People send me stuff. Lance Wallace writes:
Anthony, this short “Perspectives” report in Science seems to me to be worthy of a posting in WUWT. Not only is it a very clear indication of crucial problems with the GCMs, it appears in Science magazine, for years a dogged defender of the faith. I’m including the article (paywalled of course) because I think your readers will be blown away by the figure if you can run it.
The authors ran some extremely simplified CMIP5 GCMs, looking only at how they treated water (precipitation, cloud formation), and found extreme differences from one model to the next, as is evident from the figure.
In the final section titled Back to Basics, they make clear that the problem is a fundamental one of not understanding the coupling between water and general circulation. They specifically state it would be better to go towards numerical weather prediction rather than continue to expand the coverage of the GCMs.
By the way, they picked just two aspects–clouds and precipitation–to concentrate on, but they mention a few others, such as sensitivity and arctic amplification of temperature change. Then there are also aerosols, energy balance, and ocean circulation. I could see more examples of models simplified down to each of these aspects in turn and compared to see how they perform. – Lance Wallace
================================================================
Science 31 May 2013:
Vol. 340 no. 6136 pp. 1053-1054
DOI: 10.1126/science.1237554
What Are Climate Models Missing?
Fifty years ago, Joseph Smagorinsky published a landmark paper (1) describing numerical experiments using the primitive equations (a set of fluid equations that describe global atmospheric flows). In so doing, he introduced what later became known as a General Circulation Model (GCM). GCMs have come to provide a compelling framework for coupling the atmospheric circulation to a great variety of processes. Although early GCMs could only consider a small subset of these processes, it was widely appreciated that a more comprehensive treatment was necessary to adequately represent the drivers of the circulation. But how comprehensive this treatment must be was unclear and, as Smagorinsky realized (2), could only be determined through numerical experimentation. These types of experiments have since shown that an adequate description of basic processes like cloud formation, moist convection, and mixing is what climate models miss most.
Full text at http://www.sciencemag.org/content/340/6136/1053.summary (paywalled)
=================================================================
The figure from the article shows how four different models have wide variances on clouds and precipitation.
on the climate model, even in the simplest model confi guration. Shown are changes in the radiative effects of clouds and in precipitation accompanying a uniform warming (4°C) predicted by four models from Phase 5 of the Coupled Model Intercomparison Project (CMIP5) for a water planet with prescribed surface temperatures.
Clouds and water are central to our global atmospheric processes, and clearly, these models aren’t doing much better than dartboards at figuring out what the real atmospheric score is.
With wide variances like that, no wonder climate models can’t model reality, from Dr. Roy Spencer’s recent post: STILL Epic Fail: 73 Climate Models vs. Measurements, Running 5-Year Means
For those interested in models, here are some additional documents, this time from ECWMF:
http://www.ecmwf.int/research/ifsdocs/CY38r1/IFSPart3.pdf
http://www.ecmwf.int/research/ifsdocs/CY38r1/IFSPart4.pdf
http://www.ecmwf.int/research/ifsdocs/CY38r1/IFSPart5.pdf
Start at Chapter 2 and read everything. Again, none of the fluffy, irrelevant junk from the Met Office, as provided in previous links in this thread…
As for ‘back radiation’, a part of climate alchemy that is supposed to explain its anti-science, it does not exist as any competent professional taught Maxwell’s Equations knows from 2nd year degree level. Yet it’s taught as fact in Meteorology and Climate Sciences.
So what exactly do you call the radiation that can easily be measured coming down from the atmosphere at any point and time on Earth, looking up, including at night when there is no sun actively heating said atmosphere?
Curiously, I’ve taught Maxwell’s equations at all levels from introductory through the graduate level, have an online textbook on Maxwell’s equations on my website, have rather a library of textbooks on electromagnetism, rather understand the Poynting vector and quantum electrodynamics (with a publication or two in the latter in Physical Review) and — gasp — when somebody shows me an upward looking spectrograph with a significant integrated power in the CO_2 + H_2O absorption band, I tend to conclude that back radiation not only exists and an easily derived theoretical truth, it really exists as a simple empirical truth.
What exactly do you think of Grant Perry’s “A First Course in Atmospheric Radiation” or Cabellero’s book on Climate Physics? I had no difficulty reading or understanding either one, and found no glaring errors in the physics of either one. Perry makes no claims whatsoever for CAGW or some particular value of climate sensitivity or feedback — that’s beyond the scope of the physics — but the science presented is AFAICT rock solid and the empirical evidence for the GHE effect in action presented in the book is (again IMO) conclusive to anybody who isn’t a complete idiot and understands (as you say) Maxwell’s equations and what the flux of the Poynting vector integrated over the spectrum stands for.
I’m asking these questions rather seriously. You claim to be an expert, sufficiently expert to speak for me (a Ph.D. theoretical physicist who teaches this stuff literally all the time — it is my profession). Yet I cannot imagine why “Maxwell’s equations” in any way prohibit or disprove the greenhouse effect, especially when an ordinary broad spectrum photometer (or better yet, an actual spectrometer) can measure the downwelling radiation you claim does not exist, in the latter case in bands that precisely match the absorption spectrum of the atmospheric species that you allege cannot be producing them.
So prove it.
rgb
With the growing trend of the jet stream moving south here in europe, then there is a real treat of the gulf stream moving south as well. Because l think its was this very set up over many years is the thing what caused the ice age in europe. Because as the jet stream moved south it drew the trade wind to the south as well. Which in turn made the gulf stream flow across the Atlantic further to the south.
in the Stefan-Boltzmann equation, thermal homogeneity of the globe favors warming — spreading the available heat budget around in the heat capacity of the globe to keep it all at a single temperature makes that temperature signficantly warmer than it would be if the distribution of heat/temperature were highly localized, e.g. confined to the tropics. Conversely temperature inhomogeneity (for an equal heat flow budget) favors cooling — hot spots radiate a lot more power than cold spots once scaling is taken into account.
in the luminosity. This would suddenly and significantly drop global average temperatures, causing the northern latitudes to resume glaciation (positive cooling feedback from increasing albedo) and probably would seriously affect growing seasons and zones in the polar latitudes both north and south. I’m guessing that the additional heat in the tropics would also increase tropical humidity and instability, increasing cloud cover and heat transport up through the lower troposphere to where it can radiate away (further increasing cooling efficiency).
This set up over a number of years would of brought climate cooling in europe for the ice sheets to expand south.
One of many nonlinear, discrete changes that could completely and suddenly reset the thermostat for the entire northern polar region and, in turn, the globe.
Because of the
Any diversion of the Gulf Stream south would effectively heat the tropics while ceasing to heat the northern latitudes. The hotter tropics would actually cool more efficiently than the northern latitudes do — doubly so because of the Jacobean of a sphere (a lot more area near the equator than the poles) and because of the
Sure, I could be entirely wrong. This is a semi-heuristic argument, and we have damn all data or computable theory to support it. But the point it makes still stands — the Earth’s climate system is enormously nonlinear, and simple changes in the patterns of circulation — quite possibly changes brought about by warming — can create enormous changes in heating or cooling efficiency in the dynamical system. There is a rather large mystery in the global climate record over long times — large fluctuations in mean temperature over decadal to century time scales that we cannot explain, fluctuations of 2-3 C either way, probably even larger than this when one accounts for the fact that most of our proxies average over an interval of years and hence cannot be properly compared to the high frequency contemporary record save by noting that their probable fluctuations increase the peaks and troughs by at least some percentage over the coarse-grained means. Perhaps this — or any of the other candidates for state/attractor changes in the coupled chaotic nonlinear oscillators associated with oceanic currents and heat flow or atmospheric charge/discharge cycles and heat flow — is an explanation.
So sure, this could happen (IMO, anyway). But will it happen? I have no possible way to even guestimate the probability, and we have no data that would help us understand if this is e.g. why the MWP ended or the LIA began. Or what the proximate cause (if there is any such thing) of a diversion in the jet stream or the Gulf Stream might be. It could be as subtle as a Brazilian butterfly or as overt and obvious as a hitherto misunderstood coupling between solar magnetic state and the atmospheric flow patterns.
rgb
Can’t we start throwing out models which have outputs that are too far outside of observations.
rgbatduke
One thing l think that would cause sudden cooling as least for part of the NH.
Would be if a southern tracking Polar jet that was over the ocean pushed northwards up and over the pole and came back down on the other side of the NH. That would draw a lot of moist air over the pole and cause cold and snowy on the other side of the NH.
It appears that this report on a paper in SCIENCE, is about the same paper as these same authors, Bjorn Stevens , director of the Max Planck Institute for Meteorology, in Hamburg, and Sandrine Bony, a CNRS senior scientist at the Laboratory of Meteorology Dynamics (in French) in Paris, that I just found in Physics Today for June 2013. In the PT paper, the authors thank Kerry Emanuel, Isaac Held, John Mitchell, as well as Peter Humbug, for their thoughtful comments.
They continue to assert the 3.45 K value for the climate sensitivity that CO2 doubling has been reputed to be responsible for.
Does anyone realize how bloody cold 3.45 K is. Can’t they simply say 3.45 deg. Celsius, so we know what they are talking about. The Kelvin scale of Temperature is an absolute scale.
As for models, and modeling, they are essential tools of science and engineering. Models allow you to perform experiments, with variable values, that no experimenter ever used or observed in practice. Often, those values are within ranges, where experiments have actually been performed.
But there is an a priori assumption, that the models’ simulations, actually replicate the results of real experiments.
I have done plenty of modeling in design. For example modeling down to the bare metal, and implant or diffusion profiles of doped semi-conductor layers, as laid out in Dr. Andy Grove’s textbook, on Semi-conductor Device Physics; literally designing my own devices from the physics, and layer processes offered by the silicon foundry, who would fabricate the wafers. That also had to include SPICE modeling of the actual circuit designs, whether simply logic gate switches, or more complex modeling of analog circuits; maybe some op amps or band-gap Voltage reference circuits.
These models had to perform accurately, because, it simply is way too expensive to build a modern integrated circuit, without SPICE modeling, including parasitic device effects telling you before hand that it was going to work properly. It is one thing to SPICE model an analog circuit, that is set to its nominal operational condition, within the range of its intended use.
But what the hell is it going to do, when you turn the power supplies on, and the whole thing is going to start from zero Volts everywhere. Why should it even start up at all, instead of just sit in a corner, and sulk; unless you repeat all the modeling over the power on transient conditions as well.
You have to design the circuit so that it is forced to go to a rational state, within the normal operational range. But, you still aren’t done with it. Now you have to make sure that the startup circuitry then gracefully exits stage left, so that it no longer has any influence on the functioning circuit. One of my earliest efforts in analog circuit IC design, failed ignobly, for the simple reason, I failed to assign some circuit elements, to forcibly wake the thing up, when the power turned on. These things have to raise themselves by their own bootstraps. The circuit itself, was truly wondrous, even to this day. Fat lot of good that is if it won’t turn itself on.
So a model, that will not definitively tell you whether the global Temperature will go up, or go down, when you double the CO2 amount; and moreover, how much it will move; is about as useless as an analog IC circuit design, without a startup circuit.
Every single IC I designed, that went to mask making, functioned properly , and with high good die wafer yields. Luckily, a colleague doing a CMA audit of my enhanced sleepiness circuit, discovered the fox pass before we went to tapeout, so the wakeup call could be added.
I don’t do that any more, I model optical systems instead; both imaging and non-imaging sensor optics, as well as both imaging, and non-imaging illumination optics. You might have one of them in your hand right now, if you still have an actual computer, instead of a video game.
So far, I have never had a modeled design, ever go to manufacturing, and fail to operate pretty much exactly as modeled. It is simply not feasible to build such stuff without successful modeling.
And when you have 13 or 19 or whatever, highly touted different models; no two of which agree with each other; and none of which agree with the experimental world view of what is purportedly modeled, then you are simply selling phraud.
“””””……rgbatduke says:
June 11, 2013 at 10:59 am
“As for ‘back radiation’, a part of climate alchemy that is supposed to explain its anti-science, it does not exist as any competent professional taught Maxwell’s Equations knows from 2nd year degree level. Yet it’s taught as fact in Meteorology and Climate Sciences.”
So what exactly do you call the radiation that can easily be measured coming down from the atmosphere at any point and time on Earth, looking up, including at night when there is no sun actively heating said atmosphere?….”””””
So Robert, I hunted, but couldn’t find exactly where this mutt latched on to your ankle; so I assume it is best not to look.
For the life of me, I can’t see just where Maxwell’s Equations “from 2nd year degree level”, would throw any light on the upness or down-ness of atmospheric LWIR radiation, to the point of excluding the existence of one of the two.
One thing is for sure, the atmosphere, including clouds, does not really “reflect” LWIR radiation.
Clouds in particular, consisting partly of non-gaseous phases, very strongly absorb surface emitted radiation, and then subsequently emit new LWIR photons, of an entirely new frequency, and more importantly, with an omnidirectional isotropic, angular distribution, which in the normal course of events, would mean that about half of it was emitted upwards, and the other half downwards.
Moreover, with the highest densities being closest to the surface, one might expect the atmospheric GHG absorption to be strongest, close to the surface, which would seem to warm the surface hugging atmosphere in assistance to the conductive warming of the near surface air. So the GHG absorption bands, would seem to enhance the upward convection of heated air, which would be an aid to cooling.
But how the downward “back radiation” half of the atmosphere/cloud re-emission of LWIR radiation got disappeared, and Maxwell’s equation’s participation in that travesty, is a real stumper for me.
Seems to me that ME is/are the ultimate survivors, since not withstanding the Bohr-Sommerfeld trashing of ME’s insistence that accelerated electric charges radiate, ME today survives healthy, and with the only exact constants, in all of physics.
I do hope you can shake that pest off your ankle.
Rud Istvan comments:
Indeed. As you can see in this presentation
http://on-demand.gputechconf.com/gtc/2013/presentations/S3429-FIM-NIM-Weather-Models-GPUs.pdf
Modeling has just turned the corner on 3-4 km grids for weather models on a global basis (good enough to start modeling clouds insteead of parameterizing them) by using GPU technology. NCAR’s CAM is also being ported to GPUs. Video games rescue climate science!
My suggestion is you guys read more and post less.
I post things that should BE OBVIOUS FOR YOU TO UNDERSTAND. If you need an explanation then back to kindergarten
“””””…..David vun Kannon says:
June 11, 2013 at 6:58 pm
Rud Istvan comments:
The probable reason for this inconsistency re clouds and precipitation ( neither new news) is inherent in the grid scale of all GCMs, itself inherent in supercomputer computational limitations.
Indeed. As you can see in this presentation
http://on-demand.gputechconf.com/gtc/2013/presentations/S3429-FIM-NIM-Weather-Models-GPUs.pdf
Modeling has just turned the corner on 3-4 km grids for weather models on a global basis (good enough to start modeling clouds instead of parameterizing them) by using GPU technology. NCAR’s CAM is also being ported to GPUs. Video games rescue climate science!……”””””
That’s impressive, so how much of the total surface area of the earth, are they making these measurements on a 3-4 km rectangular grid. That’s a lot of points to be making measurements at. I presume that they must have all their equipment interconnected some how, so that they measure every one of those points, at the same instant of time.
What would be the good of having measurements at a whole bunch of points, if they aren’t all taken at the same time. I know if you try to model currents flowing in a circuit by measuring the Voltage at a whole bunch of nodes, it isn’t worth a damn, unless they are all measured simultaneously.
There must be something about modeling climate, that I don’t understand.
AlecM says:
June 9, 2013 at 11:01 pm
How many more times does this grizzled old engineer who has measured coupled convection and radiation in metallurgical plants around the World have to state the bleedin’ obvious.
The Trenberth energy balance exaggerates lower atmosphere IR energy absorption by up to 6.85 times and there is net zero absorption of ~15 micron CO2 energy. This is then offset in hind casting by using double real low level cloud optical depth. It’s a scam by the ignorant, now supported by the fraudulent and the stupid.
The double counting begins with AGWScienceFiction’s Greenhouse Effect Illusion’s energy budget giving shortwave from the Sun, mainly visible, the property of longwave infrared, that is, heat, in order to promote the science fraud that no longwave infrared from the Sun heats the Earth, so they can use real measurements of any downwelling longwave infrared and attibute it all to “backradiation by greenhouse gases from the atmosphere” and not from the Sun.
They give two really absurd reasons for the longwave infrared not reaching us, but I’ve just worked out that what I at first thought from this, that they were double counting longwave infrared is actually not quite that, it’s worse than I thought, because I’d forgotten to account for the solar constant being all thermal infrared as I was fooled yet again by working to the figures from wiki which I hadn’t double checked.., so taking the solar constant as being a split of 53% infrared and 47% visible and uv. I finally remembered here:
http://wattsupwiththat.com/2013/06/10/dr-murray-salby-on-model-world-vs-real-world/#comment-1333574
So the Trenberth and ilk cartoon is still attributing all of longwave infrared to “shortwave in”, but that is actually the whole of the amount from the Sun in their cartoon, not part of it.
The earlier post here is where I made the error and there are links to Trenberth’s cartoon and the wiki piece: http://wattsupwiththat.com/2013/06/10/dr-murray-salby-on-model-world-vs-real-world/#comment-1333343
george e. smith says:
June 11, 2013 at 6:52 pm
“””””……rgbatduke says:
June 11, 2013 at 10:59 am
“As for ‘back radiation’, a part of climate alchemy that is supposed to explain its anti-science, it does not exist as any competent professional taught Maxwell’s Equations knows from 2nd year degree level. Yet it’s taught as fact in Meteorology and Climate Sciences.”
So what exactly do you call the radiation that can easily be measured coming down from the atmosphere at any point and time on Earth, looking up, including at night when there is no sun actively heating said atmosphere?….”””””
So Robert, I hunted, but couldn’t find exactly where this mutt latched on to your ankle; so I assume it is best not to look.
For the life of me, I can’t see just where Maxwell’s Equations “from 2nd year degree level”, would throw any light on the upness or down-ness of atmospheric LWIR radiation, to the point of excluding the existence of one of the two.
One thing is for sure, the atmosphere, including clouds, does not really “reflect” LWIR radiation.
rgbatduke – where are these measurements of downward longwave infrared at night?
And, all – I’m having a real problem understanding the information that has been put out on shortwave and longwave radiation, from that last link:
———
So, back to my question, what exactly is happening here?
http://www.physicalgeography.net/fundamentals/7i_2.html
This gives shortwave (July 83-90) as
“0 to 350 W/m2. Global mean = 158 W/m2, Minimum = 0 W/m2, Maximum = 323 W/m2. (Source: NASA Surface Radiation Budget Project).
Is this including or excluding the missing heat longwave infrared direct from the Sun of the GHE energy budget?
Or is that mean what shortwave alone would be in the solar constant divided by four?
Here’s another: http://www.physicalgeography.net/fundamentals/7i_5.html
This is shortwave and longwave combined, and I don’t understand it either:
“Figure 7i-7: Average net radiation at the Earth’s surface: January 1984-1991. Total net radiation is the sum of shortwave and longwave net radiation. It is dominated by the shortwave portion. Highest values occur along the subtropical oceans of the Southern Hemisphere. Lowest values occur over areas of low solar input such as the North Pole, and areas of high surface reflection such as the South Pole. Color range: blue – red – white, light green = 0 W/m2, Values: -50 to 250 W/m2. Global mean = 114 W/m2, Minimum = -60 W/m2, Maximum = 261 W/m2. (Source: NASA Surface Radiation Budget Project).”
“It is dominated by the shortwave portion” – how is it showing this?
What difference does “Lowest values occur over areas … and areas of high surface reflection ..”
Is this or isn’t this measuring what is reaching the surface? What does what happens next matter?
Here’s another: http://www.physicalgeography.net/fundamentals/7i_6.html
This again is for the same period as the shortwave only:
“Figure 7i-8: Average net radiation at the Earth’s surface: July 1983-1990. Total net radiation is the sum of shortwave and longwave net radiation. It is dominated by the shortwave portion. Highest values occur along the subtropical oceans of the Northern Hemisphere. Lowest values occur over areas of low solar input such as the South Pole, and areas of high surface reflection such as the North Pole. Color range: blue – red – white, light green = 0 W/m2, Values: -50 to 250 W/m2. Global mean = 111 W/m2, Minimum = -65 W/m2, Maximum = 249 W/m2. ”
Again, how are they showing that shortwave dominates, and, why are the figures for combined less than shortwave alone?
——-
I should be grateful if you would take a look. Am I missing something?
@george e. smith – if you look at the slides I linked to, you’ll see that the grid is a grid of hexagons, not rectangular. The time step on a 10 day forecast is about 2 minutes. Also, a global run takes 50% of real time to complete, so a climate study of 100 years would take 50 years to run! But these models running on massively parallel GPU based systems are already helping weather forecasting.