editorial

18
Feb
editorial

I try not to pen editorials. OK here goes. I respect the attention given to this blog, as there are plenty of other great blogs on climate change, politics, finance, etc to read. I try to stay an ‘on message’ advocate for numeracy. Everyone has something to offer from their experiences though. Right at this moment, there is something to say that is important about numeracy, but takes a bit to explain.

I would encourage y’all to read the discussion on New paper on mathematical analysis of GHG in relation to VS, not because I believe in it, or because I believe in balance of probabilities it is right, but because I believe it is the way scientific progress is made. Its what I have tried to do here. Check the numbers.

This is not another ‘me too’ paper inventing there own ‘novel’ approach to affirming the cause du jour in the name of ‘research’. Its about contesting methodology of other experts in the field. Boring? No. It’s what it’s all about. Jargon? No. Its no more complex than ‘regression’. Just unfamiliar. And I am not taking a dig at anyone, as I respect everyone who posts here. Peer reviewed? Uncertain.

When Steve McIntyre started his blog 5 years ago, and I did around the same time, I sent him a email saying to the effect that he would change the way science is done. He called the FOI’s and journal processes of peer review, comments etc ‘quasi-litigation’. I agree, and acknowledge that scientists should use the available processes more. It is a natural extension of the search for truth.

The main failing of the IPCC, IMHO, is in ignoring peer-reviewed papers and comments in favor of confirmatory dreck. What to do about that? The only answer I know is by contesting the logic, models, mathematics, and results using data. Don’t just check your assumptions, check your calculations. Check the stated results are justified.

What point is there in talking about philosophy of science, when the by all accounts, most science is wrong? The overwhelming reason is because:

Moreover, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias.

Climbing into the stadium and dueling over the technical details is the only way, despite the personal cost.

Tagged as:
Published by david stockwell in: All Theory
If you like this blog please take a second from your precious time and subscribe to my rss feed!

http://www.moyhu.blogspot.com Nick Stokes

David,My problem is in finding the numbers. Not so much with GHG, which has some analysis unclearly explained. But with temperature.It's just stated to be I(1). No analysis that I could see. More seriously, no error levels. No possibility that there could be curvature. It's just I(1), and that assumption is built into everything.But it isn't. We've all pored over temp plots. David finds break points. The 40's high point is talked about. Some argue that there's no significant trend at all. But hardly anyone says that it's linear.Now I(1) doesn't exactly mean linear, but it means the differences are stationary. They're not. They are positive for a while, then negative, then positive.That's my problem, the paper is unscientific. It rattles off tests, but there's no real error analysis. Solar is spoken of, but as Leif says at WUWT, the data there is very uncertain. They just don't seem to recognise that.
davids99us

I think that temperature is not a random walk or else it would longago have gone to + or – infinity. It's probably fractionallydifferenced at about 0.9, but the analysis only applies to exactly1.0. I also think there is real 20 and 60 year periodicity related tooceanography. I would like to know what difference such uncertaintyreally makes.
http://www.moyhu.blogspot.com Nick Stokes

David,
My problem is in finding the numbers. Not so much with GHG, which has some analysis unclearly explained. But with temperature.

It’s just stated to be I(1). No analysis that I could see. More seriously, no error levels. No possibility that there could be curvature. It’s just I(1), and that assumption is built into everything.

But it isn’t. We’ve all pored over temp plots. David finds break points. The 40′s high point is talked about. Some argue that there’s no significant trend at all. But hardly anyone says that it’s linear.

Now I(1) doesn’t exactly mean linear, but it means the differences are stationary. They’re not. They are positive for a while, then negative, then positive.

That’s my problem, the paper is unscientific. It rattles off tests, but there’s no real error analysis. Solar is spoken of, but as Leif says at WUWT, the data there is very uncertain. They just don’t seem to recognise that.

Anonymous

I think that temperature is not a random walk or else it would long
ago have gone to + or – infinity. It’s probably fractionally
differenced at about 0.9, but the analysis only applies to exactly
1.0. I also think there is real 20 and 60 year periodicity related to
oceanography. I would like to know what difference such uncertainty
really makes.

sherro

In 1971 I owned a laboratory and was fascinated to see over 100 of the hand-picked analytical chemists in the world report their chemical analysis of the first Moon rocks, summarised in “Evaluation of lunar elemental analyses”George Harold MorrisonAnal. Chem., 1971, 43 (7), pp 22Aâ€“31aPublication Date: June 1971Unfortunately this is behind a paywall. The salient outcome was between the pre-advertised accuracy of separate laboratories, compared with the horrible differences between them as shown by the results. It opened my eyes to the possibility that scientists were quite capable of claiming performance beyond their capability.It made me a sceptic. It became a quest to find more examples of poor science in many disciplines. I am comfortable with the qualitative proposition that much science is wrong.Thalidomide was at first a disaster. However, new drugs are not bimodal, either brilliant or disasterous. There is a spectrum of them, with poor science contributing to what is often shrugged off as unforeseen side effects. There is some bad stats work in pharmacology development.But, science has many sub-disciplines. Some are more prone to being wrong than otheres. Error-prone fiels include those with numersoud perturbing factors that cannot be adequately offest (like dendrochroology with its special pleading to use strip bark bristlecones) and those where the scientist has no accountability for error. Some people see my blog comments as harsh and judgemental. We are products of our environment and in my environment, if you did not do good science and succeed, you went hungry. It was easy to estimate success. If you found no economic ore deposits, you had not succeeded.In a way, science needs to keep thinking of ways it can set Key Performance Indicators, to remove those who fail them and to reward those who excel. There is some of the KPI approach, but we need more. It helps with integrity.
Anonymous

In 1971 I owned a laboratory and was fascinated to see over 100 of the hand-picked analytical chemists in the world report their chemical analysis of the first Moon rocks, summarised in
“Evaluation of lunar elemental analyses”
George Harold Morrison
Anal. Chem., 1971, 43 (7), pp 22A–31a
Publication Date: June 1971

Unfortunately this is behind a paywall.

The salient outcome was between the pre-advertised accuracy of separate laboratories, compared with the horrible differences between them as shown by the results. It opened my eyes to the possibility that scientists were quite capable of claiming performance beyond their capability.

It made me a sceptic. It became a quest to find more examples of poor science in many disciplines. I am comfortable with the qualitative proposition that much science is wrong.

Thalidomide was at first a disaster. However, new drugs are not bimodal, either brilliant or disasterous. There is a spectrum of them, with poor science contributing to what is often shrugged off as unforeseen side effects. There is some bad stats work in pharmacology development.

But, science has many sub-disciplines. Some are more prone to being wrong than otheres. Error-prone fiels include those with numersoud perturbing factors that cannot be adequately offest (like dendrochroology with its special pleading to use strip bark bristlecones) and those where the scientist has no accountability for error.

Some people see my blog comments as harsh and judgemental. We are products of our environment and in my environment, if you did not do good science and succeed, you went hungry. It was easy to estimate success. If you found no economic ore deposits, you had not succeeded.

In a way, science needs to keep thinking of ways it can set Key Performance Indicators, to remove those who fail them and to reward those who excel. There is some of the KPI approach, but we need more. It helps with integrity.
hmccard

Dave,I'm unable to access the paper that you cite because I can't access Anthony's WUWT website. Moreover, I can't access Steve's CA or JefId's Air vent website either. Is anyone else having this problem or understand “what's up'?
Anonymous

Dave,

I’m unable to access the paper that you cite because I can’t access Anthony’s WUWT website. Moreover, I can’t access Steve’s CA or JefId’s Air vent website either. Is anyone else having this problem or understand “what’s up’?

Anonymous

hmccard,

all those sites are available. If you still can not access them check with your ISP or network Admin.
Anonymous

erased dupe

kuhnkat

hmccard,all those sites are available. If you still can not access them check with your ISP or network Admin.
kuhnkat

hmccard,all those sites are available. If you still can not access them check with your ISP or network Admin.
cohenite

Nick, you say: “Now I(1) doesn't exactly mean linear, but it means the differences are stationary. They're not. They are positive for a while, then negative, then positive.” Doesn't that apply to CO2 and temperature only more so? BTW VS did answer your query:VS (02:29:36) : Nick Stokes (19:32:35) on temperature (not) being I(1). Take a look at the discussion above, specifically VS (11:10:03). Also, BR2009 state that they confirm previous findings that temperature and solar irradiation are I(1), and they state that again in Table 2. The test statistics are missing though, but I suspect thatâ€™s because they didnâ€™t think this is a point of dispute (again, see discussion above .And a big : ) from me as well.
cohenite

Nearly forgot Nick; the B&R calculations for temp 1(1) appear to be here:http://www.patrickminford.net/Business_Topics/B…
cohenite

Nick, you say: “Now I(1) doesn’t exactly mean linear, but it means the differences are stationary. They’re not. They are positive for a while, then negative, then positive.” Doesn’t that apply to CO2 and temperature only more so? BTW VS did answer your query:

VS (02:29:36) :

Nick Stokes (19:32:35) on temperature (not) being I(1).

Take a look at the discussion above, specifically VS (11:10:03).

Also, BR2009 state that they confirm previous findings that temperature and solar irradiation are I(1), and they state that again in Table 2. The test statistics are missing though, but I suspect that’s because they didn’t think this is a point of dispute (again, see discussion above .

And a big : ) from me as well.

http://www.moyhu.blogspot.com Nick Stokes

Doesn’t that apply to CO2 and temperature only more so?
It’s temperature that I’m talking about. CO2 has a more regular curvature which could be seen as polynomial (quadratic, I(2) )
BTW VS did answer your query
No, what you’ve quoted is his first answer, which I then queried (more substantively) in the part I linked. I think there are contradictions in what he said, which are partly resolved by analysis that he did himself, but that’s missing in B&R.

cohenite

Nearly forgot Nick; the B&R calculations for temp 1(1) appear to be here:

http://www.patrickminford.net/Business_Topics/BusinessTopicsCOctober2009.pdf

http://www.moyhu.blogspot.com Nick Stokes

Coho, no, it’s the same as the paper – just assertion, and they don’t even say it’s their own result. But the absence of test statistics is vital. Even if you had a test that said temp is I(1), CO2 I(2), therefore no AGW, that is meaningless without the test statistics. Because all these statements are made on finite data and have uncertainty, so you could only conclude the final result at a certain probability level.

I am now convinced the paper is total crap. Firstly, there’s the lapse of elementary logic that I mentioned in an above post. These tests do not establish that anything is I(1) or I(2). It only says that those hypotheses can’t be rejected. So their statement should say, it is quite possible that temp is I(1) and CO2 is I(2). If that is true, then temp can’t be determined by CO2 (alone and unlagged).

But their hasty rejection of AGW is totally out of order. Firstly, I doubt their basic contention that I(1)&I(2) does imply a contradiction. But it certainly doesn’t rule out more complex dependences, as for example T depending on delayed CO2, and maybe other factors.

Anonymous

Nick,

“It only says that those hypotheses can’t be rejected.”

Same argument y’all used to claim the Hot Spot is really there!! Whoopsie!!

“But it certainly doesn’t rule out more complex dependences, as for example T depending on delayed CO2, and maybe other factors.”

Instead of waving your arms on this, how about some specifics about your claim of T possibly depending on delayed CO2 (got up late? stopped by Highway Patrol on the way to upper trop??) and these “other factors.”

http://www.moyhu.blogspot.com Nick Stokes

The regular notion is that it depends on the integrated (accumulated) CO2. CO2 changes heat flux, and a change in flux causes gradual, ongoing warming. The warming itself causes more IR loss, so it tapers exponentially.

But anyway it’s the job of a “proof” like this to at least consider alternative possibilities.

Anonymous

Nick,

I took the forcing numbers ( http://data.giss.nasa.gov/modelforce/RadF.txt ) and the temperature data ( http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts+dSST.txt ) we used for the Tamino two box model and did some modeling in R. The best ARIMA fit for the annual temperature anomalies was (1,0,0) with a coefficient of 0.895. The (0,1,0) model was only slightly worse in terms of estimated sigmasquared. The ghg forcings only, column 2 in the table, are indeed reasonably well fit by an (0,2,0) model or a (1,1,0) model with a coefficient of 0.97. But, I really don’t think you can pick just one forcing and look at it in isolation and draw valid conclusions. Specifically, a linear combination with independent coefficients for the forcings from CO2, methane and nitrous oxide is not physical in any sense of the word. Looking at the sum of all forcings, columns 2-11 in the table, the best ARIMA model is (1,0,0) with a coefficient of 0.81. As we have seen, you can get a fairly good fit to the temperature data using a two box model with different time constants and the forcing sum. Is that correlation spurious? I don’t think so. I do have some problems with the details and magnitudes of some of the aerosol forcings, but that’s not the topic here.

http://www.moyhu.blogspot.com Nick Stokes

Coho, no, it's the same as the paper – just assertion, and they don't even say it's their own result. But the absence of test statistics is vital. Even if you had a test that said temp is I(1), CO2 I(2), therefore no AGW, that is meaningless without the test statistics. Because all these statements are made on finite data and have uncertainty, so you could only conclude the final result at a certain probability level.I am now convinced the paper is total crap. Firstly, there's the lapse of elementary logic that I mentioned in an above post. These tests do not establish that anything is I(1) or I(2). It only says that those hypotheses can't be rejected. So their statement should say, it is quite possible that temp is I(1) and CO2 is I(2). If that is true, then temp can't be determined by CO2 (alone and unlagged). But their hasty rejection of AGW is totally out of order. Firstly, I doubt their basic contention that I(1)&I(2) does imply a contradiction. But it certainly doesn't rule out more complex dependences, as for example T depending on delayed CO2, and maybe other factors.
kuhnkat

Nick,”It only says that those hypotheses can't be rejected.”Same argument y'all used to claim the Hot Spot is really there!! Whoopsie!!”But it certainly doesn't rule out more complex dependences, as for example T depending on delayed CO2, and maybe other factors.”Instead of waving your arms on this, how about some specifics about your claim of T possibly depending on delayed CO2 (got up late? stopped by Highway Patrol on the way to upper trop??) and these “other factors.”
http://www.moyhu.blogspot.com Nick Stokes

The regular notion is that it depends on the integrated (accumulated) CO2. CO2 changes heat flux, and a change in flux causes gradual, ongoing warming. The warming itself causes more IR loss, so it tapers exponentially.But anyway it's the job of a “proof” like this to at least consider alternative possibilities.
cohenite

“CO2 changes heat flux, and a change in flux causes gradual, ongoing warming. The warming itself causes more IR loss, so it tapers exponentially.” Now Nick, that is interesting because in the aftermath of the Mockton/Lambert debate and the dispute over the correct interpretation of Pinker et al and climate sensitivity, what Pinker did find tends to contradict what you say; Pinker looks at TOA flux anomalies for the 1983-2001 period; she finds a decline rate of 0.17W/m2 for the 20S-20N TOA compared with SW surface increase rate of 0.18W/m2 [fig 4]. That is, for the period 1983-2001, over the tropics, the EEB increased by almost the same rate as the increase in SW. Surely this works against any conclusion that the greenhouse effect through increased CO2 has caused a change in flux and a temperature increase for this period.
cohenite

“CO2 changes heat flux, and a change in flux causes gradual, ongoing warming. The warming itself causes more IR loss, so it tapers exponentially.”

Now Nick, that is interesting because in the aftermath of the Mockton/Lambert debate and the dispute over the correct interpretation of Pinker et al and climate sensitivity, what Pinker did find tends to contradict what you say; Pinker looks at TOA flux anomalies for the 1983-2001 period; she finds a decline rate of 0.17W/m2 for the 20S-20N TOA compared with SW surface increase rate of 0.18W/m2 [fig 4]. That is, for the period 1983-2001, over the tropics, the EEB increased by almost the same rate as the increase in SW. Surely this works against any conclusion that the greenhouse effect through increased CO2 has caused a change in flux and a temperature increase for this period.

http://www.moyhu.blogspot.com Nick Stokes

Coho, I wasn’t there, and haven’t really followed the Pinker issue. However, on my brief reading, Fig 4 shows a drop in (reflected?) SW in the tropics at TOA. I’m not sure what that has to do with CO2. It does seem to indicate warming (less heat lost), by a small amount.

http://www.moyhu.blogspot.com Nick Stokes

Doesn't that apply to CO2 and temperature only more so?It's temperature that I'm talking about. CO2 has a more regular curvature which could be seen as polynomial (quadratic, I(2) ) BTW VS did answer your query No, what you've quoted is his first answer, which I then queried (more substantively) in the part I linked. I think there are contradictions in what he said, which are partly resolved by analysis that he did himself, but that's missing in B&R.
http://www.moyhu.blogspot.com Nick Stokes

Coho, I wasn't there, and haven't really followed the Pinker issue. However, on my brief reading, Fig 4 shows a drop in (reflected?) SW in the tropics at TOA. I'm not sure what that has to do with CO2. It does seem to indicate warming (less heat lost), by a small amount.
cohenite

Nick, I don't have the full Pinker paper to link to only the abstract;http://www.sciencemag.org/cgi/content/abstract/…The context for my point about Pinker is that AR4, Ch.9, 9.2.2.2 (pp.676-7), says about Pinker's findings â€œthe downward [sic] trend in outgoing solar radiation is consistent with the long term upward trend in surface radiation found [sic] by Pinker et al. (2005)”. But the IPCC referral to Pinker is a bad misstatement; this is a crucial point; if there is a net negative TOA flux and therefore the energy budget of the Earth is increasing, then that is the predicted result of the AGW argument; but Pinker has found a contemporaneous increase in S, surface received solar radiation , possibly due to cloud reduction, that in the tropics at least, is almost equivalent to the TOA negative flux [trend increase in surface S is 0.18Wm2/year, the decrease trend for TOA flux is 0.17W/m2 per year]. The point is that S increase of 0.18W/m2 has passed through the TOA and if cloud levels were greater would have been reflected back out as radiative flux leading to a +ve or neutral TOA flux; the reduced cloud has instead seen that S pass to the surface where absortion would have delayed the remittance through TOA; as I argued the equivalence between the increase in S and the reduction in TOA flux leaves bugger all room for an increased greenhouse effect [ie a trend of -0.01W/m2 per year]. Unless you can argue that increased CO2 is causing less cloud [which then becomes a +ve feedback as per the quaint Clement et al paper] CO2 has no measureable impact on the Earth's energy budget.
cohenite

Nick, I don’t have the full Pinker paper to link to only the abstract;

http://www.sciencemag.org/cgi/content/abstract/sci;308/5723/850

The context for my point about Pinker is that AR4, Ch.9, 9.2.2.2 (pp.676-7), says about Pinker’s findings “the downward [sic] trend in outgoing solar radiation is consistent with the long term upward trend in surface radiation found [sic] by Pinker et al. (2005)”. But the IPCC referral to Pinker is a bad misstatement; this is a crucial point; if there is a net negative TOA flux and therefore the energy budget of the Earth is increasing, then that is the predicted result of the AGW argument; but Pinker has found a contemporaneous increase in S, surface received solar radiation , possibly due to cloud reduction, that in the tropics at least, is almost equivalent to the TOA negative flux [trend increase in surface S is 0.18Wm2/year, the decrease trend for TOA flux is 0.17W/m2 per year]. The point is that S increase of 0.18W/m2 has passed through the TOA and if cloud levels were greater would have been reflected back out as radiative flux leading to a +ve or neutral TOA flux; the reduced cloud has instead seen that S pass to the surface where absortion would have delayed the remittance through TOA; as I argued the equivalence between the increase in S and the reduction in TOA flux leaves bugger all room for an increased greenhouse effect [ie a trend of -0.01W/m2 per year]. Unless you can argue that increased CO2 is causing less cloud [which then becomes a +ve feedback as per the quaint Clement et al paper] CO2 has no measureable impact on the Earth’s energy budget.

http://www.moyhu.blogspot.com Nick Stokes

Coho, if you register with Science, which is free, you get access to a lot of their papers, including this one.

I found it heavy going. But as I understand it, it just says that the TOA SW outflux decreased and the surface flux increased by corresponding amounts. That doesn’t say anything about IR, so it doesn’t constrain the total budget in any way. It does imply a small amount of warming – about 1/10 of the IPCC AGW figure.

Of course, the fact that those fluxes balance is what you’d expect. What isn’t reflected gets through.
http://www.ecoengineers.com/ Steve Short

Anthony

“Unless you can argue that increased CO2 is causing less cloud [which then becomes a +ve feedback as per the quaint Clement et al paper] CO2 has no measureable impact on the Earth’s energy budget.”

Something we do know about CO2 is that it has been causing increased plant growth on the land. This can only lead to increased Evapotranspiration (ET) and increased emission of CCN, all other things, especially annual rainfall, being equal. This can only lead to increased low cloud over the land.

Similarly, in the seas an increased partial pressure of CO2 must increase cyanobacterial primary productivity – especially on the continental shelves where, due to an increasing human population, the flux of nitrogeneous nutrient is increasing. The argument about reduced equilibrium CO2 solubility with increased SST is a presently irrelevant second order effect due to the alkalinity of water and the rate of absorption of CO2(aq) and HCO3- by cyanaobacteria (in the absence of nutrient limitation). Increased cyanobacterial primary productivity increases marine CCN production – especially DMS which converts to sulfates. Incidentally it also increases sea surface albedo. You will recall I showed, simply from the body of NOAA data that atmospheric CO2 levels over the Great Southern Ocean are slowly lagging increasingly behind the global average CO2 level i.e the mixing rate between the NH and SH is slowing and/or oceanic uptake of CO2 is increasing in the SH. You won’t find this in any literature perhaps perhaps because the AGW people are frightened of its awesome implications.

Therefore IMHO increasing CO2 can only increase the generation rate of low cloud and decrease its optical density (increase its reflectivity).This can only lead to increased low cloud over the sea – especially in the continental shelf zones.

The mean global cloud cover for the period mid-1983 to mid-2008 is 66.4±1.5%. It was at maximum of about 68.9% in about 1986 – 1988 and declined to about 64.4% in about 1999 – 2001. This period was characterized by global warming. Since about 2001 global cloud cover has recovered to be close to the 27-year average. This was a period of negligible additional warming. The 2008/10 data suggests average global cloud cover has continuing to rise above the 27-year 66.4% average.

Over the last decade from 1999-2001 to 2008-2009 mean global cloud optical thickness increased above the 27-year average of 3.9±0.3% to an annual average of about 4.4%. Over the same period the global mean cloud top pressure fell from the long term 27-year mean of 573±15 millibars to around 553 millibars.

http://isccp.giss.nasa.gov/climanal1.html

http://www.ecoengineers.com/ Steve Short

BTW Anthony, I forgot to mention that my small Excel spreadsheet empirical climate model, which I regard as essentially a teaching tool so that I can keep a clear picture of what clouds etc., covers a range of cloud covers from about 46% to 86% (Bond albedos from 0.25 to 0.35). My simple model, see here:

http://jump.fm/GOWMK

correctly scales all-sky SW cloud radiative forcing (CRF) with cloud fraction at the mean rate of approx. -0.855 W/m^2/% and LW CRF with cloud fraction at a mean rate of +0.575 W/m^2/%.

These are very realistic cloud forcing rates, e.g.:

Dong, Xi and Minnis. (2006) A climatology of midlatitide continental clouds from the ARM SGP Central Facility . Part II: Cloud Fraction and Surface Radiative Forcing. Jounranl of Climate. Vol. 19(9), 1765-1783 give values of -0.984 W/m^2/% and +0.616 W/m^2/% for the SW and LW cloud forcings respectively.

May I recommend Googling the phrase ‘cloud radiative forcing’ as there have been a lot of such papers published in recent years.

http://www.moyhu.blogspot.com Nick Stokes

Coho, if you register with Science, which is free, you get access to a lot of their papers, including this one.I found it heavy going. But as I understand it, it just says that the TOA SW outflux decreased and the surface flux increased by corresponding amounts. That doesn't say anything about IR, so it doesn't constrain the total budget in any way. It does imply a small amount of warming – about 1/10 of the IPCC AGW figure. Of course, the fact that those fluxes balance is what you'd expect. What isn't reflected gets through.
dewittpayne

Nick,I took the forcing numbers ( http://data.giss.nasa.gov/modelforce/RadF.txt ) and the temperature data ( http://data.giss.nasa.gov/gistemp/tabledata/GLB… ) we used for the Tamino two box model and did some modeling in R. The best ARIMA fit for the annual temperature anomalies was (1,0,0) with a coefficient of 0.895. The (0,1,0) model was only slightly worse in terms of estimated sigmasquared. The ghg forcings only, column 2 in the table, are indeed reasonably well fit by an (0,2,0) model or a (1,1,0) model with a coefficient of 0.97. But, I really don't think you can pick just one forcing and look at it in isolation and draw valid conclusions. Specifically, a linear combination with independent coefficients for the forcings from CO2, methane and nitrous oxide is not physical in any sense of the word. Looking at the sum of all forcings, columns 2-11 in the table, the best ARIMA model is (1,0,0) with a coefficient of 0.81. As we have seen, you can get a fairly good fit to the temperature data using a two box model with different time constants and the forcing sum. Is that correlation spurious? I don't think so. I do have some problems with the details and magnitudes of some of the aerosol forcings, but that's not the topic here.
cohenite

Nick, my point is that extra surface SW, which Pinker terms 'S' has a trend over the 1983-2001 period of 0.18W/m2; now you're numerate man and I'm only a 20 digit statistician but even I can see that a trend of +0.18W/m2 is more than sufficient to explain all the surface heating of the last 20 years; that being the case what warming has been caused by CO2 which, according to Lindzen and Choi, has been leaving at a rapid rate at the TOA. Speaking of L&C here is another look at the Harries papers on OLR which seems to bring them into the L&C fold.
cohenite

Sorry, that Harries link is here:http://www.americanthinker.com/2010/02/the_agw_…
cohenite

Nick, my point is that extra surface SW, which Pinker terms ‘S’ has a trend over the 1983-2001 period of 0.18W/m2; now you’re numerate man and I’m only a 20 digit statistician but even I can see that a trend of +0.18W/m2 is more than sufficient to explain all the surface heating of the last 20 years; that being the case what warming has been caused by CO2 which, according to Lindzen and Choi, has been leaving at a rapid rate at the TOA. Speaking of L&C here is another look at the Harries papers on OLR which seems to bring them into the L&C fold.

http://www.moyhu.blogspot.com Nick Stokes

OK, I’ll admit I’m still not on top of this. Dr Pinker did say of the 0.18 vs “CO2 warming” that “These two numbers cannot be compared at their face value.” And yes, one would like to know more.

cohenite

Sorry, that Harries link is here:

http://www.americanthinker.com/2010/02/the_agw_smoking_gun.html

Anonymous

I wrote a detailed analysis of those three Harries papers here: http://landshape.org/enm/interpretation-bias/

http://www.moyhu.blogspot.com Nick Stokes

OK, I'll admit I'm still not on top of this. Dr Pinker did say of the 0.18 vs “CO2 warming” that “These two numbers cannot be compared at their face value.” And yes, one would like to know more.
cohenite

Nick, I think Pinker's admonition about the 2 different values was in the context of Monckton's confusion between cloud forcing based just on SW reflection from the top of clouds without also considering the effect of heat trapping of IR from the underneath of clouds; she elaborates on that here:http://scienceblogs.com/deltoid/upload/2010/02/…As I argued at Deltoid Monckton appears to be wrong about cloud forcing on climate sensitivity in respect of it just being based on the extra 'S' reaching the surface as noted by Pinker's paper; however, he does appear to be correct on climate sensitivity for the reasons I've outlined based on Pinker's findings about 'S' and TOA OSR; in effect Monckton has used the tradesman's entrance to make his point that clouds are determing the climate not CO2.
cohenite

Nick, I think Pinker’s admonition about the 2 different values was in the context of Monckton’s confusion between cloud forcing based just on SW reflection from the top of clouds without also considering the effect of heat trapping of IR from the underneath of clouds; she elaborates on that here:

http://scienceblogs.com/deltoid/upload/2010/02/debate_australia_tim_lambert.pdf

As I argued at Deltoid Monckton appears to be wrong about cloud forcing on climate sensitivity in respect of it just being based on the extra ‘S’ reaching the surface as noted by Pinker’s paper; however, he does appear to be correct on climate sensitivity for the reasons I’ve outlined based on Pinker’s findings about ‘S’ and TOA OSR; in effect Monckton has used the tradesman’s entrance to make his point that clouds are determing the climate not CO2.
davids99us

I wrote a detailed analysis of those three Harries papers here: http://landshape.org/enm/interpretation-bias/
cohenite

Thanks David, I hadn't seen that; it's amazing how often Harries unexpurgated gets trotted out.
cohenite

Thanks David, I hadn’t seen that; it’s amazing how often Harries unexpurgated gets trotted out.
http://www.ecoengineers.com/ Steve Short

Anthony”Unless you can argue that increased CO2 is causing less cloud [which then becomes a +ve feedback as per the quaint Clement et al paper] CO2 has no measureable impact on the Earth's energy budget.”Something we do know about CO2 is that it has been causing increased plant growth on the land. This can only lead to increased Evapotranspiration (ET) and increased emission of CCN, all other things, especially annual rainfall, being equal. This can only lead to increased low cloud over the land.Similarly, in the seas an increased partial pressure of CO2 must increase cyanobacterial primary productivity – especially on the continental shelves where, due to an increasing human population, the flux of nitrogeneous nutrient is increasing. The argument about reduced equilibrium CO2 solubility with increased SST is a presently irrelevant second order effect due to the alkalinity of water and the rate of absorption of CO2(aq) and HCO3- by cyanaobacteria (in the absence of nutrient limitation). Increased cyanobacterial primary productivity increases marine CCN production – especially DMS which converts to sulfates. Incidentally it also increases sea surface albedo. You will recall I showed, simply from the body of NOAA data that atmospheric CO2 levels over the Great Southern Ocean are slowly lagging increasingly behind the global average CO2 level i.e the mixing rate between the NH and SH is slowing and/or oceanic uptake of CO2 is increasing in the SH. You won't find this in any literature perhaps perhaps because the AGW people are frightened of its awesome implications.Therefore IMHO increasing CO2 can only increase the generation rate of low cloud and decrease its optical density (increase its reflectivity).This can only lead to increased low cloud over the sea – especially in the continental shelf zones.The mean global cloud cover for the period mid-1983 to mid-2008 is 66.4Â±1.5%. It was at maximum of about 68.9% in about 1986 – 1988 and declined to about 64.4% in about 1999 – 2001. This period was characterized by global warming. Since about 2001 global cloud cover has recovered to be close to the 27-year average. This was a period of negligible additional warming. The 2008/10 data suggests average global cloud cover has continuing to rise above the 27-year 66.4% average.Over the last decade from 1999-2001 to 2008-2009 mean global cloud optical thickness increased above the 27-year average of 3.9Â±0.3% to an annual average of about 4.4%. Over the same period the global mean cloud top pressure fell from the long term 27-year mean of 573Â±15 millibars to around 553 millibars.http://isccp.giss.nasa.gov/climanal1.html
http://www.ecoengineers.com/ Steve Short

BTW Anthony, I forgot to mention that my small Excel spreadsheet empirical climate model, which I regard as essentially a teaching tool so that I can keep a clear picture of what clouds etc., covers a range of cloud covers from about 46% to 86% (Bond albedos from 0.25 to 0.35). My simple model, see here:http://jump.fm/GOWMK correctly scales all-sky SW cloud radiative forcing (CRF) with cloud fraction at the mean rate of approx. -0.855 W/m^2/% and LW CRF with cloud fraction at a mean rate of +0.575 W/m^2/%. These are very realistic cloud forcing rates, e.g.:Dong, Xi and Minnis. (2006) A climatology of midlatitide continental clouds from the ARM SGP Central Facility . Part II: Cloud Fraction and Surface Radiative Forcing. Jounranl of Climate. Vol. 19(9), 1765-1783 give values of -0.984 W/m^2/% and +0.616 W/m^2/% for the SW and LW cloud forcings respectively. May I recommend Googling the phrase 'cloud radiative forcing' as there have been a lot of such papers published in recent years.
cohenite

Thanks Steve, as usual, very informative. The extra CO2 does therefore have, in terms Christopher game disapproves of, a homeostatic effect in its contribution to lower cloud concentration with lower cloud, it seems to me, having either a negative feedback to temperature movement or a negative forcing, depending on your perspective.. What are your thoughts on the Pinker paper which found a decrease in low cloud from 1983-2001; that seems to dovetail with your comments.As I've said before you should publish your findings.
http://www.ecoengineers.com/ Steve Short

Hi AnthonyYes, it is hard to explain why there was a decrease in low cloud over 1983 – 2001 and an increase from the 1999 – 2001 low to now. There is no doubt that this data:http://isccp.giss.nasa.gov/climanal1.htmland Roy Spencer's excellent 9 January presentation:Clouds Dominate CO2 as a Climate Driver Since 2000http://www.drroyspencer.com/point towards a very interesting story for the 1983 – 2010 period which relates to the issue of (in particular) low cloud formation and strikes to the heart of the AGW theory.Could we be seeing the beginning of the 'revenge of the plants' as plant life on the land and cyanobacteria in the oceans ramp up their growth rates? If this were due to a unique combination of the higher pCO2 and our also unprecedented discharge of available nitrogen into the coastal shelves, then we ain't seen nothing yet.I don't yet have a copy of the Monckton talk in which I presume this Pinker paper was referenced (fig 4?). Can you point me to a copy of this paper, please?
cohenite

Thanks Steve, as usual, very informative. The extra CO2 does therefore have, in terms Christopher game disapproves of, a homeostatic effect in its contribution to lower cloud concentration with lower cloud, it seems to me, having either a negative feedback to temperature movement or a negative forcing, depending on your perspective.. What are your thoughts on the Pinker paper which found a decrease in low cloud from 1983-2001; that seems to dovetail with your comments.

As I’ve said before you should publish your findings.

http://www.ecoengineers.com/ Steve Short

Hi Anthony

Yes, it is hard to explain why there was a decrease in low cloud over 1983 – 2001 and an increase from the 1999 – 2001 low to now.

There is no doubt that this data:

http://isccp.giss.nasa.gov/climanal1.html

and Roy Spencer’s excellent 9 January presentation:

Clouds Dominate CO2 as a Climate Driver Since 2000

http://www.drroyspencer.com/

point towards a very interesting story for the 1983 – 2010 period which relates to the issue of (in particular) low cloud formation and strikes to the heart of the AGW theory.

Could we be seeing the beginning of the ‘revenge of the plants’ as plant life on the land and cyanobacteria in the oceans ramp up their growth rates?

If this were due to a unique combination of the higher pCO2 and our also unprecedented discharge of available nitrogen into the coastal shelves, then we ain’t seen nothing yet.

I don’t yet have a copy of the Monckton talk in which I presume this Pinker paper was referenced (fig 4?). Can you point me to a copy of this paper, please?

cohenite

Steve, most of the information about the Monckton/Lambert debate are here, including Lambert's slides and Moncktons plus Pinker's comments.http://scienceblogs.com/deltoid/2010/02/monckto…
cohenite

Steve, most of the information about the Monckton/Lambert debate are here, including Lambert’s slides and Moncktons plus Pinker’s comments.

http://scienceblogs.com/deltoid/2010/02/moncktons_mcluhan_moment.php
http://www.ecoengineers.com/ Steve Short

BTW Anthony, I forgot to mention that my small Excel spreadsheet empirical climate model, which I regard as essentially a teaching tool so that I can keep a clear picture of what clouds etc., covers a range of cloud covers from about 46% to 86% (Bond albedos from 0.25 to 0.35). My simple model, see here:http://jump.fm/GOWMK correctly scales all-sky SW cloud radiative forcing (CRF) with cloud fraction at the mean rate of approx. -0.855 W/m^2/% and LW CRF with cloud fraction at a mean rate of +0.575 W/m^2/%. These are very realistic cloud forcing rates, e.g.:Dong, Xi and Minnis. (2006) A climatology of midlatitide continental clouds from the ARM SGP Central Facility . Part II: Cloud Fraction and Surface Radiative Forcing. Jounranl of Climate. Vol. 19(9), 1765-1783 give values of -0.984 W/m^2/% and +0.616 W/m^2/% for the SW and LW cloud forcings respectively. May I recommend Googling the phrase 'cloud radiative forcing' as there have been a lot of such papers published in recent years.
cohenite

Thanks Steve, as usual, very informative. The extra CO2 does therefore have, in terms Christopher game disapproves of, a homeostatic effect in its contribution to lower cloud concentration with lower cloud, it seems to me, having either a negative feedback to temperature movement or a negative forcing, depending on your perspective.. What are your thoughts on the Pinker paper which found a decrease in low cloud from 1983-2001; that seems to dovetail with your comments.As I've said before you should publish your findings.
http://www.ecoengineers.com/ Steve Short

Hi AnthonyYes, it is hard to explain why there was a decrease in low cloud over 1983 – 2001 and an increase from the 1999 – 2001 low to now. There is no doubt that this data:http://isccp.giss.nasa.gov/climanal1.htmland Roy Spencer's excellent 9 January presentation:Clouds Dominate CO2 as a Climate Driver Since 2000http://www.drroyspencer.com/point towards a very interesting story for the 1983 – 2010 period which relates to the issue of (in particular) low cloud formation and strikes to the heart of the AGW theory.Could we be seeing the beginning of the 'revenge of the plants' as plant life on the land and cyanobacteria in the oceans ramp up their growth rates? If this were due to a unique combination of the higher pCO2 and our also unprecedented discharge of available nitrogen into the coastal shelves, then we ain't seen nothing yet.I don't yet have a copy of the Monckton talk in which I presume this Pinker paper was referenced (fig 4?). Can you point me to a copy of this paper, please?
cohenite

Steve, most of the information about the Monckton/Lambert debate are here, including Lambert's slides and Moncktons plus Pinker's comments.http://scienceblogs.com/deltoid/2010/02/monckto…

Niche Modeling

editorial

Categories

Blogroll

Advertising

Comments

Archives

tag cloud

Most commented

Copyright © 2008 - 2013 Niche Modeling