[lug] Anyone else hate to get rid of old equipment?

Davide Del Vento davide.del.vento at gmail.com
Wed May 26 14:43:47 MDT 2010


> How do you test AGW?  Everything is caused by it:  Droughts, floods,
> extreme cold, extreme warmth, tornadoes, hurricanes, lack of tornadoes,
> lack of hurricanes.

You confound weather with climate.
Droughts, floods, extreme cold, extreme warmth, tornadoes, hurricanes,
lack of tornadoes, lack of hurricanes are weather. Weather is
unpredictable after a few days. Let me repeat that: there isn't any
chance you can foresee weather for a period longer than "something" (I
don't remember the actual theoretical limit, but let's say a few
months). It's also very local: you might even seen a block severely
damaged by hail and the next one intact.

Climate, on the other hand, "works" on larger scales (spatial and
temporal). You can say that climate is the spatial and temporal
average under big enough spatial and temporal domains. It is very
predictable.

Now, if the climate (=average) changes, the individual items that
create that average (=weather) must change too. But you cannot state
that "this hurricane has been caused by climate change", as much as
you cannot say if the deck is stacked just drawing a single card. You
need quite a lot of statistics to be sure. And still a fair amount
even to suspect it.

So, to answer your question "how do we test climate change"? The
answer is: with computer models of the atmosphere. Note that these are
climate models, completely different from the weather models used for
the forecast. The weather forecast is trying to look at the croupier,
and how he shuffles the deck to predict the next card. If the croupier
is fast enough or if he shuffles long enough the forecast is
impossible. The climate model tells you that on average the house will
make money, on the long run. Climate is easy, weather is hard. I'm
sure you understand this well.

The climate models that we have predict very well the current and past
status of the climate, if we use everything that we know, including
but not limited to human greenhouse emissions, volcanoes, ground
reflectivity, (yes, ice reflects solar heat much more effectively than
water), etc.
Leaving out just one of these factors (e.g. volcanoes, or human
greenhouse emissions) would cause the model to predict a current
climate different from what we measure. As I said in my other email,
different enough not being explainable by measurements errors.

Thus we say that human greenhouse emissions are already causing
"measurable climate change" (difference from model runs with or
without them - which are larger than measurement inaccuracies).

Would you like to disproof this finding? Write your own model (or
modify an existing one, they're open source), using sound physics,
chemistry or even biology if you'd like to include vegetation growth
(yes, some models already do that - and it's not enough to "correct"
the CO2 that we pump into the atmosphere). Have your model able to
accurately predict the past and current, measured, status of the
climate. Have your model predict the same results (within the accuracy
of the measurements and of the computation) with or without the human
greenhouse emissions.

I'll stop here on this thread.
Bye,
;Dav



More information about the LUG mailing list