Bishop Hill +++Harris and Lewis+++
Sep 14, 2013 Climate: MetOffice Climate: Models Nic Lewis has published a detailed comment on the Met Office’s report on climate sensitivity, which was itself very much a response to the Otto et al paper of which Nic was an author. The comment is here.
There is a great deal of interest, not least of which is the fact that the Met Office seems to have made a series of misrepresentations of Otto et al, as well as making several mistakes.
One of these though is astonishing. This concerns Harris et al 2013, a paper by a group of Met Office scientists. The paper is particularly important as it is the source of the official UK Climate Projections. It examines how the virtual climate inside the Met Office model responds to having key parameters tweaked. Different parameter combinations are then weighted in the final analysis depending on how well the resulting virtual climate matches observations.
Nic explains what he found:
As HadCM3's parameters are perturbed, the resulting changes in ECS and aerosol forcing are closely linked. When significantly lower values for ECS – as suggested by recent observational studies – are obtained, HadCM3's aerosol forcing takes on highly negative values. The observational data strongly contraindicate aerosol forcing being highly negative, so parameter combinations resulting in significantly reduced model ECS levels (and thus highly negative aerosol forcing) are heavily down-weighted. As a result, whatever the actual level of ECS, HadCM3-derived ECS estimates are bound to be high
Essentially, the observations would lead you to conclude either that a high climate sensitivity is being masked by a quite strong cooling effect from aerosols, or that there is lower climate sensitivity and that aerosol cooling is a much smaller effect too. However, the Harris et al study behaves in a very different way. If you tweak the settings on the model in a way that gives you relatively low climate sensitivity, the model ends up with a very strong aerosol cooling effect too. The relationship between aerosols and climate is the opposite way round to the observational studies!
The aerosol cooling that appears in the Met Office model when climate sensitivity is low is very strong, completely inconsistent with recent observationally-based estimates of aerosol forcing. In fact it is so strong that it would also have prevented most of the temperature rise seen since industrialisation. In other words, the virtual climate produced with these settings of the model doesn’t match the real one. This means any scenario in which climate sensitivity is low – as indeed the observational studies suggest it is – gets downweighted in the final analysis to the extent that it doesn’t really show up in the final results. The effect is essentially to rule out low climate sensitivity as a possibility.
I’ll say that again in a slightly different way. The Met Office’s model, one used generate the official climate projections, has big temperature rises built in a priori.
Wow.
Reader Comments (85)
Didn't they say they use this stuff to predict weather too ?
I'm sure that Richard Betts could help here.
The trouble with predictions is that the further into the future you predict, the more likely you are to be wrong, no matter how much evidence you gather to support yourself: e.g. “Looking at that rain-cloud over there, it’s going to pour down, soon.” – high probability; “Looking at that sunset, it will be a good day tomorrow,” – reasonable probability; “The rate it has been cooling these past few days, and with what we have seen about the wind blowing over yon hillock, it will snow on the first of December.” – well, guess.
Astonishing. I don't pretend to understand how this has arisen. The complexity of individual climate models is such that most people havn't the time, energy or expertise to dig into them in the depth required to understand wht these results are obtained.
Personally, I would rather depend on specific predictions and, as the NAS has just published, the majority of climate models' predictions do not have high accuracy.
This wouldn't matter if the results of models were not so highly politicised, but now the climate modelling fraternity have to rethink their entire approach.
It doesn't matter how complex a climate model is: It is still a load of crap.
It really makes you wonder just how much the "adjustments" to temperatures and other observations are affecting the more mundane weather models. I know that the predicted high temperatures around here have been high by a degree or more almost every day this summer...
I am sure Julia Slingo will come along and tell us that all the Met Office needs are more powerful computers to do the job right!
Models that are not validated are for games only and not for the real world!
Is this just more evidence that the Met Office model is not fit for purpose? Add its failure to predict the current hiatus, and surely it must be ignored in any future policy decisions.
I have sent the link to Nic's paper to David Davies MP
Dolphinhead
I wonder if the layman's summary here might be more his thing?
Bishop Hill
The physics and Climate for the under 5s first as a primer?
This accords with my experience of a lot of the work of Government scientists. They are keen to publish without anyone thoroughly checking and verifying what they are doing. The name of the game is publish and get one's name known. In the real world of engineering, shoddy, unchecked work, results in sacking of those shoddy individuals. In Government circles it results in promotion and big bonuses all round
I'm not surprised. The Met Office is convinced that carbon dioxide driven warming completely dominates climate. Their models are always going to show massive warming as a consequence of rising CO2. They cannot believe any other scenario. Their models are constructed in keeping with this mind set.
They are unable to grasp that they are wrong.
SandyS
I don't mean any disrespect to David Davies, but Nic's article is heavy going for the uninitiated.
For best desired result please tweak any key parameter accordingly while ignoring inconvenient observational data as it only interferes with the publication timetable.
Nic Lewis must be wrong. After all Harris et al 2013 is a peer-reviewed paper. ;<)
Why do they want to make their models so complex. Surely starting at a basic level would be far more beneficial.
Start with solar radiation, then add in water vapour, then clouds (high, medium then low). See how this matches observations. If it doesn't, tweak the input values for each until it does, because these are the major influences on the climate.This then becomes the base model.
Once that has been achieved to a high level of accuracy, add in the effects of other particulates bit by bit to see if this improves the model or otherwise. You reject the particulates which make the base model worse and keep the ones which improve the model match to observations.
They must know this, surely? Are we supposed to believe that this is mere incompetence?
The HadCM3 result is inherent, explicable, and applies to virtually all GCMs. Nic identified it precisely, a negative lapse rate feedback ( humidity as a function of altitude). AR4 WG1 black box 8.1 is clear about constant UTrH. That means there is no lapse rate feedback. Yet multiple observation methods have shown UTrH declined with warming. That is an observed negative lapse rate feedback. Upshot is positive water vapor feedback is overstated and the model runs too hot. So it gets artificially cooled to mach past history by overstating observed aerosols. When that proved observationally incorrect, the Trenberth hidden heat nonsense emerged to paper over this fundamental GCM flaw.
Lindzen hypothesized the physical mechanism back in 2001 as an 'adaptive iris'. More warmth, more surface humidity, more tropical convection (thunderstorms). These produce more precipitation, which removes humidity that would otherwise reach the UT. Also release more latent heat to radiate away as OLR. A classic negative feedback mechanism damping any warming effect, including (shown in the 1990s) El Nina. Explains the lack of an actual equatorial troposphere hotspot that the CMIP3 archive all model. Problem is inherent and insoluble, since the best supercomputers do not enable sufficiently small grid scales to model these convection cells. The impossibility of adequately reflecting this climate fundamental is probably not something MET modelers would want to admit, lest they risk defunding. Yet AR5 SOD itself acknowledged the problem in WG1 at 7.2.1.2 with respect to clouds generally ( and cloud feedback generally), for which Lindzens adaptive iris is a special and doubly important case because it also produces the negative lapse rate feedback.
"I don't mean any disrespect to David Davies, but Nic's article is heavy going for the uninitiated."
Maybe I'm having a bad day, but even the Bish's summary seems like spaghetti to me apart from the very last paragraph.
Your Grace
I have also sent DD MP a link to this post
kind regards
Delfino
The Met kiddies need to create hysteria so they can keep their precious seats on Glowball Warming Gravy Train. Every one of these seats comes with liplock on the public teat, cushy working conditions, wonderful vacations attending Conferences in places like Bali and a very short work career before generous taxpayer funded pensions kick in.
Of course they are going to torque up their models to create climate hysteria. It is in their personal interest.
Time for Cameron to pull an Ozzie.
Fire.
Them.
All.
Leave the troll’s comments, Your Bishness. Like the outside broadcaster pestered by the odd-balls who insist on being in the picture, it can be quite amusing (even if you know the broadcaster wishes to assassinated the prat!).
I remember when measured temperatures began to deviate from modelled ones. The Met Office introduced aerosol cooling to counteract the CO2 driven warming.
Of course, as we have always known, much of the warming and cooling have been due to natural variability, but the Met Office have always regarded natural drivers as trivial, so their models have not adequately included them.
So the Met Office model is probably based on CO2 warming with some cooling from aerosols and the rest is tinkering around the edges or beyond their modelling skills.
I remember when measured temperatures began to deviate from modelled ones. The Met Office introduced aerosol cooling to counteract the CO2 driven warming.
Of course, as we sceptics have always known, much of the warming and cooling has been due to natural variability, but the Met Office have always regarded natural drivers as trivial, so their models have not adequately included them.
So the Met Office model is probably based on CO2 warming with some cooling from aerosols and the rest is tinkering around the edges or beyond their modelling skills.
It seems to me that the Met Office adopted a set of beliefs and constructed their models around these beliefs and today they are finding that their initial assumptions were flawed. It is now late in the day to start introducing natural factors such as the ocean oscillations, solar effects and cloud albedo so it is difficult to see how the Met Office can convert from a belief based religion to objective science without being caught out.
Their current public briefings are all about extreme weather, local climate impact, heat going into the deep oceans and this decade being the hottest ever, so they are obviously desperate to distract gullible politicians and prevent them from learning that the models on which government policy is based are beginning seriously to unravel.
Brilliant. The Met Office should contract Nic Lewis to correct papers and statements before publication. This is not a joke, they owe this to the public.
One of these though is astonishing
OR deliberate !!
Time for Cameron to pull an Ozzie.
Cameron couldn't his proverbial.
"the observations would lead you to conclude either that a high climate sensitivity is being masked by a quite strong cooling effect from aerosols, or that there is lower climate sensitivity and that aerosol cooling is a much smaller effect too"
The dust levels (especially Log[DUST]) in the ice cores are negatively correlated with temperature. It is relatively trivial for the modelers to use the ice-core temperature/[CO2]/log[Dust] with the numbers their models spit out. Their problem is that dust levels change by more than three orders magnitude between war/ice ages and so it would suggest that the climate sensitivity of CO2 is very low.
I’ll say that again in a slightly different way. The Met Office’s model, one used generate the official climate projections, has big temperature rises built in a priori
that is the least shocking news since, bears found to us wood as toilet .
In reality the MET office models have a built in prior which best ensures money to the MET office , and political reality means that is warming .
Not to worry.
"Support for the European Union’s climate and energy policy eroded further Friday as the Czech Republic became the latest member to denounce subsidies for clean but costly renewable energy and pledged to double down on its use of fossil fuels.
It followed Poland’s declaration that it would use its abundant domestic coal supplies for power generation rather than invest in costly renewable energy facilities. Spain abolished subsidies for photovoltaic power generation in July and the U.K.’s power markets regulator last month froze solar power subsidies for the rest of the year".
http://stream.wsj.com/story/latest-headlines/SS-2-63399/SS-2-326504/
Anyone who claims that they can model a massive non-linear chaotic system subject to extreme sensitivity to initial conditions and driven by a practically infinite number of feedbacks, of which even of the few we know we often don't know the signs - which, as in the case of some types of clouds can change from positive to negative and back every 24 hours, for example - and on the basis of those models make predictions decades out into the future is either a liar or a fool.
Possibly (probably?) both.
And anyone who bases economic policy to the value of literally trillions and affecting practically every human and a considerable number of non-human inhabitants of the planet Earth on their prognostications is an even bigger fool.
Quem deus vult perdere, prius dementat.
The Met Office’s model, one used generate the official climate projections, has big temperature rises built in a priori.
well, what a surprise. Who'd have thought it.
I'm a bit sceptical about this (at first glance, without having read anything properly).
The ECS is an output of the model, whereas aerosol forcing is an input.
So saying that low ecs forces negative aerosol forcing doesn't make much sense to me.
Bishop Hill
point taken.
So what are the tweakable parameters?
I bow to no one in admiration for His Grace's work in making things accessible to us unwashed, but I still can't quite understand what could be meant by "If you tweak the settings on the model in a way that gives you relatively low climate sensitivity, the model ends up with a very strong aerosol cooling effect too."
Surely the ratio of forcing to aerosol concentration is a directly controllable parameter. I'm assuming that aerosol concentration belongs to the set of stimuli to which the model calculates a response, while the resultant forcing is computed by applying that parameter to that stimulus. So the only meaning I've so far been able to assign to the quoted passage is that the model is written in such a way that only when the modelers directly make that parameter--watts/m^2 / unit aerosol concentration--highly negative do the resultant values of other, less-directly-settable parameters result in a low ECS.
But I'm pretty sure I've misunderstood. Could I impose upon someone explain that in even simpler terms?
Paul Mathews
"The ECS is an output of the model, whereas aerosol forcing is an input."
Not so here; they are both outputs. The Met Office climate models and GCMs generate aerosol loading and the forcing it causes from estimated emissions of sulphates, organic and inorganic carbon particles and other pollutants. The modelled aerosol forcing from any given level of emissions is strongly affected by the settings of the same model parameters that have most effect on the model's ECS - mainly parameters that affect clouds.
Joe Born
"So what are the tweakable parameters?"
There is a list of the 31 atmospheric model parameters that were tweaked in Table 1 of the Rougier 2009 paper, available at http://journals.ametsoc.org/doi/pdf/10.1175/2008JCLI2533.1 . The effect on ECS of tweaking them is shown in Fig.7.
" Surely the ratio of forcing to aerosol concentration is a directly controllable parameter."
I think not. There are separate aerosol model parameters in HadCM3, but tweaking them has substantially less effect on the resulting aerosol forcing than tweaking those of the 31 main atmospheric model parameters that have a strong effect on ECS - mainly cloud parameters. So the ECS - aerosol forcing relationship appears not to be breakable by adjusting any parameters. See Figure S2 in the Harris et al 2013 Supplementary Information.
Well, this is good isn't it. It means their projections will continue to diverge from reality and as time goes on, look even more absurd than they already do. Once enough grains have fallen onto this pile of sand, eventually an avalanche will occur.
Not necessarily, Robinson.
"Who controls the past controls the future. Who controls the present controls the past."
The ecofascists and their useful idiots will simply fabricate the temperatures they want, and deny the real ones. Easy.
In their May 2013 paper Oikawa et al derive all-sky aerosol forcings of -0.58WM^-2 and -0.61W^M-2 from two different analysis techniques.
http://onlinelibrary.wiley.com/doi/10.1002/jgrd.50227/abstract
I am struck by the similarity of this value to the O.58W^M-2 quoted in AR4 as the net energy imbalance between surfance insolation and outward longwave radiation driving AGW.
If the two effects have been equal and opposite in recent years, it would provide a possible explaination for the pause in atmospheric warming.
Sep 15, 2013 at 12:11 AM | entropic man
Wow. What a coincidence! To me it seems at least 97% likely.
My reading of this is that the Met Office is incompetent in a really embarrassing kind of a way. Sort of a student level of incompetence...
What's worse, is that research on aerosol effects on temperatures has been ignored, or more likely actively discouraged, in order to maintain aerosols as the magic control knob in the climate models.
In the real climate, it is notoriously difficult to tie an effect back to a specific cause, but aerosols are an exception, because pretty much every city in the world has a regular weekly aerosol cycle. Yet only a handful of papers relating temperature differences to day of the week have been done and I am not aware of any, in the last 20 years, that include aerosol measurements.
This one from Tokyo clearly shows aerosols cause net warming, because weekdays are warmer than weekends (and Mondays).
http://www.ide.titech.ac.jp/~icuc7/extended_abstracts/pdf/375665-1-090515145054-004.pdf
"Quem deus vult perdere, prius dementat." --Catweazle
He whom God would destroy is crazy about the Prius. Right? : ]
Sep 15, 2013 at 2:30 AM | Unregistered Commenter jorgekafkazar
That's how I understand it too!
I totally agree with Catweazle.
I think it needs to be made more explicit that (if my understanding is correct) various trials (simulated climate runs) with distinct starting parameters (e.g. feedback coefficients) are attempted and only those which match the observed reality are taken as true.
The model is designed such that only those with high CO2 sensitivity will approximate reality, even though other possibilities could and should be closer to reality, but the model is hard wired to produce a junk result for those parameters.
"So what are the tweakable parameters?"
with that model there are something like 30-32 paramaters to be tweeked.
A full factorial examination of the parameter space is not feasible so they run certain corner cases, then they can build a statistical emulator and investigate the parameter space with that.. note any weird results and then run the full blown model.
Hmm somewhere on the Cambridge climate series seminar there is a video of a presentation of this approach.
Some paramters are things like with and without a slab ocean.. with and without sulfer cycle..
That Nic Lewis is a bloody troublemaker. The Met Office promised it was going to turn out nice and I've booked my holidays.
http://www.dailymail.co.uk/news/article-2420783/Global-warming-just-HALF-said-Worlds-climate-scientists-admit-computers-got-effects-greenhouse-gases-wrong.html