Alistair Davidson: aerosols cause warming not cooling

Posted: November 14, 2010 by tallbloke in solar system dynamics

I picked up this tidbit on Judy Curry’s blog:

alistairmcd | November 13, 2010 at 11:43 am | Reply

When I posted on Real Climate pointing out the glaring error in the optical physics used in most if not all the climate models to predict cloud albedo from optical depth**, my post was taken down in less than a minute.

**Mie solved Maxwell’s equations for a plane wave so the assumption of constant ‘Mie asymmetry factor’ is correct only when light first enters a cloud. Also, substantial direct backscattering at the upper cloud boundary is ignored yet it has an opposite dependence on droplet size than diffuse scattering. Therefore, above a threshold ‘optical depth’, pollution causes a reduction of albedo, another form of AGW.

So, at the very least, the IPCC’s predictions of CO2-AGW should be reduced by a factor of about three, possibly much more if ‘cloud albedo effect’ heating explains most recent warming.

Was it the result of Asian industrialisation [the ‘Asian brown Cloud’]? As it’s self-limiting, this might explain why according to ocean heat content, global warming ceased in 2003 and will reverse once Asian aerosol pollution is reduced.


A bit more info on this from elsewhere on the net. The Author is Alistair  Davidson. Hopefully he’ll call by and tell us a bit more about his discovery.

“the ‘two-stream’ approximations used to predict cloud albedo from optical depth and, in particular, the effect of aerosol pollution. There is no doubt whatsoever that they’re fundamentally wrong because light attenuation is from two processes, backscattering at the upper cloud surface shielding the cloud interior and backscattering due to internal diffuse scattering. Unfortunately, the approximations assume just the latter and that excess light from the upper surface of the cloud is steered out by predominately forward scattering.

However, that’s, a mistake apparently dating to Dutch physicist Van de Hulst, who made the assumption to fit experimental data from liquid sols. To do so, he assumed constant ‘Mie asymmetry factor’, using the value Mie got from solving Maxwell’s equations, and got a reasonable fit. However, he and apparently none of the successor physicists who used the idea, including Sagan, studied the fine print: Mie assumed a plane wave front boundary condition, only met as the wave enters the surface of the sol. In reality, for diffuse scattering, g is by definition operationally 0.5, giving symmetrical exit streams.

So, the fit of the equation is fortuitous and whilst it can be used to predict albedo from tau, use to predict change of albedo when pollution changes tau goes the wrong way.”

This also fits with the article on WUWT a couple of months back saying ‘black carbon’ or ‘soot’ to you and me, in the atmosphere causes warming.

  1. More groping in the dark. I was watching yesterday at a science program on cable TV, where it was shown that while flying across some clouds in the Florida coasts, these showed, in some cases, an electric charge of more than 100 KV. Then, what are we doing watching at PC screens while describing meteorological phenomena only in terms of Hot Air and Cold Air, convection and so on…?, that’s silly OK!, that’s the way of being accepted: we do not opposed “settled science”-what we could call “self calming science”-and everyone is HAPPY!)
    Let’s stop fooling around and let’s begin considering all FIELDS together. Take into account that from other parts of the world you have already become a JOKE, which qualifies your “science” as “Hollywood Science:”

  2. tallbloke says:

    It’s all about cause and effect and the magnitudes of the energies Adolfo. Read some stuff by Tinsley on the global electrical circuit. I agree that electrical effects are no sufficiently well integrated into our understanding of climate yet. However, unless some mechanisms and quantification are forthcoming, that isn’t going to change much.

    Today, I was up on the moors near my home watching a temperature inversion produce a cloud over the valley bottom below me. It formed, thickened, then melted away over the course of an hour or so.

    There was no lightning or strong static electricity charge that I could detect. It was just to do with temperature, adiabatic lapse, moisture and solar energy (electromagnetic radiation) being turned to heat in water vapour.

  3. paintitblack says:

    APD: perhaps you could provide answers to the following questions:
    – you say that your theory better models thick cloud behaviour: do you have any real life data to support this?
    – your theory differs from mainstream science only in the case of thick clouds. Such clouds do not spring into existence spontaneously, but develop gradually and not all clouds become so thick. What percentage of the time do clouds exist in the form that your theory deals with? What percentage of the life time of the clouds is that?
    – What percentage of the Earth’s surface is affected by this theory and what latitudes?
    – You claim that your theory reduces climate sensitivity by a factor of three. How does this explain earlier climate change when aerosols were not present?

  4. Father William3 says:

    Attention: Paintitblack

    Dear sir and/or madam,

    I am a rather ancient person with challenged vision (age-related macular degeneration problems in both eyes), and this may possibly account for my being unable to locate the word “thick” in the statements that your comment above addresses. It would be helpful if you would provide a quote containing the word “thick” for my benefit. Thanking you in advance.

  5. Alistair Davidson says:

    Father William3 and paintitblack: my definition of thick clouds!

    Lacis-Hansen eq 19: , and its variants assume one process, Internal Diffuse Scattering preferentially directed upwards by high Mie asymmetry factor, g, c. 0.85, assumed constant. It appears to fit alpha/apparent tau data but in reality there are two processes: IDS and Direct Backscattering at the upper boundary so the fit is in my view fortuitous.

    Furthermore, the latter process can be extreme, e.g.: explains how agglomerated ice crystals form at the top of cumulo-nimbus clouds giving high DB, tau ~1,000, alpha ~0.9. Since IDS is by definition symmetrical because operational g from random photon direction is 0.5, what light enters is split into two parts: up and down: 80% shielding, 10% diffuse up and down.

    For thick water clouds, you don’t get agglomeration so total albedo is lower. 0.7 means 40% DB, 30% IDS up and down. The origin of part of the DB is easy to understand. Since g~0.85, light backscattered by droplets in line of sight is ~[1=-g], or 15%. The origin of the rest is less certain. However, go into Mie theory and after the first scattering, the relative peak intensity of the coherent forward scattered peak is 10^7 for 15 micron droplets [unpolluted] and 10^5 for 5 microns.

    The maths is hideous and as far as I can tell never solved but what I suspect is that at the second interaction, perhaps a few more, the deviation of the wave from being plane is so extreme, g may be <<0.5. So, a practical definition of 'thick' is a stereological problem: a high probability of second interaction within the physical depth of the cloud. The effect of pollution is clear: this process is attenuated, cooling and, very importantly, there is no first-order number per unit volume effect.

    NASA also thinks there’s a second process: they replace Twomey’s correct Mie analysis which predicts a maximum hemispherical albedo of 0.5 [no absorption, symmetrical] with enhanced ‘reflection’ from higher water surface area in polluted clouds : . [Take care with Twomey – he gives spherical albedo asymptoting to 1.]

    In other publications NASA claim albedos up to 0.9. However, by about 2003, there had been no experimental proof of ‘cloud albedo effect’ cooling and there’s no such physics. There is some evidence however for the kind of DB I’m proposing. Satellites show an angle-dependent radiance when Lambert’s law predicts no such effect. As a cloud gets thinner, the proportion of albedo from IDS decreases and the effect of pollution switches over to cooling.

    So, we have an alternate form of AGW. Knock off 0.15 from a start albedo of 0.7 and the transmitted energy increases by 50%: it could be extraordinarily powerful.

    A further interesting factor is that when experimentalists use ‘ships’ tracks’ to measure the effect of pollution, angular DB albedo can easily be missed and a pyranometer which is designed to integrate over all angles, can miss it completely.

  6. Alistair Davidson says:

    Correction: 3rd para from bottom: NASA claims polluted cloud albedos up to 0.9.

    [I’m very suspicious about why they did this when they had just commissioned work to find why polluted clouds didn’t have the expected higher albedo: ]

  7. tallbloke says:

    Hi Alistair and welcome, many thanks for dropping by to add further detail to the comment I picked up.

    Forgive us, but we’ll struggle a bit with acronyms, abbreviations and research references at first. Can you point us to any synoptic articles freely accessible online? So much of the specialist literature is behind paywalls these days.

    On first reading, it looks like one of those fields where large degrees of uncertainty concerning the magnitude of the net effect if not the physics wouldn’t go away even if a large effort was employed to work on this. Is that right?

    To help build a context for back of cig packet WAG’s could we get an approximation of the relative quantities of anthropogenic aerosols vs natural aerosols from lightning ignited forest fires, volcanism etc?

  8. Alistair Davidson says:

    Thanks for replying TB. I’m not a climate scientist but with 40 post PhD years under my belt, I have an awful lot of experience in dealing with complex problems. There is no literature in what I am proposing. It’s so off the wall as to be loony tunes. However, just remember this. In 2004, NASA published fake physics. Why?

    Also, the L-H model and all the other two-stream approximations that don’t consider the change from a plane wave front may be wrong. It’s a fact that the phase of the forward-scattered intense beam peak predicted by Mie scattering theory is set by the last molecule releasing energy so the next interaction is almost like that with a laser. Also, the strong angle-dependence of measured low-angle albedo suggests an internal, pseudo-geometrical scattering process.

    The bottom line is that I don’t believe in the ‘cloud albedo effect’ part of global dimming. To reverse the sign is a game changer.

  9. tallbloke says:

    NASA is a law unto itself so who knows what their motivation might have been. Support the NASA GISS modeling effort with a convenient source of cooling to offset an inflated sensitivity calculation, obtain more funding, I don’t know, speculations on a postcard to the usual north pole address.

    If dirty clouds are trapping more heat through more scattering due to increased nuclei abundance, I can see that might lower the co2 sensitivity, and screw up water vapour feedback assumptions dear to the heart of Gavin and the Team. Either way, I doubt the effect is drastic, as my own reearch (amateur though it is) suggests to me the high solar activity in the later C20th accounts for more of the warming than the IPCC says it does.

    Interesting stuff!

    Here’s another warming soot article, different mechanism though.

  10. Alistair Davidson says:

    There’s also a lot of politics here with Lord Stern warning the USA of a trade war if the Americans don’t tax carbon. Take away the CO2-AGW hypothesis and two decades of Marxist politics goes down the pan.

  11. tallbloke says:

    Stern is a buffoon. But this is all just pre-Cancun hystrionics anyway.

  12. Alistair Davidson says:

    Just found a reply to my comment on another blog where a guy wrote that he has come to a similar conclusion to me, even to the extent of the same reduction of IPCC-predicted CO2-AGW. The reason i know it’s genuine is that he writes that the physics’ error predates Sagan:

    Looks like there’ll be a second Spartacus when I’m due to be crucified by the UN/EU Deniers’ Court…..

  13. tallbloke says:

    Heh, don’t shoot your mouth off in California either Alistair!

    That comment looks to me like someone has just done a potted summary of your own work, without bothering to provide the source.

  14. Alistair Davidson says:

    Could be, but he must have had a good knowledge of optical physics to know how it developed from the 1950s.

  15. suricat says:

    Hi Tallbloke.

    Interesting post, but how can ‘Asian Brown Cloud’ be considered as an ‘aerosol’? OK, so the pollutant ‘may/may not’ be encapsulated within a water droplet, but it’s still a ‘black body’ that’s floating in the air.

    For the life of me I can’t understand how ‘optical formula’ can resolve the scenario of a ‘floating black body’! Well, perhaps with the reduced absorption of a ‘floating brown body’.

    I’m not qualified in that field, but surely an optical resolution can only offer ‘forward/back scattering’ for an ‘optical medium’, but neither brown (~soil humus), or black (~IC engine exhaust), carbon is an optical medium. They’re ‘opaic’! Thus, they are ‘black body’ entities that exhibit an ‘albedo’ with all the absorption and emission paraphernalia that this invokes. 🙂

    Best regards, Ray Dart.

  16. suricat says:

    Alistair Davidson.

    “It’s a fact that the phase of the forward-scattered intense beam peak predicted by Mie scattering theory is set by the last molecule releasing energy so the next interaction is almost like that with a laser.”

    If it’s to do with ‘cloud’, and in the troposphere, it’s nothing like “a laser”. This is ‘atmospheric lensing’ and the main point of interest with this phenomenon is the redirection of incident UV below cloud at high Earth latitudes.

    Best regards, Ray Dart.

  17. tallbloke says:

    Hi Suricat,
    thanks for the obs. I’m going to leave these for Alistair to answer because I’m out of my optical depth. 🙂

  18. suricat says:


    Nice pun. )

  19. suricat says:


    Question. Can I edit my posts here?

    Best regards, Ray Dart.

    [reply] Unfortunately not, but you can always repost an amended version and rquest the previous be removed.

  20. suricat says:


    How often does Alistair call in here?

    It would be nice to get ‘some’ response to my queries before I die. While I realise that Alistair’s post isn’t your ‘statement’, it’s been made on your blog!

    Thus, I take it as read that Alistair’s post and your blog are both bullshit. I’ve been reluctant to come to this conclusion because your blog agenda contains many elements that are of interest to me. However, if the dialogue is only in one direction, it’s only a blog that dictates state of the art science and not a blog that I want to be associated with. Without the interactivity of dialogue a blog is only a news broadcast!

    Regards, Ray Dart.

    [Reply] Hi Ray, he only joined this one thread after I picked up his comment elsewhere. People lead busy lives and for most climate is a spare time thinking activity. The blog is for free, and I do my best to juggle a lot of incoming and a lot of other priorities falling on me. I’ll email him for you when I get a moment ok?

  21. Alistair Davidson says:

    Sorry suricat.

    My posts are not wind-ups, they’re work in progress by a retired scientist. I’m not a pukka physicist so it’s taken a bit of time to catch up on the detail but having been top of my PhD year at imperial, I’ve easily got the horesepower.

    What I’m suggesting is that 50 years of atmospheric physics is wrong because of the assumption of constant Mie asymmetry factor. The coherence of the forward scattering after the first interaction could be important but not necessarily. However, I think that the methodology in the climate models which claims enhanced diffuse backscattering by that forward concentration is wrong.

    This is because for random diffuse scattering there is, by definition, no wave front unless you invoke a mechanism akin to phonons in the physics. Someone with that background could educate me. In the absence of that explanation, operational g must be 0.5

    The other explanation is to consider a cumulo-nimbus cloud with optical depth c. 1,000 and albedo >0.9. That ‘s because agglomerated ice at the top is a snow-cap and shields the interior. The models think of it as high, internal, directed, diffuse scattering.

    Similarly, a water cloud with 0.7 albedo comes from a shielding effect. You can see this with the naked eye when a cloud is near to raining. The dark interior isn’t absorption or enhanced scattering, it’s optically shielded indicating that that process is enhanced for larger droplets.

    I’m currently going into the Mie physics for the first interaction and trying to work out how the upward diffuse flux modifies the scattering process.

  22. As light we got scattered in details (the devil works on that: He is in the details)…

  23. suricat says:


    I take it all back! Thank you for your timely assistance. I extend my humble apology for any doubting of your integrity, but I still don’t see how ‘Asian brown cloud’ has anything to do with ‘Tau’ (other than a possible water droplet inclusion).

    Best regards, Ray Dart.

  24. suricat says:


    Thank you for your response here, but ‘reading between the lines’ it seems your original hypothesis doesn’t include either ‘black, or brown’ carbon for your calculation of Tau.

    I say this because ‘particulate’ carbon inclusions to the atmosphere don’t give a value of Tau, they offer a ‘black body albedo’.

    Best regards, Ray Dart.

  25. suricat says:


    As I said, “this isn’t my subject”, but after some thought on this:

    The depth to extinction of ‘insolation’ (INcoming SOLar radiATION) through *pure water* varies greatly. IR ~3 metres, red spectrum a bit further, and so on to the blue/UVa spectra at ~700 metres. Some ref. media:

    So if cloud shadowing was caused by spectral absorption we would observe a thickening cloud becoming blue as its shadow increases (similar to an increasing ocean depth scenario). This isn’t the case, but what we do observe is an increasing ‘greying’ (due to an increasing absence of all light) as a cloud thickens. Thus, absorption of insolation by ‘cloud water droplets’ (aerosols) is a false scenario!

    If we take a look at the reflective and refractive properties of a *pure water* droplet under incident insolation, it becomes obvious that both reflective and refractive ‘scattering’ of the incident insolation must occur. I believe you are correct with your assumption that only the ‘first scattering’ depth can be predicted from the ‘droplet size’ and ‘droplet density’ (units per cubic metre) ratio. This ‘first depth to diffusion’, together with ‘total cloud depth’, must proportionately determine the ‘cloud shadow’, and inversely, the ‘reflective cloud albedo’, respectively.

    That’s all well and good, but cloud droplets don’t consist of *pure water* alone and it’s rare that a ‘particulate’ isn’t included somewhere within a cloud droplet from the ‘seeding’ of the droplet as local ‘water vapour’ (WV) condensed from the atmosphere. These ‘particulates’ are precursors to condensation in that they provide a ‘surface’ for WV to condense upon, but also have a ‘BB’ (black body) albedo that may inhibit condensation by the ‘CC’ (Clausius Clapyron) relationship with temperature (this may seem to be getting complicated, it is).

    The CC relationship determines whether or not WV is able to condense to water on a surface that can absorb (at least some of) the latent energy of condensation, and the BB albedo of an atmospheric particulate determines the temperature of that particulate. Can you see where I’m going with this?

    Perhaps this paper will help you to understand the problems with defining an Earth energy budget. 🙂

    The definition of the mediator is most important.

    Best regards, Ray Dart.

  26. Alistair Davidson says:

    As with everybody else, I start with the premise of no absorption. However, if you put it in, there’s no real change in energy transmission so long as you total Vis and IR.

    The reasoning is that multiple scattering increases the probability of conversion to IR but because optical scattering is inversely proportional to the fourth power of the wavelength, the IR then gets out very quickly. There is of course an increased heating of the cloud and this will be greater for black carbon.

    However, in terms of the effect on my prime reasoning, which is that pollution may significantly reduce direct backscattering at the top of a cloud thus allowing more energy to penetrate it, it’s a secondary issue.

  27. suricat says:


    “As with everybody else, I start with the premise of no absorption.”

    Unlike you (and everyone else it seems), I realise that there are so many energy attractors out there that I start with the premise that ‘absorption shall occur, somewhere’ (the main problem is identifying the energy attractors and their significance).

    “However, if you put it in, there’s no real change in energy transmission so long as you total Vis and IR.”

    I’m not sure what you mean here. Is this absorption of insolation, OLR, or ‘entropy to enthalpy’ (energy to chemical alteration) that you imply? That may seem an odd question, but as an engineer I’d expect to account for energy from whatever source to whichever sink :).

    “The reasoning is that multiple scattering increases the probability of conversion to IR but because optical scattering is inversely proportional to the fourth power of the wavelength, the IR then gets out very quickly. There is of course an increased heating of the cloud and this will be greater for black carbon.”

    As I said, I believe your conclusion is correct for “scattering”. The greater the depth of cloud (dependant on the wavelength of the observed energy), the greater the possibility of ‘depth to extinction’ from a lesser attractor. Thus, an increased cloud depth positively facilitates the ‘depth to extinction’ of whatever radiation wavelength. This is why I question your understanding of “multiple scattering”. The two may well be the inverse of themselves!

    I’m sure you didn’t read my second link. If you did you’d realise that C[brown] has a greater impact as an attractor to energy than C[black], so you would have mentioned C[brown] as the major mediator and not C[black]! This is a ‘surface albedo’ scenairo at altitude. Please keep up with the postings. 🙂

    “However, in terms of the effect on my prime reasoning, which is that pollution may significantly reduce direct backscattering at the top of a cloud thus allowing more energy to penetrate it, it’s a secondary issue.”

    I disagree! This issue is paramount to the energy equilibrium formula for Earth’s energy ‘budget’.

    Hope you understand my feeble reply. 🙂

    Best regards, Ray Dart.

  28. Alistair Davidson says:

    Hi Suricat,

    What you are trying to do is something similar to the ‘two-stream’ approximation which lumped various aspects of the problem together. However, it was in effect a sophisticated curve fit.

    It’s too damned complex for me so I have split the problem into two optical processes. The energy that gets into the cloud is scattered symmetrically because it’s diffuse.

    At the upper surface there’s a still unknown process, although I have some ideas, which directly backscatters light. The evidence of this is the angular dependence of albedo.

    I have worked on ‘brown effects’ before. You can see it as the coloured anodised aluminium cladding on prestigious buildings. What you have to understand is that the light you see is coloured as the complement of what is absorbed.

  29. tallbloke says:

    This seems relevant here:

    Click to access cess.pdf

    It seems GCM’s improve if ‘unphysical’ ad hoc increases are made to cloud shortwave absorptivity

  30. tallbloke says:

    David Springer says:
    March 30, 2012 at 2:16 pm (Edit)

    Stephen Wilde says:
    March 30, 2012 at 1:10 pm

    You will probably be interested in this:

    I subscribe so I got the whole article in the dead tree edition this year.

    It describes, among other things, how clouds are modeled in GCMs. They use a small number of water droplet sizes and calculate reflectivity from that. But there’s a reflection phenomenon from clouds that has never been satisfactorily explained. If the sun is at your back and you cast a shadow onto a cloud below you see a rainbow of colors like a halo surrounding your shadow hence the name “glory”. It was first observed and documented hundreds of years ago. The rainbow cannot be observed by others much outside your 180 degree viewing angle. They just see a shadow.

    The difficulty in explaining this is that 180 degrees is well outside the angle of refraction in a water droplet. The author of the article explains the effect through quantum tunneling. Quantum tunneling of light is a known phenomenon that is employed in some high tech gadgets so that’s not controversial. In effect when a photon from the sun passes within a wavelength of a water droplet it can tunnel into the water drop then tunnel out the opposite side moving in exactly the opposite direction. So the GCM cloud model is incomplete and clouds are reflecting more solar energy than is accounted for in the model. IMO it’s probably enough energy to change the sign of the feedback from slightly positive to slightly negative. A slightly negative feedback from clouds changes catastrophic global warming into beneficial global warming. Not good for the CAGW weenies or the polticians who tried to leverage the panic into a much greater concentration of power in poltical hands.

  31. Mydogsgotnonose says:

    Hi Tallbloke. yes, I have seen this article.

    I have another explanation………!

  32. tallbloke says:

    Hi Mydog, did you get the copy of Wayne’s energy budget I sent you?

    And,,, do tell. 🙂