For various reasons (see bottom) I've become interested in the sensitivity
of weather to initial conditions. I was hoping any experts here could help
me with the following question:

Suppose an air parcel of about 10 km size suddenly (magically?!) changed its
density, humidity and temperature, while staying within normal atmospheric
values. (Energy and mass conservation etc. don't apply at the instant of
transformation.) How long would it take till global weather was completely
scrambled from "what would have been" so that it would have no similarity
to the unperturbed system? (Apart from seasonal variations, large scale
climate etc.)

The question can of course not be answered experimentally, but are there any
results from computer simulations pointing towards a timescale? (Just order
of magnitude).

Be warned: I've had a course in atmospheric physics, but I forgot most of
it. I'm ignorant enough to be dangerous, and educated enough to be unaware
;-)

regards,

Niels
You may contact nvj -- fys ku dk

Why do I ask? From http://1632.org :

"In May of the year 2000, a six-mile sphere, centered on Grantville, West
Virginia, was displaced in space and time to Germany and 1631. Cardinal
Richelieu is not amused. This story, by Eric Flint, is the basis for a
great deal of lively discussion on http://bar.baen.com"

One of the questions being discussed is how long after the "displacement" it
is realistic to assume major meteorological events (storms etc.) developed
as they did in the REAL 1630ies. Expert help needed!

The story itself is better than the cheesy plot device. If you're
interested, go read it at
http://www.baen.com/library/0671319728/0671319728.htm
There's several sequels written by Flint and Fans too...

In article , Niels Vestergaard Jensen writes:
Suppose an air parcel of about 10 km size suddenly (magically?!) changed its
density, humidity and temperature, while staying within normal atmospheric
values. (Energy and mass conservation etc. don't apply at the instant of
transformation.) How long would it take till global weather was completely
scrambled from "what would have been" so that it would have no similarity
to the unperturbed system? (Apart from seasonal variations, large scale
climate etc.)

The question can of course not be answered experimentally, but are there any
results from computer simulations pointing towards a timescale? (Just order
of magnitude).


First of all. Such an event would not scramble the global weather,
it would hardly perturb it at all, although you can always
point to the butterfly effect and develop a couple of storms.

An example: Detonations of atomic bombs in the atmosphere certainly
scrambled the temperature, density and humidity in a more than 10 km
wide sphere, yet there are no indications that it affected
the global weather very much.


Timescales:
Within the local area a few hundred km: A few hours
Europe: A couple of days.
Northern or Southern temperate regions: A week
Tropic areas: A month.
The hemisphe 2-3 months
Globe: 1-2 years.

Øyvind Seland

Øyvind Seland wrote:

In article , Niels
Vestergaard Jensen writes:
Suppose an air parcel of about 10 km size suddenly (magically?!) changed
its density, humidity and temperature, while staying within normal
atmospheric values. (Energy and mass conservation etc. don't apply at the
instant of transformation.) How long would it take till global weather
was completely scrambled from "what would have been" so that it would
have no similarity to the unperturbed system? (Apart from seasonal
variations, large scale climate etc.)

The question can of course not be answered experimentally, but are there
any results from computer simulations pointing towards a timescale? (Just
order of magnitude).


First of all. Such an event would not scramble the global weather,
it would hardly perturb it at all, although you can always
point to the butterfly effect and develop a couple of storms.

Are you talking about large-scale climate here? That's not what I had in
mind. What I'm interested in knowing is how long it'd take till the
pertubation had grown enough to change the timing and strength of
particular fronts appreciably. If the atmosphere really is a chaotic system
(isn't it?) won't a small perturbation grow?

An example: Detonations of atomic bombs in the atmosphere certainly
scrambled the temperature, density and humidity in a more than 10 km
wide sphere, yet there are no indications that it affected
the global weather very much.

Good comparison. No, I don't expect nuke-size events to start global
warming. (If we ever drop that many of them we'll have bigger problems :-/)
I'm looking for them to contribute the initial pertubation. Unfortunately
we can't compare a nuked atmosphere to an unnuked experimentally, so it's
got to be done by computer which is poor, but all we've got. Any ideas
where to find references?

Timescales:
Within the local area a few hundred km: A few hours
Europe: A couple of days.
Northern or Southern temperate regions: A week
Tropic areas: A month.
The hemisphe 2-3 months
Globe: 1-2 years.

That's about the timescales I presumed for the pertubation to grow enough to
scramble fronts. Is that how I'm supposed to read the table?

Thanks for your interest,

Niels

In article , Niels Vestergaard Jensen writes:
Øyvind Seland wrote:

In article , Niels
Vestergaard Jensen writes:

What I'm interested in knowing is how long it'd take till the
pertubation had grown enough to change the timing and strength of
particular fronts appreciably. If the atmosphere really is a chaotic system
(isn't it?) won't a small perturbation grow?

No, it might grow or it may not, and it will be difficult to find out.

Unfortunately
we can't compare a nuked atmosphere to an unnuked experimentally, so it's
got to be done by computer which is poor, but all we've got. Any ideas
where to find references?

I do not know whether anyone have tried to simulate the effects Anyone?


Timescales:
Within the local area a few hundred km: A few hours
Europe: A couple of days.
Northern or Southern temperate regions: A week
Tropic areas: A month.
The hemisphe 2-3 months
Globe: 1-2 years.

That's about the timescales I presumed for the pertubation to grow enough to
scramble fronts. Is that how I'm supposed to read the table?

More or less.

If I were to give a guess on how your sphere would affect the weather I
would say that you could have some nasty winds and possibly intense
precipitation, locally 100 km away. A front very close might be affected,
but no guarantee. At mid-latitudes I would say that the strongest possible
imaginable effect on a longer timescale would be "removing or setting"
up a blocking high. In other words you would get a wet summer instead of a dry or
vice versa. I do not think it would be very likely however.


Øyvind Seland

yvind Seland wrote:
In article , Niels Vestergaard Jensen writes:

Øyvind Seland wrote:


In article , Niels
Vestergaard Jensen writes:


What I'm interested in knowing is how long it'd take till the
pertubation had grown enough to change the timing and strength of
particular fronts appreciably. If the atmosphere really is a chaotic system
(isn't it?) won't a small perturbation grow?


No, it might grow or it may not, and it will be difficult to find out.

Yes, it really would depend on the environment in which
the perturbation is introduced. If it's a cold, dry, winter
air mass, for example, under strong NW'ly flow -- oops, I see
it's in May. Okay, still, nice cP airmass, low humidity,
very stable, then the perturbation really won't grow much
with time, and any effects will damp out. If there's a weak
storm moving through an unstable environment, then
it is possible that the perturbation will affect things.


scott