The climate sensitivity and the surface temperature record question – answers from major players

mkelly says: (April 27, 2011 at 8:09 am)
“The weather channel gives a “feels like” temperature when it is humid. So even the weather channel knows, but won’t admit that water is far more important than CO2.”
Perhaps there should also be a “should be” temperature given, one that tells us what the temperature would be if CO2 was back at the 280 ppm level. It should be easy enough to do, after all, the believers claim to know how much heating we have added since that era.

“”””” JohnB says:
April 27, 2011 at 2:27 am
I have to agree with DRs Schmidt and Trenberth. “Climate Sensitivity” is defined as the equilibrium temperature change for a doubling of CO2. It’s not like the term is just now being defined, it was defined in the past and as such is what it is. “””””
Well John B, that’s the problem; it was “defined” to be that, and that is exactly what I take those two words “Climate Sensitivity” to mean; nothing more and nothing less.
BUT !!! implicit in that quite arbitrary definition is the very definitive statement that : T-T0 = (CS) log2(CO2/CO2/0) where my resource constrained terminology is self evident.
That means that T(560) – T(280) is identically equal to T(2) – T(1) where the numbers in brackets are PPM atmospheric molecular abundance of CO2.
Now if I choose to DEFINE a newly discovered Martian rock mineral sample to be “Amalgamese Tobunganate”, then from now and henceforth, that is exactly what that mineral shall be called.
BUT !!! I do NOT have license to simply define laws of Physics to be just whatever I want them to be; and expect to get general acceptance of my practice. The establishment of “Laws of Physics” requires experimental observations of the pertinent phenomena, and verification that the observed values of variables, is in agreement with those values predicted from application of those laws.
So humans have been “reliably” measuring the atmospheric molecular abundance of CO2 at Mauna Loa, now since about 1957/58; basically the Geo-Physical year, and those pretty much universally accepted results, have see the CO2 value go from 315 PPM in 1957/58 to a present value of 390 PPM; a ratio of 1.238 whose log base two is 0.308.
So we have observed 0.3 of one doubling in CO2 so the increase in Global Mean Temperature is 0.36 deg C for the blue line, and 0.9 deg C for the red line; based on the Schmidt/Trenberth insisted Definition of “Climate Sensitivity” and its IPCC mandated 3:1 fudge factor ratio of uncertainty (=/- 50%).
Well since 1957/58, the Global Mean Surface Temperature has gone up and down all over the place; so for anyone to define “Climate Sensitivity” based on that observed behavior of the real actual operational Physical Laws; the ones that Mother Gaia adheres to; is simply farcical.
I’m perfectly happy to defend the Schmidt/Trenberth claim to that definition of “Climate Sensitivity”, since it is exactly the one I use myself; but Unlike them; it is for me just a definition of words; I acknowledge no proof of any validity to the implied Physical Law behind that definition.

It seems clear that, Mr. Schmidt notwithstanding the problem is semantic, and it is a serious semantic problem in that having infused a particular meaning into the term, alarmists such as him then use that metric to draw conclusions not entailed by it in any way.
He is quite right that one does not measure impact of “climate change” on surface ecosystems and human civilization by metrics involving upper-atmosphere or oceanic changes — you measure this impact by what directly impacts it: surface temperature. He, and apparently some others such as Mr. Trenberth, have chosen to take this as the definition of climate sensitivity. Ok, I’m fine with this. As a mathematician I am quite comfortable with the idea that terminology is arbitrary; you can use whatever labels you like.
The problem comes when you attempt to use this particular metric, as they appear wont to do, as a predictive variable. Surface temperature is an effect, not a cause. The number cannot be fed back into the system or climate models as if it has some important feedback effects. And here is where the semantics of this particular choice of terminology becomes severely problematic.
The term “climate sensitivity” suggests that the variable in question is an important determiner of how the overall climate responds to inputs. It is a serious mistake to infer such a relationship, if surface temperature is the only, or primary, datum involved in the metric. Yes, it is a datum in which we have some practical interest. But, no, it is not a number with any particularly important predictive value; nor does it represent in any direct way any important element of the evolution of our climate system. Ocean heat content, in contrast, does both.
Perhaps I’m conflating terms, but I’m afraid none of your responders’ answers jived very closely with my understanding of sensitivity. I have always understood “sensitivity” in this context to refer to the ratio of actual, observed response to derived, theoretical response.
Thus, if (for example) radiative physics predicts (on whatever simplistic assumptions are invoked) that the change in atmospheric CO2 content over the last 100 years should lead to 0.2 C in warming, and 0.6 C is observed, then one infers a sensitivity of 0.6/0.2 = 3, which is specific to the CO2/temperature relationship, and has no broader meaning than this. I have understood this to be a deliberately naive, back-of-the-envelope calculation that involves far-too-simplistic assumptions such as that the atmosphere is a brick with no internal dynamics and that ALL temperature change can be attributed to CO2 content variation. The factor, 3 (in my example) represents some effects unknown in the sense that they have not been accounted for in the theoretical model used. Perhaps they ARE known, however — one might attribute the amplification by a factor of 3 to H2O responding to CO2, for example.
But none of this is important in understanding the long-term evolution of the climate. I do think these guys are talking past each other.

Speaking both sarcastically and seriously, a more useful metric for climate change is DISTANCE per doubling of CO2. Most of the people being asked to reduce GHG emissions live northern temperate zone and many of these people are somewhat familiar with how the climate changes as one moves north and south. For example, mean annual temperature in North Dakota and Montana is about 41 degF, while about 1000 miles to the south the mean annual temperature is about 60 degF in Oklahoma and Arkansas. Using this scale, a 1 degC temperature change is equivalent to moving 100 miles or 150 kilometers to the south, a trend which is reasonably accurate for most of the US (except Oregon and Washington). So the IPCC is telling us that climate sensitivity is 150-450 miles or 230-690 kilometers south for 2X CO2. (We can adapt by moving that the same distance to the north. Despite warming over the past 30 years, more people are moving south than north in the US.)
Given that global warming is anticipated to be greatest at night and in the winter – when high temperature is usually not usually a problem – the effective climate sensitivity is probably about half of this distance.
For the environmentally-aware residents of coastal California – who live where the ocean moderates summer temperature so that the daily high rises about 1 degF for every for every mile one moves away from the ocean – climate sensitivity is about 3-8 miles for 2X CO2. In the atmosphere or mountains, where temperature tends to drop about 6.5 degC/vertical kilometer, 2X CO2 is equivalent to 230-700 vertical meters or 750-2300 ft (which would be significant for some mountain snowpacks).

Sky says;
“For basic physical reasons that few in climate science seem to understand properly, the entire question of “climate sensitivity” to CO2 changes is largely an empty one. Because they produce no energy, GHG’s cannot “force” anything in any strict physical sense. What we should be asking is what the effect of different concentrations may be upon the rate of thermal energy transfer from atmosphere to space, when the principal mode of transfer from surface to atmosphere is moist convection. In other words, is the climate system’s Lyapunov exponent being affected? Globally averaged surface temperatures do not address that crucial question.”
Thank you, as I have said a few times the important factor at play here is the “speed of heat” through the climate system. Since GHG’s CANNOT produce
energy the only remaining question is if they can slow the flow of heat through the system enough to affect the “equilibrium temperature”. I posit that since increases in GHG’s displace the amount of NON-GHG’s in the system the “speed of heat” through the system becomes more weigthed towards the speed of light (IR radiation vs convection) against the speed of heat which is quite a bit slower.
The Earth (including the gases in the atmosphere) are most acurately called a “Blackbody RE-RADIATOR” which is only capable of changing the rate at which energy flows from a TRUE Blackbody Radiator (i.e. the SUN) through the Earth/Atmosphere System to the cold vacuum of space.
Cheers, Kevin.

Okay, it seems clear to me that if you want to pin the tail on human caused CO2 then you certainly want to use GAAST as the primary measurement.
What if you want to measure the forcing from another first order human influence such as land use? Would you still want to use GAAST, or would you want to use some other measurement?
My point is that there’s quite a bit of gamesmanship going on here. Depending on what agenda you are trying to push, the ideal frame of conversation will change. If you are trying to guide the conversation towards man made CO2, then you don’t want to talk about the ocean because it’s too slow and visa versa.

To George E. Smith;
You wrote;
“That would seem to imply that if you know one you know the other.”
With respect I would like to suggest that we could rephrase this as;
“If we know something about one item we also know something about another item”
Do we know EVERTHING about the other item? Perhaps, but possibilty not! For example electrical current is certaintly related to electrical voltage, but as the the frequency increases many interesting things occur (i.e. look up “skin effect”). So if we know some things about an effect do we indeed know everything about that effect ?
So, in summary, temperature IS related to energy, but there are many other factors involved including the the thermal capacities of the materials involved.
Cheers, Kevin.

“”””” KevinK says:
April 28, 2011 at 8:42 pm
To George E. Smith;
You wrote;
“That would seem to imply that if you know one you know the other.”
With respect I would like to suggest that we could rephrase this as;
“If we know something about one item we also know something about another item” “””””
What is the point in rephrasing my words ? They then become NOT my words, so they don’t have my meaning.
But YOU are free to say anything you want to say, and use your own words saying it. It just wouldn’t be my words, or have my meaning.
So on the Absolute Thermodynamic scale of Temperature the mean kinetic energy per degree of freedom of all of the particles present IS the measure of the Temperature.
E = kT = h(nu) =mc^2
What else is there to know about the relationship between energy and Temperature; their numerical values differ only in the factor, that converts units of energy (Joules) into units of Temperature (Kelvins).
Some Physicists even use a system of units in which c = h = k = 1 (actually they use hbar, and measure frequency (nu) in radians per second). In that case E = T = nu = m
Ask anna v about that system.
I actually have a Major in Radio-Physics, so I’m hep to Volts and Amps, and even skin effect. None of those are related to either Temperature or energy.

Philip,
It’s a habit, quoting trenberth’s numbers from KT97, even though they tend to be screwed up quite often. Older albedo values tend to be pushing 0.40 while more recent ones claim around 0.30, give or take a bit. Over 80% of the total albedo is going to be atmospheric, clouds, scattering, etc. and only a little is surface contribution.
However, albedo is definitely a variable.

In terms of Albedo, the number which is used to control satellites (and they do need to take into account all radiation levels in the satellite’s environment including reflected solar to maintain proper orbits) is 29.83%.
This happens to be the exact same number used by Trenberth in his latest Earth Energy Budget paper.
Clouds make up about 16 to 17 percentage points of this number and the Surface makes up about 13 to 14 percentage points.
The Earth is about 65% cloud covered on average (including haze from very thin sirrus clouds which we might not count as cloudy) but on average 75% of the sunlight (including UV) either gets through the clouds or is still reflected to the Earth.
Generally, the surface which is most important to Albedo is Ice and Snow and Sea Ice which is the farthest from the Earth average and can vary the most over time.
The Earth’s Albedo in time looks like it can vary from about 24.5% (Pangea and no Ice on the planet) to about 50.0% (in Snowball Earth conditions). If you want to crunch the numbers on these estimates, you can get pretty close to the historical temperature estimates with just Albedo alone.
Then put 33.3% in for the Last Glacial Maximum – Hansen and climate theory have used 31.0% which is clearly too low in my opinion – this value is tuned so that CO2/GHGs produce 3.0C per doubling. The map of the extent of glaciers and snow and sea ice at the Last Glacial Maximum show it was much higher.

Bill,
the numbers I see are somewhat along the lines of KT97, 10-30 w/m^2 surface albedo out of over 100 W/m^2. Clouds and atmosphere are the vast majority of albedo. the numbers you present are twice to three times the surface albedo of these values. There is not that much cryosphere and it is way off to the edge where there is not that much solar power. Your 17% attribution to clouds (and must include atmospheric scattering too) leaves cloud reflectivity at under 30%. What type of cloud reflects (scatters) under 35% at the very least? Your 14% surface contribution, considering only 35% of the surface is exposed (out from under the clouds), results in an average surface albedo of 40%, way higher than Mars or the Moon, and this is with 70% of the surface covered in water, which runs less than 0.04 albedo. That means the 30% land coverage must have an albedo of 0.40 / 0.3 = 1.3. Show me anything with an albedo above 0.8. Essentially, the conditions you present would be for at least a full glaciation period or even snowball Earth conditions for clouds to have that little effect and for the surface to have that great an effect. In fact, you’d have to have a substantial glaciation covering a lot of the Earth’s oceans to achieve this, roughly 50% of the surface assuming fresh snow at maximum snow albedo.