A Correction – And Much More of the Answer
Guest essay by Mike Jonas
I think this post is a big deal. It’s not quite the answer to everything but, if I’ve got it right, it solves a lot of the climate riddle. It also shows that CO2’s contribution to late 20th century global warming was very minor. So here’s a request: please can the best brains on WUWT check it all very carefully – a serious online peer-review. If I’ve stuffed up, I want to know that right away, so please get a critical comment in asap. Most of the relevant data is in the spreadsheet absorptioncalcs_upper (.xlsx see also Appendix B).
1. A Correction from Nick Stokes
What seems like a long time ago, when I saw Nick Stokes’ comment on my How Climate Works – Part 1 post, I thought it looked significant and I should check it carefully. Well, silly me, I didn’t check it until after my last post, which asked where El Niño’s heat came from. In that post, I looked into the ocean at 10-100m depth, and found enough extra energy absorption caused by a 1983-2009 cloud cover decline to match the global warming claimed for CO2. There’s enough energy for that alright, but the trouble is, when I examined it further, there’s too much water down there and not enough extra energy.
What I had been doing in the post was to check whether the extra heat going into the deep ocean was enough to match the amount claimed for CO2. All the analysis in the post was correct for the part of the ocean I was looking at, so what had I missed? [Actually, one formula I was using was wrong; I use the hopefully correct one in this post.].
I went back and checked Nick’s comment, and it supplied the answer. It showed the daily cycle of temperature in the upper ocean:
Figure 1. The diurnal (day-night) cycle in the top few metres of the ocean. From Nick Stokes’ blog Moyhu. NB. The two panels have different scales on the x-axes (that’s not an issue at all, just be careful to see the panels correctly).
What this chart shows is that, on a daily basis, the solar energy absorbed into the top fraction of a millimetre of the ocean then mixes (conducts and convects) into the top 5-10m only, and nearly all of it stays in just the top 1m. But overnight, it is all lost, back into the atmosphere.
I was looking far too far down into the ocean. The answers are much nearer the surface. That’s the correction from Nick.
So the first clue that put me back on track was that the daily mixing was in the top 1m, not the top 10m that I had been allowing for. There’s longer-term mixing too, but that basically spreads the heat across a wider band of ocean. The heat retained from the top 1m will still be there, it will just be diluted.
The second clue, which confirmed that I had been basically on track, was that the surface skin is cooler, day and night, than the subskin. What that tells you is that the net heat flow between the skin and the subskin is one-way – upwards from subskin to skin. So no matter how much the skin may affect the rate at which the subskin warms or cools, it cannot ever give it a higher temperature than its own. If the subskin’s temperature is higher, then the subskin’s other heat sources have to be capable of providing.all of its temperature.
I’ll make one more comment on Nick’s arguments, before I move on: It is unreasonable to extrapolate, as Nick implicitly does, from a single day’s data, arguing that the net zero effect over one day applies to longer periods. The net effect over one day may simply be too small to measure, or disappear in “noise”. To see what happens over longer periods, you need to find a way to work with longer periods.
NB. While you’re reading this article, please don’t think that I’m criticising Nick. Nick’s information and ideas have been invaluable. It’s just that I have a different interpretation, and obviously it’s the differences that get most attention.
2. The CO2 argument
Nick’s argument re CO2 is simple: “On average, the surface loses heat, by evaporation and radiation (and some conduction to air). Incoming IR does not generally need to be absorbed. It simply offsets some of the emission.  In a very technical sense, the sea is heated by sunlight rather than downwelling IR, as is the land. That’s just trivial arithmetic – the sun is the heat source.  But downwelling IR does add joules to the sea just as effectively as SW “.
Here, Nick confirms that the sun is the heat source, but skates over the mechanism saying it’s “just trivial arithmetic“. We’ll do some of the arithmetic in a while, and see if it’s trivial.
First, I want to establish how effective CO2 is.
The IPCC say that a doubling of CO2 increases downward IR by 3.7 Wm-2, and that without feedbacks this would increase global temperature by about 1.2 deg C [at equilibrium]. I’ll use these numbers, and use average surface temperature 290 deg K, to relate temperature to radiation: for radiation R, temperature T, some k, we have R=k*T^4 so k*291.2^4-k*290^4=3.7 hence k=3.7/( 291.2^4-290^4)=3.14E-8. That’s different to the black body figure, but presumably we’re not dealing with a black body. NB. I’m not looking for extreme accuracy, just the ball-park.
From 1983 to 2009, the increased CO2 delivered a downward RF increase of +0.20 Wm-2 (see previous post Appendix A). That would raise temperature by dT where k*((290+dT)^4-290^4) = 0.2 which gives dT ~= +0.07 deg C. That’s only reached at equilibrium, and as Nick says, ““Equilibrium Climate Sensitivity. What has happened after everything has settled down, which takes a very long time.“. [my bold].
I think that +0.07 deg C at equilibrium is probably about right. And its contribution from 1983 to 2009 would have been much less.
3. The cloud effect
The argument about clouds is even simpler: Clouds affect upward and downward radiation roughly equally, so cloud changes have negligible effect on atmospheric temperature. NB. We’re looking at averages, not day vs night.
I think that’s probably about right, too. As I said in my last post, “Clouds have a minor overall effect on average atmospheric temperature“.
But does that argument extend to the ocean ?
Nick Stokes argues that solar radiation that penetrates the ocean is just re-radiated away. He puts it this way: “A large part of the insolation that penetrates the sea, to a depth of several meters, is later radiated in this way. It’s a big part of the thermal balance. Somehow, that heat is conveyed to the surface, and is emitted by the surface layer.“. And that one word that I have highlighted is the third clue – what is this “Somehow“?.
Let’s look at those “several metres” and see what goes on there:
Using all the same data as in the El Niño’s heat post , here is the absorption profile for the ITO down to 15m depth:
Most of the solar radiation is indeed absorbed in the top 1 metre, but for the ITO virtually none is in the top 1mm. The total energy absorbed from thee ITO in the top say 5m is 88 Wm-2. That definitely doesn’t look trivial.
What we are interested in is the effect of clouds. Clouds have little net effect on IR, but for the ITO the equation is different. A reduction in cloud cover lets more ITO into the ocean well below the surface skin. There is now an additional heat source inside the ocean. Using the same figures as before, but correcting the formula (see Appendix A), the 4 percentage-point reduction in global cloud cover from 1983-2009 would see an additional 2.0 Wm-2 absorbed in top 1m, and 1.6 Wm-2 absorbed into the 1-15m band.
That’s a lot more than the +0.2 Wm-2 from CO2 over the same period. Sure, it’s going to get to the surface Somehow, but in the meantime it is going to heat the ocean below the surface. In fact, it can’t (net) release any of its heat to the surface unless it has warmed to a higher temperature. Just like the CO2 change, this extra wattage isn’t a one-day wonder, it’s coming in every day while the cloud cover stays down.
The exact equations from here onwards get difficult, because the situation in the real ocean is fluid – pun intended – ie, the water can move, horizontally or vertically, and heat conducts through it, too, so there’s a lot more going on than just radiation. But the bottom line is that the radiation balance – the “Somehow” – comes from the top few metres of the ocean getting warmer. And if the the top few metres of the ocean get warmer then the globe gets warmer. So …..
4. How much of the Late 20th Century Global Warming was Natural?
We now have the necessary data to start to calculate how much of the late 20th century global warming was natural, and how much was from CO2.
Downward RF from CO2 increased by about 0.2 Wm-2 over the 1983-2009 period. All of that went into the ocean surface skin. Nick’s data shows that it then mixed into the top 5-10m of the ocean, but mainly into the top 1m.
Over the same period, because of cloud cover changes, the ITO increased by 4.5 Wm-2. About 80% of this was absorbed into the top 15m in the following distribution (the rest went into the deeper ocean).
Figure 3. Where the ITO’s 1983-2009 increase of 4.5 Wm-2 was absorbed.
The greater the depth of the mixing, the greater the contribution from the ITO, as shown in column “Cumul.” in Figure 3.
If we assume mixing to 1m only, the proportion of the extra RF provided by CO2 is 9% (0.2/(0.2+2.0). For mixing to 5m it’s 6.1% (0.2/3.2), for mixing to 10m it’s 5.5%.
Those percentages for CO2 are still too high, because
· CO2 increased linearly, while the ITO increase had all occurred by 2000. The ITO then stayed high.
· The ITO is supplying some additional RF to the next depths too, while CO2 is not.
We can state with confidence that the data shows clearly that CO2 contributed less than 9%, and probably less than 6%, of the global warming that occurred from 1983-2000.
The global temperature increased from 1983 to about 2000, but then stalled. This matches the pre-2000 / post-2000 pattern of the ITO as controlled by clouds. The relationship is surely worth investigating very thoroughly. The next 5 years of ISCCP cloud data is due out later this year, and that should present a testing opportunity. By contrast, the CO2 pattern was quite different, CO2 levels increased steadily over the period.
We have seen, above, how clouds are a major driver of ocean surface temperature, and hence of climate, though more data over a longer data period is probably needed before the process can be understood in detail. All of the ocean oscillations (ENSO, AMO, PDO, etc) have a big impact on climate over timescales that range from a year to a few decades. Solar variation appears to have a long term effect on climate, and a possible mechanism has been shown to be via GCRs and clouds. It had been thought that clouds had only a minor effect on temperature, but by looking specifically at the ocean not the atmosphere I have shown how clouds do have a significant impact.
The ocean oscillations are not, as far as I am aware, caused by clouds. Clouds can affect the temperature of the water going into the oscillations, and this I think is the likely cause of the 20th century global temperature pattern. This pattern could be seen as a ~60yr cycle on a rising trend. The cycle seemed to match reasonably well to the Atlantic and Pacific oscillations, and the rising trend related quite well to solar activity for most of the 20th century (increased solar activity => less GCRs => less clouds). The sun is not the only driver of clouds, and clouds continued to decline in the late 20th century even though the sun then started weakening. We need to find why the clouds behaved as they did, but it is clear that the late 20th century global warming was driven mainly by clouds. We know that El Niño affects clouds, so it is likely that the other ocean oscillations affect clouds too. Climate is non-linear, which adds another obstacle to analysis – the same factors can have different results at different times, depending on the state of other factors.
I would postulate that the ocean’s chief influence on the weather (periods of days, weeks) is from the top few mm. That is the layer whose heat is lost fastest into the atmosphere. CO2 would have no influence here, because it doesn’t vary on those short timescales, but clouds certainly would. Factors such as winds would be be important too. An El Niño operates on a slightly longer timescale, and winds have been identified as a (or the) major factor. An El Niño can lift local ocean surface temperature by around 10 deg C, and there is a limit to the depth of water that can be heated by that amount. It therefore seems likely that the pool of warm water that feeds El Niño is not very deep. The multi-decadal ocean oscillations, such as the AMO and PDO, change temperatures by less but over longer periods, so their pools of warm water are probably deeper, and consequently are released over a longer period. But if the pool of warm water is deeper, then the proportion attributable to CO2 over the 1983-2009 period would be at the low end of the range. Over even longer timescales, solar variation via clouds would appear to be a (or the) major factor.
Climate scientists would benefit massively, in my considered opinion, by abandoning their absurdly detailed and desperately manipulated computer models driven mainly by CO2 and the atmosphere. I have explained in previous posts why, as currently structured, they can never work. Energy in Earth’s system is basically a one-way street : sun – ocean – atmosphere – space. (Please note that the ocean surface is warmer than the atmosphere, on average, so net energy transfer is indeed from ocean to atmosphere. See here, eg.). Clouds have a major influence, as they control solar energy entering the ocean. The sun, in turn, has a long term impact on clouds. Until we learn how to predict activity of the sun, clouds and the ocean, we will not be able to predict future climate.
I need to re-write some of my earlier documents, in light of what I have learned since. This will take a while. The concepts are generally unchanged, but some of the detail has changed.
Appendix A. Cloud Formula
It’s not very clear in the Kiehl and Trenberth energy budget diagram (Figure A.1 in the El Niño’s heat post) what the effect of a change in cloud cover on the various radiation components would be. So I’ve made some assumptions. NB. I’m not looking for extreme accuracy, just the ball-park. The assumptions:
· At 71.2% cloud cover (the 1983-2009 average), solar radiation entering the ocean is 168 Wm-2. (Figure A.1 in the El Niño’s heat post).
· At zero cloud cover, solar radiation entering the ocean would be 72.5% of total solar radiation. (Figure A.2 in the El Niño’s heat post, “70-75% transmitted“).
· Clouds’ effect is linear with cloud cover.
From this, a 1 percentage-point decrease in cloud cover would increase solar radiation entering the ocean by ~1.12 Wm-2. For a 4 percentage-point increase, it’s 4.5 Wm-2.
This formula is a correction to the one I used before. The previous one gave too high a figure.
Appendix B. A very brief guide to AbsorptionCalcs_Upper.xlsx
Download: absorptioncalcs_upper (.xlsx)
Atmosphere and Ocean data is digitised from graphs shown in worksheets AtmosphereGraphs, OceanGraphs respectively. Atmosphere and Ocean calcs are in worksheets Atmosphere, Ocean respectively. Some energy budget data is taken fron Kiehl and Trenberth’s energy budget chart in worksheet EnergyBudget. The full combination is then calculated in worksheet 2003.
I use SORCE data for 2003. All years are almost identical.
In the SORCE data, total solar radiation is 330.5 Wm-2 based on Earth’s surface area – cell ‘2003’!D4 divided by 4. Total ITO is 236.5 Wm-2 – cell ‘2003’!E4 divided by 4. The 88 Wm-2 absorbed in the top 5m comes from cells ‘2003’!BC1:BG1. Cells ‘2003’!BC2:BQ2 give the percentagess in Table 3 above.
In the previous worksheet, absorptioncalcs (.xlsx), there was an error in a section labelled Bands for graph only: at cell ‘2003’!AD8009 (now cell ‘2003’!AZ8009). This section is not used for any calcs. The error is that the z axis in the Absorption graph in worksheet Graphs is out by a factor of 10. The section is not used anywhere else, so no calcs are affected.
AMO – Atlantic Multidecadal Oscillation
CO2 – Carbon Dioxide
ENSO – El Niño Southern Oscillation
GCR – Galactic Cosmic Ray
IPCC – Intergovernmental Panel on Climate Change
IR – Infra-Red radiation
ISCCP – International Satellite Cloud Climatology Project
ITO – Into The Ocean [Band of Wavelengths approx 200nm to 1000nm]
PDO – Pacific Decadal Oscillation
RF – Radiative Forcing
SORCE – Solar Radiation and Climate Experiment
SST – Sea Surface Temperature
SW – Short Wave
Wm-2 or W/m2 – Watts per square metre
WUWT – wattsupwiththat.com