On 2011-12-15, Wojciech Derechowski (aka Bruce)
was almost, but not quite, entirely unlike tea:
> And all that even before the modelling starts. Yes I know the models
> are run all the time. But that is only the temperature and pressure
> field mometum that can be extrapolated on what, 10 days? Apparently
> after more than 10 days the simulation becomes too complex and it is
> anybody's guess if it will rain on a hot summer day. The models seem
> rather imperfect in predicting the future.
As Stuart said, weather != climate. Weather is a horribly complicated
non linear chaotic system. The improved forecasts over time have been
due to increased grid and time resolution so atmospheric features can
be better tracked and extrapolated.
Climate is a simple energy balance heat transfer equation with
positive and negative feedbacks. The positive feedbacks make things
really bad, and the negative feedbacks at best can mitigate only some
of the warming (or cooling, if the heat transfer equations went that
way, which they don't -- the Sun is on the outside of the Earth's
atmosphere). Relatively easy to forecast decades into the future,
with outcomes only really dependant upon the unknown scale of the
positive and negative feedbacks, but bad outcomes whichever way you
look at it.
Make no mistake - the climate will affect the weather. More energy in
the atmosphere == more
interesting.cn weather.
> So if it is impossible to control the climate, why not try to manage it?
> Why not try to make some clouds? My guess is that only 3% more clouds today
> will bring down the temperatures to the level observed before industrial
> revolution. And if it doesn't, let's try someting else. But never pretend
> we understand the climate and can predict what it will do in respose to
> our actions. It's a complex system.
I've worked on control systems before. Simple 500 tonne objects
working in 2 dimensions. It takes a long time to tune a control
system so it doesn't oscillate wildly about. And most control systems
we have experience with have very reasonable response times, so it is
easy to learn how the system responds, and easy to hit the emergency
stop when it starts to go wrong. And they don't have any chaotic
components. And they're also of human scale. It will take us years
to realise that we've just pushed a component of the atmosphere or the
oceans in the wrong direction or at the wrong speed. By which time we
have some other climate crisis on our hands that will take decades to
resolve with the next push of the climate in another direction. I
wonder what is happening down-weather of China who are currently cloud
seeing to the tune of something like 30 gigalitres per year?
You *really* don't want to go down the planetary engineering route.
That way lie dragons.
--
TimC
"The application failed to fail"