Guest Post by Willis Eschenbach
I love thought experiments. They allow us to understand complex systems that don’t fit into the laboratory. They have been an invaluable tool in the scientific inventory for centuries.
Here’s my thought experiment for today. Imagine a room. In any room dirt collects, as you might imagine. In my household I’m responsible for keeping the dirt down, so I get out my vacuum cleaner to clean up.
But suppose I had a magic way to handle that dirt. Suppose there was a class of beings such that whenever there was a concentration of dirt in some small part of the room, one of these beings would pop into existence, clean up the dirt, and then disappear. We’ll say this being is a rare relative of the Tasmanian Devil, call it a “Tasmanian Dirt Devil” (TDD).
Figure 1. One of the few authenticated photos of the elusive Tasmanian Dirt Devil (TDD) cleaning a floor in its natural habitat. PHOTO SOURCE: Annals of Cryptozoology, Vol. 6, 1954
After observing the room for a while, we realize that the the TDD only appears when there is some small area of the floor with more than a certain concentration of dirt. However, we see that the TDD does not limit itself to that small concentration of dirt. It moves around and cleans out any other smaller concentrations of dirt around it as well. Once the area is cleaned to a certain level, the TDD vanishes, leaving the room somewhat cleaner. We also see that on days when traffic is heavy, often there are a number of Tasmanian Dirt Devils working on the room at once. No single TDD cleans the whole floor, but the floor is never dirty anywhere for long.
Now, here’s the question for our thought experiment: can we use a computer to model the effect of the TDDs, for example to calculate the rate at which dirt is being added to the floor, based on the average amount of dirt on the floor?
I say we cannot model it adequately if our only input to the model is the average level of dirt on the floor. Here’s two circumstances to explain part of why the problem is ugly.
1. Someone spills a very small bit of dirt on one corner of the floor. Because the dirt is concentrated in one area, a TDD materializes, cleans up the dirt and the surrounding area, and vanishes.
2. Four people simultaneously spill a very small bit of dirt in all four corners of the floor. Four TDDs materialize, clean up the dirt and the surrounding areas, and vanish.
If all we have is the average dirtiness of the floor, a few bits of dirt which are rapidly cleaned up will make little difference in a daily average of floor conditions. Despite those small fluctuations, in one case there is four times as much dirt being added to the system as in the other case.
So I think we can agree that in our thought experiment, the average dirt level of the floor is not linearly related to amount of dirt being added to the system. If we want to model what’s going on, it is very difficult to do it based on the average dirt level. We need much more detailed information in both time and space.
Here’s an illustration of a different problem. Again, two conditions.
1. Someone spills a very small bit of dirt on one corner of the floor. Because the dirt is concentrated in one area, a TDD materializes, cleans up the dirt and the surrounding area, and vanishes. Average dirt level on the floor ends up slightly below where it started.
2. Someone spills a very small bit of dirt evenly all over the floor. There is no concentration of dirt above the threshold level, so no TTD appear. Average dirt level on the floor ends up slightly above where it started.
Again, as you can see, average dirt levels and amount of dirt added show no correlation.
So what do Tasmanian Dirt Devils have to do with the climate? If we saw something like a TDD in our kitchens, we’d be amazed. However something just as amazing exists in the climate. We’re not astounded by it all purely because are so familiar with it. However, let me take a small digression on the way to explaining the relationship between climate and Tasmanian Dirt Devils.
Emergent phenomena are a special class of things. They can be recognized by certain traits that they have in common. In general, emergent phenomena arise spontaneously at a certain time and place. Typically they exist for a definite duration, and eventually dissipate at another time and place. Their appearance is often associated with some natural variable exceeding a threshold. Often they can move about somewhat independently. If so, although they have general tendencies, their specific actions are usually very difficult to predict.
One clear characteristic of emergent phenomena is that the properties of emergent phenomena are not apparent in the underlying stratum from which they arise.
Examples of natural emergent phenomena with which we are familiar include sand dunes, the behaviour of flocks of birds, vortexes of all kinds, termite mounds, consciousness, and indeed, life itself.
Regarding climate, there is one particularly important class of natural emergent phenomena. These are the natural “heat engines”. Heat engines are able to turn heat into work. Examples of these natural emergent heat engines include hurricanes, thunderstorms, dust devils, tornadoes, and the Hadley Circulation itself.
The most common and most important of these heat engines are thunderstorms. Thunderstorms do two kinds of mechanical work. First, they power the deep tropical convection that is the driving force for the circulation of the entire ocean and atmosphere.
Second, thunderstorms drive what can be thought of as a sophisticated air conditioner, using a variation of the standard refrigeration method. This method, used in your home air conditioner, uses ambient heat to evaporate a liquid. This removes the heat from the area where the evaporation is taking place.
Then you move the evaporated liquid (and the latent heat it contains) to another location. In the new location, you condense the liquid, releasing the latent heat of condensation. The heat is then transferred to the surroundings, and the condensed liquid is returned to start the cycle over.
In the natural Hadley air conditioner we call a thunderstorm, the same process takes place. Water is evaporated at the surface, cooling the surface. The water vapor rises to the clouds. There it is condensed. The latent heat it contained is released, rises, and is radiated out to space.
Meanwhile, in addition to losing latent heat through evaporation, the surface is further cooled by the fall of cold rain from the thunderstorm. This is accompanied by an entrained cold wind, which assists in the cooling.
In both cases (Hadley circulation and refrigeration) the net effect of a thunderstorm is to remove energy from the surface and move it up into the troposphere.
Having digressed, I return to what climate has to do with Tasmanian Dirt Devils.
Consider our thought experiment. If you replace TDDs with thunderstorms, and replace the room with a climate model gridcell of the tropical ocean, and replace dirt with energy, you have an excellent description of the action of the climate system at the hot end of the climate heat engine, the Tropics.
Whenever there is a “hot spot” on the tropical ocean or land, if it is hot enough, a thunderstorm springs up and starts pushing huge amounts of energy vertically. As the thunderstorm moves across the surface, it moves towards the warmest area in its path. This preferentially cools the warmest areas. In addition, it continues to do so until the local surface temperature is a few degrees below the initiation temperature.
There are some conclusions that we can draw from this thought experiment:
1. Increasing the rate at which dirt is added does not commensurately increase the average dirtiness of the floor. Similarly, increasing the rate at which energy is added to the Tropics does not commensurately increase the surface temperature.
2. Attempting to model our thought experiment using room-wide averages won’t work because Tasmanian Dirt Devils are driven by local conditions, not average conditions. Similarly, attempting to model our climate using gridcell-based averages won’t work because thunderstorms are driven by local conditions, not average conditions.
3. Modeling a system which contains simple linear feedback is not too difficult. In that case, average changes in the response variable are linearly related to changes in the forcings. Modeling a system with an active governor, like TDDs or thunderstorms, requires a much different type of model. As I showed above, in that case the response variable is not linearly related to the forcing.
4. Thunderstorms preferentially cool the warmest areas. Although the average temperatures might be the same, this has a different effect than a gridcell-wide uniform cooling. Again, this makes the modelling of the system more complex.
Let me be clear about what I am saying about models. I’m not saying that we can’t model the climate. I think we can, although it won’t be easy. But we have to model it the way it really is.
It is not a system with a linear relationship between forcing and temperature as conventional theory claims. It is a dynamic governed system with a complex, nuanced, non-linear response to forcing. Yes, we can model that. But as I show above, we can’t do it under the assumptions made by the climate models.
Could we model it parametrically, without having to model individual thunderstorms? Probably … but the model has to be designed to do that. And the current climate models either are not designed to do it, or are not doing it successfully.
How do I know that they are not doing it successfully? Drift. Consider the room with the Tasmanian Dirt Devils. If there is no change in the amount of dirt being added per day, the system will rapidly take up a steady-state condition.
The models are subjected to a very similar test. In this test, called a “control run”, every one of the forcings of the model is held exactly steady. Then the models are run for a number of model years. Figure 2 shows the results from the Coupled Model Intercomparison Project (CMIP) control runs. We would expect the models to rapidly take up a steady-state condition.
Figure 2. Results of control runs for 16 coupled atmosphere-ocean climate models. SOURCE
Notice the drift in the surface air temperature in a number of runs over the 80-year simulation. The CERFACS model is the worst, but even a mainstream model like the NASA GISS model of James Hansen and Gavin Schmidt shows drift over the 80 years.
How much drift? Well, the trend in the NASA GISS model control run is a warming of about 0.7°C per century. This is about the same as the IPCC estimate of the warming over the last century, which is 0.6°C.
Now, you could look at that GISS model 0.7°C per century inherent warming drift with no forcing change as a bug. I prefer to think of it as a feature. After all, it lets Hansen and Schmidt simulate the warming of the 20th century without the slightest change in the forcings at all, and how many models can do that?
However, that drift does strongly suggest that they are not modeling the climate correctly …
As always, the quest for understanding continues. My best regards to all,
w.
Brilliant analogy! Thank-you.
See http://climateaudit.org/2010/12/11/remember-gavins-taunts-about-steig-et-al-2009/#comment-248841
In that CA thread, I quoted from a post made by Bender, in which he offered his own translation of Realclimate’s criticisms of the various CA and Air Vent analyses of the Steig 2009 Antarctic warming paper, as follows: “ ……. Because once we figure out what regulates natural climatic variability we’ll be in a much better position to estimate GHG contributions to warming, which are pretty uncertain ..….”
I asked the following question of the CA readership:
Focusing on the phrase “what regulates natural climatic variability”, is it possible to state that within some range of temperature values, apparent changes in the earth’s global mean temperature could be the result of system drift from the current long-term GMT trend, drift which has no identifiable driving mechanism whatsoever?
If this is so, if a long-term trend in GMT can drift within certain temperature ranges, but with no identifiable driving mechanism whatsoever being present to cause that drift, is it possible to state with any reasonable assurance what the temperature range of the system’s inherent drift might be?
On that CA thread, I also remarked that perhaps this is a question Dr. Koutsoyiannis might be interested in discussing, if he is following that particular CA thread.
So far, as of December 13th, there has been no response from anyone on CA in response my question. Willis, perhaps you would like to take a stab at it.
Party pooper. I was working upon a sequence of posts on Judith Curry’s blog to say much the same thing. You’ve sprung my trap.
The TD, the native Sarcophilus harrisii , you might note, starts with sarc on.
Willis,
Are you on a warm island in a hammock sipping on something? Whatever it is, I want some.
This post is like a shaggy dog story. http://en.wikipedia.org/wiki/Shaggy_dog_story
You went a long and circuitous route to the punch line, which is very good. I trust it will hold up under the withering attack of warmest scolds.
Willis likes thought experiments.
I like Willeses articles, from which I get thought experiments. 2 for the price of 1. I win.
How can I trade in my super-turbo-self-cleaning vacume cleaner for a bunch of those TTD’s? I’m guessing I’m gonna have to trade a bunch of carbon credits to make up the difference. I hope the CCX can assist.
Stop it with these these theories about how the climate works!
You ‘ll get much more palatable “AGW Style” results if you’re more dispassionate about the data and apply analysis that ignores any semblence of reality.
Very interesting, a good job of making it so. I strikes me that nature of these thought games is more then slightly associated with the nature of deterministic models of dynamic systems. Interesting and more fiction then reality.
I was hoping you would carry the thought experiment to left field and suggest that the TDDs somehow add more dirt to the average using the dirt that already exists. Excellent analogy Willis!
Wonderful posting… music to my ears.
I prefer to think of it as a feature.
Unfortunately I think you are totally correct here… my guess is that this temperature drift has a non-linear forcing relationship that drives budget drift exponentially upwards… and when this pool of excess monetary resources reaches a critical point the system spontaneously generates The Warming And Terrifying Scientists that converts the budget surplus into a hot air feedback that drives the temperature drift ever higher.
My own post reminds me of this from George Carlin:
“A crumb is a magical thing. It’s the only thing on Earth that when you split it in half, you don’t have two half-crumbs, you have two crumbs.”
Are these local conditions high local temperatures, or local high temperature gradients?
(Or some mixture, of course…) The first condition implies Global Warming is self-limiting.
As I understand it, the Climate models forecast greater temperature increases at the Poles, from which I infer a reduced temperature gradient across the globe. Consequently, in the second case there is more heat/dirt before a thunderstorm/tdd, not less.
The above question seems to me to be important. On it’s answer depends either a negative feedback or a positive one. Unless the question is too simplistic, of course.
The SOURCE also shows that GISS has an upward precipitation drift of 0.1 mm per day over 80 years… which, according to my maths, equates to an additional 45 mm of precipitation per year after 100 years… but I guess this precipitation must be composed of magic water © that doesn’t cool the planet but manages to accumulate in the oceans and drift sea levels ever higher.
http://www-pcmdi.llnl.gov/projects/cmip/overview_ms/control_tseries.pdf
So it would seem that what appear to be random, chaotic events at the climate model grid scale are in fact predictable phenomena when examined at the proper scale.
FWIW, it looks like some folks have found ENSO to be an ’emergent behaviour’ in a computer model. No external events need apply. Just a natural instability.
http://chiefio.wordpress.com/2010/12/13/enso-a-natural-oscillation/
Dear Willis. Brilliant analogy, Have you considered developing it into a short educational cartoon?
“Vortex” has a Latin root. The plural is “vortices”, not “vortexes”
Well done Sir.
May I add something I think might be relevant to your piece, it’s a simple thought experiment and it stumps most Warmists…they get that ‘look’.
I ask them to recall the last time they took a commercial jet flight.
To picture themselves coming up through the clouds, rising above them – to cruising altitude.
I ask them if the Captain at any point mentioned the temperaturere…” we’re flying at 29,000 feet and the temperature is….” or if perhaps there was a Temperature Display in the cabin (I’ve seen this on several airlines.)
I then ask them if they remembered the outside temperature at cruising altitude.
(At this point I am amazed by the percentage of people who ardently defend the ‘science’ of the greenhouse theory but have no idea of this fairly simple fact.)
A few know the temperature range I am discussing and we proceed.
We establish quickly that there is a vast resevoir of thin cold air just above our heads, every time we look up and see a contrail we are looking at frigid air.
I then ask, “What does hot air do, in nature?”.
“Rises.” can be the only only answer.
So imagine yourself looking out of the window of your plane…on the display it says minus 30 degrees. Down below are clouds…some are flat but some are high, in fact some are towering magnificently into the chilled atmosphere – indeed the Captain has told you to strap your seatbelts because of turbulence caused by high altitude air currents.
Now given that hot things rise, and hotter things rise higher/faster- if….if for some reason our planet was to become ‘over heated’ surely all the clouds would just rise a few metres higher and lose, through radiation whatever excess heat carried them there.
I conclude by asking them if they can honestly argue, given the vast resevoir of coldness that begins a mere five miles above our heads and extends upwards towards the void, that the earth could ever lose its ability to shed its heat into space?
Thank you for the illustrative thunder/refrigeration cycle comparison, never thought it that way before.
But for the models, in my view 80 years (cycles) do not yet prove that models are bad (that they leak energy), but leave possibility that modelled system has multiple solutions beweeen which it oscillates, or have bad initial conditions and very long time constant (are these atmosphere only?). They should run models long enough to have them settle, and with more than one set of forcings.
Multiple degree offset discrepancy between models begs explanation.
I like he explanation of how storms take heat away from the earth and allow it to be expelled to space. That neatly explains why raising CO2 levels do not cause the temperature to rise.
I am not quite as comfortable with the idea of drift.
The climate generally and temperature & rainfall in particular are chaotic systems.
Drift in a chaotic system infers high Hurst numbers (above 0.5 – Brownian randomness). Such systems are auto correlated (eg the stock market) and tend to drift up or down untill Hurst’s “Joker” suddenly intervenes, the system is reset and goes off in a completely new direction.
Both rainfall and temperature in a number of Australian locations have extremely low Hurst numbers (Sydney Observatory Hill approximately 0.02). Such systems have a strong tendency to revert to the mean and are trendless if left to themselves.
The only interference that I have found is UHI. It’s commencement can be clearly identified in both Sydney and Adelaide temperatures for example.
In each case I have positively identified the massive change in the built environment the allowed UHI to begin to raise the temperature. (Rainfall is imune so it is the ideal control for UHI).
So I can’t see that random drift is the answer.
At least in many well dispersed locations of Australia.
In the case of thunderstorms, it is more than temperature. For example you don’t have many thunder storms in the Sahara. You also need humidity, and an atmospheric profile that creates instability, along with some mechanism to initiate lifting of the air mass to start the process.
Instability is created when a mass of warm moist air is over laid by dryer air. This creates a situation that if a parcel of that warm moist air is forced to rise due to any one of several causes it will continue to rise due to it being lighter than the surrounding air. Lift can be due to such things as being pushed over rising terrain by winds, or convergence where two directions of flow collide forcing the air upward where they meet. A frontal passage (rapid changes in barometric pressure) is often a trigger to initiate lift, or in cases of severe instability, just random motions generate the upward movement which becomes self reinforcing.
As the warm moist air rises, it cools more slowly than the dryer air above and around it due to its higher heat capacity, so the higher it gets lifted, the more it buoyant the parcel becomes. At some point the moisture begins to condense releasing latent heat of condensation. This extra heat then drives the upward motion ever faster creating the strong updraft that builds the convective column of the thunder storm.
You need heating, moisture and the proper air temperature profile, and usually some trigger to initiate lift, with the proper local wind flow (shear) .
The local winds need to be such that the updraft is fed more warm moist air as the updraft forms, without the local winds being too strong. If too strong, they can tear the convection apart and re-mix the rising parcel with dryer surrounding air killing the convection.
Some of these variables are captured in indices like.
LI – (Lifted index)
CAPE – (convective available potential energy)
dew point (humidity) vs local air temperature and air temperature at higher levels.
Skew T profiles (plots of the atmospheric conditions at different altitudes.
Storm watchers pay more attention to these variables than they do just temperature.
At certain combinations of LI and CAPE thunderstorm development is almost guaranteed, as long as there are not competing effects like unfavorable winds (very high winds) or a lack of lift to initiate the development of the storm.
http://en.wikipedia.org/wiki/Lifted_index
http://en.wikipedia.org/wiki/Convective_available_potential_energy
http://airsnrt.jpl.nasa.gov/SkewT_info.html
Larry
If you believe feedback (eg from water vapour) is positive the you are asserting that the climate is inherently unstable. Any model built with these assumptions is going to drift in the absence of any forcing. Indeed drift is too mild a term as the behavior of such a system could be quite wild; massive oscillations, aymptotes, all bets are off in an unstable chaotic system with positive feedback.
It has always puzzled me that the rhetoric of climate change is all about nicely predictable linear responses to forcing. This much reduction of CO_2 needed to reduce that much warming, and so forth. Whereas the assumption of positive feedback which is needed to get alarming results posits an unstable climate where the response to forcing is likely to be highly non-linear and indeed where the climate may well behave chaotically even in the complete absence of forcing.
Without positive feedback the temperature change caused by increased CO_2 is too small to be a problem. With positive feedback the temperature response might indeed be alarming, however there would be no point in trying to control CO_2 emissions because such a system is inherently so unstable that nothing we could do would ensure climate stablity.
Thank you Willis, stick to the heat pump it is the only valid and measurable system that makes sense. Chaos tending always to simplicity and beauty in a non linear system, even vortices are perfection and beauty.
But….. but….. there is only one graph.
What happened?
Run out of graph papar?
Very good, though.
Willis.
Follow the drift into the attribution studies.
AR4 chapter 9. Attribution studies. Supplmental material. page 9.5
FAQ 9.2
only 14 models are selected for attribution studies. Those with drift of .2C per century or lower.
Question: when trying to do an attribution study, they select models with small drift.
Small drift means tighter CIs. Tighter CIs on the models means you have an easier time rejecting the null. On the other hand when projecting the future they select all the models, including those with higher drift. Higher drift means wider CIs. Wider CIs on the forecast means lower probability of conflciting with Obs.
I’ve be bugged by this for a couple years. have a look.
Given the scarcity of TDD’s, the solution to global warming would then be maid service.
@ Grumpy
The Latin plural of vortex is vortices, but illis writes in English, so “vortexes” is good.
@ Ian H
It is the net feedback which governs the system, not any one component. It seeems to me bleeding obvious that the net feedback on the warming side is strongly negative, since the climate has been bouncing around for millions of years under an upper bound a few degrees higher than todays. On the cooling side, again there must be negative feedback, or we would slide into the ‘snowball earth’ stable solution, but the Ice Ages show us that the lower bound on temperature is a lot lower than today’s temps, and given the long ‘dwell time’ in Ice Ages, the feedback seems to be weaker near the lower bound than near the upper bound. For me the challenge is not to develop more precise models based on assumptions about feedbacks, but to elucidate the cause-and-effect relationships, which will allow the feedbacks to be estimated / calculated, rather than assumed.
charles n.;
Not tea bags!
But it’s “reservoir”, not “resevoir”. And yes, that 2nd ‘r’ IS pronounced.
So here’s what “settled” means: “we stopped thinking”.
Robbo;
unfortunately, you’re correct. Either is acceptable. But “mouses” is NOT! >:(
About the bounds and feedbacks:
Willis Eschenbach, ably assisted by Charles Nelson (December 13, 2010 at 11:48 pm), have between them most entertainingly put this global warming conundrum to bed.
Sensible skeptics start from the twin observations that:
(1) Contrary to uninformed doom mongering over the past 30 years, at only 0.5degC or so per century the Earth’s temperature is steadfastly failing to rise in correspondence with the sharp (~30%) post-WW2 upturn in anthropogenic GHGs.
(2) The straightforward physics of CO2 indicates that an increase in its atmospheric concentration should have a significant and potentially worrying warming effect.
Consequently, we reason that there must be something else holding the temperature down. What is it?
As I have blogged many times both here and elsewhere, my best thought experiment is my 40kW house central heating system which keeps the temperature inside my house in the winter very close to 21degC simply because of the strong feedback effect provided by its thermostat, which constantly measures the house temperature against a fixed physical set point.
Suppose I then turn on the 3kW electric fan heater that I keep only for occasional use and emergencies. Surprisingly (to some people), my house does not increase in temperature from its 21degC set point. Why is this? It is because any additional heating effect from the 3kW heater is exactly offset by the house thermostat which shuts down the central heating boiler commensurately.
The CO2 warming effect has no (efficient) thermostat. The hydrological cycle, via convection, thunderstorms, cloud formation, etc., has a very effective one. Therefore the CO2 warming effect is completely enclosed within the highly regulated hydrological feedback loop, in exactly the same way that the warming effect of my unregulated fan heater is completely enclosed within the warming effect of my highly regulated central heating system.
Warmists can’t seem to get this idea of one warming effect being enclosed within another and therefore being entirely neutralised.
If only they were control system engineers!
End of story.
As usual, Willis takes an initially reasonable analogy – and misuses it. His details are the devil. In this case, his “thought experiment” does not give the result he wants it to. One could indeed successfully model “the rate at which dirt is being added to the floor, based on the average amount of dirt on the floor”. I don’t “agree that in our thought experiment, the average dirt level of the floor is not linearly related to amount of dirt being added to the system”, because it is. Contrary to his claim that “average dirt levels and amount of dirt added show no correlation”, they will indeed be correlated, with the specific correlation dependent on the algorithm employed by the dust devils. The correlation is certainly not perfect, and the linear relationship is not the only relevant phenomenon, because, as he points out, the actual ratio depends on the local distribution of dirt; but, for typical scenarios in which the average manner in which dirt is deposited is fairly consistent, a kitchen-wide (“global”) linear model is likely to work pretty well, as a predictor of the overall dirtiness of the kitchen. I’d actually be quite surprised if the manufacturers of dust devils (which are now starting to appear on the market in reality) don’t use some such computer model.
The reason that Willis’s thunderstorm thermostat doesn’t do what he thinks it should is that there are still underlying broadly linear (or at least monotonic) dependencies on temperature and insolation that the local feedback from his “governor” does not eliminate. For example, thunderstorms are driven by a temperature difference between surface and stratosphere, not the absolute temperature; and there is evidently a variable latitude, or range of latitudes, by which the tropical mechanism ceases to operate. He’s had this pointed out to him several times before, but prefers to ignore it.
Great post and analogy Willis, which reminds me: I’ve been meaning to say the following for some time now:
On negative/positive feedbacks:
Place a small sphere inside a smooth round bowl and give it a small kick. This will roll up and down the sides and through the centre of the bowl until the sphere stops to rest at the centre, bottom of the bowl, where it finds its stable natural position inside the bowl after the energy is dissipated. The oscillation will be repeated when the sphere is acted upon by another force. This is analogous to negative feedback.
Now turn the bowl over and rest the small sphere on the vertex of the smooth bowl and let it go free. The sphere will hockey-stick down, never to be able to go up again to the top of the bowl. This is positive feedback.
If the earth’s climate had positive feedback due to anything, there would not be any life on earth, me thinks. Consider the quantity and magnitude of all the forcings that our planet has gone through during its billionial (can I say this word? My first language is not English, so pardon me) life.
Now, the IPCC wants us to spend hundreds of billions of dollars/euros and put back society a hundred years back in time so as to keep the sphere inside the bowl 2mm above its natural stable position.
Meanwhile millions of people are suffering in Haiti, Africa, Asia, but what the heck? Who cares/ As long as we save the planet from 2C warming.
Alex the skeptic says:
December 14, 2010 at 2:05 am
“If the earth’s climate had positive feedback due to anything, there would not be any life on earth, me thinks.”
There are both negative and positive feedbacks in the Earth’s climate. Note that even a net positive feedback does not necessarily imply that a system is unstable – it may simply amplify fluctuations by a finite but sustainable factor.Conversely, even with net negative feedback, a system can still be destroyed if you hit it hard enough or find a resonance (the ball will still jump out of the bowl if you jiggle the bowl sufficiently violently or at the right frequency).
@Paul Birch: “for typical scenarios in which the average manner in which dirt is deposited is fairly consistent”
I don’t understand. What typical scenarios are you assuming where the dirt is deposited consistently? My kitchen has many local regions. By the door, a lot of dirt comes in from outside in big amounts at certain times of day. Meanwhile under the furniture, in quiet spots, dirt can accumulate very slowly over a long period of time. Some of it comes from the door so it also depends on how quickly the door area was cleaned up. If the door area was subjected to lots of dirt suddenly (muddy say) then it gets cleaned up immediately. If it is a dry day, then the dust can accumulate a bit and spread to the rest of the kitchen. So I don’t understand what you’re assuming? And I don’t understand what you’re assuming about the planet?
Willis,
Very interesting article.
You said: “Meanwhile, in addition to losing latent heat through evaporation, the surface is further cooled by the fall of cold rain from the thunderstorm. This is accompanied by an entrained cold wind, which assists in the cooling.”
This localized cooling from the cold rain water dosen’t change the earth’s heat content, as the heat is just moved from one place within the biosphere to another (not radiated to space), right? Also, by simple observation, it appears “visually” to me that many of the control run models show no drift. True?
Paul Birch says:
December 14, 2010 at 2:40 am
Alex the skeptic says:
December 14, 2010 at 2:05 am
“If the earth’s climate had positive feedback due to anything, there would not be any life on earth, me thinks.”
There are both negative and positive feedbacks in the Earth’s climate. Note that even a net positive feedback does not necessarily imply that a system is unstable – it may simply amplify fluctuations by a finite but sustainable factor.Conversely, even with net negative feedback, a system can still be destroyed if you hit it hard enough or find a resonance (the ball will still jump out of the bowl if you jiggle the bowl sufficiently violently or at the right frequency).
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
It was expected that someone would come up with the ‘kick hard enouhg to throw the sphere out of the bowl”, but as I said in my original comment:
“Consider the quantity and magnitude of all the forcings that our planet has gone through during its billionial life.” Fact is that our planet has never experienced such a kick during its geological histroy, not due to CO2 increases, solar forcings, not even the catastrophic direct hit that presumably killed the dinosaurs. The planet still managed to find its equilibrium following this global devastating event. The climate bowl must be very very deep, so deep that not even a celestial direct hit could manage to throw the ball out of the bowl.
I think it would be interesting to see Figure 2 presented as departures from some reference period. The entire time period, or the first 20 or 30 years.
Any extraneous gases in an unbounded heat pump using H2O as a refrigerant will make three fifths of five eighths of sweet FA of a difference to the null set of the thermostat. The only possible variation to the thermostat is the heat input, this is obviously varied by solar and cosmic influences beyond our control. Internal upsets such as extra heat from internal sources or particulate heat blocking from volcanic activity or industry and forest fires, are the only other variables. This is predicated on the hope various large rocks from the cosmos do not feel an urge to upset our peaceful existance, for they also tend to upset our thermostat. Those who believe we can control the thermostat by living in bark huts and caves are delusional.
90% of the time for the last odd million years our thermostat has been set on deep freeze, this setting is not particularly conducive to the life style we have come to enjoy.
It is not global warming that will cause us pain, it is cooling that will see suffering and death. I therefore ask those of scientific bent on this site to forget the nonsense of warming and investigate the reasons why in the past our thermostat was set to deep freeze for most of the time. Huge quantities of research and information are out there, they may have been distorted to prove a fallacy but they remain as information.
Brian thanks for the spelling tip…I did actually know that-honest.
And Socrates dear boy, we definitely need more practical engineers on here
I tire of voodoo physics.
I’ve got another one for you, shorter.
One evening after a great a dinner party I was confronted by a mellow wall of friends, all comfortably off and in senior positions who tut tutted my ardently skeptical position on the subject of climate change. (This is a few years back, obviously.)
And one of them took the lead with talk of uv spectra and infra red absorption, tropospheric warming, increased hurricanes, droughts, icecaps melting, feedback
loops….I let him go on a bit.
Eventually and in a good natured way he explained that he had looked at the science and had come down on the side of the warmists, he believed what they had to say. I chose to believe ‘other’ sources of information. I knew what he meant by other but I let it go. I waited a moment and then asked him.
‘Do you know how your fridge works?’
I saw a flicker in his eye, his composure was gone.
‘What’s that got to do with it?.’ He laughed and everyone formed a little protesting chorus behind him.
‘Come on surely somebody knows how a fridge works – that one in the corner, that kept this beer cold?’ One by one their glares were an admission of ignorance.
‘That’s below the belt.’ I heard someone mutter.
” So you would, discuss, spectra, parts per million, infra red, uncle tom cobbly and all with me? You would explain the workings of the planetary climate system with relaxed confidence but have no knowledge of the inner workings of a ubiquitous household object which uses gases and presures to keep things cool….
My missus gave me one of her looks. Something had changed in the atmosphere…and it wasn’t CO2 levels.
Paul Birch says:
December 14, 2010 at 2:01 am
If we can add dirt and end up with more dirt, and we can also add the exact same amount of dirt and end up with less dirt, no, it’s not linear. If we can add three piles of dirt and end up with the same floor conditions as when we add one pile of dirt … no, it’s not linear.
In a system with linear feedback, there is a one-to-one relationship between dirt added and dirt levels. But in a system with a governor, there is a many to one relationship between dirt added and dirt levels. We get the same result if we add three units of dirt and three TDDs clean it up, or if we add one unit of dirt and one TDD cleans it up.
Mathematically, this kind of system is termed to be not “invertible”. An average, for example, is not invertible. Knowing the algorithm, we can say that the average of three and five is four. However, if we know the average is four, we can’t determine what the two numbers were that made up the average. Could be six and two, or seven and one. This is the “many-to-one” relationship that we get in a system with a governor.
Which is why I said that we can’t calculate the rate at which dirt is added from the average amount of dirt on the floor.
And as you point out, to go the other way and predict floor conditions requires a knowledge of the dirt devils’ algorithm. But it requires something more than that. It requires that we know the distribution of the dirt that’s being added. But in a climate model, we don’t know that, because we are working both from and to gridcell averages. So even if we know the algorithm, it won’t help much if we don’t know the distribution, we can only get a crude picture.
Huh? That’s what I said in my head post.
Now, that’s what I said. So why are you acting like I said it couldn’t possibly be modeled? My point is that the current crop of models aren’t designed to model that kind of system. They’re designed to model a linear one-to-one system, not a governed system.
I’m afraid I’m not following you here. You speak of:
That’s not clear. Seems like you’re saying that thunderstorm formation depends on other temperature and insolation factors. Your example is that “thunderstorms are driven by a temperature difference between surface and stratosphere, not the absolute temperature.” Well, kinda. You are correct that there are a host of factors that affect thunderstorm formation. And? Exactly which of my claims does that somehow negate?
As to your claim that there are places where a “tropical mechanism ceases to operate”, well, yeah, that’s called the extra-tropics. Again I’m not following your argument. How does that negate something I’ve said?
And you were doing so well, too, but then you decided to blow all your gains by getting nasty. So it goes. It’s useful to know, though. I find that kind of unwarranted anti-social behavior to be a pretty reliable gauge of a man’s inner belief in the strength of his arguments …
Now, you may be right, it might have been pointed out to me before. But if they did it the way you just did, by mumbling something unintelligible about the “broadly linear (or at least monotonic) dependencies” of something that you disdain to name “on temperature and insolation”, I didn’t “ignore it”. I simply didn’t understand it. What is it that you are referring to, that is dependent on insolation? Where did I say that the thunderstorm governing system was supposed to eliminate that mystery dependency? What are you on about?
Paul, here is a primer on how to tear my work to shreds.
1. Quote my exact words that you are disagreeing with, and cite where they came from. Most important. Quote what you think is wrong.
2. Explain as clearly as you can exactly which parts of that statement of mine you think are incorrect. Be as specific as you can be. Saying “it’s wrong” is meaningless. You need to say something more along the lines of “Your estimate of the value of X is three times the real value” or “Your claim of the evolutionary descent of monopoles from monotremes is highly improbable” or “You have made a mathematical error in the third step, E is actually equal to MC squared, not unequal”. That way, we both know exactly where you think I went off the rails.
3. Explain as clearly as you can exactly why you think what I said is incorrect. Again, be specific. Include citations and details. You want to say something along the lines of “Your estimate is high because you have neglected the first order vestigial discriminant factor” or “Higgbotham (1977) showed that monopoles have very different genetic makeup from monotremes”. Bring out the big guns, show precisely how my facts are wrong or my math sucks, post links to support for your ideas, then we’ll know exactly why you think my ideas are hokum …
If you do that, you’ll get some traction with your ideas and your claims. But just making vague assertions about some unknown dependencies that someone is supposed to have told me about sometime or other , and then leaving with an insult?
Not so much …
Why do the warmist lot get the high paying jobs and will it stop? I am a lowly engineer, just employed to fix practical things, in the real world, I have a memory and can read, this stopped me believing in AGW from the start I saw it was a scam. Also the “fix” sending all the wests money and jobs to China seemed ineffectual and unlikely to have any effect that was good so tell me why do the liars get the good wages?
This ming game was good but I went there many years ago, we need to become a powerful group, worldwide with a voice, the trouble is Im a lot busy right now coping with the green taxes to keep my family warm and fed, maybe thats the real plan.
It is a nonlinear chaotic system and thus can not be predicted. Your only hope is to work out the major bifurcation modes and prepare responses for those conditions based on their probability.
Then comes the once-in-a-century “perfect storm”….oh well.
Excellent article; thanks, Willis!
An English point: The possessive of it is its. “It’s” means “It is” or “It has.” (One would never write hi’s meaning “belonging to him.”) No pronoun, regardless of case, has an apostrophe; the apostrophe, if present, means a contraction (i.e., two words expressed as one).
David Socrates says:
December 14, 2010 at 1:54 am
Willis Eschenbach, ably assisted by Charles Nelson (December 13, 2010 at 11:48 pm), have between them most entertainingly put this global warming conundrum to bed.
Sensible skeptics start from the twin observations that:
(1) Contrary to uninformed doom mongering over the past 30 years, at only 0.5degC or so per century the Earth’s temperature is steadfastly failing to rise in correspondence with the sharp (~30%) post-WW2 upturn in anthropogenic GHGs.
(2) The straightforward physics of CO2 indicates that an increase in its atmospheric concentration should have a significant and potentially worrying warming effect.
########
A. you’d have to net out negative forcings ( like volcanoes ) to get the isolated impact of GHGs
B. Time lag. The response to C02 addition is not instananeous. Its not like a blanket despite stupid metaphors to that effect
What if the nature of those “dirt devils” is electrical?, “a la Vukcevic”, and those pesky Solar storms, outside the “room” where those devils work, affect its behavior, as Piers Corbyn contends, and worse, if those are “She devils” then they are affected too by the Moon cycles, and every 28 days, they have an irritating character…. 🙂
The modified warmist position, best I can tell, is that increased warming leads to more unstable weather. This thought experiment seems to lead to the same conclusion. I don’t think the two claims are quite the same, but I’m not clear on the differences. Would someone care to help clear that up?
So it requires dropping the gridcell concept in order to liberate the models from their constraints. So simple, yet so difficult. The supercomputers ought to have enough horsepower to handle it now.
Willis Eschenbach says:
December 14, 2010 at 3:43 am
“If we can add dirt and end up with more dirt, and we can also add the exact same amount of dirt and end up with less dirt, no, it’s not linear. If we can add three piles of dirt and end up with the same floor conditions as when we add one pile of dirt … no, it’s not linear.”
Sorry, but this does not follow. If adding dirt in a particular distribution increases or decreases the overall amount of dirt in approximate proportion to the added amount – or even leaves it the same – that is a linear relationship.
“In a system with linear feedback, there is a one-to-one relationship between dirt added and dirt levels.”
No, this does not follow, because there can be other things (like noise) going on at the same time. A one-to-one relationship is not required for there to be a correlation. Quite the opposite. We use correlations to link varibables precisely when there isn’t a one-to one relationship.
“But in a system with a governor, there is a many to one relationship between dirt added and dirt levels.”
There may or may not be. It depends on the system. A simple AGC is a governor, with non-linear feedback, having an asymptotically limited output that is monotonic and in one-to-one-relationship with the input. A thermostat with hysteresis, on the other hand, does not have a one-to-one relationship of input and output; but in general there will still be a linear relationship between the average room temperature and the external temperature, the radiator temperature and the heating power. Contrary to popular belief, thermostats seldom if ever prevent room temperatures from varying with external conditions – they merely reduce the impact of such changes (in some cases with a change of sign).
“We get the same result if we add three units of dirt and three TDDs clean it up, or if we add one unit of dirt and one TDD cleans it up.”
Not in general. A similar result, very probably. But exactly the same? Almost certainly not.
“Mathematically, this kind of system is termed to be not “invertible”.”
This is nothing to do with the governor or feedback mechanism. It is a simple consequence of the fact that the single output is derived from multiple inputs; there is insufficient information to separate out the inputs, given only the output figure. However, this does not debar one from measuring the sensitivity of the output to each of those inputs in turn. And in a well-behaved physical system, those dependencies will be linear, for small changes about a norm.
“Which is why I said that we can’t calculate the rate at which dirt is added from the average amount of dirt on the floor.”
And you’re wrong. We can. It will only be an estimate, subject to various fluctuations and errors, but on average it will be correct (assuming that the deposition mechanism has not altered significantly since the observations upon which the model is based).
“And as you point out, to go the other way and predict floor conditions requires a knowledge of the dirt devils’ algorithm.”
No, it doesn’t. The details of the algorithm will indeed change the correlation, but it is not necessary to understand the algorithm in order to measure the correlation. An empirical relationship can be obtained and used as a predictor (in either direction), whether or not one understands the physical mechanism.
“Let me be clear about what I am saying about models. I’m not saying that we can’t model the climate. I think we can, although it won’t be easy. But we have to model it the way it really is.”
No, we really don’t. We can empirically or analytically model those parts of the system and those mathematical relationships that interest us, without bothering with all the complexities that we’re not interested in, or don’t know much about, or can’t predict.
“They’re designed to model a linear one-to-one system, not a governed system.”
They’re designed to model the linear dependencies that exist in any well-behaved system, including “governed” ones (whether with linear or non-linear feedback).
“Seems like you’re saying that thunderstorm formation depends on other temperature and insolation factors. Your example is that “thunderstorms are driven by a temperature difference between surface and stratosphere, not the absolute temperature.” Well, kinda. You are correct that there are a host of factors that affect thunderstorm formation. And? Exactly which of my claims does that somehow negate?”
A thermostat that measures or regulates the difference in temperature between two rooms cannot set the temperature for the whole house.
“As to your claim that there are places where a “tropical mechanism ceases to operate”, well, yeah, that’s called the extra-tropics. Again I’m not following your argument. How does that negate something I’ve said?”
Because a thermostat that regulates the temperature in a variable fraction of a room cannot regulate the temperature of the whole house. The latitude band over which the thunderstorm mechanism operates itself depends on external conditions (such as insolation, greenhouse warming, non-tropical albedo, etc.) in an essentially linear fashion (to first order or for reasonably small changes). So even if the thunderstorms were a perfect regulator (which of course they’re not), their effect on the global average would still be such a linear one, within the scope of a simple semi-empirical model.
“I find that kind of unwarranted anti-social behavior to be a pretty reliable gauge of a man’s inner belief in the strength of his arguments … ”
Yes, that’s why I’m afraid I have to consider you a cowardly poster – you seem to respond readily enough to the ignoramuses who don’t understand basic physics, but if ever someone comes up with a cogent criticism, then as soon as you realise you can’t answer it, you ignore them and pretend it wasn’t said. You have done this to me more than once. For example, I demonstrated conclusively on a previous thread why your claim that a one-shell greenhouse cannot produce enough warming was false; I supplied not one but three distinct mechanisms capable of generating unlimited warming. You ignored the comment completely. I have also seen you do the same to other commenters, who have shown you quite explicitly how your thermostat model breaks down. You didn’t ask them for clarification; you just ducked out of the discussion. If this was merely inadvertance on your part, then I apologise.
Stefan says:
December 14, 2010 at 3:10 am
@Paul Birch: for typical scenarios in which the average manner in which dirt is deposited is fairly consistent
“I don’t understand. What typical scenarios are you assuming where the dirt is deposited consistently? ”
On whatever timescale you’re interested in observing the average dirtiness of the kitchen, that’s the timescale on which the deposition mechanism has to be reasonably consistent (for a simple empirical model). The kitchen “climate” rather than its “weather”. It might be a day, or a week, or a year. Obviously, the more you know about the specifics of how the dust devils operate, and how dust accumulates in the kitchen, the better the handle you can get on the “weather” too.
Alex the skeptic says:
December 14, 2010 at 3:24 am
“… ‘kick hard enouhg to throw the sphere out of the bowl”, but as I said in my original comment:
“Consider the quantity and magnitude of all the forcings that our planet has gone through during its billionial life.” Fact is that our planet has never experienced such a kick during its geological histroy, not due to CO2 increases, solar forcings, not even the catastrophic direct hit that presumably killed the dinosaurs. ”
This is somewhat debatable, since mass extinctions arguably should count as “sphere out of bowl” (even if it subsequently found its way back in). However, my point was that this does not tell you whether or not you have net positive or negative feedback. The Earth could have survived with the former – or suffered catastrophe with the latter.
“Whenever there is a “hot spot” on the tropical ocean or land, if it is hot enough, a thunderstorm springs up and starts pushing huge amounts of energy vertically. As the thunderstorm moves across the surface, it moves towards the warmest area in its path.”
Hello Willis,
A nice analogy that ties in perfectly with your analysis on the models, but I have to somewhat disagree with the statement above as being too simplistic and sometimes incorrect. A better term than “hot spots” might be “hot corridors” but that still wont remove the oversimplification.
From my long term observational experience in my neck of the woods (Florida), given the absence of a strong synoptic pattern summer thunderstorm activity is governed strictly by daily mesoscale events, i.e. sea breezes. There are no “hot spots” per se, but rather collision corridors created by two air masses traveling in opposite directions with nowhere to go but up when they collide. The position of the “corridors” for these collisions is set by the differential wind speed between the competing sea breezes (mainly the easterly component) and is accurately predicted on a daily basis by local meteorologists. Where the thunderstorms go from there is a result of which sea breeze wins the daily fight with the easterly component winning the large majority of them, so thunderstorms usually move to the west.
In your defense though, this too is kind of simplistic because there are many other variables that come into play. For example, this is a sunrise image made on the West coast of Florida:
http://www.josesuroeditorial.com/Nature/Clouds/Clouds/1228906_ry6rV#1024192053_eXxNC-O-LB
The image depicts an unusual but not rare event when the wind pattern is reversed and storms are born early in the morning, instead of in the afternoon/evening, and they move to the East rather than to the West.
Below are two images showing the more typical east to west summer pattern – evening storms:
http://www.josesuroeditorial.com/Landscapes/Scenics-2007/2348997_u6avH#179117325_6suat-O-LB
http://www.josesuroeditorial.com/Landscapes/Scenics-2008/4113375_y8Sqs#374868986_kbbD5-O-LB
Best,
Jose
Paul Birch says:
December 14, 2010 at 2:01 am
For example, thunderstorms are driven by a temperature difference between surface and stratosphere, not the absolute temperature; and there is evidently a variable latitude, or range of latitudes, by which the tropical mechanism ceases to operate. He’s had this pointed out to him several times before, but prefers to ignore it.
I distrust thought experiments because they are too Aristotelian for my taste. I want real data. But the quote from your post above has a contradiction in real data evaluation.
You say that thunderstorms are driven by temperature differences, not absolute temperatures.
And then “there is a range of latitudes by which the tropical mechanism ceases to operate” !!!
Higher latitudes mean lower SST s , i.e. absolute temperatures because of lower energy inputs from the sun.
Real data contradicts you.
Re: Figure 2 – WOW – look at the range of absolute temperatures! Apparently, they all start with different initial conditions. Oh well, it IS climate “science”…
Chris Edwards says:
December 14, 2010 at 3:49 am
Why do the warmist lot get the high paying jobs and will it stop?
Chris – my own research has show that government employees like Hansen, Schmidt, Karl, Lacis, et al. earn very comfortable six figure salaries – not including benefits. They also received salary increases in 2009 when people in the private sector were being laid off. And I’m sure the notoriety of people like Jim Hansen earn them additional incomes from their “outreach” efforts…
Jose Suro says:
December 14, 2010 at 7:38 am
Fabulous photographs!
/Mr Lynn
Paul Birch wrote:
In replying to Willis assertion that we have to model climate the way it is, you replied:
“No, we really don’t. We can empirically or analytically model those parts of the system and those mathematical relationships that interest us, without bothering with all the complexities that we’re not interested in, or don’t know much about, or can’t predict.”
This doesn’t make much sense to me. If you are saying we can model only those parts that interest us or we understand, then what we are doing is modeling a subset of a system. But I thought the idea is to model the climate as a whole – in other words, to predict the global averaged temperature at some future date. If you are saying that if all we want to do is model, say, the radiative forcing from CO2 levels, and ignore the effect on the climate as a whole, then I agree with you. But if you are implying that we can then go on and model the whole climate even though we have missed out bits of the whole, then I disagree.
Maybe I am misuderstanding your point here.
anna v says:
December 14, 2010 at 7:42 am
“Higher latitudes mean lower SST s , i.e. absolute temperatures because of lower energy inputs from the sun. Real data contradicts you.”
So you are saying that higher latitudes, with there relatively lower temperatures from surface to stratosphere, prove that thunderstorms are driven by absolute temperatures and not the surface-stratosphere temperature differential? I don’t see even one baby step in your chain of (bad) logic for that. Is there a lower temperature differential, yet more thunderstorms, at higher latitudes?
Thanks, Willis. At the risk of calling down the wrath of some, when I read the first part of your post, I thought you were heading down the path of showing why it is not possible to know exactly how much CO2 is being added to the atmosphere (or removed), simply based on the amount currently measured in the atmosphere. I think that would be a fair point as well and one that follows from your thought experiment.
steven mosher says:
B. Time lag. The response to C02 addition is not instananeous. Its not like a blanket despite stupid metaphors to that effect
Say the atmosphere doubles in CO2 concentration instantly. When do we see the effects? A year? Ten? A hundred? We see the effects of reduced W/M^2 everyday, not instantly sure. Why is CO2 forcing special, needing a lag?
“Now, you could look on that GISS model 0.7°C per century inherent warming drift with no forcing change as a bug. I prefer to think of it as a feature. ”
In programming, bugs are features. They are features we do not desire, but they are features.
Awww, Willis, now you’ve gone and given the AGWers a clue.
@ Paul Birch
“However, my point was that this does not tell you whether or not you have net positive or negative feedback. The Earth could have survived with the former – or suffered catastrophe with the latter.”
Err, no. Net positive feedback means that temperature is unbounded. In the “ball and bowl” analogy, net positive feedback is the inverted bowl. No matter how shallow the curve, the ball will roll off. We have a billion years or so of history to show this hasn’t happened. Therefore we have negative feedback in the system. The challenge to climate scientists should be to elucidate the cause and effect relationships which give rise to the negative feeedbacks both in the heating and cooling phases.
@Paul Birch:
“No, we really don’t. We can empirically or analytically model those parts of the system and those mathematical relationships that interest us, without bothering with all the complexities that we’re not interested in, or don’t know much about, or can’t predict.”
Huh? Are you sure you didn’t mean ‘can’t’? If not, why wouldn’t the exact opposite also be true? Isn’t this akin to arguing that being able to say with 100% certainty ‘the ball will stay within the wheel and roll around a bit’ means you can accurately model the outcome of a roulette game? The stock Market operates within defined boundaries that can be modelled, however if it were possible to dismiss inconvenient complexity with a hand wave, there would be some very rich software engineers out there.
It seems to me you are determined to set up a straw man in implying that Willis is arguing that because it is impossible to model a certain climatic system with any useful accuracy given the crudity of current models, it is impossible to construct any kind of model. The ‘useful accuracy’ part is vital to justify however if you are going to make prognostications of doom based on flimsy virtual evidence and circular reasoning. Hey, my climate model’s just told me it’s going to rain sometime. It works!
Aye – the Hadley Heat Pump.
There is a reason why heat pumps use a phase change, too – because it carries a whole lot more heat (without even requiring an actual change in temperature!) than is possible otherwise. No heat pump uses CO2…lol
Please note, also, that H2O gas requires no convection- it will rise because it is considerably less dense than the other gasses – addition of heat only makes it rise faster than fast.
I don’t see this as any kind of ’emergent’ phenomenon, though.
Paul Birch says:
December 14, 2010 at 6:52 am
“Sorry, but this does not follow. If adding dirt in a particular distribution increases or decreases the overall amount of dirt in approximate proportion to the added amount – or even leaves it the same – that is a linear relationship.”
Willis is stating that if the exact current distribution of dirt is known and the exact future deposit of dirt is known (all the devils), then yes we can model the kitchen floor’s “dirt climate”. You just need to know where all the dirt is now and where all the new dirt is going. From that, you know where TDD’s will appear and perform their clean up (after some experimental testing).
But if all you know are averages (the average depth of dirt on the floor now, and the average deposit of dirt coming in the future) you will not know where TDD’s will appear, so you cannot model the future “dirt climate”. Depending on how the dirt falls, future dirt could go up or down. There is no simple linear relationship between average dirt deposit and future average dirt level. You could certainly create a model, and over time it may give you average predictions that match average results! But for the particular result of any point in time, odds are high that the model will be wrong.
“(Willis) We get the same result if we add three units of dirt and three TDDs clean it up, or if we add one unit of dirt and one TDD cleans it up….(Paul) Not in general. A similar result, very probably. But exactly the same? Almost certainly not.”
Now that’s just damn funny. Willis creates a thought experiment with a fantasy creature called a Tasmanian Dirt Devil, and you tell him that he doesn’t get to say how his fantasy creation works! If the inventor of the thought experiment states that his creations follow a specific algorithm, guess what – you have to follow the thought experiment according to that algorithm!
I mean really, disagreeing on the physics of the Tasmanian Dirt Devil – priceless.
Steve says:
December 14, 2010 at 8:50 am
anna v says:
December 14, 2010 at 7:42 am
“Higher latitudes mean lower SST s , i.e. absolute temperatures because of lower energy inputs from the sun. Real data contradicts you.”
So you are saying that higher latitudes, with there relatively lower temperatures from surface to stratosphere, prove that thunderstorms are driven by absolute temperatures and not the surface-stratosphere temperature differential? I don’t see even one baby step in your chain of (bad) logic for that. Is there a lower temperature differential, yet more thunderstorms, at higher latitudes?
No, you are saying that.
Let me spell out what I am saying:
He (Birch) said that “the tropical mechanism ceases to operate at higher latitudes”. This is an observational fact. Also an observational fact is that the absolute temperature in the tropical region is larger than in the higher latitudes, as the link shows. Data again. The stratosphere has the same temperature, so it is the high absolute seas surface temperature that is responsible for creating the storms and not the large differential, considering that energy goes with T^4, and not with ( energy differential)^4, and it is heat energy that is feeding the tropical thunderstorms. The large differential is due to the absolute sea surface temperatures and is irrelevant to the problem.
I’d like to thank Charles Nelson for his marvelous analogy and thought game on rising air and its limits, or, more likely, lack of them.
anna v says:
December 14, 2010 at 7:42 am
“Let me spell out what I am saying:…The stratosphere has the same temperature, so it is the high absolute seas surface temperature that is responsible for creating the storms and not the large differential … The large differential is due to the absolute sea surface temperatures and is irrelevant to the problem.”
Ahhh, I see. Since it is the troposphere that is receiving the lion’s share of the heat, and increasing in absolute temperature, you have decided that the causal mechanism for thunderstorms can be seen as independent of the troposphere-stratosphere temperature differential.
Since Earth’s climate is relatively stable, over the geologically short term you may be correct (I don’t know). Given the average absolute temperature of the troposphere during any point in time, you are saying someone could accurately predict the average number of thunderstorms at any given latitude. Maybe Andy could speak on that – what do meteorologists do? Yeah, I know it’s weather, not climate, but in predicting thunderstorms do meteorologists compare troposphere-stratosphere temperature differentials, or do they just care about pressure, (absolute) temperature and latitude? If they knew the troposphere-stratosphere temperatures, would their thunderstorm predictions be more accurate (i.e., how relevant is the data)?
Which doesn’t change the fact that thunderstorm’s, causally, do rely on the troposphere/stratosphere temperature differential. The stratosphere does not have an evenly distributed temperature throughout all latitudes, by the way (I assume that’s what you meant by “the stratosphere is the same temperature”).
anna v says:
December 14, 2010 at 10:25 am
Anna – like you, I am amazed at the people who show up here and seem to believe that absolute temperatures have no place in atmospheric physics. Let’s take a look a just basic thermodynamics. There are, for example:
* boiling and freezing points of water
* the ideal gas law
* the Stefan-Boltzmann relation for radiation heat flux, q” = sigma*T^4
* Thermodynamic and transport property variations with temperature (e.g. specific heat, viscosity, thermal conductivity)
* saturation pressure versus temperature
* entropy
This is why comparisons like Figure 2 are interesting. Why is it that climate scientists are apparently OK with GCM numerical solutions where the mean absolute temperature levels between different codes vary by a full 5 deg C!
@ Frank K. “government employees like Hansen, Schmidt, Karl, Lacis, et al. earn very comfortable six figure salaries”
OK, maybe they are paid six figures, but are you really saying they earn them ?
I think Paul Birch is a meanie and has poor manners; but I haven’t the slightest idea what he’s talking about.
Steve says:
December 14, 2010 at 11:15 am
Which doesn’t change the fact that thunderstorm’s, causally, do rely on the troposphere/stratosphere temperature differential. The stratosphere does not have an evenly distributed temperature throughout all latitudes, by the way (I assume that’s what you meant by “the stratosphere is the same temperature”).
OK,have a look at the temperatures in the stratosphere.
at 30.000 meters it plays between -43 and -45C
the tropical sea is +28 – +30C difference 75C
higher latitudes +16 – +19C difference 64C, taking upper numers
My claim is that it is not the 9 degree change in the difference, but the 11 degree hotter sea surface temperature in the tropics compared with higher latitudes that starts the process that leads to tropical thunderstorms.
oh dear, that should be 11 not 9 🙁 .
Let us see the details, but closely:
In the beginning of time, man produced heat by friction between two wood sticks. What does it happen at the end of those sticks?, does it happen the same in every phenomenon where heat/energy appears?
We need to get back to the basics to understand, otherwise we just put names….
Temperature is a “necessary” condition, but not “sufficient” by itself to initiate thunderstorm development. Regardless of temperature, you cannot have thunderstorm development with out adequate moisture (humidity), the proper temperature humidity profile with altitude (instability) and some mechanism to initiate lift.
The large temperature differential does provide more convective energy once things get kicked off, but you can have thunderstorms even in much lower temperatures if you have the proper atmospheric temperature/humidity profile and a strong enough trigger to initiate lifting of the unstable air far enough for it to cross the convective barrier and begin rising due to buoyancy.
In the mid latitudes where the Hadley cell circulation is downward, you have a strong bias against lifting and convection because the air mass as a whole is sinking, warming and drying out. You would need a very strong injection of moisture at low levels and a very strong lifting mechanism such as a front passage to over come this general sinking of the atmospheric column.
Larry
Steven Mosher says:
December 14, 2010 at 4:39 am
David Socrates says: December 14, 2010 at 1:54 am: Willis Eschenbach, ably assisted by Charles Nelson (December 13, 2010 at 11:48 pm), have between them most entertainingly put this global warming conundrum to bed. Sensible skeptics start from the twin observations that:
(1) Contrary to uninformed doom mongering over the past 30 years, at only 0.5degC or so per century the Earth’s temperature is steadfastly failing to rise in correspondence with the sharp (~30%) post-WW2 upturn in anthropogenic GHGs.
(2) The straightforward physics of CO2 indicates that an increase in its atmospheric concentration should have a significant and potentially worrying warming effect.
########
A. you’d have to net out negative forcings ( like volcanoes ) to get the isolated impact of GHGs
B. Time lag. The response to C02 addition is not instananeous. Its not like a blanket despite stupid metaphors to that effect
##########
Steven,
What on EARTH are you on about?
I write a considered piece showing how Earth’s thermostat (the hydrological cycle) will always exactly compensate for every relatively minor perturbation including increases in CO2.
And what do you do? You completely ignore the main thrust of my argument (which you would be perfectly entitled to argue with, constructively, if you disagreed) and instead talk obliquely about another minor perturbation, the cooling effect of volcanoes, which (if you had understood what I was saying) you would have realised would also be compensated for exactly by the same mechanism.
Then, just for good measure, you introduce the issue of time lags and “blanket” warming metaphors. Did my analysis mention blankets? No. Is my analysis affected in any way by time lags? No. So why not address my analysis head on if you disagree with it rather than trying to be uber>-clever by mentioning factors that are quite irrelevant?
This is not a complicated issue. Why turn it into one? Obviously you are no engineer!
WOW!
Paul Birch says:
December 14, 2010 at 6:52 am
Paul, I make a sincere attempt to answer honest scientific questions. Sometimes I don’t answer because I don’t see them. Sometimes I’m tired of repeating myself.
And sometimes I get tired of people who repeat the same point over and over, and want to call me a coward in the bargain …
For example, I said that if an input of x give you a response of +1 sometimes and -1 other times, that’s not linear. Linear is something of the form of
y = m * x + b
Your response was
We have already established that we don’t know the distribution of the added dirt. That’s the meaning of a gridcell average. You can’t figure out the distribution. I said that at the start. I repeated that at the middle. I said that again at the end.
But here you still are, after all of that repetition, going on about how it is linear if we know the distribution. Maybe it is, maybe not, but that is MEANINGLESS in our situation because we don’t know the distribution. As I have said. Over and over. And over again.
So no, Paul, I’m not going to go on trying to convince you. It’s not cowardice that makes me do that. It is exasperation and frustration. Like they say, you can lead a snake to school, but it’s a bitch to make one sit up and take notes. Please apply elsewhere for your further education, you’ve proven here that you don’t listen to anything I say, and want to be unpleasant in the bargain. Once again you have made vague allegations of what I am supposed to have done to someone, sometime, somewhere … don’t you get tired of polluting the air like that?
Look, Paul, I’m just another fool trying to hack my way through a blizzard of mostly execrable science and tinkertoy models. I’m sorry if the way that I’m doing it doesn’t meet your standards. If you want it done better, I encourage you to start your own blog.
But that does not entitle you to accuse me of bad faith in the bargain. I’m doing my honest best in this deal, and I find your recurring accusations unwarranted, unpleasant, and as I said before, very revealing about your point of view in all of this. It doesn’t paint a very flattering picture of you, I’m afraid.
Yes, as someone else said, “party pooper. ” I need to turn off the computer and go clean house, Christmas is coming. Sure wish those TDD’s were for real.
Willis
I’m immensely enjoying your thought experiments. Simple, to the point, and highly rational. I becoming a real fan.
To today’s subject, I don’t see Paul’s argument. To the point — I believe Paul makes his “dust devil” algorithm too complicated. At a simplistic level, the “devil” needs to know only three things: 1) see dirt, 2) my turn, and 3) done. The devil does not need to “know” much more and the algorithm be not more complicated than necessary to fulfill those requirements.
When one extends your logic to the proposed Hadley Cycle “heat engine”; an individual “dust devil” clearly become’s a random sequence of metrological events that produce a thunderstorm. The actual sequence of events required are nicely outlined by “hotrod (Larry L” @ December 14, 2010, at 12:01 am and later just above. I see no conflict exists with your chaotically created thunderstorms generally appearing in particular latitude band; simply because the confluence of events necessary for thunderstorm creation are simply more likely to occur in that area. In this case:
The random events leading to thunderstorm formation = See Dirt & My Turn
Depleting the available energy able to sustain formation = Done
In my view, Paul is trying to establish excessive “order” — where none exists or needs to exist.
Well done, Kforestcat
P.S. Paul Birch: where you wrote to Willis stating “you seem to respond enough to the ignoramuses who don’t understand basic physics”. You remind me of a particularly anti-social Phd in my employee. I didn’t fire the lad and counseled temperament; because, I knew his peers would run him off if he didn’t learn to respect even our lowliest janitor. Even a casual observation of the comments on this site show the “average” commenter has a better than average knowledge of physics, engineering, statistics, and/or meteorology. A typical respond on purely “technical” subjects tends to bear this out. Incidentally; I have a chemical engineering degree, 25 plus years of professional experience, I am paid a six figure salary, and I routinely advise executives on how to spend $5-7 billion in process equipment, fuel, and supplies. Never-the-less there are times on this blog when I know I’m out-smarted, out-gunned, and out-classed. Part of my success has been in knowing when to shut-up and listen to a broad category of specialist and opinions. Courtesy, mutual respect, and good manners are everything. Highly recommend you keep that in mind.
Paul Birch,
As one who was right there beside you arguing against Willis’ metal shell and it’s radiative effects, I can honestly advise you that you (we) were wrong. And, not just a little wrong… big time foolishly wrong.
Just as Kforestcat above states, I too have an advanced degree (Medical Physics), and quite successful in my field, and am considered well educated by my peers. And, I too am made to look foolish sometimes when I venture into a somewhat related field that I am not so comfortable with.
As to the current topic, Willis’ analogy holds up under scrutiny, and your arguments do not actually contradict anything in his thought experiment.
For example, Willis never that the distribution of the added dirt was known. In fact, he makes it clear that its not known. He clearly seems to be arguing that this corresponds to climate models and their grid averaging methodology. If you disagree with that analogy, then you would be better off showing that climate models do not bin that way. I suspect, however, that Willis is correct, and that they do bin that way. And if they do, then he is also correct that they ignore the non-linear nature of the essential problem.
You also state:
**************************************************************************
Willis: “Let me be clear about what I am saying about models. I’m not saying that we can’t model the climate. I think we can, although it won’t be easy. But we have to model it the way it really is.”
Paul: “No, we really don’t. We can empirically or analytically model those parts of the system and those mathematical relationships that interest us, without bothering with all the complexities that we’re not interested in, or don’t know much about, or can’t predict.”
*************************************************************************
(sorry… dont know how to add quotes)…
Well Paul, if you can, you and the entire climate change industry haven’t done it yet… or at least not very well. In case you haven’t been paying attention, the climate models suck.
As far as Willis being a cowardly poster… I would argue just the opposite. It takes real courage and patience to deal with people that are stubbornly wrong… and to do so with grace and courtesy. I for one, thank Willis for those qualities. If he didn’t exhibit them, I would not have followed through with his offered knowledge, and I would have missed out on an opportunity to educate myself. Thanks Willis.
David Socrates.
“Sensible skeptics start from the twin observations that:
(1) Contrary to uninformed doom mongering over the past 30 years, at only 0.5degC or so per century the Earth’s temperature is steadfastly failing to rise in correspondence with the sharp (~30%) post-WW2 upturn in anthropogenic GHGs.”
1. There is nothing in the theory that predicts a monotonic increase in temp, or an immediate increase. The effect is
A. sometimes masked by short term negative forcings
B. lagged to the input.
For example, if you are steering a jetski and you apply a forcing at the tail you will
see an immediate response. As in a sharp turn to the left. Even if the current is going the other way, even if the wind is going oppoiste to your control input. The time constant is short. apply forcing, see reactio. If you are steering an oil tanker you apply the forcing and you may not see any response for a while. Think inertia. But eventually that forcing will turn the boat. The post war 30% increase is nothing:
1. the response is a log response
2. the response has to overcome thermal inertia
3. the effect needs to be large enough to rise above background noise ( like shot noise from a volcano) and long term (15-30 year) oceanic cycles.
Now double the C02 and you’d see an effect of 1-3C at equillbrium. takes decades.
Anton Eagle says:
December 14, 2010 at 7:48 pm
Dear friends, I hold this gentleman up as an inspiration. He is a true scientist, and an honest man. Well done, sir, I doff my hat to you.
Anton eagle shows courage, the kind of courage most of us have trouble summoning.
steven mosher says:
December 14, 2010 at 10:31 pm
Steve I always read your posts with interest and often learn something from them.
But I’m afraid my “reason centre/BS detector” just won’t accept the “takes decades” claim.
Just because an analogy sounds good, doesn’t mean it applies to the intended target.
Radiation happens at the speed of light. How is it that a molecule of CO2, just released into the atmosphere, doesn’t take part in the absorption/emission game for decades?
I don’t wish to sound like a smart a$$ but do these molecules wait in the sidelines?
p.s. Hop on a jet ski the size of an oil tanker and see how sharp you can turn. I would have thought it’s the ratio of force to size. tankers are not designed to turn sharp.
anna v says:
December 14, 2010 at 7:42 am
Paul Birch says: For example, thunderstorms are driven by a temperature difference between surface and stratosphere, not the absolute temperature; and there is evidently a variable latitude, or range of latitudes, by which the tropical mechanism ceases to operate.
“I want real data. But the quote from your post above has a contradiction in real data evaluation.
You say that thunderstorms are driven by temperature differences, not absolute temperatures.
And then “there is a range of latitudes by which the tropical mechanism ceases to operate” !!!
Higher latitudes mean lower SST s , i.e. absolute temperatures because of lower energy inputs from the sun.
Real data contradicts you.”
No, it doesn’t. One can only have a tropical thunderstorm if there is a sufficient temperature difference between surface and stratosphere and a high enough absolute humidity. If the overall temperature is too low, or the air too dry, you don’t get them. If the temperature is high, but there is insufficient temperature difference to drive them, you don’t get them. At higher latitudes, the temperature difference is too small. The absolute temperature helps determine the strength of the effect, but it is the temperature difference that the “governor” is regulating.
Vince Causey says:
December 14, 2010 at 8:26 am
Paul Birch wrote: We can empirically or analytically model those parts of the system and those mathematical relationships that interest us, without bothering with all the complexities that we’re not interested in, or don’t know much about, or can’t predict.
“This doesn’t make much sense to me. If you are saying we can model only those parts that interest us or we understand, then what we are doing is modeling a subset of a system. But I thought the idea is to model the climate as a whole – in other words, to predict the global averaged temperature at some future date. ”
Yes, but that is a subset of the system. We can have a viable model that predicts global climate (average temperature, rainfall, etc.) without necessarily being able to predict or understand local weather. Or vice versa. The idea that a model is no good unless it includes everything is misguided. It’s all too easy to go into such detail that you can’t see the wood for the trees. (I’m not claiming that the global circulation models currently employed are good models – they’re not – but the mere fact that eg., they don’t explicitly include tropical thunderstorms, does not in itself make them inadequate or invalid).
Robbo says:
December 14, 2010 at 9:03 am
@ Paul Birch However, my point was that this does not tell you whether or not you have net positive or negative feedback.
“Err, no. Net positive feedback means that temperature is unbounded.”
No, it really doesn’t. It means that temperatures vary more than they would have done in the absence of the feedback, that’s all. The inverted bowl is not a example of normal positive feedback, but of instability. One can have a stable system with either positive or negative feedback; and one can have an unstable system with either positive or negative feedback.
JJB MKI says:
December 14, 2010 at 9:14 am
Please see my reply above on: December 15, 2010 at 1:46 am
“It seems to me you are determined to set up a straw man in implying that Willis is arguing that because it is impossible to model a certain climatic system with any useful accuracy given the crudity of current models, it is impossible to construct any kind of model. ”
Quite the opposite. I am saying that you can get useful models without having to understand the underlying mechanisms. Indeed, for many purposes, an empirical model, by virtue of its simplicity, may well be superior to more sophisticated ones. Willis seems to be trying to argue that if the model doesn’t explicitly include regulation by tropical thunderstorms, etc., it’s no good. And that simply doesn’t follow. Whether or not deliberately including such thunderstorms would improve the predictive power of a global temperature model is an open question.
Steve says:
December 14, 2010 at 10:19 am
“But if all you know are averages (the average depth of dirt on the floor now, and the average deposit of dirt coming in the future) you will not know where TDD’s will appear, so you cannot model the future “dirt climate”.”
This does not follow. It is not necessary to know such details to be able to create an empirical model with predictive power. If you can plot the overall amount of dirt against the amount of dirt being added, then find a correlation between the two, then you have a working model. Even if the dust devils themselves are a complete mystery to you.
“Depending on how the dirt falls, future dirt could go up or down. There is no simple linear relationship between average dirt deposit and future average dirt level. You could certainly create a model, and over time it may give you average predictions that match average results! But for the particular result of any point in time, odds are high that the model will be wrong.”
But it’s the average results that you want! The “climate”, not the “weather”.
“(Willis) We get the same result if we add three units of dirt and three TDDs clean it up, or if we add one unit of dirt and one TDD cleans it up….(Paul) Not in general. A similar result, very probably. But exactly the same? Almost certainly not.
“Now that’s just damn funny. Willis creates a thought experiment with a fantasy creature called a Tasmanian Dirt Devil, and you tell him that he doesn’t get to say how his fantasy creation works! If the inventor of the thought experiment states that his creations follow a specific algorithm, guess what – you have to follow the thought experiment according to that algorithm!”
Fair enough, except that he didn’t specify the algorithm. He gave a general description only; and my point is that, in general, under Willis’s own description, the two cases will not give exactly the same result. He would have considerable difficulty finding an algorithm that would do so.
Willis Eschenbach says:
December 14, 2010 at 6:14 pm
“Paul, I make a sincere attempt to answer honest scientific questions. Sometimes I don’t answer because I don’t see them. Sometimes I’m tired of repeating myself.”
Then kindly answer the questions and points I have raised in response to your previous posts, which at the time you discourteously ignored. The excuse of repeating yourself won’t wash. No one in those threads had raised those points previously.
“We have already established that we don’t know the distribution of the added dirt. That’s the meaning of a gridcell average. You can’t figure out the distribution. I said that at the start. I repeated that at the middle. I said that again at the end.
But here you still are, after all of that repetition, going on about how it is linear if we know the distribution.”
I said, correctly, “If adding dirt in a particular distribution increases or decreases the overall amount of dirt in approximate proportion to the added amount – or even leaves it the same – that is a linear relationship”. I said nothing about knowing what that distribution is. It doesn’t matter what the distribution actually is, so long as the deposition mechanism stays sensibly similar over the period of interest.
“I’m doing my honest best in this deal, and I find your recurring accusations unwarranted, unpleasant, and as I said before, very revealing about your point of view in all of this. ”
My point of view about you is unfortunately based upon hard experience of how you have behaved to me (and others) in the past (I gave you a specific example), and how you are still behaving now. You are deliberately refusing even to address the specific points that I raised above, and which, despite your protestations, you have never answered at all. The one point you did address, you misread or misunderstood, then used that misunderstanding as an excuse for ignoring the rest.
Kforestcat says:
December 14, 2010 at 7:06 pm
“To today’s subject, I don’t see Paul’s argument. To the point — I believe Paul makes his “dust devil” algorithm too complicated. At a simplistic level, the “devil” needs to know only three things: 1) see dirt, 2) my turn, and 3) done. The devil does not need to “know” much more and the algorithm be not more complicated than necessary to fulfill those requirements. ”
My argument does not depend on what the algorithm is. I am saying that in general you can model the system empirically and find simple linear relationships in the correlation between overall dirt levels and rates without knowing anything about the dust devils or their algorithms. This is what Willis seems to be denying.
“Courtesy, mutual respect, and good manners are everything. Highly recommend you keep that in mind.”
I’m afraid that Willis has behaved discourteously to me from my earliest posts on this site, so you should not be surprised that I dislike him and am not inclined to show him much respect.
My point about “ignoramuses” was not meant to insult any commenters (I meant that that seemed to be Willis’s opinion of them, though perhaps I could have phrased it better!), but rather to indicate how very possible it is to give the misleading appearance of being open to criticism by cherry-picking only those points you think you can easily answer, and skipping over the ones you can’t.
Anton Eagle says:
December 14, 2010 at 7:48 pm
“As one who was right there beside you arguing against Willis’ metal shell and it’s radiative effects, I can honestly advise you that you (we) were wrong. And, not just a little wrong… big time foolishly wrong.”
The arguments you were making were not the same as mine, by a long chalk. Willis did not address mine at all, which demonstrated that even a single shell can, in principle, generate an arbitrarily high degree of greenhouse warming.
“For example, Willis never [said] that the distribution of the added dirt was known. ”
Nor did I.
“Well Paul, if you can, you and the entire climate change industry haven’t done it yet… or at least not very well. In case you haven’t been paying attention, the climate models suck.”
I don’t entirely disagree. But the reason is that the “climate industry” models are not empirically derived but are ideology driven. There are (or used to be) genuine climatalogical models – but since what they predict is no significant change beyond natural fluctuations, the industry won’t use them.
“As far as Willis being a cowardly poster… I would argue just the opposite. It takes real courage and patience to deal with people that are stubbornly wrong… and to do so with grace and courtesy. ”
Sorry, I disagree. It does not take courage to argue against people who (you think) are “stubbornly wrong”. It takes courage honestly to consider and evaluate cogent criticisms that genuinely threaten your theories.
@Paul Birch
Many thanks for your reply, and my apologies if my post came across as somewhat snarky- I was under the impression you were labelling commenters here as ‘ignoramuses’.
Could you elaborate on why, in your opinion, the global circulation models currently employed are not good?
Thanks, J Burns
steven mosher says:
Now double the C02 and you’d see an effect of 1-3C at equillbrium. takes decades.
————–
Why?
The Earth responds every day to the loss of forcing from the Sun. The effect can even be quite pronounced under a cloud. Or in an eclipse.
http://eclipse99.nasa.gov/pages/faq.html
The main effect is in the ‘radiant heating’ component which goes away suddenly at the moment of eclipse and produces a very fast temperature decrease.
And then, returns to normal following the eclipse. What possible explanation is there for a molecule of CO2 not interacting with such a fast responding system for decades?
What Dave F said.
Baa Humbug;
Sorry to say that your bit about turning tankers is not right – they take a long time to slow down in a straight line, but to stop in a hurry all you do (if you have the space to the side) is put the rudder hard over, and by the time you have gone round 360 degrees you will be just about stopped.
For 22 years I was a ship pilot.
Dave F says:
December 15, 2010 at 9:48 am
” (Steve Mosher) Now double the C02 and you’d see an effect of 1-3C at equillbrium. takes decades…. (Dave) Why? What possible explanation is there for a molecule of CO2 not interacting with such a fast responding system for decades?”
Steve’s key phrase is “at equilibrium”. The climate would indeed react instantaneously to an instantaneous doubling of atmospheric CO2 content. But that instantaneous reaction would not give you the mean increase in temperature that you would see after the sum total of all climate feedbacks that (according to Steve) would take decades to play out. The instantaneous change does not equal the change at equilibrium.
Do you think that someone could take the instantaneous change in temperature based on a 10 minute eclipse of the sun and extrapolate it to a new global mean temperature if that eclipse were to last 100 years? How long into that 100 year eclipse would it be before the new equilibrium point is reached?
Paul Birch says:
December 15, 2010 at 2:21 am
“It is not necessary to know such details to be able to create an empirical model with predictive power. If you can plot the overall amount of dirt against the amount of dirt being added, then find a correlation between the two, then you have a working model. Even if the dust devils themselves are a complete mystery to you.”
Well that’s the whole point – you won’t find a statistically sound correlation between the two! Finding the correlation requires knowing the details, not the averages.
But I agree that you can create a model. Even the wildest random process can be modeled. The question is – how useful is the model? What will your margin of error be for your prediction of overall floor dirt level at any point in time? We know the minimum and maximum dirt levels for the floor: minimum is zero, maximum is the level that summons a TDD. The average dirt level at any point in time will fall between those two extremes. I could say that the average dirt level is exactly halfway between those two extremes with a margin of error of +/- 100%. BLAM – a 100% accurate model that is 0% useful.
So I agree that you can create an empirical model, but I do not agree that it will have any predictive power, with power being the key word.
Steve says:
How long into that 100 year eclipse would it be before the new equilibrium point is reached?
How many licks does it take to get to the center of absolute zero? 🙂
I would think that it wouldn’t take a long time for Earth to lose the energy it holds. Still, the Earth fluctuates wildly every day between receiving, using round figures, about 1000 W/M^2 and practically nothing at night. Why would the comparatively small additional forcing cause the train to come off of the tracks wrt equilibrium?
Paul Birch doesn’t understand what “empirical” means in this thread. Birch opines:
“It is not necessary to know such details to be able to create an empirical model with predictive power.”
There is empirical data, and there are models. The first is evidence, the second is not.
Since Birch doesn’t understand those basic details, it’s surprising that Willis even bothers to try and explain anything to him.
Dave F says:
December 15, 2010 at 2:52 pm
“How many licks does it take to get to the center of absolute zero? :-)…I would think that it wouldn’t take a long time for Earth to lose the energy it holds.”
Jesus Haploid Christ, what kind of eclipse is it that we’re talking about?! You do realize that a huge portion of the Earth would still be receiving sunlight, right?
My question referred to calculating the time taken for the climate to re-equlibrate to suddenly having a shadow cast upon it 24/7, such as the one linked below.
http://apod.nasa.gov/apod/ap990830.html
Oops, haha, I went with the idea we were blocking out all of the Sun, not just one part of the Earth from receiving Sun. Sorry for the misunderstanding.
If the shadow were to stay in one spot, I have no idea.
Still, CO2 isn’t really analogous. You have huge changes in the amount of insolation in one location over a day, to which the atmosphere responds everyday. I don’t think it is reasonable to suspect that CO2 is going to destabilize the equilibrium when you look at the Δs for WM^2 on a daily basis and then look at the Δ expected for doubling CO2.
Paul Birch
Please tell us how you can have a positive feedback that is bounded?
Paul Birch :
December 15, 2010 at 1:28 am
No thunderstorms without humidity, you are right.
Humidity depends on the absolute temperature of the water. The hotter the water the more steam, which you can observe in your kitchen. This disproves your statement that absolute temperatures are irrelevant to the creations of thunderstorms.
Therefore hunderstorms depend to first order on the absolute temperature of the water, and to second order to the difference in temperatures between surface and stratosphere.
oldseadog says:
December 15, 2010 at 10:24 am
Thanx for the heads up sea dog. I have no doubt you are correct.
I guess my erraneous analogy proved my earlier statement (“Just because an analogy sounds good, doesn’t mean it applies to the intended target”) to be true.
I just felt comparing the massive size of a tanker to that of a jet ski to be off the mark, I still maintain that.
Leaving analogies aside, having read the comments, I’m still not convinced about the ‘decades” claim.
A large volcano goes off near the equator, it doesn’t take decades for T’s to fall. A strong El Nino develops in the Eastern Pacific, it only takes few months for T’s to rise.
But I may at some stage read a convincing argument and change my mind, until then….
Steve says:
December 15, 2010 at 2:17 pm
Paul Birch says: If you can plot the overall amount of dirt against the amount of dirt being added, then find a correlation between the two, then you have a working model. Even if the dust devils themselves are a complete mystery to you.
“Well that’s the whole point – you won’t find a statistically sound correlation between the two!”
How do you know? Have you tried it? I suspect that if you did you would find quite a strong correlation on at least some timescales. Certainly some control systems of this type do show such correlations – though sometimes the dependency can be opposite in sign to what you might expect (ie, the more dirt added, the lower the average dirt level).
“Finding the correlation requires knowing the details, not the averages.”
It really doesn’t. Empirical correlations are found by plotting the empirical data.
“So I agree that you can create an empirical model, but I do not agree that it will have any predictive power, with power being the key word.”
If it predicts either of the variables from the other at even slightly better than chance, then it has predictive power. And if what you are wanting is a model not of the immediate local “weather” but the long-term overall kitchen “climate” (which surely is the point of this analogy) then the model is likely to be quite good.
Smokey says:
December 15, 2010 at 3:03 pm
“Paul Birch doesn’t understand what “empirical” means in this thread. … There is empirical data, and there are models. The first is evidence, the second is not.”
An empirical model is a model based on empirical data. As distinct from an analytic model, which is based upon physical or mathematical analysis from first principles. Or a semi-empirical model (like the highly useful empirical mass formula for atomic weights) which uses a bit of both. Or, I suppose, an ideological model (like AGW), based upon what you want the results to be.
James Macdonald says:
December 15, 2010 at 6:33 pm
“Paul Birch
Please tell us how you can have a positive feedback that is bounded?”
OK. Consider a simple amplifier circuit with two additive inputs and unity gain on each. Apply a signal S. The output signal O=S. Then feed half of the output back to the other input. This is positive feedback. The output signal is now S+S/2=3S/2. But this means that 3S/4 is now fed back. So the output increases again to S+3S/4=7S/4. Then again to S+7S/8=15S/8. Then 31S/16, 63S/32 … to the limiting value of 2S. Positive feedback has doubled the initial signal. In general, if the feedback fraction is f, the output is multiplied by 1/(1-f). This remains finite (stable) so long as f is less than 1.
anna v says:
December 15, 2010 at 10:37 pm
“No thunderstorms without humidity, you are right. Humidity depends on the absolute temperature of the water. The hotter the water the more steam, which you can observe in your kitchen. This disproves your statement that absolute temperatures are irrelevant to the creations of thunderstorms.”
I made no such claim. On the contrary, what I said was this: “One can only have a tropical thunderstorm if there is a sufficient temperature difference between surface and stratosphere and a high enough absolute humidity. If the overall temperature is too low, or the air too dry, you don’t get them. If the temperature is high, but there is insufficient temperature difference to drive them, you don’t get them … The absolute temperature helps determine the strength of the effect, but it is the temperature difference that the governor is regulating.”
As heat engines, thunderstorms are fundamentally driven by the surface-stratosphere temperature difference. With a sufficient temperature gradient, they can even arise in polar and mountain regions at quite low absolute temperatures, albeit less powerfully than the tropical thunderstorms we are talking about here. But once the thunderstorm has hammered down that temperature difference it switches off. The thermostat switch is not working off the absolute temperature, but off the temperature difference. So changes in the absolute temperature are not regulated (ie, held within fixed bounds), only ameliorated by the negative feedback from the thunderstorms. The system dynamics are quite different.
JJB MKI says:
December 15, 2010 at 5:01 am
“Could you elaborate on why, in your opinion, the global circulation models currently employed are not good?”
The fundamental reason is that they’re not scientific models at all – they’re propaganda tools intended to “justify” an essentially irrational dogma, endlessly fudged and refudged to give the desired results.
More specifically, if you create a model supposedly based upon the underlying physics, then you have to include all the relevant phenomena. Not every last detail, but all the stuff capable of producing effects within the range of interest. In a climate model, you will, for example, have to include the feedback from variable cloud cover, tropical thunderstorms, vegetation, etc., or your model will be garbage. The AGW models leave out much of the crucial material, or fudge it with invalid approximations or worse. Much of the information that would be necessary for an adequate analytic climate model is simply unknown or not yet understood.
Alternatively, you can base your model on what you actually observe; then you don’t need to address all the individual phenomena explicitly, because they are built into the empirical data automatically by nature. However, the AGW models never had any real correspondance with observation; they couldn’t even retrodict the climate of the previous century. Reluctantly recognising this (if never openly admitting it), the AGW ideologists found ways of “adjusting” the models and the data so that the failure of the theory was less obvious.
So what we have now is a mish-mash that is neither empirically nor analytically sound nor complete. Garbage, in fact. However, like garbage, it is not totally useless; if you hunt about in the muck long enough you can find grains of recyclable truth amid the lies. Perhaps, though, it’s not worth the bother; might as well chuck it all away.
Moderator: my post of
December 15, 2010 at circa 3 am ,
directly responding to
[Willis Eschenbach says:
December 14, 2010 at 6:14 pm]
seems to have gone missing.
This was perhaps the most important of my replies yesterday, so I’d be grateful if you could recover it.
PB.
******
steven mosher says:
December 14, 2010 at 10:31 pm
Now double the C02 and you’d see an effect of 1-3C at equillbrium. takes decades.
******
Nonsense. The GHG effect is essentially instantaneous. Some effects lag, of course, maybe up to 1000 yrs (time for oceans bottom-currents to complete a “cycle”), but unless one assumes some fantastically huge positive feedback, then the immediate effect is by far the largest.
beng December 16, 2010 at 5:54 am said: Nonsense. The GHG effect is essentially instantaneous. Some effects lag, of course, maybe up to 1000 yrs (time for oceans bottom-currents to complete a “cycle”), but unless one assumes some fantastically huge positive feedback, then the immediate effect is by far the largest.
Beng, believe me, it’s absolutely no use debating with Steven Mosher.
This all started as a result of my original comment here on December 14, 2010 at 1:54 am to which he replied on December 14, 2010 at 4:39 am to which I replied on December 14, 2010 at 2:15 pm.
Originally I had suggested that the hydrological cycle stabilizes the Earth’s temperature because the fixed physical environment of the Earth involves some very strong negative feedbacks that (in the absence of any other perturbations) cause the temperature to remain broadly constant. But, in contrast, adding an extra slug of CO2 is an open-ended perturbation possessing no feedback compensation mechanism of its own.
The result? The hydrological cycle does what it always does when presented with any upward (or downward) uncompensated perturbation, it simply adjusts its own strong negative feedbacks so as to exactly compensate for the CO2 perturbation.
To illustrate the significance of this, I drew an analogy between the Earth’s hydrological cycle and my 40kW house central heating boiler system which similarly keeps my house at a more-or-less fixed 21degC due to the fixed physics of the boiler and the very simple negative feedback loop involving its house thermostat.
The house analogy to adding CO2 to the atmosphere is that I turn on a 3kW electric fan heater (which has no thermostat) which just continues indefinitely to add heat to the interior of the house.
Does the house temperature go up? Does it heck! Why? Because the central heating system’s thermostat exactly compensates for the added heat from the 3kW heater by closing down for proportionately longer periods.
So, presented with this powerful analogy as to why we haven’t seen any sign of a warming effect due to the sharp upturn in post-WW2 man-made CO2 emissions, instead of addressing it head on he chooses to go off at a complete tangent by suggesting that:
(1) The effect of the post-WW2 upturn in emissions will take decades to show up in the temperature record. This is unsubstantiated rubbish, as you and others here have quickly pointed out (why, for example, doesn’t water vapor take that long?)
(2) Negative forcings from volcanoes etc. might have masked the positive forcing of the added CO2 for 40 years (and why not earlier? why not in the future?). This simply shows he didn’t understand my proposed mechanism which exactly compensates for all forcings whether positive or negative.
The bottom line is that Steven is a convinced warmist, which is absolutely his right and good luck to him. But if he really wants to take part in a useful debate with the highly intelligent skeptics that inhabit this blog, he should engage with their arguments just as they try to engage with his.
But this he always fails to do, so nobody learns anything.
It’s true, Mosh is like Gore in this respect. He just delivers his pronouncement, and ignores logical argument against it.
Very statesmanlike…
😉
Paul Birch says:
December 16, 2010 at 1:37 am
“How do you know? Have you tried it? I suspect that if you did you would find quite a strong correlation on at least some timescales. Certainly some control systems of this type do show such correlations – though sometimes the dependency can be opposite in sign to what you might expect (ie, the more dirt added, the lower the average dirt level).”
Per your own words, “I suspect” is an ideological model, not an empirical model. You will not find a strong correlation between future average dirt levels and average dirt level added because there is no strong correlation! The strong correlation is between future average dirt levels and specific dirt level added per unit area. If you add a lot of dirt to a small area, odds are high that future average dirt levels will go down a little. If you add the same amount of dirt to a wide area, future average dirt levels may go up or down. If you add a huge amount of dirt to the entire area, odds are extremely high that future average dirt levels will go down.
“(Steve)Finding the correlation requires knowing the details, not the averages….(Paul)It really doesn’t. Empirical correlations are found by plotting the empirical data.”
Well, a correlation can always be found, however weak it is. If your empirical data shows a weak correlation between two variables, you can create a weak model.
“(Steve) So I agree that you can create an empirical model, but I do not agree that it will have any predictive power, with power being the key word….(Paul) If it predicts either of the variables from the other at even slightly better than chance, then it has predictive power.”
Well then you have given us your definition of an predicatively powerful model – anything slightly better than chance (over any time scale, I assume). That is not my definition of a powerful model.
tallbloke says: December 16, 2010 at 9:15 am It’s true, Mosh is like Gore in this respect. He just delivers his pronouncement, and ignores logical argument against it. Very statesmanlike… 😉
Tallbloke, You’ve just given me a hell of an idea. Perhaps Steven Mosher is Al Gore.
Paul Birch says:
December 15, 2010 at 2:08 am
Paul, let me go over this again. I have said, very explicitly, that I do think it is possible to model the thunderstorm situation parametrically.
You have made this identical ludicrous claim before. I have pointed out before that your claim was nonsense, viz:
Seems pretty clear to me. I said it in the head post, and I repeated it in response to you the first time you made your current asinine claim. In response, you simply make the claim again.
This why I said that you will have to apply elsewhere for your further education. Here, you not only pay no attention to what I say. It appears you also pay no attention to what you have said.
Paul Birch says:
December 15, 2010 at 2:54 am
Paul, see my immediately previous post. I am tired of repeating myself, particularly to an unpleasant inquisitor who seems to not notice that he is asking the same question over and over DESPITE IT BEING ANSWERED!
So no, Paul, I will not answer your questions. You want answers from someone? Fine. Find someone who knows something, treat them decently and LISTEN WHEN THEY ANSWER!
Paul Birch says:
December 16, 2010 at 1:37 am
Ummmm … err … Paul, it’s a THOUGHT EXPERIMENT. How would someone “try it”???!??
Steve says:
December 16, 2010 at 9:22 am
“Per your own words, “I suspect” is an ideological model, not an empirical model.”
It is neither. It is a suspicion.
“You will not find a strong correlation between future average dirt levels and average dirt level added because there is no strong correlation!”
Prove it. My experience as a scientist and systems engineer leads me to suspect that there would in fact be quite a strong correlation for most viable algorithms and realistic deposition patterns. This is because of the lags that arise before the deposition is detected and the time it takes for the devils to work. If you believe (ideologically?!) that there is no such correlation, then you will have to show mathematically how your chosen algorithms produce no correlations for any reasonable deposition pattern. Neither Willis nor you has done this. Alternatively, go out and buy some actual dust devils (they do exist, though I’m not sure whether they’re on the consumer market outside Japan yet) and measure what happens empirically.
“The strong correlation is between future average dirt levels and specific dirt level added per unit area…”
These are not global “kitchen climate” correlations; they are the specific programmed responses of the devils’ algorithm locally. Quite different.
“Well, a correlation can always be found, however weak it is. If your empirical data shows a weak correlation between two variables, you can create a weak model.”
There are two ways in which a correlation can be “weak”. One is through being statistically insignificant. The other is through being noisy. In a climate model correlations are always going to be noisy, because of the effects of local weather, but that does not make them insignificant or useless.
“(Steve) So I agree that you can create an empirical model, but I do not agree that it will have any predictive power, with power being the key word….(Paul) If it predicts either of the variables from the other at even slightly better than chance, then it has predictive power.
“Well then you have given us your definition of an predicatively powerful model – anything slightly better than chance (over any time scale, I assume). That is not my definition of a powerful model.”
No, I have given you the standard scientific definition of predictive power. Just like physical power, you can have little or lots, pW or GW. How “powerful” a model may be depends upon what you want it for. A model may be next to worthless for predicting day to day local weather, yet powerful at predicting mean global temperature over timescales of a century or more. Or vice versa.
Willis Eschenbach says:
December 17, 2010 at 12:11 am
Paul Birch says: … you can get useful models without having to understand the underlying mechanisms. Indeed, for many purposes, an empirical model, by virtue of its simplicity, may well be superior to more sophisticated ones. Willis seems to be trying to argue that if the model doesn’t explicitly include regulation by tropical thunderstorms, etc., it’s no good. And that simply doesn’t follow. Whether or not deliberately including such thunderstorms would improve the predictive power of a global temperature model is an open question.
“Paul, let me go over this again. I have said, very explicitly, that I do think it is possible to model the thunderstorm situation parametrically. ”
I never said you couldn’t. I said that it’s an open question whether that would improve the predictive power of a global temperature model.
“Could we model it parametrically, without having to model individual thunderstorms? Probably … but the model has to be designed to do that. And the current climate models either are not designed to do it, or are not doing it successfully.”
I did not dispute this either. However, an empirical model does not have to include thunderstorms explicitly (whether “parametrically” or “individually”), and if our physical understanding or detailed knowledge of the phenomena is poor it may be superior to any sort of analytic model.
“You have made this identical ludicrous claim before. ”
What is this claim you consider “ludicrous”? That one can get useful empirical models without understanding all the underlying mechanisms? That’s basic scientific method and practise. It shouldn’t even be contentious. I even gave a specific well-established example: the empirical mass formula.
If it’s something else you think I’ve said that is “ludicrous”, “nonsense” or “asinine”, then please tell me what and why you think so. Simply repeating big chunks of your post doesn’t tell me what your problem is.
“This why I said that you will have to apply elsewhere for your further education. Here, you not only pay no attention to what I say. It appears you also pay no attention to what you have said.”
Pot. Kettle.
Willis Eschenbach says:
December 17, 2010 at 12:26 am
Paul Birch: If you can plot the overall amount of dirt against the amount of dirt being added, then find a correlation between the two, then you have a working model. Even if the dust devils themselves are a complete mystery to you. … Have you tried it?
“Ummmm … err … Paul, it’s a THOUGHT EXPERIMENT. How would someone “try it”???!??”
By putting representative algorithms into a computer model of the kitchen and running simulations. Then plotting the results. Alternatively, by performing a real experiment on the same lines.
Willis Eschenbach says:
December 17, 2010 at 12:16 am
Paul Birch says: Then kindly answer the questions and points I have raised in response to your previous posts, which at the time you discourteously ignored. The excuse of repeating yourself won’t wash. No one in those threads had raised those points previously.
“Paul, see my immediately previous post. I am tired of repeating myself, particularly to an unpleasant inquisitor who seems to not notice that he is asking the same question over and over DESPITE IT BEING ANSWERED!”
You did not respond to those points or the comments in which they were made at all. You completely ignored them. If you deny this, please direct me to the specific comment in which you claim to have answered my demonstration that even a single shell can in principle provide an arbitrarily large greenhouse warming factor.
“So no, Paul, I will not answer your questions. You want answers from someone? Fine. Find someone who knows something, treat them decently and LISTEN WHEN THEY ANSWER!”
You haven’t answered! You have responded – with a blatant misreading of what I actually said – to only a single one of the numerous distinct points I raised on December 14, 2010 at 6:52 am.
Paul Birch says:
December 17, 2010 at 4:10 am
“(Paul) Prove it. My experience as a scientist and systems engineer leads me to suspect that there would in fact be quite a strong correlation for most viable algorithms and realistic deposition patterns. (Steve) The strong correlation is between future average dirt levels and specific dirt level added per unit area…(Paul) These are not global “kitchen climate” correlations; they are the specific programmed responses of the devils’ algorithm locally. Quite different.”
Ding ding ding!
Willis explained that their is no direct causal relationship between average amount of dirt now, average dirt level added and future dirt levels (the “global kitchen climate”). There is no viable algorithm between these variables. The viable algorithm is between the current distribution of dirt, the distribution of dirt added and the future distribution of dirt. That was the entire point of his article – this climate can’t be (usefully) modeled without empirical data on the details.
“(Steve) Well then you have given us your definition of an predicatively powerful model – anything slightly better than chance (over any time scale, I assume). That is not my definition of a powerful model. (Paul) No, I have given you the standard scientific definition of predictive power. Just like physical power, you can have little or lots, pW or GW. How “powerful” a model may be depends upon what you want it for. A model may be next to worthless for predicting day to day local weather, yet powerful at predicting mean global temperature over timescales of a century or more. Or vice versa.”
Well, in your “experience as a scientist and systems engineer”, I suppose you didn’t learn that there actually is a standard scientific definition of predictive power. And your definition isn’t it. http://en.wikipedia.org/wiki/Predictive_power
If I can guess the average dirt level of the floor for a given period of time and be correct 1 time in 1,000 , and a model comes along and correctly predicts the average dirt level correctly 5 times in 1,000, that model is 5 times more powerful than guessing. But it still would not be considered a powerful model, because 95% of the time it’s wrong!
Your comments haven’t clarified or added to the original article in any way. I understand that you strongly believe some “global kitchen climate” correlations could be found if you analyze enough data, and you could use those correlations to create a model that’s wrong most of the time. Wonderful.
Steve says:
December 17, 2010 at 10:01 am
Thanks, Steve. You have pointed out (once again) the flaw in Paul’s argument.
Here’s the thing. Suppose we have positive feedback, such that any change in a variable ends up getting doubled in the output. We can see this and measure it. Then, as Paul suggests, we can use this to estimate the underlying algorithm. This method works great for a system with feedback.
But now let us suppose that we are in a “governed” system, that is to say a system with a governor. The example most people are familiar with is “cruise control” on your car. This is a governed system, as is the Tasmanian Dust Devil system above.
Here’s the problem. We get in a car, and we want to figure out the engine algorithm that relates the fuel use to the speed that the car is travelling. We note that when the fuel use goes up the speed goes up, and that when the fuel use goes down the speed goes down. We use that information to construct an algorithm for what’s happening regarding fuel and speed.
But when we repeat our experiment in a governed system like a car with cruise control, we see something surprising. Fuel use goes up and fuel use goes down, but the average speed doesn’t change at all! (Note that this also happens with the Tasmanian Dirt Devils, where dirt input goes up and down, but the average dirtiness of the floor stays constant.)
Now, if an input variable (fuel or dirt) changes and the output variable (speed or dirtiness) doesn’t change, we’re out of luck. At that point, it is not simply difficult (as Paul claims) to construct an approximation of the function relating the input to the output. Why?
Because they are not related. In a governed system, output (speed, dirtiness) is not a function of input (fuel or dirt added).
That is why the question of whether the earth has a governor should be the very first question we ask. I say yes here, but YMMV …
Finally, once again, hopefully for the last time to put this to bed, do I think we can model a governed system like the climate or the Tasmanian Dust Devils? I wrote my first computer program in 1963, and I have extensive experience with many types of computer models, including writing some. I say yes, of course in theory we can model the climate.
But to model a governed system like the Tasmanian Dirt Devils’ room or the global climate, you need to explicitly model the governor, and I know of no current climate that does that.
Steve says:
December 17, 2010 at 10:01 am
“Willis explained that their is no direct causal relationship between average amount of dirt now, average dirt level added and future dirt levels (the “global kitchen climate”). There is no viable algorithm between these variables. The viable algorithm is between the current distribution of dirt, the distribution of dirt added and the future distribution of dirt. That was the entire point of his article – this climate can’t be (usefully) modeled without empirical data on the details.”
No, he did not “explain” it. He claimed it. And he’s wrong. In general there is a (noisy) causal relationship between those variables. I have actually performed a simulation that proves it. I will provide details later.
“Well, in your “experience as a scientist and systems engineer”, I suppose you didn’t learn that there actually is a standard scientific definition of predictive power. And your definition isn’t it. http://en.wikipedia.org/wiki/Predictive_power”
My definition of predictive power is in fact quite standard. Your reliance upon wiki to contradict me smacks of desperation, and is in any case futile, since it is here entirely in agreement with my explanation. From wiki:”The predictive power of a scientific theory refers to its ability to generate testable predictions.” As I stated. Any prediction at better than chance is statistically testable. “Theories with strong predictive power are highly valued …”, because obtaining statistical significance is easier; but note that this entails (as I pointed out) that predictive power is a matter of degree. It can be weak or strong. “The predictive power of a theory is closely related to applications”. Again, as I stated. A theory or model could have strong predictive power for global climate while having only weak predictive power for local weather, or vice versa.
“If I can guess the average dirt level of the floor for a given period of time and be correct 1 time in 1,000 , and a model comes along and correctly predicts the average dirt level correctly 5 times in 1,000, that model is 5 times more powerful than guessing. But it still would not be considered a powerful model, because 95% of the time it’s wrong!”
You still don’t seem to have grasped the difference between “predictive power” and a “powerful model”. You also don’t seem to understand the nature of predictions under noisy conditions. The strength of the prediction is measured by the accuracy of the estimates – the magnitude of the typical errors – not on whether any given estimated value will be “right” or “wrong”, because with continuous (or finely quantised) variables they will never be “right” (there will always be some error) even for very well understood and highly robust theories like ballistics.
Willis Eschenbach says:
December 17, 2010 at 12:39 pm
“Here’s the thing. Suppose we have positive feedback, such that any change in a variable ends up getting doubled in the output. We can see this and measure it. Then, as Paul suggests, we can use this to estimate the underlying algorithm. This method works great for a system with feedback.”
I did not claim that “we can use it to estimate the underlying algorithm”. We might or might not. What I have said – repeatedly – is that we do not need to know the underlying algorithm to get a useful empirical model. All we need is empirical data.
“But now let us suppose that we are in a “governed” system … We use that information to construct an algorithm for what’s happening regarding fuel and speed.”
That’s not a control algorithm. It’s an observational relationship connecting these two (or more) variables.
“But when we repeat our experiment in a governed system like a car with cruise control, we see something surprising. Fuel use goes up and fuel use goes down, but the average speed doesn’t change at all! (Note that this also happens with the Tasmanian Dirt Devils, where dirt input goes up and down, but the average dirtiness of the floor stays constant.)”
This is one of your unproven claims, which turns out to be false. Even fully regulated systems do not in general maintain a constant average output for different inputs. The behaviour of such control systems is much more subtle and surprising than you imagine. Your Tasmanian Dirt Devils will not hold the average dirtiness of the floor constant under different dirt inputs. Try it yourself – as I already have.
Even if your tropical thunderstorms did constitute a temperature governor (they don’t, they constitute a temperature difference governor) this would not prevent the global average temperature from varying over potentially quite a wide range.
Here are the results of a numerical experiment on the properties of systems utilising “dirt devils”.
Take an 8×8 grid (chessboard). Let units of dirt (draughts pieces) be placed randomly on the board, and choose a simple detection threshold of dirt (pieces) present in any two or more contiguous or superposed squares. Then remove the dirt from those squares.
For a deposition rate of one unit of dirt per move, the average total dirt = 5.8+/-0.2 units (filling factor 9.0%).
For a deposition rate of two units of dirt per move, the average total dirt = 3.7+/-0.4 units (filling factor 5.8%).
This is a gradient, between those values, of -2.1 moves. This is a predictive empirical model (approximately linear within this range) with differences of the same order as the move to move fluctuations, and thus quite a strong predictor even over surprisingly short timescales of say ten moves or less. Over longer periods of course it should be considerably more robust.
At lower and higher rates, the correlation asymptotes to about 10% at the bottom end and 5% at the top, with increasingly strong fluctuations as the rate increases. In other words, there’s about a factor of two in output as a function of changes in the input rate.
This analysis is based on a scenario in which the devils work (infinitely) fast. If we modify it to let each devil remove only a single unit of dirt per move, this increases the average level accordingly; the greater the deposition rate, the greater this increase.
So, for 1 unit/move, the average total dirt becomes 6.3+/-0.2 (10%),
for 2 units/move, the average total dirt becomes 5.1+/-0.4 (8.0%),
and for 3 units/move, the average total dirt becomes 7.4+/-0.6 (11.6%).
Note how the curve has turned up again at the high end. It does not asymptote (except at ~100%).
This assumes an unlimited number of devils (one per simultaneous detection). If, more realistically, the number of devils is limited, then the curve will rise even more steeply, becoming vertical when the deposition rate equals the total number of devils (who will then be unable to keep up even when working flat out).
A pseudo-random number generator on my calculator was used to generate the coordinates, to give a random distribution of dirt. If instead the dirt is more clustered, this has the effect of moving the curves upwards and to the left (as if from a higher deposition rate). Similarly, if the dirt is anti-clustered, the curves move down and to the right (as if from a lower deposition rate).
Note that if one selects a particular algorithm to minimise the dependence of the output on the input in a given range, one can, with only modest changes, also produce one that will have either a positive or negative dependence (decreasing the speed or reaction time of the devils increases the gradient, increasing them decreases it).
This is all in line with the sort of behaviour I expected; as indeed I stated earlier in the thread, only for what has turned out to be my sound scientific intuition to be pooh-poohed.
Here is the raw data for rates 1 and 2:
Omit starting transients (5 moves):
Delta=+1 – (0 0 0 0 0) 2 0 2 0 2 0 0 0 2 2 0 0 2 0 2 2 0 2 0 0 0 2 0 0 0 2 2 2 3 2 0 2 0 0 0 2 2 0 2 2 0 2 0 2 0 0 2 0.
Delta=+2 – (0 2 0 2 2) 0 3 2 0 2,2 0 2,2 4 2 2 3 2 2 2 3 0 0 0 3 2 2 2,3 0 2,2 0 2 2 2 3 3 3 2 2 2,2 0 0 2,2 2 0 2 0 2 2 0 2,3 2 2 2 0 2,2 2 2.
Paul Birch says:
December 18, 2010 at 7:40 am (Edit)
I take a car with cruise control. I drive sixty miles up Pike’s Peak. I note the mileage, I see I’ve burned three gallons. My average speed is 30 miles an hour.
I take the same car and drive sixty miles on the flat. I note the mileage. I see I have burned two gallons. My average speed is 30 miles an hour.
That is analogous to the climate situation. We don’t have detailed moment by moment observations. We have observations at the start and at the end of a period. From that we need to see if we can develop some kind of algorithm (which I seem to use in a much general sense than you do) that relates the fuel used to the speed.
OOPS. Can’t do that. Since 30 mpg and 20 mpg both give us the same speed, the situation is non-invertible. So we can’t address it directly, we have to use other information to try to figure out the situation.
Now, as you point out, in a real system there are small signals from the transient response to the control forcing. And you keep saying, over and over, that using those small signals we can construct a rough model of the system.
And you are right about that, you can do amazing things with very small measurements … but here in climate science we have nowhere near the number and accuracy of measurements that we would need to try to tease out such a small relationship. If you truly think it can be done, perhaps you could tell us the frequency of observations and the amount of detail that we would need to do it … my point is that while you are right in theory, in practice at present it’s not possible. To make it worse, computer models have a “black box” called a gridcell. Within that box, there is absolutely NO DETAIL. The gridbox may be 100 km on a side, and within it, we don’t know where anything is. We don’t know if the heat is in the north or south of the gridbox. We don’t know if the wind is stronger in the east or west.
So yes, as you point out, if we had fine detailed measurements of all the relevant variables we might be able to tease out the tiny signal that you truthfully say is there.
But in the real world, we have nothing like that. We have few measurements, scattered in time, with plenty of gaps and missing data. The data to date, for example, is not sufficient to answer questions like how long carbon resides in the atmosphere. Answering questions like yours? Doubtful … and we can’t use the current generation of models to figure it out.
In addition, you have missed the main point. This is that if there is a feedback system governing the planet, we will neither find it nor understand it using the current climate models.
Finally, regarding your comment:
It is that kind of comment, the sly dig to try to establish just how dumb we all are and how much smarter you are, that makes it so unpleasant to read what you post.
My friend, you don’t have a clue what other people know or what they imagine. You’ve never met me. You don’t know me. You don’t know what experience I have had. You don’t have a clue what “subtle and surprising” thing I can imagine.
The truth is that it is other people that are “much more subtle and surprising than you imagine”. You’d get more traction if you kept that in mind.
Willis Eschenbach says:
December 18, 2010 at 3:22 pm
“I take a car with cruise control. I drive sixty miles up Pike’s Peak. I note the mileage, I see I’ve burned three gallons. My average speed is 30 miles an hour. I take the same car and drive sixty miles on the flat. I note the mileage. I see I have burned two gallons. My average speed is 30 miles an hour.”
This is yet another unwarranted assertion. In general, you will not get the same average speed in the two situations. You would not do so unless the governor (cruise control) were able to react infinitely rapidly, infinitely strongly and with infinitesimal hysteresis, which it won’t. Typical regulated systems are likely to show quite significant – and predictable – changes in average output as a function of input conditions – as much as a factor of two, or more. In my previous comment, I proved that experimentally for systems utilising your dirt devils.
If the dirt devils were indeed an appropriate analogy for your tropical thunderstorms, then we might expect global temperatures to range over a similar factor as the dirt level. Or perhaps, making the analogy more directly with energy fluxes, by only the fourth root of it – which would still mean that average global temperatures could vary over a 55K range! This is not a negligibly small effect.
“Now, as you point out, in a real system there are small signals from the transient response to the control forcing. And you keep saying, over and over, that using those small signals we can construct a rough model of the system.”
No, this is not what I’ve said. First, the inputs are not in general “forcings”. A forcing is an oscillating signal at a frequency other than the resonant frequency of the system forcing the system to oscillate at that frequency. Turning on a tap or blocking a drain is not a forcing. This is another of those terms, like anomaly, that the AGW “climate scientists” misuse. Second, I wasn’t talking about transients, but about the long-term averages. Third, there is no reason to suppose that these signals will be “small” and, as I have proved for the dirt devils scenario, often they are not.
“To make it worse, computer models have a “black box” called a gridcell. Within that box, there is absolutely NO DETAIL. ”
If all we want is the average global temperature, we simply don’t need detail. Any more than we need to know the location and variety of every plant in every field in the country to predict the harvest. The details don’t much help and don’t much matter.
“So yes, as you point out, if we had fine detailed measurements of all the relevant variables we might be able to tease out the tiny signal that you truthfully say is there.”
I said no such thing! What I said was almost the diametric opposite: that we do not need detailed measurements to get a useful model, and that the dependence of even regulated (or “governed”) systems on input conditions is often considerable (far from tiny, and sufficient for good empirical models).
“In addition, you have missed the main point. This is that if there is a feedback system governing the planet, we will neither find it nor understand it using the current climate models.”
I have not missed this point. I’m saying that your claim is false. There are many feedbacks present and they can all – in principle – be included in the models. None of them holds the average global temperature constant. Overall, they simply reduce (or in some cases increase) the amount it varies. There is no need to assume that any of those feedbacks comprises a global temperature governor; and tropical thunderstorms certainly do not (because the regulated variable is a temperature difference, not the absolute temperature; and because their range is not global but is itself variable); nor does the tropics to polar “heat engine” (for similar reasons).
PB: The behaviour of such control systems is much more subtle and surprising than you imagine.
“It is that kind of comment, the sly dig …”
It wasn’t a “sly dig”, it was a directly pertinent remark (perhaps with a touch of frustration). You yourself had demonstrated in your post and comments that you were unable to imagine how a governed control system could show a strong dependence of average output on average input, or how general observations could lead to useful empirical models of the overall system behaviour even without specific knowledge of the control algorithm or underlying physics. You still seem unable to imagine this, even after I have proved the point by experiment.
Paul Birch;
That the speeds are not identical (to how many decimal places?) is actually irrelevant. Up front we design the feedback system to react to variance of a given amount, just as a thermostat allows temperature to drop a degree or two below its setting before firing up the furnace, and raises it a degree or two above the setting before shutting it off.
As for your overwhelmed Devils, you could attempt to drive a car up too steep a grade for it to sustain 30 mph, but that is also irrelevant.
Brian H says:
December 21, 2010 at 5:36 pm
“That the speeds are not identical (to how many decimal places?) is actually irrelevant. Up front we design the feedback system to react to variance of a given amount, just as a thermostat allows temperature to drop a degree or two below its setting before firing up the furnace, and raises it a degree or two above the setting before shutting it off.
As for your overwhelmed Devils, you could attempt to drive a car up too steep a grade for it to sustain 30 mph, but that is also irrelevant.”
These features are not in the least “irrelevant”. Not only do they directly contradict Willis’s false and simplistic claims, but as common control system behaviours they are also likely to be found in the global climate systems that he wishes to explain by analogy with his dirt devils and cruise control. Even if there exist global temperature governors (which I rather doubt), as distinct from merely ameliorative negative feedback, it does not follow that their hysteresis range must be small; it could easily be ~50K or more. Nor does not follow that they cannot be close to overload; indeed, systems such as tropical thunderstorms are quite likely to be near overload near (some of) the margins of their geographical extent.
Note, by the way, that the cruise control analogy is considerably poorer than the dirt devil one, because there is only a single, accurately known (measured) speed of the car, under a known and fairly restricted range of conditions, being regulated by a single governor (control system). It would not be particularly hard to make the hysteresis range quite small (for other than improbably extreme stresses – like running into a brick wall!). By contrast, dirt devils, and tropical thunderstorms, are regulating dirt levels and temperature differences in multiple locations simultaneously. One should not expect these to have the same system dynamics as a single control loop.