36 replies to this topic

[Iceberg]

VillagePlank, on 17 February 2010 - 18:19 , said:

The missing data is a point to note about the paper, itself - I just typed it into Excel.

Arrhenius₁ made some observations himself that seem, on the face of it, reasonable - but, for me, not necessarily obvious.

Firstly, the nebulosity (I had to look it up - it means cloudiness) of the southern hemisphere, presumably on account of the southern oceans, means the effect will be less than the Northern hemisphere (important implications in measuring global temperature: should we, or do we, account for this?)

Secondly, an increase in C0₂ will reduce diurnal range of temperatures. He fleets over this, and I don't understand why, and I can't find any source that explains it, either.

Thirdly, that the difference between the equator and the poles are different because "those places ... alter their albedo by the extension or regression of the snow-covering" So temperature increases at the poles are larger with the same C0₂ differential₂.

Add that lot together, and, it seems to me, at least one of the constants, probably alpha, needs to be some function of albedo. Not a great surprise, but I didn't expect to see it.

(Sorry if this is excess to requirements - this is the sort of thing that, ahem, gets me going, one might say. The hunt for the alpha function is on!)

1: On the Influence of Carbonic Acid in the Air upon the Temperature of the Ground, Svante Arrhenius, Philosophical Magazine and Journal of Science, Series 5, Volume 41, April 1896, pages 237-276. page 265
2: As above, page 257

Going back to this post (sorry a bit of jumping around), as I rarely get time to look in on this thread.

I think it's important to note that Svante Arrhenius's work is old and predates alot of what is known nowadays about global temperature re-distribution, so although I have a lot of respect for him we have moved on lot, with atmospheric cell placement (i.e hemisphere triple cell, Jet Stream, Ocean conveyors etc).

albedo was pretty much the only mechanism known back then, not so now.

WRT to climate sensitivity (Lambda?), we are not talking about a simple formula IMO, which is why climate modelling is so important everything from the percentage of GHG to the age of the GHG (atmospheric lifetimes reduce GHG effectiveness on a range of scales depending on the molecule.), through to the height of the GHG in the atmosphere through to distribution within the atmosphere, CFC's are a good example here.

you can then role in various factors such as albedo, WV etc.

[Captain_Bobski]

Iceberg, on 19 February 2010 - 12:34 , said:

I think it's important to note that Svante Arrhenius's work is old and predates alot of what is known nowadays about global temperature re-distribution, so although I have a lot of respect for him we have moved on lot, with atmospheric cell placement (i.e hemisphere triple cell, Jet Stream, Ocean conveyors etc).

Hectic, hectic, hectic at the moment, hence my absence, but time for a quick post!

Arrenhius's work is old, granted, but the whole of AGW science stems from his original paper - to the extent that his Greenhouse Gas law is still used to this day (this one: ΔF = α ln(C/C₀) ).

The validity, or otherwise, of Arrenhius's paper is one of the fundamental building blocks of the theory of AGW. Although we have come a long way since that paper was written it is still a crucial part of AGW theory. If it is wrong in any way then there may be knock-on effects further down the line.



CB

[Boar Wrinklestorm]

Captain_Bobski, on 19 February 2010 - 14:04 , said:

[/i]The validity, or otherwise, of Arrenhius's paper is one of the fundamental building blocks of the theory of AGW. Although we have come a long way since that paper was written it is still a crucial part of AGW theory. If it is wrong in any way then there may be knock-on effects further down the line.

Hi CB, good to see you back.

Indeed, such that this paper is relevant is that the IPCC say that the relationship is logarithmic. Indeed, it is, but why? I can find no 'audit' trail back of it's heritage - and, therefore, I am working on the notion that is so obvious that an idiot like me must have missed it.

I am doing the background reading at the moment; and some of it seems extremely tricky. As far as I can ascertain the family heritage is something like this:

Maxwell's equations (without the Lorentz abstraction) ->
Beer's law (made me laugh, too) ->
The Arrhenius approximation.

More later next week (am stuck on vector field divergence, and curling at the moment)

Iceberg, on 19 February 2010 - 12:34 , said:

albedo was pretty much the only mechanism known back then, not so now.

That may well be so, but the Arrhenius approximation still describes what we are observing to some magnitude or another. If you know of an analysis of a paper that describes Arrhenius' flaws, I am pretty sure that everyone who takes the time to read this thread would be delighted if you'd share the link.

After, we don't ditch Pythagoras simply because it's old - or do we?

Edited by VillagePlank, 19 February 2010 - 14:27 .

[Captain_Bobski]

VillagePlank, on 19 February 2010 - 14:24 , said:

Hi CB, good to see you back.

Indeed, such that this paper is relevant is that the IPCC say that the relationship is logarithmic. Indeed, it is, but why? I can find no 'audit' trail back of it's heritage - and, therefore, I am working on the notion that is so obvious that an idiot like me must have missed it.

I am doing the background reading at the moment; and some of it seems extremely tricky. As far as I can ascertain the family heritage is something like this:

Maxwell's equations (without the Lorentz abstraction) ->
Beer's law (made me laugh, too) ->
The Arrhenius approximation.

More later next week (am stuck on vector field divergence, and curling at the moment)



That may well be so, but the Arrhenius approximation still describes what we are observing to some magnitude or another. If you know of an analysis of a paper that describes Arrhenius' flaws, I am pretty sure that everyone who takes the time to read this thread would be delighted if you'd share the link.

After, we don't ditch Pythagoras simply because it's old - or do we?

Hi VP - nice to be back - my head's a bit crowded at the moment so I've not been able to really focus on in-depth technical discussions, but it's starting to clear slightly...!

With regards to Arrhenius's flaws, I've dug up this old PDF which refers to Angstrom's rebuttal of the original Arrhenius paper:

http://docs.lib.noaa...29-06-0268a.pdf

In particular is this comment: "The remainder of Angstrom’s paper is devoted to a destructive criticism of the theories put forth by the Swedish chemist, S. Arrhenius, in which the total absorption of CO[2], ie quite inadmissibly inferred from data which include the combined absorption of CO[2], and the vapor of water."

Subsequent to this, Arrhenius's supporters discredited Angstrom's rebuttal (though whether this was valid or not I don't know) by claiming that his experiments were not a legitimate approximation of the actual atmosphere, and that there were errors in the taken measurements.

I'm not sure how those objections invalidate Angstrom's assertion that Arrhenius had not appropriately distinguished between CO2 effects and water vapour effects, so I shall have to do more reading up on the subject - perhaps if I could find Angstrom's actual paper it might help!

CB

[Captain_Bobski]

Realclimate give their view on the debate here:

http://www.realclima...gassy-argument/

I've got to go and check on something before I say more...

[Iceberg]

Don't get me wrong I am not attacking Arrhenius (SA's) work.

Just like like all simple things such as Newtons gravity theory, it's not ditched, just that it's refined and improved upon, which is what's happened to SA's work.

A very good example is Methane and CFC's, CFC's tend to go to the poles, Methane tends to stay in concentration near source(depending on source location).

Any equation describing a simple log relationship, is just that simple.

A couple of papers that help to build on climate sensitivity.

http://www.agu.org/p...5JD006713.shtml

http://www.springerl...6q0m0m13p26042/

Feel free to post up the none paying versions if they are found.

I have to admit it's not something I've spent alot of time looking at, the assumed forcing of a doubling of CO2 = 1.5C or so of GW I am happy to accept as are most others, even skeptics generally, it tends to be the more contentious WV and feedbacks that effect climate sensitivity.

[Captain_Bobski]

I haven't read it all yet (!) but here is a link to Frank Very's 1900 work "Atmospheric Radiation: A Research" in PDF format (be warned, it's a scanned-in book and the download is 24MB!!).

http://www.archive.o...radia00veryrich

[Boar Wrinklestorm]

Iceberg, on 19 February 2010 - 15:31 , said:

Any equation describing a simple log relationship, is just that simple.

I'm afraid I do not agree with that. The reeaon is why does it hold?

Can you please post peer-reviewed papers that show the physical reason with reference to the CO₂/temp relationship?

EDIT: My objection is based on the notion that it is not well known why there is a logarithmic relationship, and it is extremely difficult (it seems to me) to track it down. Nothing personal.

Edited by VillagePlank, 19 February 2010 - 16:49 .

[jethro]

This isn't peer reviewed but it does a fair job of explaining the basics of why the relationship is logarithmic, it also has some references at the bottom.

http://brneurosci.org/co2.html

[sunny starry skies]

jethro, on 19 February 2010 - 17:28 , said:

This isn't peer reviewed but it does a fair job of explaining the basics of why the relationship is logarithmic, it also has some references at the bottom.

http://brneurosci.org/co2.html

I'm a bit concerned that one appears to have been written from a certain perspective (particularly in their assertions of linearity for some concepts), though maybe some of the material may be sound. No time (nor probably the ability) to do a proper critique though, so I'll leave any criticisms to others.

Looking around, I found this one, which seems very intuitively right for me and explained a lot of what I didn't get about CO2 concentrations:
http://chriscolose.w...fect-revisited/
It has a rather good explanation of why the effect of CO2 is logarithmic, in relation to the spectrum of outgoing longwave radiation. It thus explains beautifully why methane is far more 'effective' a greenhouse gas by the same principle. It explains why small quantities have such a strong impact, and how the saturation effect does not prevent CO2 from being a continuing effective GHG at ever-increasing quantities.
sss

[Gray-Wolf]

sunny starry skies, on 25 February 2010 - 16:37 , said:

Looking around, I found this one, which seems very intuitively right for me and explained a lot of what I didn't get about CO2 concentrations:
http://chriscolose.w...fect-revisited/
sss

Good find SSS!

Might take me longer to absorb than CO2 does heat though!!!

[jethro]

A step closer to accurately reconstructing past temperatures?

http://www.nature.co...ontent=Netvibes

[sunny starry skies]

jethro, on 12 March 2010 - 10:41 , said:

A step closer to accurately reconstructing past temperatures?

http://www.nature.co...ontent=Netvibes

It's a good technique isn't it! I saw a presentation about it, and the micromilling of the shells required, a few years ago, but it looks like they've made some good progress. I've just had a read of the PNAS paper, and I have some issues with their interpretation of their own graphs. They speak of a 6C drop in winter temperatures post-settlement of Iceland, which is simply not present in their graphs. The 2-3C drop in summer maximum temperatures (not another 6C drop that they talk of) would be quite enough to cause the famine in Iceland however. It's odd, because the data they present in Figures 3 and 4 of the PNAS paper are superb, but their interpretation is poor, but... the outcome of most of their interpretations of that part of their record is the same, despite their own misrepresentation of their figures! I also think they misquote their own data on summer/winter temperatures for their most recent bivalves (they say summer/winter of 8.2 and 5.5C, which is the range of summer temps according to their own graph, winter temps are 1-3C on their graph). Confused yet?? I am!

So it looks like the method has loads of potential, especially with a more detailed dataset, but I wonder if the interpretations of this specific paper will be corrected/criticised in future? Other interesting observations - muted MWP/LIA signal, despite it being most well developed in the North Atlantic, and it will be still more interesting to make comparisons with ice core and tree ring data (but difficult given the lack of either in Iceland, and the challenges of teleconnecting individual temperature series in the North Atlantic).
sss

[sunny starry skies]

A quick last one from me before I go away for a bit (some will be happy to hear ). I stumbled across this - it's a 7-part blog, and may be an accessible walkthrough of CO2 science from the bottom upward, called "CO2 - an Insignificant Trace Gas?". Quite involved, lots of maths, and so perhaps VP in particular will be interested in it (notably parts 3 and 4 perhaps), and the references within.

Other things of note, and I don't have time to reference them entirely, but two interesting posts at RealClimate, covering recent papers discussing sea level rise (underestimated), solar activity influence (around the 10% level), and the scariest of the lot is:
http://www.sciencema...ct/326/5955/984
Greenland's mass balance looks like it has dropped off a cliff since about 2000, based on GRACE data, and the authors say it would have been 100% worse but for some incresed snowfall. Definitely worrying times. I know some might construe these various papers as 'alarmist', but the data is getting ever more solid, and we are observing the effects all the more each year. I just hope the paper that suggested much higher climate sensitivity was wrong (http://www.sciencema...bstract/1178296)...
sss

[sunny starry skies]

VP asked where on the log curve we were:
If
ΔF = 5.35 * ln(C/Cbase)
ΔT = ΔF * λ.
where ΔF is radiative forcing relative to pre-industrial, C is desired CO2 concentration and Cbase is preindustrial (278ppm), then the forcing can be calculated. This gives us our 3.7W/sq m for a doubling, and a 1.7W/sq m for present concentrations, compared to preindustrial levels.
The equation is from the IPCC AR3 and AR4, derived from Myhre et al (1998). This is directly based on the radiative forcing driven by absorbtion of longwave radiation as described in links I've put up elsewhere (Jethro doesn't like linking to blogs in the technical thread but the relevant blogs are duly referencing the literature on radiative transfer). The relationship is logarithmic because of the gradual saturation of the CO2 absorbtion band.

Temperature change (ΔT) is related to forcing by the variable λ, which is very close to a constant of 0.5.
So to answer VP's question, x = 387ppm, y = 1.77W/sq m. Add another 100ppm and we get to 3W/sq m, demonstrating the logarithmic effect.
sss

[Boar Wrinklestorm]

sunny starry skies, on 10 April 2010 - 04:48 , said:

VP asked where on the log curve we were:
If
ΔF = 5.35 * ln(C/Cbase)
ΔT = ΔF * λ.
where ΔF is radiative forcing relative to pre-industrial, C is desired CO2 concentration and Cbase is preindustrial (278ppm), then the forcing can be calculated. This gives us our 3.7W/sq m for a doubling, and a 1.7W/sq m for present concentrations, compared to preindustrial levels.
The equation is from the IPCC AR3 and AR4, derived from Myhre et al (1998). This is directly based on the radiative forcing driven by absorbtion of longwave radiation as described in links I've put up elsewhere (Jethro doesn't like linking to blogs in the technical thread but the relevant blogs are duly referencing the literature on radiative transfer). The relationship is logarithmic because of the gradual saturation of the CO2 absorbtion band.

Temperature change (ΔT) is related to forcing by the variable λ, which is very close to a constant of 0.5.
So to answer VP's question, x = 387ppm, y = 1.77W/sq m. Add another 100ppm and we get to 3W/sq m, demonstrating the logarithmic effect.
sss

The lambda variable - how was it derived?

(Thanks for the paper reference)

Edited by VillagePlank, 10 April 2010 - 07:20 .

[sunny starry skies]

VillagePlank, on 10 April 2010 - 06:23 , said:

The lambda variable - how was it derived?

(Thanks for the paper reference)

Afraid I won't be at the office for quite a while to check the Myhre paper. Dickinson, 1982; WMO, 1986; Cess et al., 1993 are what the TAR refers to in relation to the second equation, so you could look there.

Dickinson, R.E., 1982: In: Carbon Dioxide Review [Clark, W.C. (ed.)]. Clarendon, New York, NY, USA, pp. 101-133.
WMO, 1986: Atmospheric Ozone: 1985, Global Ozone Research and Monitoring Project, World Meteorological Organization, Report No. 16, Chapter 15, Geneva, Switzerland.
Cess, R.D., M.-H. Zhang, G.L. Potter, H.W. Barker, R.A. Colman, D.A. Dazlich, A.D. Del Genio, M. Esch, J.R. Fraser, V. Galin, W.L. Gates, J.J. Hack, W.J. Ingram, J.T. Kiehl, A.A. Lacis, H. Le Treut, Z.-X. Li, X.Z. Liang, J.-F. Mahfouf, B.J. McAvaney, K.P. Meleshko, J.-J. Morcrette, D.A. Randall, E. Roeckner, J.-F. Royer, A.P. Sokolov, P.V. Sporyshev, K.E. Taylor, W.-C. Wang, and R.T. Wetherald, 1993: Uncertainties in CO₂ radiative forcing in atmospheric general circulation models. Science, 262, 1252-1255.