Is global average temperature a useful or thermodynamically valid concept?

[swansont]

15 hours ago, VenusPrincess said:

Sorry, I read the text on the graphic that referenced the pre-industrial average and assumed that it pertained to the graphic. The point still stands.

That was the graph that clearly shows temperature increasing over the last ~140 years. So no, your point doesn't stand.

The animation shows that looking at only a few locations wouldn't give us an accurate read on what is happening, since there are fluctuations (i.e. weather happens)

Water and land each have some specific heat capacity, so their temperature will rise or fall if the absorb or release energy.  

Q = mc∆T

If I measure in enough places to be representative of the whole, I can sum up the Q for all those areas and figure out if heat was absorbed or emitted overall. The worldwide average ∆T is representative of that value, which (as Area54 pointed out) is easier for non-experts to grasp.

Saying the global average increased by 1ºC is saying we absorbed enough energy for the whole surface to increase by 1º even though some areas saw a larger increase and some saw a smaller increase, or possibly a decrease, because this is not a system in steady-state

 

[VenusPrincess]

8 hours ago, joigus said:

Thermodynamics is the study of equilibrium.

  This is just wrong. There are many branches of thermodynamics and there is no strict assumption of equilibrium unless you choose to study a branch that makes that assumption.

8 hours ago, joigus said:

IOW, there is no chance that a small region of phase space can store big quantities of energy making local averages non-robust, as you are suggesting.

Also wrong. Convection currents in the atmosphere reach altitudes far beyond the placement of our thermometers, yet they also pass through areas near the surface. You cannot make conclusions about the heat content of an atmosphere with turbulent flow and convection currents by merely averaging measurements at the surface. The global average temperature anomaly is not a robust estimate of the atmosphere's heat content.

8 hours ago, studiot said:

You have failed to respond.

I didn't respond since your reply was overly wordy, fairly irrelevant and pedantic. What was your point? That some obscure equation that incorporates the average surface temperature could be useful?

22 hours ago, studiot said:

In the same way the average surface temperature can form part of another dimensionless environment number ( Xi) that indicates whether a planet has an atmosphere at all and, if so, what the gases are likely to be.

Well we know Mars and Earth have atmospheres, and we know which gases they are composed of. So I guess we don't need the obscure equation you mentioned.

7 hours ago, swansont said:

That was the graph that clearly shows temperature increasing over the last ~140 years. So no, your point doesn't stand.

Reading comprehension? My point was about the uselessness of calculating a global average temperature, that it has no thermodynamic validity, and more specifically I mentioned afterwards that local trends may not match the global trend. The point does stand. Not sure what "the graph that clearly shows temperature increasing over the last ~140 years" has to do with my point.

[studiot]

2 hours ago, VenusPrincess said:

I didn't respond since your reply was overly wordy, fairly irrelevant and pedantic. What was your point? That some obscure equation that incorporates the average surface temperature could be useful?

So you thought that the rules of both politeness and this forum don't apply to you ?

2 hours ago, VenusPrincess said:

Well we know Mars and Earth have atmospheres, and we know which gases they are composed of. So I guess we don't need the obscure equation you mentioned.

A response typical of those ignorant of the subject they are espousing.

 

[joigus]

  

3 hours ago, VenusPrincess said:

This is just wrong. There are many branches of thermodynamics and there is no strict assumption of equilibrium unless you choose to study a branch that makes that assumption.

You seem to be overly interested in words and moot points. Saying just "thermodynamics" suggests either classical, chemical or equilibrium thermodynamics; all of them based on equilibrium. There's also statistical mechanics, but that's almost never called thermodynamics. And there's non-equilibrium thermodynamics, but that's such a misnomer (it's not about just heat, temperature, and the like) that everybody referring to it always mentions it by the whole name, "non-equilibrium thermodynamics," only to make clear that it's not thermodynamics (T, Q, etc.)

Here are all as covered by Wikipedia:

https://en.wikipedia.org/wiki/Thermodynamics#Branches_of_thermodynamics

https://en.wikipedia.org/wiki/Non-equilibrium_thermodynamics

You tell me which one you're referring to. There's also kinetics, but I'm not interested in climbing the tower of Babel.

3 hours ago, VenusPrincess said:

Also wrong. Convection currents in the atmosphere reach altitudes far beyond the placement of our thermometers, yet they also pass through areas near the surface. You cannot make conclusions about the heat content of an atmosphere with turbulent flow and convection currents by merely averaging measurements at the surface. The global average temperature anomaly is not a robust estimate of the atmosphere's heat content.

I meant (and I said) regions of phase space. Regions of phase space are not regions of 3-dimensional space. You're confusing both. I mean volumes in the way of,

\[d^{3n}xd^{3n}p\]

IOW, regions in a humongous 3Nx3N-dimensional space. It is in that space of huge dimension where sampling is robust.

Macroscopic systems are ergodic, meaning that time averages give you a very good idea of phase-space averages (averages to all momenta and all positions) when systems are at equilibrium or going round and round in cycles. The molecules you describe as going up to the outer reaches of the atmosphere have to go back and recycle, participating in the overall thermal and dynamical processes, and exchanging the energy. That's the key.

But all this is quite academic and, if pressed, I wouldn't be too sure of anything, the way you seem to be. Here's a much more intuitive explanation of why sampling in this way works even for chaotic systems:

Tim Palmer's lecture at Perimeter Institute Lecture: Climate Change, Chaos, and Inexact Computing

11' 40''-16' 37''

(I copied the link starting at about the time when he explains the point.)

I'm not an expert in climate change. @Area54 or @Ken Fabian can probably give much more accurate information and point out my excesses. I just wish to argue with my toolkit. And with my toolkit at hand, the arguments about "global warming" (however much of just a catchphrase that is) make a lot of sense to me.

[studiot]

8 hours ago, joigus said:

You seem to be overly interested in words and moot points...................but I'm not interested in climbing the tower of Babel.

........................

I meant (and I said) regions of phase space. Regions of phase space are not regions of 3-dimensional space. You're confusing both. I mean volumes in the way of,

+1

[Snape]

@VenusPrincess

The volume of water in the ocean is determined by the water’s mass, but also in part by the water’s temperature.

Obviously, ocean temperature varies by location, which means a weighted average is a better gauge of thermal expansion/volume than the temperature at single location or depth.

****

Tbe ambient temperature of air above an area of ocean has an influence on the temperature of the water below. 

And like the above, atmospheric temperature varies by location, meaning a weighted average is better gauge of the atmosphere’s influence on ocean temperature than the temperature at a singular location.

 

[BillyFisher]

I still haven't found an answer to your question: What's really going on?

Edited January 12, 2021 by BillyFisher

[timo]

2 minutes ago, BillyFisher said:

I still haven't found an answer to your question: What's really going on?

There are lots of answers to that question. For some, the answer is that there is a pedophile ring drinking children blood in the backroom of a pizzeria (*). If you haven't found your answer in this thread you should invest the time to be a bit more specific about the question.

 

(*) Rumor has it that this actually is some people's answer about the question what climate science is about, too.

[Ken Fabian]

On 11/3/2020 at 10:00 AM, Snape said:

Tbe ambient temperature of air above an area of ocean has an influence on the temperature of the water below.

It goes both ways - I think more a case of sea surface temperatures determining air temperature than the other way around.

 

13 hours ago, BillyFisher said:

I still haven't found an answer to your question: What's really going on?

My answer is/was that global average temperature was and is an arbitrary choice for providing a simple, single indicator of change to the climate system; others might do as well or better (I would nominate Ocean Heat Content but the record doesn't go back so far), but weather records are our longest running direct measurements from which change can be observed. So, addressing part 1 of original question, yes it is useful. Is it a thermodynamically valid concept? I think that is probably a pointless question - temperatures are not the same as heat but when they are going up it clearly indicates gain of heat. How closely that temperature reflects overall change to the global energy balance? My understanding is... reasonably well. The extent and nature of local and regional change is a whole lot more complex.

The implied question of whether an average temperature is a valid concept - ie the use of averaging samples from some or many places in place of measuring every place - seems to rest on the proposition that the places not sampled could have temperatures and trends of temperatures very different from those of the places sampled. In conspiracy theory argument, chosen for showing the trend wanted. In reality pretty much every place sampled around the world over a long period shows warming temperatures and that could not be random.  I am not sure I can do this question justice but will say it is not just statistics, but about climate and weather processes - we can observe that nearby locations experience similar conditions, such that e.g. a cold air mass blowing across a region is not observed to cause specific locations to result in one place being cold, but nearby locations are not - other than where other geographic factors come into play like elevation, orientation, vegetation. And where those differences exist it is evident in the weather records of those places; it won't be randomly changing. The local conditions make their own local baseline from which change and trends of change can be observed.

The number of samples matters of course - a single sample, or too few, for the whole planet would not give a valid result, but for these purposes there are (IIUC) more than 30,000 that qualify as long running and reliable. If I understand correctly averaging comes with uncertainty range - that decreases with the number of samples. It would be possible to select from weather station data in order to be deliberately biased - but because those showing cooling trends are so few (are there any?) you would have to exclude most records to NOT get a warming trend. I think basic honesty and a professional attitude is the default; presumption of incompetence or bias built into the supposed "skepticism" opponents of climate  responsibility and action use are vile in my view.

[Snape]

VenusPrincess said that if you averaged the air temperature inside a car with the car’s oil temperature, the result wouldn’t be useful.  Duh.    You obviously want to look at each measurement separately.

Which is why, for example, the GMST and temperatures at the bottom of the ocean are not combined into an average.  

(She needs to come up with a better analogy)