Re: OT: is the AGW bubble about to burst?

In article <1187977301.347605.309510@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>, James
Arthur wrote:
On Aug 23, 8:21 pm, bill.slo...@xxxxxxxx wrote:
On Aug 23, 5:32 am, James Arthur <dagmargoodb...@xxxxxxxxx> wrote:

On Aug 22, 4:31 pm, bill.slo...@xxxxxxxx wrote:

On Aug 22, 8:37 pm, James Arthur <dagmargoodb...@xxxxxxxxx> wrote:

<SNIP stuff quoted about 5-plus times>

You misunderstand. To predict the future of the system, it's
essential to know the initial state: temperatures, compositions,
directions, and flows throughout the oceans, ice caps, and atmosphere,
as well as cloud cover, and surface temperatures of the earth. You
have to know where the heat is, any inputs and outputs, and where it's
all going.

Consider the question "At what angle must I weld my rudder to in New
York in order to arrive at Bristol?" Knowledge of all the initial
rates and flows and how they'll change over time is clearly essential
to predicting the ship's actual course, of course. Change to question
to Liverpool or Manchester and the effect of initial state is even
more evident!

So, you need an accurate snapshot of initial conditions. Those, in
turn, must be input to a sufficiently accurate model in order for the
model to predict the future.

Likewise for climate. For example, how can we have ice in the world
when it's so hot in the Sahara?

Currents move heat. You can't predict where a current will wind up
unless you can predict the other currents it will interact with. To
predict those you need to know their initial state.

If you can't predict the heat flows, you can't predict the climate.

We know that you can't predict the weather that way, let alone the
climate.

Incorrect. Weather *is* predicted that way: intial conditions input
into a finite element analysis (FEA) modelling program.

The results are surprisingly good these days, but the farther you
project the results, the greater the model diverges from reality. The
divergence point is getting farther and farther out as the dataset
gets more detailed, and the FEA resolution is improved with faster
computers.

The equations you have to solve to do detailed predictions are too
sensitive to intial conditions for it to be possible to predict the
weather more than four or five days in advance.

AIUI, weather can be predicted decently well as much as two weeks(?)
in advance by the above method. The chief limitations, it was
explained to me, are the resolution and accuracy of the initial
conditions input to the model, and compute power.

Take current global system with past week or two data or however far
back such data is relevant for weather forecasting, use the various
weather prediction algorythms from *now* (a twice-a-day-exercise)...

Extend for many years...

Repeat with corrections according to atmospheric CO2 content increased
by 20 or 100 PPMV, as best as possible...

I suspect high lack of agreement within a month projected into the
future as to where it is raining or snowing or sunny.

I give a good chance that most to all of the
various-above-mentioned/implied tracks of global/regional weather
forecasting will do poorly at forecasting where gets flooded or impacted
by droughts, where gets hit by more hurricanes or where gets unusually
spared hurricanes, or for that matter year-by-year hurricane count in the
"North Atlantic Basin".

What I do suspect is good agreement on regional and global trends for
time scale at least a couple to a few decades.

In order to predict the climate, you have to lose the fine detail and
set up lumped approximations that capture the average behaviour of the
system - it isn't precise or exact, but it does give you a better
understanding of the system than does throwing your hands up in the
air and denying that any kind of prediction is possible.

That would be kind of useless, wouldn't it? How do you propose to
know the climate without knowing the course and temperature of the
Gulf Stream, that huge moderating influence to the U.K.'s (and
Europe's?) weather?

Surely we've not forgotten the large affect El Niño has on our
weather? Or the jet stream, for that matter. How can one make claims
about the future climate without knowing these?

And how can you project these surface effects without knowing the deep
ocean currents?

How can you intelligently project the influence and future of clouds--
enormously important to global warming/cooling--if you don't know
humidities and temperatures in the mountains / deserts / etc.?

Plenty of weather forecasting models take all of those into
consideration and consider and forecast this as well as possible to same
extent as to-as-well-as-possible in New Jersey, eastern Maryland, southern
England, Louisiana, Florida, or 800 miles west of the Cape Verde Islands!

The global climate *is* a heat transport problem--heat (sunlight) is
input at the equator, and circulated to the poles by various air and
ocean currents--and I believe (but can't swear) it's modelled that
way. That understanding is supported by this quote from Rich's link:

"The measurements also show that we have a flaw in some of our global
atmospheric circulation models. Now we can go back and improve those
models to better predict the temperatures in the middle and upper
atmospheres throughout both hemispheres."

I have the advantage of, through a friend, knowing one of THE
professionals who does this. I'll have to inquire next time I get the
chance.

- Don Klipstein (don@xxxxxxxxx)
.

Relevant Pages

  • Re: And I thought it would be cold!
    ... The weather at home has ... So it's always at positive temperatures ... Actually to get to the core of the problem, the ocean is a fluid, ... waters sink down, while the warmer waters below moves ...
    (uk.sci.weather)
  • Re: SPD Weather
    ... No offense meant but where I live the forecasting here is probably the ... are experienced understand how the ocean affects our weather. ... And make no mistake about it, the ocean does affect our weather even ...
    (microsoft.public.pocketpc)
  • OT Current Climate Models May Underestimate Future Warming
    ... evidence that tropical Atlantic Ocean temperatures may have once reached ... The study suggests that climate models ... Hole Oceanographic Institution through the Ocean and Climate Change ...
    (alt.sports.basketball.nba.la-lakers)
  • Re: cap-and-trade
    ... Record low in Rochester, NY – Ties record set in 1926 - NATIONAL ... WEATHER SVC BUFFALO NY… THE TEMPERATURE AT THE ROCHESTER AIRPORT ... Temperatures in Concord, NH fell to 47-degress, the lowest ... INCHES OF SNOW... ...
    (alt.machines.cnc)
  • Re: How do you say Global Cooling in Swedish? NBC
    ... over 100 years this week as winter weather takes hold of the country, ... "It is far below average temperatures, ... With more snow on the way in from the Baltic Sea, ... Probably the same way you say Global Warming in Cleveland Ohio. ...
    (rec.music.artists.springsteen)