk, short
channel effects, etc. None of these were a great surprise from a physics
perspective but they required more iron to do the simulations. The same
holds true with climate modelling. When I look at the climate models I see
the same well-formed models. In fact, I would have killed to get such good
correlations at so many corners as these models (and I would like access to
their computers too :-) ).

One item specific to climate modelling also gives me confidence. The
satellite data was not meshing with the models in that the high altitude
temperatures were not as predicted. An error in the data collection was
fixed and then the models matched. This discrepency caused a literal act of
Congress to determine why the models were not so good. Since then much has
been done to make these models much, much better. There is still work to do
in order to more accurately predict the small-scale implications of global
warming because these are important for good public policy. But, the models
are accurate enough now for the upcoming IPCC AR4 to have a 90% confidence
that we are experiencing global warming and that it's primarily caused by
anthropogenic greenhouse gases.

Models matching the current (and now extensive) climate data are a necessary
condition for any alternative hypotheses now. None of Janice's skeptics have
put forth their own computer model (you can check for this youself by doing
a Google scholar search for the author) and show that matches either the
current or paleoclimate data better. It's not that these individuals have
presented research results that got poo-pooed by the mainstream they have
offered literally NOTHING. If I was Pim I would call it vacuous. :-) From
now on, Janice will need to reference papers that have these models in them.
For example, she needs to show that a model that has solar forcing more than
5X smaller than GHG forcing fits the curve better than the current model.
Climate science is now quantitative and a qualitative hand waving no longer
cuts it.

To unsubscribe, send a message to majordomo@calvin.edu with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Wed Jan 24 11:35:17 2007