[geeks] More on global warming

[Shannon Hendrix]

On Dec 24, 2007, at 12:59 PM, Dr Robert Pasken wrote:

> Model verification is done at several levels. First thing to do is  
> discard
> the notion that different models MUST produce the same results.

If they are modeling the same thing, they certainly should.

If they are not, then that's like saying that chocolate doesn't taste  
like Spam(TM).

Of course it doesn't.

> differencing to compute dereivatives. A simple assesment of possible
> combinations of methods produces a lot of different models with  
> possibly
> different results, but this is exactly the point of the different  
> models.

We also have to breeak things up like that because we have a tiny  
fraction of the power needed to do much else.

That runs the risk of not being able to account for the interaction of  
the parts.

Maybe the effect is small, or maybe it would change everything.

The longer the time period being modeled, the more likely the effect  
will be great.

Of course, my interest in this kind of thing is that it shows how much  
computational power there is in the world around us.  I'm interested  
in looking at natural events in terms of how much computation is  
involved in it.

Just for example, in its lifetime a tree will learn how to fight off  
diseases.  That represents computation power, and currently we could  
spend 100 years of worldwide computer power and not do but a fraction  
of what an oak figures out in the same time period.

All this, and computational power isn't even the biggest hurdle: data  
is.

> Models are verified in a lot of different ways. A simple method is to
> start with todays data and run the model backwards several hundred  
> years
> to determine if it reproduces history. The biggest flaw with this  
> method
> is verification data doesn't extend back in time far enough nor in  
> detail
> enough to do a good comparison.

...which seems to indicate that we don't have the needed knowledge.

Maybe we are on the right track, but my view of history, especially  
that related to computation, says we have at best a 50/50 chance of  
being right.

If you look back over the last 100 years, a lot of ideas have changed  
about climate models, some of it 180 degree reversals.

I expect the same for the future.

We are probably learning, but some of the stuff we are trying to do  
isn't a matter of power or knowledge, but time.

What we lack most of all is literally centuries of well recorded  
observation and verification of our models, and only time will give us  
that.

> limited surface observations in the US. How do I know if I missed a  
> little
> bit in space and time when verifying the model.

You don't.  You also don't know that your model might be accurate for  
99 hurricanes, and miss the big one that wipes out New York.

> and run the model forward to compare the forward results against
> observations. This method allows me to determine if a "busted  
> forecast"
> was due to spatial or temporal errors. The dust bowl was forecasted,  
> but
> to far north/south/east/west or earlier/latter than what actually

Why only spatial or temporal?

How about lack of data, incorrect programming, faulty models, invalid  
assumptions, and a few dozen other problems even the most basic  
computer programs might have?

BTW: Who forecast the dust bowl from a mathematical model?

I've read a lot about farmers and native Americans who warned it was  
coming, but they said that because they knew were were over farming  
and depleting aquifers, which I've always read was the primary cause,  
not other external events.

A little off topic I guess, but sometimes I wonder if we paid a little  
more attention to ancient lessons, we'd even do better with our  
modeling.

> happened. Hansen et al (1988) described a model verification that  
> with the
> exception of the timing of temperature declines due to vulcanic  
> eruptions
> forecasted the global average temperature very well. Hansen et al.  
> (1988)
> assumed vulcanic eruption would occur, but assigned the timing of  
> these events
> at 50 years apart. The model predicts to 2020 and so far the global  
> average
> temperatures are within the error bars. That model is over 20 years  
> old.

Hansen believes that CO2 is not yet a factor, and won't be until  
another ten years have passed.  He says current warming was caused by  
aerosols, and this will shift to CO2 as the largest factor during the  
next ten years.

His worry is that CO2 changes will have a greater effect, especially  
when China surpasses the US, and that will push us past a tipping  
point as he calls them.

According to predictions, US output might drop in half soon, some say  
it already has if you compared to the 1970s, when our industrial  
pollution was at disaster-like levels.

During the same time, China will reach our 1970s levels and pass them  
like a bullet, and I suppose other countries will too, but China is  
the one everyone is looking at.

The net result is that our reductions will have no net effect, and  
will continue long term CO2 cycles rather than us seeing them decrease.

It seems to make sense, except:

I have read many times that mankinds CO2 output is not even at the  
magnitude of noise in the signal compared to nature, so I can't quite  
understand how it has such a large effect.

The numbers I've seen seem to be around 9 billion tons per year  
manmade, while nature fluctuations more than that, and its output is  
300 billion and up.

If nature can change by 20 billion tons (or a lot more according to  
some), why does our 9 billion tons have such a huge effect?

> The biggest problem is that the global warming nay sayers are cherry
> picking results.

So are their counterparts.

There is far too much profit from pushing conflict rather than  
solutions right now.

Hansen has heavy ties with people who have a vested interest in his  
models being considered correct.

Does that corrupt him?  I have no idea, but I wish that we could limit  
the stakes to only the important things, rather than short term profit  
and power.

I still say: I see no consensus, just a huge mob fight, and no one  
doing much to plan for whatever is coming, which in the end is far  
more important.

Even without global warming, the native Americans say that huge  
hurricanes like we've never seen before hit the US east coast.

The last time the population was probably 20K at most.

The next time it will hit an east coast that currently has 60% of the  
total national population.

We know a lot of what we are doing will make an event like that worse,  
but we do pretty much nothing at all about it.

What difference does it make what climate model is accurate, if we  
plan to do nothing in reaction to what we learn?

-- 
Shannon Hendrix
shannon at widomaker.com