climate models
Dave Halliwell
It has been about a month since this was originally posted. I was sick
for a week at the time, and then Christmas intervened. Since Shearer
has now resumed posting material on the subject, a response is long
overdue. Obviously, he has taken silence as an indication of victory.
In <19951214....@almaden.ibm.com>, jshe...@VNET.IBM.COM writes:
| I continue to argue with Halliwell about stability.
| I had posted.
|> I continue to discuss a model in which the ability of the
|>atmosphere to absorb shortwave radiation is temperature dependent.
Given the time since that model was posted, I will append Shearer's
description of the model at the bottom of this post.
|> I had posted an analysis of (among other things) when this
|>situation would be stable (using a two component atmosphere, earth
|>model suggested by Pierrehumbert). Halliwell responded by disputing
|>my stability conclusions based on his numerical calculations showing
|>instability.
|> Briefly Halliwell's calculations prove nothing because they
|>are using too big a timestep (or equivalently too large values for Ha
|>and He). This is apparent from the negative temperature values he
|>obtains.
| Halliwell responded:
|> If you examine the output for b=4 in my tabulated results, you will
|>notice that the "correction" value sends the temperatures further _away_
|>from the equilibrium value. No change in Ha, He, or time step will alter
|>that fact.
| So what? A trajectory which converges to 0 may be moving away
|from 0 at some times (even in linear systems). Think of spiraling
|in along near ellipses.
Trajectories that move _away_ from equilibrium are unlikely (to say
the least) to eventually converge on equilibrium, and that was the sort
of behaviour that was observed in my post. Your description applies
to systems that are _not_ already at equilibrium. If you can't tell
the difference between a system at equilibrium and one that is not,
then there isn't much I can do for you.
| I had also posted:
|> Hence I believe if Halliwell repeats his calculations
|>with a much smaller timestep the instability he is seeing will go
|>away.
| Halliwell responded:
|> It can't possibly go away in the case where a positive departure from
|>equyilibrium leads to a further positive departure. All that a smaller
|>time step will do is make the process take longer for the instability to
|>show up.
|
| As noted above this is wrong.
Not from a position already starting at equilibrium.
|I tried your b=4, d=.5 case with
|Ha=.5, He=1.0. A timestep of 1.0 is quite unstable. .1 appears to be
|stable (although not a good approximation to the differential equations).
|.01 seems to be stable and approximate the differential equations. (I
|calculated that timesteps less than .1068 will be stable).
| Btw what values of Ha and He were you using? I was unable to
|reproduce your numbers. Please try smaller timesteps (or equivalently
|smaller Ha and He values). If you still see instability please post
|enough details so your results can be reproduced.
(As a refresher, Shearer's model looks at surface and atmospheric
temperature, Te and Te, in a radiative equation where absorptivity
(a) is dependent on Ta, and rates of temperature response are
related to the values of He and Ha.)
I was using Ha=0.5, and He=1.0, but there was an error in my
spreadsheet formulae (a relative instead of a fixed cell address).
When the error is corrected, I find the system of equations to be
stable, so my original analysis of instability was incorrect. (I reduced
Ha to 0.005 and He to 0.01, with a time step of 1, which is equivalent
to leaving Ha and He at 0.5 and 1.0 and reducing the time step to 0.01).
I then proceeded to test the behaviour of the system when
inititalized with non-equilibrium temperatures. The behavour of the
system in question is one in which Ta and Te (If plotted as an XY pair)
spiral gradually towards the equilibrium value, as Shearer discusses
above.
If you examine the time series for either Ta or Te, you find that
either will oscillate back and forth about the equilibrium value,
gradually getting closer to equilibrium. (Again, rather than
cluttering up the text of this post with data, I will append a
portion of the results to the end of this post.) The system is very
_slow_ to approach equilibrium, and a great number of oscillations
are required. The example at the bottom of this post starts out 1 degree
warmer than equilibrium (for both Ta and Te), and shows oscillations
with an amplitude considerably _larger_ than the initial perturbation.
The system _amplifies_ the initial perturbation.
What does this mean? It means that Shearer has described an AC
system, and not one in which a DC electrical analogue is appropriate.
The original position, as described by Tobis and supported by myself
and Pierrehumbert is that the climate system is one in which DC
analogues are appropriate. In such a system (DC), feedbacks
CANNOT reverse the sign of a forcing. Shearer's assertion to the
contrary is still just an assertion.
In AC systems, a perturbation leads to responses which alternate
in sign over time. The concept of feedback "reversing" the sign
of the perturbation doesn't mean the same thing when the
perturbation's sign is time-dependent.
All Shearer has done is show that he can't tell the difference
between AC and DC systems.
Now, given Shearer's system as it stands, there are a few other
characteristics of it that are worth mentioning. First, the value
of Ta must remain between 269.5 and 287K for absorptivity to
remain between 0 and 1. This is a narrow range. For a solar input
of 342 W/sq.m, a situation where Ta and Te are 5C warmer than their
equilibrium values leads to a time-dependent solution where Ta reaches
the maxiumum.
Secondly, the system is highly sensitive to cyclic variations in
solar input (S). For example, varying S sinusoidally, with an
amplitude of +/-10%, at a frequency given by sin(t/10) also leads to
the temperature reaching the maximum, along with a very complex pattern
for Ta and Te (i.e. it is no longer a simple spiral). This is the
result of the system's very slow approach to equilibrium: a changing
value of the forcing cannot be compensated for quickly, and the
system is likely to reach an alternate fixed state before it can
zero in on the intermediate "equilibrium" state.
Why can we say that the climate system doesn't show this sort of
AC behaviour? Responses to daily and seasonal variations in solar
forcing don't show it. The entire atmosphere responds fairly quickly to
seasonal variations, and the lower portion responds quickly to daily
variations.
When it comes to deep ocean temperatures, time reponses become
a factor (as has been discussed many times in this group), and
the DC analogy may be problematic. However, such systems are prone
to shifts in state, which would likely lead to major changes in
climate. Shearer may think that the possibility of such major shifts
is a reason to *not* worry about CO2-induced climate changes, but I
do not share that position. A climate in which DC analogues are
appropriate is a lot less problematic than one in which they are not.
Now, if Shearer would restrict his future "examples" to ones that
actually apply to the topic at hand, the discussion might be worth
continuing.
The remainder of this post is Shearer's original description of the
model, followed by a lot of lines of temperature response in my
implimentation of it. If you don't want to read that, skip the rest
of the post.
-------
| Newsgroups: sci.environment
| Subject: climate models
| Message-ID: <19951207....@almaden.ibm.com>
| From: jshe...@VNET.IBM.COM
| Date: Thu, 7 Dec 95 19:34:15 EST
| Lines: 185
|
| I continue to discuss the following model which I had
| proposed in an earlier post.
[...preamble deleted - DH]
| I should have chosen a means of heating the
| surface that only has indirect effects on the atmosphere. Assume
| say a network of nuclear power stations or as Halliwell suggested in
| a later post increased geothermal heat flow. In what follows I will
| discuss such a revised model (ie keep the solar constant constant and
| force the surface directly in some way).
| First we will look at the equilibrium behavior. Following
| Pierrrehumbert let:
| Ta - temperature of isothermal slab atmosphere
| Te - temperature of surface
| a(T) - shortwave absorption of atmosphere at temperature T
| S - solar flux
| c*T**4 - blackbody radiation at temperature T
| f - surface forcing
| Assume surface albedo 0 and atmospheric emissivity 1.
| Assume radiative energy transfer only. Then the energy leaving
| the atmosphere will be 2*c*Ta**4, the energy leaving the surface
| will be c*Te**4, the energy absorbed in the atmosphere will be
| a(Ta)*S+c*Te**4 and the energy absorbed in the surface will be
| (1-a(Ta))*S+c*Ta**4+f. At equilibrium the energy emitted and
| absorbed will balance in the atmosphere and at the surface. So
| we have the following equations.
|
| a(Ta)*S + c*Te**4 = 2*c*Ta**4 <1>
|
| (1-a(Ta))*S + c*Ta**4 + f = c*Te**4 <2>
|
| Adding <1> and <2> we may solve for Ta**4 ie:
|
| Ta**4 = (S+f)/c <3>
|
| Note Ta is independent of the form of a(Ta). We now choose a(Ta)
| as follows:
|
| a(Ta) = b*((c/S)*Ta**4 - 1) + d <4>
|
| Note when f=0, Ta**4=(S/c) and a(Ta)=d. a(Ta) must lie between
| 0 and 1 which will be true for values of Ta near (S/c)**.25 if
| 0<d<1.
| Now plug <4> and <3> into <1> and solve for Te. We
| obtain:
|
| Te**4 = ((2-d)*S + (2-b)*f)/c <5>
|
| Hence when b is greater than 2 the surface temperature at
| equilibrium moves in a direction opposite to the forcing f.
[stability analysis deleted]
The following is the first 110 time steps of my implementation of
the model. Equilibrium values for Te and Ta are 1 degree cooler than
the values at time step 0.
S = 342
b = 4
c = 5.67E-08
d = 0.5
He,a 0.01 0.005
t Te Ta
0 309.4 279.7
1 309.2 279.8
2 309.0 279.8
3 308.8 279.9
4 308.6 280.0
5 308.4 280.1
6 308.2 280.1
7 308.0 280.2
8 307.8 280.3
9 307.6 280.3
10 307.4 280.4
11 307.2 280.4
12 307.0 280.5
13 306.9 280.5
14 306.7 280.5
15 306.5 280.6
16 306.4 280.6
17 306.2 280.6
18 306.1 280.7
19 305.9 280.7
20 305.8 280.7
21 305.7 280.7
22 305.5 280.7
23 305.4 280.7
24 305.3 280.7
25 305.2 280.7
26 305.1 280.7
27 305.0 280.7
28 304.9 280.7
29 304.9 280.7
30 304.8 280.7
31 304.7 280.7
32 304.7 280.7
33 304.6 280.6
34 304.6 280.6
35 304.6 280.6
36 304.5 280.5
37 304.5 280.5
38 304.5 280.5
39 304.5 280.4
40 304.5 280.4
41 304.5 280.3
42 304.5 280.3
43 304.5 280.2
44 304.5 280.2
45 304.6 280.1
46 304.6 280.1
47 304.6 280.0
48 304.7 280.0
49 304.7 279.9
50 304.8 279.9
51 304.9 279.8
52 304.9 279.7
53 305.0 279.7
54 305.1 279.6
55 305.2 279.6
56 305.2 279.5
57 305.3 279.4
58 305.4 279.4
59 305.5 279.3
60 305.6 279.2
61 305.7 279.2
62 305.8 279.1
63 305.9 279.0
64 306.0 279.0
65 306.2 278.9
66 306.3 278.8
67 306.4 278.8
68 306.5 278.7
69 306.6 278.7
70 306.8 278.6
71 306.9 278.5
72 307.0 278.5
73 307.1 278.4
74 307.2 278.4
75 307.4 278.3
76 307.5 278.3
77 307.6 278.2
78 307.7 278.2
79 307.9 278.1
80 308.0 278.1
81 308.1 278.0
82 308.2 278.0
83 308.3 277.9
84 308.4 277.9
85 308.6 277.9
86 308.7 277.8
87 308.8 277.8
88 308.9 277.8
89 309.0 277.7
90 309.1 277.7
91 309.2 277.7
92 309.3 277.7
93 309.4 277.6
94 309.5 277.6
95 309.6 277.6
96 309.6 277.6
97 309.7 277.6
98 309.8 277.6
99 309.9 277.6
100 309.9 277.5
101 310.0 277.5
102 310.1 277.5
103 310.1 277.5
104 310.2 277.5
105 310.2 277.5
106 310.3 277.5
107 310.3 277.6
108 310.3 277.6
109 310.4 277.6
110 310.4 277.6
--
Dave Halliwell I don't speak for my employers, and you
Edmonton, Alberta shouldn't expect them to speak for me.
jshe...@vnet.ibm.com
which I posted some time ago. I will split my reply to Halliwells
last post into several parts (posted as time permits).
Dave Halliwell posted:
> I was using Ha=0.5, and He=1.0, but there was an error in my
>spreadsheet formulae (a relative instead of a fixed cell address).
>
> When the error is corrected, I find the system of equations to be
>stable, so my original analysis of instability was incorrect. (I reduced
>Ha to 0.005 and He to 0.01, with a time step of 1, which is equivalent
>to leaving Ha and He at 0.5 and 1.0 and reducing the time step to 0.01).
This is a good illustration of one reason why I am skeptical
that large complicated climate models have much value. Even if you
stipulate that the physics of the model is correct there remains the
question of how you can be confident that the model has been correctly
implemented on the computer. It is almost impossible to debug a large
computer code when you don't know what the correct answers are.
The model Halliwell and I are discussing is trivial compared
to current large climate models. Yet Halliwell's first attempt to
implement it was incorrect. In this case the error was detected because
I did a completely independent implementation of the same model and
obtained different results. I doubt very many of the current large
climate models have been implemented independently by several groups.
I suspect if this were done routinely it would often uncover bugs.
There are psychological factors involved also. If the program
results match what the implementer expected he is likely to search less
vigorously for bugs than when the program results are unexpected. This
creates a bias which may cause results of several groups to converge
to a faulty consensus. (This has been observed to occur in the
determination of physical constants by several groups.)
James B. Shearer (email j...@watson.ibm.com)
jshe...@vnet.ibm.com
I posted in which feedback reverses the sign of the response to
forcing.
Dave Halliwell posted:
> If you examine the time series for either Ta or Te, you find that
>either will oscillate back and forth about the equilibrium value,
>gradually getting closer to equilibrium. (Again, rather than
>cluttering up the text of this post with data, I will append a
>portion of the results to the end of this post.) The system is very
>_slow_ to approach equilibrium, and a great number of oscillations
>are required. The example at the bottom of this post starts out 1 degree
>warmer than equilibrium (for both Ta and Te), and shows oscillations
>with an amplitude considerably _larger_ than the initial perturbation.
>The system _amplifies_ the initial perturbation.
>
> What does this mean? It means that Shearer has described an AC
>system, and not one in which a DC electrical analogue is appropriate.
>The original position, as described by Tobis and supported by myself
>and Pierrehumbert is that the climate system is one in which DC
>analogues are appropriate. In such a system (DC), feedbacks
>CANNOT reverse the sign of a forcing. Shearer's assertion to the
>contrary is still just an assertion.
>
> In AC systems, a perturbation leads to responses which alternate
>in sign over time. The concept of feedback "reversing" the sign
>of the perturbation doesn't mean the same thing when the
>perturbation's sign is time-dependent.
>
> All Shearer has done is show that he can't tell the difference
>between AC and DC systems.
You appear to be a bit confused. The perturbation we are
interested in is directly heating the surface (say by a network of
nuclear power stations). This is introduced by changing the equations
(adding the f term in my previous posts). This forcing term is not
time-dependent and therefore has an unambiguous sign. Changing the
equations changes the position of the equilibrium. If the system
starts at the old equilibrium when the forcing is turned on it will
converge to the new equilibrium. Therefore as time goes to infinity
the response has an unambiguous sign relative to the original
equilibrium. In this example Te (the temperature of the surface) will
decrease, a response of opposite sign to the forcing, which is the
point of the example.
As for your complaint that the system approaches equilibrium
in an oscillatory fashion, I don't see what this has to do with
anything. However if it bothers you, you can look at the system with
Ha/He =.1 which does not.
Dave Halliwell
> Halliwell and I have been discussing a simple climate model
>which I posted some time ago. I will split my reply to Halliwells
>last post into several parts (posted as time permits).
> Dave Halliwell posted:
>> I was using Ha=0.5, and He=1.0, but there was an error in my
>>spreadsheet formulae (a relative instead of a fixed cell address).
>>
>> When the error is corrected, I find the system of equations to be
>>stable, so my original analysis of instability was incorrect. (I reduced
>>Ha to 0.005 and He to 0.01, with a time step of 1, which is equivalent
>>to leaving Ha and He at 0.5 and 1.0 and reducing the time step to 0.01).
> This is a good illustration of one reason why I am skeptical
>that large complicated climate models have much value.
This is a good illustration why I think that your criticism of climate
models in this group is based on a position of ignorance, and that you
are more interested in scoring debating points than actually discussing
the science.
>Even if you
>stipulate that the physics of the model is correct there remains the
>question of how you can be confident that the model has been correctly
>implemented on the computer. It is almost impossible to debug a large
>computer code when you don't know what the correct answers are.
Yet in your original post, you never actually bothered to implement
the model at all. You just set up the equations, and looked at the
equilibrium values and initial trajectory. You never ran the model to
show the time-dependence that I ended up with, which showed that your
system did not fit the DC electrical analogy in question.
Even after you questioned my initial results and performed such a
simulation, you didn't bother to post your results. Perhaps because you
didn't want to show the oscillatory nature of the temperatures? Perhaps
because you realized the error that you had made?
When I implemented the model originally, I posted results. When I
detected the error, after you questiuoned the results, I acknowledged the
error and posted further results. If I had done as you had done - argue a
position without actually providing the output - I could have hidden
anything I didn't like. I did not hide the results because that is the
way that science is done.
> The model Halliwell and I are discussing is trivial compared
>to current large climate models. Yet Halliwell's first attempt to
>implement it was incorrect. In this case the error was detected because
>I did a completely independent implementation of the same model and
>obtained different results.
Which you did not post.
>I doubt very many of the current large
>climate models have been implemented independently by several groups.
Argument from personal incredulity. Can you back up that assertion?
Several groups have broadly _similar_ models (similar in principles of
physics) that _are_ implemented pretty much independently. Do they all
have the same "bugs"?
Do you really think that scientific models undergo no more validation
than I performed for a Usenet post? No, you just use my post as an excuse
to make unwarranted attacks on the climatological community. All you want
to do is score debating points.
The fact is that components of the complex models are verified
individually, with comparison to real data wherever available.
>I suspect if this were done routinely it would often uncover bugs.
Here we have it: Shearer is so determined to reject the idea of
CO2-induced warming that he dismisses over 30 years of work by hundreds
or thousands of researchers in climatology as "bugs in the code".
> There are psychological factors involved also. If the program
>results match what the implementer expected he is likely to search less
>vigorously for bugs than when the program results are unexpected. This
>creates a bias which may cause results of several groups to converge
>to a faulty consensus. (This has been observed to occur in the
>determination of physical constants by several groups.)
Of course, Shearer is the only person that isn't affected by bias.
Regardless of what anybody comes up with in the way of scientific
evidence supporting CO2-induced warming, Shearer is going to reject it as
"bug-ridden" or "biased". He's got his head stuck in the sand.
If CO2-induced warming is all just "bugs in the code", then why is it
that the coal and gas industry, and other "skeptical" groups have never
produced a model of global climate that doesn't show warming?
Rich Puchalsky
: Dave Halliwell posted:
: > I was using Ha=0.5, and He=1.0, but there was an error in my
: >spreadsheet formulae (a relative instead of a fixed cell address).
: This is a good illustration of one reason why I am skeptical
: that large complicated climate models have much value. Even if you
: stipulate that the physics of the model is correct there remains the
: question of how you can be confident that the model has been correctly
: implemented on the computer. It is almost impossible to debug a large
Here Halliwell takes time, purely out of the goodness of his heart, to
actually set up a spreadsheet to test one of Shearer's propositions. No
publication is expected; no special rigor is desired for this off-hand
Usenet-inspired little task. Yet when Halliwell makes an error setting
up the spreadsheet -- which anyone could make, given the non-importance
of the matter and therefore the lack of careful checking involved --
Shearer has the *gall* to hold this up as analogous to the work of
professional modellers that has been checked and reviewed prior to
publication in a journal.
Shearer's entire response (at least initially -- he promises more later)
consisted of generalizing from this error of Halliwell's to science at
large. A pages-long response.
This kind of crap has to be contributing to the exodus of scientists from
sci.environment. I note that Pierrehumbert just posted that he no
longer has time to participate; Grumbine also recently left. I'll refrain
from going through the full list.
I was going to end this post with some kind of scathing conclusion, but
none is really needed. The Shearers reading this group wouldn't understand;
everyone else already does.
Dave Halliwell
> I continue to respond to Halliwell's comments on a model
>I posted in which feedback reverses the sign of the response to
>forcing.
> Dave Halliwell posted:
>> What does this mean? It means that Shearer has described an AC
>>system, and not one in which a DC electrical analogue is appropriate.
>>The original position, as described by Tobis and supported by myself
>>and Pierrehumbert is that the climate system is one in which DC
>>analogues are appropriate. In such a system (DC), feedbacks
>>CANNOT reverse the sign of a forcing. Shearer's assertion to the
>>contrary is still just an assertion.
>> In AC systems, a perturbation leads to responses which alternate
>>in sign over time. The concept of feedback "reversing" the sign
>>of the perturbation doesn't mean the same thing when the
>>perturbation's sign is time-dependent.
>>
>> All Shearer has done is show that he can't tell the difference
>>between AC and DC systems.
> You appear to be a bit confused.
No, _you_ appear to be a bit confused.
>The perturbation we are
>interested in is directly heating the surface (say by a network of
>nuclear power stations).
There are two aspects to the increase in surface heating: one is the
simple process of dissipating heat from the surface, which would apply to
instantaneous or temporary heating as well as continuous heating. The
other is the approach to a new equilibrium, resulting from continuous
heating.
>This is introduced by changing the equations
>(adding the f term in my previous posts).
Or by making a previously-zero value non-zero. It's just a change in
the value of f, not a new system.
>This forcing term is not
>time-dependent and therefore has an unambiguous sign. Changing the
>equations changes the position of the equilibrium. If the system
>starts at the old equilibrium when the forcing is turned on it will
>converge to the new equilibrium. Therefore as time goes to infinity
>the response has an unambiguous sign relative to the original
>equilibrium. In this example Te (the temperature of the surface) will
>decrease, a response of opposite sign to the forcing, which is the
>point of the example.
An instantaneous (transient) release of heat will also cause a shift in
surface temperature, without any change in long-term equilibrium. In that
case, Te will increase to begin, and must eventually decrease in order to
get back to equilibrium. That will be the case _regardless_ of any
possible feedback mechanisms.
> As for your complaint that the system approaches equilibrium
>in an oscillatory fashion, I don't see what this has to do with
>anything.
...which suggests that you don't understand the distinction between AC
and DC systems. In AC systems, the question of phase (and phase lag) is
critical. It's easy to get a response that is out of phase with the
forcing, causing an apparent "reversal". In DC systems, there is no such
thing.
>However if it bothers you, you can look at the system with
>Ha/He =.1 which does not.
The system with Ha/He=0.1 is still an AC system. Take a look at how it
responds to a _transient_ pulse of surface heating. I used He=0.01 and
Ha=0.001, with a time step of 1. Turn on "f" for _one_ time step, and
watch what happens: first Te increases, then decreases _below_ the
original equilibrium, then increases again, etc., oscillating in ever-
decreasing amounts around the original (=new) equilibrium.
Dave Halliwell
>jshe...@VNET.IBM.COM wrote:
>: This is a good illustration of one reason why I am skeptical
>: that large complicated climate models have much value. Even if you
>: stipulate that the physics of the model is correct there remains the
>: question of how you can be confident that the model has been correctly
>: implemented on the computer. It is almost impossible to debug a large
>Here Halliwell takes time, purely out of the goodness of his heart, to
>actually set up a spreadsheet to test one of Shearer's propositions. No
>publication is expected; no special rigor is desired for this off-hand
>Usenet-inspired little task. Yet when Halliwell makes an error setting
>up the spreadsheet -- which anyone could make, given the non-importance
>of the matter and therefore the lack of careful checking involved --
>Shearer has the *gall* to hold this up as analogous to the work of
>professional modellers that has been checked and reviewed prior to
>publication in a journal.
Not only that, but it's _Shearer's_ model. I was implementing the
model because his original post did not examine the overall
time-dependence of the system, from initial conditions to equilibrium: he
only looked at the initial trajectory and the final equilibrium value.
In a "publication" analogy, Shearer is complaining because the reviewer
didn't do a perfect job of completing work that the author failed to
provide. Most reviewers would probably just pull out the REJECT stamp,
and not bother going into detail.
>Shearer's entire response (at least initially -- he promises more later)
>consisted of generalizing from this error of Halliwell's to science at
>large. A pages-long response.
He's now posted another reply, to which I have also responded, in
which he again fails to examine in sufficient detail the time-dependence
of his model. It's come back to me for a _second_ review, and he's still
not getting it right.
>This kind of crap has to be contributing to the exodus of scientists from
>sci.environment. I note that Pierrehumbert just posted that he no
>longer has time to participate; Grumbine also recently left. I'll refrain
>from going through the full list.
Tobis hasn't been around much, and I'm pretty damn close to giving up
as well. If Shearer actually wants to learn something about climatology,
instead of proudly displaying his ignorance as if it is wisdom, then I
might continue the thread. Otherwise, it's getting a little tiresome to
constantly pursue his incomplete meanderings. He wants "debate" instead of
"discussion".
>I was going to end this post with some kind of scathing conclusion, but
>none is really needed. The Shearers reading this group wouldn't understand;
>everyone else already does.
Amen.
jshe...@vnet.ibm.com
> As for your complaint that the system approaches equilibrium
>in an oscillatory fashion, I don't see what this has to do with
>anything.
> ...which suggests that you don't understand the distinction between AC
>and DC systems. In AC systems, the question of phase (and phase lag) is
>critical. It's easy to get a response that is out of phase with the
>forcing, causing an apparent "reversal". In DC systems, there is no such
>thing.
You mean like in the climate system the seasons are out of
phase with solar insolation. At this point I am going to have to ask
you to define AC and DC systems. Please explain in detail why the
climate system with its periodic (ie AC) forcing is a DC system while
my example with its constant (ie DC) forcing is an AC system.
I continued:
>However if it bothers you, you can look at the system with
>Ha/He =.1 which does not.
Dave Halliwell responded:
> The system with Ha/He=0.1 is still an AC system. Take a look at how it
>responds to a _transient_ pulse of surface heating. I used He=0.01 and
>Ha=0.001, with a time step of 1. Turn on "f" for _one_ time step, and
>watch what happens: first Te increases, then decreases _below_ the
>original equilibrium, then increases again, etc., oscillating in ever-
>decreasing amounts around the original (=new) equilibrium.
Not exactly. Te increases at first. It then decreases until
it moves below the original equilibrium Te (at timestep 30). It then
starts to increase again but it never returns above the equilibrium Te,
instead it approaches the equilibrium Te from below. It does not
oscillate around the equilibrium Te in "ever-decreasing amounts".
Marc Levy
other, and how differences among them get treated will be interested in this
web site:
http://sedac.ciesin.org/sedac-2.3/mvas.home/mvas-home.html
It has a very well-done guide to climate models in general, gives you
access to some scenarios of three of the major models, and even lets you do
some runs of your own.
----------------------
Marc Levy
ml...@wws.princeton.edu
jshe...@vnet.ibm.com
>I suspect if this were done routinely it would often uncover bugs.
> Here we have it: Shearer is so determined to reject the idea of
>CO2-induced warming that he dismisses over 30 years of work by hundreds
>or thousands of researchers in climatology as "bugs in the code".
This is untrue. I do not reject the idea of CO2 induced
warming. Some time ago I posted the following to this group:
> I will assume your questions are based on a hypothetical
>2*CO2 atmosphere. I will admit that this group of experts expects
>some global warming to occur and that there appears to be about as
>much expert support for major warming as minor warming.
> For what it's worth my current estimates are:
> sensitivity < 0 - 1%
> sensitivity < 1.5C - 30%
> 1.5C < sensitivity < 4.0C - 60%
> 4.0C < sensitivity - 10%
>These estimates are not firmly held and are subject to change. I
>would be interested in the estimates of other posters.
These are not the figures someone "determined to reject
idea of CO2-induced warming" would give. I do not believe I have
posted anything rejecting the idea of CO2-induced warming. I
suggest Halliwell substantiate his claim or withdraw it.
I had continued:
> There are psychological factors involved also. If the program
>results match what the implementer expected he is likely to search less
>vigorously for bugs than when the program results are unexpected. This
>creates a bias which may cause results of several groups to converge
>to a faulty consensus. (This has been observed to occur in the
>determination of physical constants by several groups.)
Dave Halliwell responded:
> Of course, Shearer is the only person that isn't affected by bias.
I have never claimed to be unaffected by bias. I claim
everybody is affected by bias, sometimes in quite subtle ways.
Some pervasive ways in which people are influenced by bias are the
following:
1. People tend to believe what it is in their interest
for others to believe. For example that an auto accident was not
their fault.
2. People tend to interpret new information in terms of
a preexisting mental model. They are reluctant to discard a model
before the evidence against it becomes overwhelming. For example a
pilot may fly a plane into a mountain while thinking he is somewhere
else entirely. He will have incorporated what might seem like ample
cues that something is wrong into his erroneous mental model of where
he is.
3. People tend to adjust their opinions so as not to stray
too far from the opinions of their peers. For example analysts
estimates of company earnings cluster more tightly than justified
by the actual predictability of such earnings. (Ie the actual
earnings are often far outside the range of expert opinion.)
Scientists can try to minimize the effects of these biases
on their work. However in my view they can never be entirely
successful. Furthermore a blind denial that potential problems
exist does not inspire confidence.
Dave Halliwell continue:
> Regardless of what anybody comes up with in the way of scientific
>evidence supporting CO2-induced warming, Shearer is going to reject it as
>"bug-ridden" or "biased". He's got his head stuck in the sand.
This is again untrue. I accept that what is popularly called
the "greenhouse effect" warms the surface of the earth. I accept that
CO2 is a greenhouse gas. I accept that the observed increase in the
CO2 content of the atmosphere is almost surely of anthropogenic
origin. I accept that there is a plausible argument that increasing
the CO2 content of the atmosphere will increase the magnitude of the
greenhouse effect thereby warming the surface. I accept that simple
climate models allow crude estimates of the amount of warming that
can be expected.
I reject (or at least am extremely skeptical of) claims that
very elaborate climate models requiring large computer codes to
implement allow (or will allow anytime soon) us to significantly
refine estimates of climate change due to CO2 available from simpler
models. I believe it is the people suggesting such models be used
to evaluate policy options who have their heads stuck in the sand.
Consulting the entrails of a goat would be as sensible.
Dave Halliwell
> I posted:
>> As for your complaint that the system approaches equilibrium
>>in an oscillatory fashion, I don't see what this has to do with
>>anything.
> Dave Halliwell responded:
>> ...which suggests that you don't understand the distinction between AC
>>and DC systems. In AC systems, the question of phase (and phase lag) is
>>critical. It's easy to get a response that is out of phase with the
>>forcing, causing an apparent "reversal". In DC systems, there is no such
>>thing.
> You mean like in the climate system the seasons are out of
>phase with solar insolation.
The seasons (in terms of warm and cool months) are _in_ phase with the
forcing function of net energy input.
In the seasons, the _response_ is the result of the _current_ cycle:
if the seasonal forcing were to suddenly stop and become a _steady_
forcing, there would not be a continuing "seasonal cycle" in the
response.
>At this point I am going to have to ask
>you to define AC and DC systems.
It has to do with how they _respond_ to forcings, not the forcings
themselves. In an AC system, changings in forcings lead to time-dependent
responses (oscillations).
As an example, let's take an electric motor hooked up to a voltage
source. If the voltages cycles (an AC input), and the motor responds by
reversing itself for each cycle (and shows steady motion as soon as the
input steadies), then that motor is showing a DC response to varying
input. (Typically, a DC motor hooked up to AC line power would probably
actuualy just not move at all, since it can't respond quickly enough and
the net effect of the alternating input is a DC current of zero.)
> Please explain in detail why the
>climate system with its periodic (ie AC) forcing is a DC system
...because of the way it _responds_ to the forcing: e.g. the seasons,
where there is a direct correspondence between the forcing and the
response.
Another example is the response to a transient perturbation, such as
Mt. Pinatubo: it causes cooling, followed by a slow rise back to
"normal". It does not lead to a cooling, followed by warming back above
"normal", followed by further cooling, etc.
>while
>my example with its constant (ie DC) forcing is an AC system.
...because steady inputs, or single transient pulses, lead to
oscillatory behaviour. (This behavour is more obvious in some situation
than in others.) Your system overshot the approach back to equilibrium,
after a transient pulse. I could take advantage of that reversal, and time
my transient pulses to be in _phase_ with that oscillation (add more heat
during the "rebound" warming), and get it to do all sorts of things with
minimal input that wouldn't happen if it were a DC system.
Christoph Müller
Chris
Dave Halliwell
> I reject (or at least am extremely skeptical of) claims that
>very elaborate climate models requiring large computer codes to
>implement allow (or will allow anytime soon) us to significantly
>refine estimates of climate change due to CO2 available from simpler
>models. I believe it is the people suggesting such models be used
>to evaluate policy options who have their heads stuck in the sand.
>Consulting the entrails of a goat would be as sensible.
Since I see no evidence that you have read any of the IPCC reports,
which give a great deal of discussion regarding model validation and
reasons for thinking that the models are at least somewhat successful in
emulating real climate, you are invoking both argument ad ignoratum and
argument from personal incredulity.
Dave Pettingill
jshe...@VNET.IBM.COM writes:
> I reject (or at least am extremely skeptical of) claims that
> very elaborate climate models requiring large computer codes to
> implement allow (or will allow anytime soon) us to significantly
> refine estimates of climate change due to CO2 available from simpler
> models.
What would be more impressive is if you actually compared
some outputs. They started building models in the 1970's
There's probably some stuff that would run on a pc.
> I believe it is the people suggesting such models be used
> to evaluate policy options who have their heads stuck in the sand.
> Consulting the entrails of a goat would be as sensible.
The first part is testable. Do the model comparison and I'll
do the goat thing.
Dave Pettingill <iso...@igc.org>
Paul Farrar
Dave Pettingill <iso...@igc.apc.org> wrote:
>
>
> jshe...@VNET.IBM.COM writes:
>
> > I reject (or at least am extremely skeptical of) claims that
> > very elaborate climate models requiring large computer codes to
> > implement allow (or will allow anytime soon) us to significantly
> > refine estimates of climate change due to CO2 available from simpler
> > models.
>
> What would be more impressive is if you actually compared
> some outputs. They started building models in the 1970's
> There's probably some stuff that would run on a pc.
There is a small instructional model in Henderson-Sellers's and
McGuffie's _A Climate Modeling Primer_. It is in PC-basic. It is a
1-D model with latitude, in 10deg bands as the variable.
>...
--
Paul Farrar
http://www.datasync.com/~farrar/
far...@datasync.com
70053,3464
JoshuaHalpern
JoshuaHalpern
: I posted (regarding large complicated computer models):
: >I suspect if this were done routinely it would often uncover bugs.
: Dave Halliwell responded:
: I reject (or at least am extremely skeptical of) claims that
: very elaborate climate models requiring large computer codes to
: implement allow (or will allow anytime soon) us to significantly
: refine estimates of climate change due to CO2 available from simpler
: models. I believe it is the people suggesting such models be used
: to evaluate policy options who have their heads stuck in the sand.
: Consulting the entrails of a goat would be as sensible.
Does this also apply to the econometric models which are used for
budgeting purposes by CBO, OMB, etc, which have even greater impact
on policy options and the models used for program trading which
have enormous impact on our economic well being (well, at least
my hoped for retirement income)? If so, what is your position
as to the posturing in Washington over whose budget proposal is
honest? If you hold the econometric models to be useful, then
why are they more useful for policy decisions than the GCMs. The
later appear (at least to me) to be based more on knowledge than
the former.
As you might guess, the purpose of asking you this question is to
point out that most policy decisions today are based on models,
and comparatively the Global Climate Models are based on a lot
better data and knowledge than the economic models that we depend
on.
Almost without exception, policy decisions are made in the absence
of perfect (or even good) knowledge of the future consequences of
that action, and, as has been pointed out elsewhere, not to
decide is to decide.
Regards
Josh Halpern
:w
:q
: James B. Shearer (email j...@watson.ibm.com)
Scott Nudds
: Almost without exception, policy decisions are made in the absence
: of perfect (or even good) knowledge of the future consequences of
: that action, and, as has been pointed out elsewhere, not to
: decide is to decide.
Three months into the 1990 recession, the head of the U.S. federal
reserve stated that there was no convincing evidence that the U.S. was in
recession.
** If you line up all of the worlds economists, and point them all north,
they will still all point in different directions.
So much for economics.
--
Mark A. Friesel
It's necessary to be very careful to understand what your model is
doing. In a nutshell, valid models of the phsyical world must both
describe and predict the behavior of the systems they represent.
Although economic and financial models must do the same if they are used
to describe and predict trends, when a given model is widely accepted it
creates trends that validate it post hoc. This is because human behavior
patterns are mutable whereas the laws of nature are fixed. Physical
models attempt to describe the latter, and there is no effect of
the predictions of the model on the behavior of nature.
On 5 Feb 1996, JoshuaHalpern wrote:
> jshe...@VNET.IBM.COM wrote:
> : I posted (regarding large complicated computer models):
> : >I suspect if this were done routinely it would often uncover bugs.
> : Dave Halliwell responded:
> <massive snip, as I have only one question>
>
> : I reject (or at least am extremely skeptical of) claims that
> : very elaborate climate models requiring large computer codes to
> : implement allow (or will allow anytime soon) us to significantly
> : refine estimates of climate change due to CO2 available from simpler
> : models. I believe it is the people suggesting such models be used
> : to evaluate policy options who have their heads stuck in the sand.
> : Consulting the entrails of a goat would be as sensible.
>
> Does this also apply to the econometric models which are used for
> budgeting purposes by CBO, OMB, etc, which have even greater impact
> on policy options and the models used for program trading which
> have enormous impact on our economic well being (well, at least
> my hoped for retirement income)? If so, what is your position
> as to the posturing in Washington over whose budget proposal is
> honest? If you hold the econometric models to be useful, then
> why are they more useful for policy decisions than the GCMs. The
> later appear (at least to me) to be based more on knowledge than
> the former.
>
> As you might guess, the purpose of asking you this question is to
> point out that most policy decisions today are based on models,
> and comparatively the Global Climate Models are based on a lot
> better data and knowledge than the economic models that we depend
> on.
>
> Almost without exception, policy decisions are made in the absence
> of perfect (or even good) knowledge of the future consequences of
> that action, and, as has been pointed out elsewhere, not to
> decide is to decide.
>
> Regards
> Josh Halpern
> :w
> :q
>
>
> : James B. Shearer (email j...@watson.ibm.com)
>
>
jshe...@vnet.ibm.com
> I reject (or at least am extremely skeptical of) claims that
> very elaborate climate models requiring large computer codes to
> implement allow (or will allow anytime soon) us to significantly
> refine estimates of climate change due to CO2 available from simpler
> models. I believe it is the people suggesting such models be used
> to evaluate policy options who have their heads stuck in the sand.
> Consulting the entrails of a goat would be as sensible.
Josh Halpern asked:
>Does this also apply to the econometric models which are used for
>budgeting purposes by CBO, OMB, etc, which have even greater impact
>on policy options and the models used for program trading which
>have enormous impact on our economic well being (well, at least
>my hoped for retirement income)? If so, what is your position
>as to the posturing in Washington over whose budget proposal is
>honest? If you hold the econometric models to be useful, then
>why are they more useful for policy decisions than the GCMs. The
>later appear (at least to me) to be based more on knowledge than
>the former.
I don't trust any elaborate computer model that can't be
convincingly validated. As I have posted before in this group this
certainly includes many econometric models.
Josh Halpern continued:
>As you might guess, the purpose of asking you this question is to
>point out that most policy decisions today are based on models,
>and comparatively the Global Climate Models are based on a lot
>better data and knowledge than the economic models that we depend
>on.
I don't agree, I believe most policy decisions today are
based on politics.
Josh Halpern concluded:
>Almost without exception, policy decisions are made in the absence
>of perfect (or even good) knowledge of the future consequences of
>that action, and, as has been pointed out elsewhere, not to
>decide is to decide.
It is true that often important decisions have to be made in
the presence of uncertainty. Policy makers find agonizing over such
decisions painful. For this reason they are often very susceptible
to claims that there is a way to reduce or eliminate the uncertainty.
In ancient times this might involve consulting an oracle. Today it
might involve running an elaborate computer model. In my view the
value added is often about the same.
Steinn Sigurdsson
In a "publication" analogy, Shearer is complaining because the reviewer
didn't do a perfect job of completing work that the author failed to
provide. Most reviewers would probably just pull out the REJECT stamp,
and not bother going into detail.
Hmm, there seems to be some confusion here as to who
precisely is in the position of reviewing whose work.
jshe...@vnet.ibm.com
exception to my using errors in computer calculations by Dave
Halliwell as an excuse to expound on the hazards of buggy computer
calculations in general. I also respond to Halliwell's comments on
Puchalsky's post.
Rich Puchalsky posted:
>Here Halliwell takes time, purely out of the goodness of his heart, to
>actually set up a spreadsheet to test one of Shearer's propositions. No
>publication is expected; no special rigor is desired for this off-hand
>Usenet-inspired little task. Yet when Halliwell makes an error setting
>up the spreadsheet -- which anyone could make, given the non-importance
>of the matter and therefore the lack of careful checking involved --
>Shearer has the *gall* to hold this up as analogous to the work of
>professional modellers that has been checked and reviewed prior to
>publication in a journal.
While posts to this group are not the same as publication
in a professional journal, they may well be seen by as many people.
In my view this means posters should take some care about what they
post, particularly when they are declaring another poster was wrong.
It is true anyone can make programming errors. (This is
because writing correct programs is hard, much more so for the more
complicated climate models than for the simple model Halliwell
attempted.) However not everyone has Halliwell's blind faith in
computer calculations. In the case at issue Halliwell declared
my mathematical argument showing stability was invalid based on a
numerical calculation in which the temperature of the earth went
from 306.6 K to -2659.8 K in one time step. A more humble individual
might have considered this as an indication that he was doing
something wrong and looked for bugs.
Finally many published papers dependent on computer
calculations are not in fact checked in any meaningful way prior to
publication. If Puchalsky thinks the results of buggy computer
programs never make it into published papers, he is extremely naive.
Puchalsky continued in part:
>This kind of crap has to be contributing to the exodus of scientists from
>sci.environment. I note that Pierrehumbert just posted that he no
>longer has time to participate; Grumbine also recently left. I'll refrain
>from going through the full list.
I have never attempted to drive anyone out of this group by
vilifying them. I wonder if Puchalsky can say the same.
Dave Halliwell posted:
> Not only that, but it's _Shearer's_ model. I was implementing the
>model because his original post did not examine the overall
>time-dependence of the system, from initial conditions to equilibrium: he
>only looked at the initial trajectory and the final equilibrium value.
>
> In a "publication" analogy, Shearer is complaining because the reviewer
>didn't do a perfect job of completing work that the author failed to
>provide. Most reviewers would probably just pull out the REJECT stamp,
>and not bother going into detail.
Ok, let us consider a publication analogy. I submit a paper
showing (using mathematical arguments) what conditions are needed for
a certain simple model to have a stable equilibrium solution.
Halliwell, as reviewer, rejects the paper based on his totally bogus
numerical calculations which indicate instability. Eventually I
convince the reviewer that he is wrong. However he still doesn't
accept the paper. Instead he invents a new and equally spurious
ground (that a discussion of under what conditions a model has a
stable equilibrium is obligated to include a description of the
trajectories leading to the equilibrium) for rejecting it. At this
point I would be justified in asking the editor for a different
reviewer.
Robert Keith Fullerton
: It's necessary to be very careful to understand what your model is
: doing. In a nutshell, valid models of the phsyical world must both
: describe and predict the behavior of the systems they represent.
: Although economic and financial models must do the same if they are used
: to describe and predict trends, when a given model is widely accepted it
: creates trends that validate it post hoc. This is because human behavior
: patterns are mutable whereas the laws of nature are fixed. Physical
: models attempt to describe the latter, and there is no effect of
: the predictions of the model on the behavior of nature.
Human behavior is just as preditable as any other natural system, especially
for large populations. The problems with modeling such behavior is that
the complexity of the subject makes developing a correct model very
difficult. It doesn't take much of a mistake to give erroneous results.
This also applies to physical models. If critical variables are not included,
results are questionable. Assumptions are crutial.
Steve Emmerson
> I don't trust any elaborate computer model that can't be
>convincingly validated.
Neither do climatologists, which explains why they're always seeking
validation of the models.
The members of IPCC WG I also don't trust unvalidated models.
--
Steve Emmerson st...@unidata.ucar.edu ...!ncar!unidata!steve
JoshuaHalpern
: It's necessary to be very careful to understand what your model is
: doing. In a nutshell, valid models of the phsyical world must both
: describe and predict the behavior of the systems they represent.
: Although economic and financial models must do the same if they are used
: to describe and predict trends, when a given model is widely accepted it
: creates trends that validate it post hoc. This is because human behavior
: patterns are mutable whereas the laws of nature are fixed. Physical
: models attempt to describe the latter, and there is no effect of
: the predictions of the model on the behavior of nature.
True, but if the economic model is false, although it may have
an effect, the effect will not necessarily be that the model
predicts, and will not necessarily validate that model. What you
say is true in some cases, but not an intrinsic property of
economic models. It is especially not clear that wide acceptance
of an invalid model will even tend to produce results that validate
that model.
Moreover, many behaviorists would question how mutable human behavior
is and it is my understanding that economic models do not place a
strong emphasis on accounting for the vagaries of human behavior
(even if they knew how to include it!)
On the other hand, widely held but false "physical"
models can lead people to find confirming evidence which is
very hard to displace, for example the geocentric
system, which placed the earth at the center of the universe.
This is especially true if the physcial model is not totally
false, but only partially true. In that case, the tendancy is
to search for confirming evidence, and explain away any
contraditory information.
I would also hasten to add, that what is generally accepted is
more often than not correct or at least substantially correct
(perhaps with the caveat of within the measurement possiblities
available at the time).
As far as "physical models" go, quantum mechanics has taught us
that the measuring instrument is necessarily part of the system
being measured, which I gather was the ground on which you seek
to distinguish economic and physical models.
?1;2
Finally, and somewhat mischeaviously, I would remind you of the
strong anthropic principle, which holds that the universe is
what it is because we are here to observe it.
Best regards
Josh Halpern
: On 5 Feb 1996, JoshuaHalpern wrote:
: > jshe...@VNET.IBM.COM wrote:
: > : I posted (regarding large complicated computer models):
: > : >I suspect if this were done routinely it would often uncover bugs.
: > : Dave Halliwell responded:
: > <massive snip, as I have only one question>
: >
: > : I reject (or at least am extremely skeptical of) claims that
: > : very elaborate climate models requiring large computer codes to
: > : implement allow (or will allow anytime soon) us to significantly
: > : refine estimates of climate change due to CO2 available from simpler
: > : models. I believe it is the people suggesting such models be used
: > : to evaluate policy options who have their heads stuck in the sand.
: > : Consulting the entrails of a goat would be as sensible.
: >
: > Does this also apply to the econometric models which are used for
: > budgeting purposes by CBO, OMB, etc, which have even greater impact
: > on policy options and the models used for program trading which
: > have enormous impact on our economic well being (well, at least
: > my hoped for retirement income)? If so, what is your position
: > as to the posturing in Washington over whose budget proposal is
: > honest? If you hold the econometric models to be useful, then
: > why are they more useful for policy decisions than the GCMs. The
: > later appear (at least to me) to be based more on knowledge than
: > the former.
: >
: > As you might guess, the purpose of asking you this question is to
: > point out that most policy decisions today are based on models,
: > and comparatively the Global Climate Models are based on a lot
: > better data and knowledge than the economic models that we depend
: > on.
: >
: > Almost without exception, policy decisions are made in the absence
: > of perfect (or even good) knowledge of the future consequences of
: > that action, and, as has been pointed out elsewhere, not to
: > decide is to decide.
: >
: > Regards
: > Josh Halpern
: > :w
: > :q
: >
: >
: > : James B. Shearer (email j...@watson.ibm.com)
: >
: >
Mark A. Friesel
On 14 Feb 1996, JoshuaHalpern wrote:
> Mark A. Friesel (mfri...@beta.tricity.wsu.edu) wrote:
>
> : It's necessary to be very careful to understand what your model is
> : doing. In a nutshell, valid models of the phsyical world must both
...del
>
> True, but if the economic model is false, although it may have
> an effect, the effect will not necessarily be that the model
> predicts, and will not necessarily validate that model. What you
> say is true in some cases, but not an intrinsic property of
> economic models. It is especially not clear that wide acceptance
> of an invalid model will even tend to produce results that validate
> that model.
I agree that little or none of what I described is certain to occur, and I
also agree with what you say above except for your statement on intrinsic
property. I propose that most people and institutions will act in
response to their perceptions of the economic climate in order to maintain or
increase their profits, or minimize their losses. A model which is
perceived as effective will attract its adherents and detractors
depending primarily on whether they see it as being able to aid their
quest for profit, or obstruct that quest. If the model is good and
provides reliable predictions, then if it is generally accepted at least
a portion of those obstructed by its predictions will work to overthrow
it. The way to do this, perhaps the only way, is to make the predictions
of the model false. Even some of those who profit from it will attempt to
manipulate it in order to profit more, and again it is quite likely that
large profit can be obtained by occurances which counter prediction.
Others will do their utmost to make the model succeed if they are
favorably affected by it. Those with the greatest amount of resources will
succeed.
An economic model which places at a disadvantage any individual or group
with sufficient resources, or which provides opportunity for manipulation by
such a group or individual will fail. Yet by attempting to incorporate
this observation, the modeller is responding to externalities, i.e. human
psychology. A difficult task, and one that I don't think most people are
up to. What we see rather is the use of politics and propaganda to attempt to
control human behavior and make it predictable, allowing 'favorable' economic
models to succeed, at least in the short term. How extrinsic is human
behavior to a model that cannot work independently of its political context?
Now, of course I can't prove
my contention, but I think it's valid none-the-less because it is the way
my mind works 8^)!
>
> Moreover, many behaviorists would question how mutable human behavior
> is and it is my understanding that economic models do not place a
> strong emphasis on accounting for the vagaries of human behavior
> (even if they knew how to include it!)
>
> On the other hand, widely held but false "physical"
> models can lead people to find confirming evidence which is
> very hard to displace, for example the geocentric
> system, which placed the earth at the center of the universe.
> This is especially true if the physcial model is not totally
> false, but only partially true. In that case, the tendancy is
> to search for confirming evidence, and explain away any
> contraditory information.
>
Yes, there is no conclusive proof that the earth is not the center of the
solar system. Yet the point is that the position of any planet with
respect to another can be calculated with rather good accuracy, and this
relative position is pretty much unaffected by anyone's opinions, or any
model used to describe it. The methods used to calculate the planet's
location will always be able to calculate the planet's location. An
entire 'model', such as Newtonian mechanics, contains an approach which
permits the prediction of a large number of physical phenomenon, and will
always be able to do so - the phenomenon themselves occur as they do because
of immutable principles which Newtonian mechanics effectively describes.
> I would also hasten to add, that what is generally accepted is
> more often than not correct or at least substantially correct
> (perhaps with the caveat of within the measurement possiblities
> available at the time).
>
> As far as "physical models" go, quantum mechanics has taught us
> that the measuring instrument is necessarily part of the system
> being measured, which I gather was the ground on which you seek
> to distinguish economic and physical model
My contention is that the basis of economic 'phenomenon' is mutable, i.e.
the economic system is sensitive to human behavior e.g. manipulation.
> ?1;2
> Finally, and somewhat mischeaviously, I would remind you of the
> strong anthropic principle, which holds that the universe is
> what it is because we are here to observe it.
...which, as written above, is rife with ambiguity. What do you mean by
'what it is'?
Scott Nudds
: Human behavior is just as preditable as any other natural system, especially
: for large populations.
Oh, I don't think so. If this were the case, markets would be easily
predictable. Since they are not, I must reject your statement.
--
Scott Nudds
: Finally many published papers dependent on computer
: calculations are not in fact checked in any meaningful way prior to
: publication. If Puchalsky thinks the results of buggy computer
: programs never make it into published papers, he is extremely naive.
Are you kidding? Model varification is and always has been an
important aspect of all modeling. Many techniques are available for
providing such verification.
As to programming errors creaping into the calculations, this
possibility is excluded by many factors, including the most obvious fact
that all models produce essentially the same results. IE. they show a
warming of roughly equal magnitude. None show a cooling.
In order to explain the warming predicted by models by appealing to the
possibility of computational error, one would have to assume that all
models contain errors and that all of the errors produce the same result
- warming, and that none of the errors produce a cooling.
For this reason alone, programming error as the source of warming in
models is a virtual impossibility.
--
Len Evens
>
>As far as "physical models" go, quantum mechanics has taught us
>that the measuring instrument is necessarily part of the system
>being measured, which I gather was the ground on which you seek
>to distinguish economic and physical models.
This would have gone over better if you had added a smiley
after it. As you well know, climate models don't have to take
the Heisenberg uncertainty principle into effect since they operate
at the macrosopic level of classical physics. (Or course there
are aspects of climatoogy which ultimately involve quantum mechanics
such as determining spectra and the like, but that is another
matter.)
>Finally, and somewhat mischeaviously, I would remind you of the
^^^^^^^^^^^^^^
>strong anthropic principle, which holds that the universe is
>what it is because we are here to observe it.
>
You said it yourself, although you apparently mistyped it. I
don't think anyone claims that the climate is what it is today because
we are here to observe it. I know you are only pulling his leg,
but be careful. Someone out there is going to take you seriously
much to your consternation.
>Best regards
>Josh Halpern
>
>
Leonard Evens l...@math.nwu.edu 708-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
Steinn Sigurdsson
jshe...@VNET.IBM.COM wrote:
: Finally many published papers dependent on computer
: calculations are not in fact checked in any meaningful way prior to
: publication. If Puchalsky thinks the results of buggy computer
: programs never make it into published papers, he is extremely naive.
...
As to programming errors creaping into the calculations, this
possibility is excluded by many factors, including the most obvious fact
that all models produce essentially the same results. IE. they show a
warming of roughly equal magnitude. None show a cooling.
In order to explain the warming predicted by models by appealing to the
possibility of computational error, one would have to assume that all
models contain errors and that all of the errors produce the same result
- warming, and that none of the errors produce a cooling.
For this reason alone, programming error as the source of warming in
models is a virtual impossibility.
I don't think there is a fundamental error in the programming of the
GCM models, however, Mr Nudds overstates the case for consistency
a little bit. The scatter among the three main models (Hadley,
Princeton and Max Planck) is of the same magnitude as the net
warming predicted by the model predicting the smallest response.
Somewhat more worryingly, the models have been predicting
systematically smaller responses as they improve, the predicted
response to doubling is down by over a factor of two in the
last couple of decades, back to about the level a "naive" 1-D
calculation predicts.
There are conceivable mechanisms by which there could be cooling,
the physics necessary to see those is not in the GCMs yet. The gross
response seen in the models has probably converged to something close
to the the true response on decadal time scales, IMHO.
Jeff Candy
How can you agree or disagree until you specify exactly what is to be
predicted? Classical problems like population growth, and more interesting
questions regarding disease transmission, epidemic growth, etc., can be
analyzed with some success (in the sense of prediction). Obviously, in
various other cases -- like the one you mention -- reliable prediction is
virtually impossible. One can see a direct physical analogy in the
physics of a fluid. Macroscopic quantities such as (P,V,T) can be
computed in very complex situations, but knowledge about the motion of
individual particles or, say, the dynamics of turbulence, is fundamentally
more challenging.
Jeff Candy ----------------------------------
Analytic Theory Group http://hagar.ph.utexas.edu/~candy/
JET Joint Undertaking ----------------------------------
===============================================================================
The above article is the personal view of the poster and should not be
considered as an official comment from the JET Joint Undertaking
===============================================================================
Mark A. Friesel
Just for the sake of presenting relavent information, there was an
article in the American Journal of Physics some time ago that showed
historic results, with estimated error bars, of measurement of the speed
of light. Prior to a late 1800's (? - probably related to clacualtions
based on Maxwell's equation for the speed of electromagnetic waves in
vacuum) measurement, there was general agreement on the speed of light.
There was also general agreement after this single measurement, but they
all were some 10% or so higher than previously.
From my experience, I don't think you're on firm ground by claiming
errors are a virtual impossibility.
On 15 Feb 1996, Scott Nudds wrote:
> jshe...@VNET.IBM.COM wrote:
> : Finally many published papers dependent on computer
> : calculations are not in fact checked in any meaningful way prior to
> : publication. If Puchalsky thinks the results of buggy computer
> : programs never make it into published papers, he is extremely naive.
>
> Are you kidding? Model varification is and always has been an
> important aspect of all modeling. Many techniques are available for
> providing such verification.
>
> As to programming errors creaping into the calculations, this
> possibility is excluded by many factors, including the most obvious fact
> that all models produce essentially the same results. IE. they show a
> warming of roughly equal magnitude. None show a cooling.
>
> In order to explain the warming predicted by models by appealing to the
> possibility of computational error, one would have to assume that all
> models contain errors and that all of the errors produce the same result
> - warming, and that none of the errors produce a cooling.
>
> For this reason alone, programming error as the source of warming in
> models is a virtual impossibility.
>
> --
>
>
jshe...@vnet.ibm.com
> For this reason alone, programming error as the source of warming in
>models is a virtual impossibility.
Warming is not the issue, simple models predict warming. The
issue is whether programming errors could affect predictions of things
like the amount of rainfall in Kansas in the more complicated models.
Btw I doubt that "all models produce essentially same results" for
things like that.
JoshuaHalpern
Len Evens (l...@schur.math.nwu.edu) wrote:
: In article <4frqus$b...@news2.ios.com>, JoshuaHalpern <j...@haven.ios.com> wrote:
: >
: >As far as "physical models" go, quantum mechanics has taught us
: >that the measuring instrument is necessarily part of the system
: >being measured, which I gather was the ground on which you seek
: >to distinguish economic and physical models.
: This would have gone over better if you had added a smiley
: after it. As you well know, climate models don't have to take
: the Heisenberg uncertainty principle into effect since they operate
: at the macrosopic level of classical physics. (Or course there
: are aspects of climatoogy which ultimately involve quantum mechanics
: such as determining spectra and the like, but that is another
: matter.)
First, I was not specifically talking about climate models here. The
distinction was made between economic and physical models in
general. I would both agree with your statement, that climate models
are not quantum in nature (although they do depend on quantum
parameters such as absorption coefficients, etc.) and maintain
that my answer was a sufficient counter example to falsify the
original claim.
Second, although GCMs are not quantum in nature, they do have
to account for the "chaotic" (see nice discussion in one of the
threads above) nature of weather to at least some extent. In
such a case, where the outcomes from arbitrarily close initial
conditions can be widely separated, the act of measurement
may affect the outcome arbitrarily
Third, I dislike smileys. Then again, the only thing I like
about Bob Dole is his sense of humor :).
: >Finally, and somewhat mischeaviously, I would remind you of the
: ^^^^^^^^^^^^^^
: >strong anthropic principle, which holds that the universe is
: >what it is because we are here to observe it.
: >
: You said it yourself, although you apparently mistyped it. I
: don't think anyone claims that the climate is what it is today because
: we are here to observe it. I know you are only pulling his leg,
: but be careful. Someone out there is going to take you seriously
: much to your consternation.
Well, _actually_, humans can only exist in a narrow range of
temperature, pressure and atmospheric compositions. So in that sense,
although (scouts honor) I had not thought of it when I originally
posted, yes, the climate is what it is today, because we are here to
observe it. Even more pleasing is that fact that this can be argued
either from a fundamentalist biblical point of view: God created
the earth for mankind to live on, or a an evolutionary biologists
point of view: humans have evolved to fit into their environmental
niche(s). Metaphysics is fun.
Again
: >Best regards
Scott Nudds
: af...@freenet.hamilton.on.ca (Scott Nudds) wrote:
: >Robert Keith Fullerton (rfullert@unilab) wrote:
: >: Human behavior is just as preditable as any other natural system, especially
: >: for large populations.
: >
: >Oh, I don't think so. If this were the case, markets would be easily
: >predictable. Since they are not, I must reject your statement.
: How can you agree or disagree until you specify exactly what is to be
: predicted?
I can do so since a blanket statement was made. "Human behavior" which
is the set of all such behaviors was not refined in any way by the
statement. Therefore as a blanket statement all I must do to disprove it
is to provide a counter example, which I did.
: Classical problems like population growth, and more interesting
: questions regarding disease transmission, epidemic growth, etc., can be
: analyzed with some success (in the sense of prediction).
In some instances, human behavior can be "predicted" with a reasonable
degree of accuracy. I don't deny that. But this is a far cry from
saying that "like any other natural system", human behaviour is predictable.
: Obviously, in various other cases -- like the one you mention -- reliable
: prediction is virtually impossible.
Systems of interaction like the stock market must by necessity be
unpredictable. If they were predictable then everyone could make money.
Since this can not be the case, the system must devolve into
unpredictability.
: One can see a direct physical analogy in the physics of a fluid.
Generally no.
--
Scott Nudds
: Warming is not the issue, simple models predict warming.
If "Warming is not the issue" then I take it that you are now in
agreement that warming will take place? Previously you had held the
position that there was no evidence for warming and that the models could
not be trusted.
Have you changed your position? You will not be damaged for admitting
to having done so.
: The issue is whether programming errors could affect predictions of things
: like the amount of rainfall in Kansas in the more complicated models.
Don't fret. You can pretty much guarantee that any predictions with
this level of detail are significantly wrong in these details.
--
JoshuaHalpern
: On 14 Feb 1996, JoshuaHalpern wrote:
: > Mark A. Friesel (mfri...@beta.tricity.wsu.edu) wrote:
: >
: > : It's necessary to be very careful to understand what your model is
: > : doing. In a nutshell, valid models of the phsyical world must both
: ...del
: > Finally, and somewhat mischeaviously, I would remind you of the
: > strong anthropic principle, which holds that the universe is
: > what it is because we are here to observe it.
: ...which, as written above, is rife with ambiguity. What do you mean by
: 'what it is'?
The anthropic principle tries to answer the question of why we obtain
the answers we get for measurements of fundamental constants such
as the charge of the electron, electron mass, gravitational constant,
etc. In its weak form, the anthropic principle says that if the
fundamental constants (the universe) were even a little bit
different, human beings could not exist in that universe, thus
the fact that we are here to make the measurements, means that
the fundamental constants (and the universe that results) must
be within certain very narrow bounds. In its strong form the
anthropic principle implies that the beings making the observation
have somehow by their existence forced the universe to take on
its measured shape.
There is a post a few below this one, advertising a talk on
causality and conciousness in modern physics, a closely
related topic.
Again
: >
: > Best regards
: > : >
: >
: >
JoshuaHalpern
: : >Robert Keith Fullerton (rfullert@unilab) wrote: : : >: Human behavior
is just as preditable as any other natural system, especially : : >: for
large populations. : : > : : >Oh, I don't think so. If this were the
case, markets would be easily : : >predictable. Since they are not, I
must reject your statement.
: : How can you agree or disagree until you specify exactly what is to be
: : predicted?
<snip>
: Systems of interaction like the stock market must by necessity
be : unpredictable. If they were predictable then everyone could make
money. : Since this can not be the case, the system must devolve into :
unpredictability.
Well, program (computer) trading works well enough that it makes a lot of
money. So the stock market is predictable enough
that an algorithm can be used to play the market successfully.
Regards
Josh Halpern
: : One can see a direct physical analogy in the physics of a fluid.
: Generally no.
: --
Scott Nudds
: Well, program (computer) trading works well enough that it makes a lot of
: money. So the stock market is predictable enough
: that an algorithm can be used to play the market successfully.
Great. Lets solve the worlds financial problems by giving everyone this
program and allowing them to make money in the stock market.
Clearly, any market predictability is transient.
--
Mark A. Friesel
On 20 Feb 1996, JoshuaHalpern wrote:
> Mark A. Friesel (mfri...@beta.tricity.wsu.edu) wrote:
>
>
> : On 14 Feb 1996, JoshuaHalpern wrote:
>
...del
>
> : > Finally, and somewhat mischeaviously, I would remind you of the
> : > strong anthropic principle, which holds that the universe is
> : > what it is because we are here to observe it.
>
> : ...which, as written above, is rife with ambiguity. What do you mean by
> : 'what it is'?
>
> The anthropic principle tries to answer the question of why we obtain
> the answers we get for measurements of fundamental constants such
> as the charge of the electron, electron mass, gravitational constant,
> etc. In its weak form, the anthropic principle says that if the
> fundamental constants (the universe) were even a little bit
> different, human beings could not exist in that universe, thus
> the fact that we are here to make the measurements, means that
> the fundamental constants (and the universe that results) must
> be within certain very narrow bounds. In its strong form the
> anthropic principle implies that the beings making the observation
> have somehow by their existence forced the universe to take on
> its measured shape.
>
Thanks, I know what the anthropic principle is and was looking for
clarification about the phrase 'what the universe is'. Personally, I
have never accepted the anthropic principle in its strong form due to
lack of evidence, and the weak form, i.e. we exist therefore conditions must
have been right for our existence, contains effectively no useful
information due to its being trivial.
> There is a post a few below this one, advertising a talk on
> causality and conciousness in modern physics, a closely
> related topic.
> Again
> : >
> : > Best regards
> : > : >
> : >
> : >
>
>
jshe...@vnet.ibm.com
models continues.
I had posted:
> Warming is not the issue, simple models predict warming.
> If "Warming is not the issue" then I take it that you are now in
>agreement that warming will take place? Previously you had held the
>position that there was no evidence for warming and that the models could
>not be trusted.
>
> Have you changed your position? You will not be damaged for admitting
>to having done so.
Nudds is misstating my previous position. As I posted a few
weeks ago when Halliwell misstated my position:
> This is again untrue. I accept that what is popularly called
>the "greenhouse effect" warms the surface of the earth. I accept that
>CO2 is a greenhouse gas. I accept that the observed increase in the
>CO2 content of the atmosphere is almost surely of anthropogenic
>origin. I accept that there is a plausible argument that increasing
>the CO2 content of the atmosphere will increase the magnitude of the
>greenhouse effect thereby warming the surface. I accept that simple
>climate models allow crude estimates of the amount of warming that
>can be expected.
> I reject (or at least am extremely skeptical of) claims that
>very elaborate climate models requiring large computer codes to
>implement allow (or will allow anytime soon) us to significantly
>refine estimates of climate change due to CO2 available from simpler
>models. I believe it is the people suggesting such models be used
>to evaluate policy options who have their heads stuck in the sand.
>Consulting the entrails of a goat would be as sensible.
As I have previously indicated one reason I am skeptical of
large models is that it is hard to be sure that the large computer
codes on which they rely are bug free.
I had continued in my later post:
>The issue is whether programming errors could affect predictions of things
>like the amount of rainfall in Kansas in the more complicated models.
Scott Nudds replied:
> Don't fret. You can pretty much guarantee that any predictions with
>this level of detail are significantly wrong in these details.
Fine, we agree on this. So where is the value added (relative
to simpler models) in the large complicated models?
jshe...@vnet.ibm.com
to simpler models) in the large complicated models.
Scott Nudds
: As I have previously indicated one reason I am skeptical of
: large models is that it is hard to be sure that the large computer
: codes on which they rely are bug free.
Bugs occurr at random, and can be expected to produce random errors.
Presuming that these random errors do not cancel themselves out, would
you mind telling us how many models, all of which you must think contain
bugs, give essentially the same result?
: > Don't fret. You can pretty much guarantee that any predictions with
: >this level of detail are significantly wrong in these details.
: Fine, we agree on this. So where is the value added (relative
: to simpler models) in the large complicated models?
These models predict long term averages, and do not attempt to predict
short term details.
You may not know precisely where your children are at 1 am. But you
almost certainly know what city they are in.
--
JoshuaHalpern
: JoshuaHalpern (j...@haven.ios.com) wrote:
: : Well, program (computer) trading works well enough that it makes a lot of
: : money. So the stock market is predictable enough
: : that an algorithm can be used to play the market successfully.
: Great. Lets solve the worlds financial problems by giving everyone this
: program and allowing them to make money in the stock market.
Well, you would also have to give them a lot of capital (let the
so and so's EARN it) and a seat on the exchange. The margin of
profit is of the order of the cost per trade to a retail customer
and to make any money the programs have to trade a lot of shares.
They also don't work when there are too many computers in the
market. One of the causes of the huge market fall in 87 (?)
was too many programs trading against each other. The
SEC introduced restrictions on program trading so that it
would not happen again.
:
: Clearly, any market predictability is transient.
well as Keynes said, in the long run we are all dead.
Regards
Josh Halpern
:w
:q
: --
Scott Nudds
: Scott Nudds (af...@freenet.hamilton.on.ca) wrote:
: : JoshuaHalpern (j...@haven.ios.com) wrote:
: : : Well, program (computer) trading works well enough that it makes a lot of
: : : money. So the stock market is predictable enough
: : : that an algorithm can be used to play the market successfully.
: : Great. Lets solve the worlds financial problems by giving everyone this
: : program and allowing them to make money in the stock market.
: Well, you would also have to give them a lot of capital (let the
: so and so's EARN it) and a seat on the exchange.
And you don't see the glaring logical inconsistancy in your position?
--