Guest Post by Willis Eschenbach
Since the late Nineties the US has had seven industrial-strength stations that measure a variety of climate variables every minute, 24/7. These are called “SURFRAD” stations. As a data junkie I’ve been wanting to look at their results for a while … but the data is in an ugly format. They have a single data file for each station for each day of the last 17 years … not my idea of a party.
Anyhow, I finally bit the bullet and downloaded a year’s worth of data, about a quarter of a gigabyte. For no particular reason I picked the SURFRAD station in Goodwin Creek, Mississippi, and the year of 2010. For each minute they have no less than 21 different measurements (see end notes) … so I sorta started digging around in the data to see what stuck out. Here was the first oddity I came across:
Figure 1. Average 10-metre surface air temperature (black, °C) and average downwelling infrared radiation (blue, W/m2) for the year 2010. Measured at the SURFRAD station in Goodwin Creek, Mississippi. Average covers the entire year, and is shown repeated twice (two days) for clarity.
I don’t know why, but I wasn’t exactly expecting that … which is the best part of science. I love surprises, the unexpected, and climate science is chock full of those. I mean, I knew that downwelling radiation was a function of air temperature … I just didn’t expect the alignment with the underlying surface temperature to be so exact. Other than the atmosphere starting to cool a bit earlier in the day than the ground (as we’d expect from the relative masses) they match up perfectly.
Now, seeing how good that match was, I got to wondering how well that fits the theoretical profile that we’d expect from the Stefan-Boltzmann (S-B) relationship. This relationship says that infrared radiation is equal to emissivity times the Boltzmann constant times the temperature to the fourth power. I figured that using that formula, I could calculate an approximate value for the emissivity from the data with a simple linear analysis.
Now, here’s the curious part. When I did that, I got an emissivity of 0.590 … which from everything I’ve read is too low.
So I thought, well, that kinda makes sense, because the temperature up where the radiation is coming from is cooler. But how much cooler? That depends on what altitude the radiation is coming from. Now my bible in these matters is “The Climate Near The Ground”, by Rudolph Geiger, which anyone interested in climate science should read. Geiger gives the following table for downwelling radiation (called “counterradiation” in those days):
Table 5-1 Contribution of various atmospheric layers to counterradiation received at the surface Layer thickness (m) % share of counterradiation 87 72.0 89 6.4 93 4.0 99 3.7 102 2.3 108 1.2
I figured that I could use that to give me at least a first cut at the temperature of the overlying atmosphere at altitude, using the lapse rate of one degree C per each hundred metres of altitude. For the six layers given by Geiger, this gives mid-layer temperature drops of 0.4°, 1.3°, 2.2°, 3.2°, 4.2°, and 5.2° degrees C. A weighted mean of these (allowing for the fourth power relationship) gives an average temperature drop of 0.85°C. This makes sense, because about three-quarters of the downwelling radiation comes from the bottom hundred metres of atmosphere, which is not much cooler than the surface.
However, this doesn’t solve the conundrum. Remember that I got an emissivity of 0.590 using the surface temperature. IF in fact on average the radiation is coming from a temperature which is 0.85°C cooler, then using that temperature it only brings the emissivity up to 0.595 … hmmm.
So that’s my puzzle for today. Is Geiger wrong about the source of the downwelling radiation? Is the emissivity of the atmosphere really on the order of 0.6? Is something else going on?
Inquiring minds wonder …
My best to everyone,
w.
AS USUAL: if you disagree with someone, please quote the exact words you disagree with. This lets all of us understand the exact nature of your objections.
CODE AND DATA: The R code, the functions, and the hundreds of daily files for 2010 are in a zipped folder called “SURFRAD Analysis”. WARNING: 21 megabyte file.
{UPDATE] Prompted by a typically detailed and interesting comment below from Dr. Robert Brown (rgbatduke), here is a scatterplot of the complete temperature and downwelling IR datasets:
[UPDATE 2] The same graph, but for Boulder, Colorado.
SURFRAD VARIABLES:
# station_name character station name, e. g., Goodwin Creek
# latitude real latitude in decimal degrees (e. g., 40.80)
# longitude real longitude in decimal degrees (e. g., 105.12)
# elevation integer elevation above sea level in meters
# year integer year, i.e., 1995
# jday integer Julian day (1 through 365 [or 366])
# month integer number of the month (1-12)
# day integer day of the month(1-31)
# hour integer hour of the day (0-23)
# min integer minute of the hour (0-59)
# dt real decimal time (hour.decimalminutes, e.g., 23.5 = 2330)
# zen real solar zenith angle (degrees)
# dw_solar real downwelling global solar (Watts m^-2)
# uw_solar real upwelling global solar (Watts m^-2)
# direct_n real direct-normal solar (Watts m^-2)
# diffuse real downwelling diffuse solar (Watts m^-2)
# dw_ir real downwelling thermal infrared (Watts m^-2)
# dw_casetemp real downwelling IR case temp. (K)
# dw_dometemp real downwelling IR dome temp. (K)
# uw_ir real upwelling thermal infrared (Watts m^-2)
# uw_casetemp real upwelling IR case temp. (K)
# uw_dometemp real upwelling IR dome temp. (K)
# uvb real global UVB (milliWatts m^-2)
# par real photosynthetically active radiation (Watts m^-2)
# netsolar real net solar (dw_solar – uw_solar) (Watts m^-2)
# netir real net infrared (dw_ir – uw_ir) (Watts m^-2)
# totalnet real net radiation (netsolar+netir) (Watts m^-2)
# temp real 10-meter air temperature (?C)
# rh real relative humidity (%)
# windspd real wind speed (ms^-1)
# winddir real wind direction (degrees, clockwise from north)
# pressure real station pressure (mb)
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
The emissivity off the atmosphere have allways been 0,598-0.6, it never changes and thats why more co2 doesnt raise temperature.
Thanks, Willis.
Interesting. Needed to re-read it twice. These are a sample of two. Normally if I saw the picture on the net I’d imagine “ahh, clouds!”
The S-B formulae gives the temperature at which the radiation is generated. The radiation at max. from your graph (365W/m2) gives a temperature of +10C, some 12C lower than the surface (10m0 level). With a standard lapse rate this is over 1000m above station level.
The actual fornulae ha another term on the right, e, but not knowing the material emitting the IR this term is unknown. Without e a black body emission is calculated.
Q=sigma e T^4 T being the absolute temperature. sigma 5.67^-8
Downwelling solar also needs to be in this picture. Then upwelling solar and upwelling IR and then net radiation in and net radiation out versus air temperature
I spent some time looking at the SurfRad station at Table Mountain Colorado. There are a number of radiation flows one needs to look at. And there is a really important piece missing from the measurements – how much net energy (joules) is the ground below the instruments absorbing/releasing as a 24 hour day goes by.
Generally, the surface air temperature is changing by extremely miniscule amounts per second (I mean like 0.002 joules/second) while very high radiation levels are flowing in and flowing out every second. The ground surface could be the shock absorber keeping the air temperature so steady while 800.000 joules/second is coming in from the Sun in mid-afternoon. Either that or the energy is flowing out as fast as it is coming in. Note one can see the influence of clouds (during the day and during the night).
Note that Upwelling IR is always “higher” than Downwelling IR which means this is really misdirection by the warmers. Downwelling IR should not heat up the air because more is actually Upwelling every second than is Downwelling. It should be thought of as a parcel of air where energy is constantly going up and down and all around as would be expected given this is a gas and there are photons constantly travelling through all gases all the time 24/7. Generally, IR energy is always moving up and out of this air parcel all the time. Sun comes up, solar radiation comes in and energy levels increase and the energy flows back up the air cloumn back to space as IR (every second 24/7).
http://s18.postimg.org/s66p367ft/Table_Mountain_All.png
Bill, Willis,
I live at 41N, just south of Lake Erie. I routinely measure sky temps 80-100F colder that ground temps, -40 to -60F or colder. Humidity reduces the difference between the surface and sky temps (as do clouds). So Willis I think the Mississippi Down welling IR is a large part from water vapor, While you see Bill’s Colorado station is much lower, which I believe is due to differences in water vapor between the two locations.
Bill, I think if you could get an IR read of the ground, you’d see air temps track the ground, with all the caveats. Like grass because it traps air, acts as an insulator, and on cold clear night is 10F colder than the ground and has frost, while the ground is still over 32F. Air temps over 20F grass is colder than air over 32F concrete. But I’ve also noticed 5F to 10F / hour cooling rates after sunset, drops to under a couple degrees/hour once rel humidity gets up over 80-85% as the air cools (which this wringing the water out of the air is where the moisture for dew and frost comes from).
This is why a change in Co2 has so little effect. When you clear the water out, it has 2-4F difference in clear sky temps, they coldest they get. Clouds are 40F to +70F warmer than clear skies. Water vapor adds from a few degrees to 80-90F to the clear sky temp. Because the difference in Miss and Colorado DW IR isn’t from Co2, it’s remained the same.
Mi Cro November 25, 2014 at 8:05 am Edit
Thanks, Mi, but presumably that’s using an IR thermometer which ASSUMES an emissivity … whereas I’m trying to calculate the emissivity from a known (or at least estimated) temperature.
w.
I understand. I had the same question (what e should I use, since I can change it). I found a study that said ~.7 (.72? so you’re not so far off). It was cold that night, so I had been measuring sub -60F temps, dropped the e down to .7 and it read somewhere closer to -100F.
I decided I could be more conservative, leave e at .95, it might very well be reading quite warm, but I figured it was still cold enough to show how little Co2 could possibly be contributing even being overly generous.
Isn’t the really important question the radiation over the ocean in the tropics, and how this varies with cloud. I’d like to sea something like this with also a measure of cloud directly overhead.
The really good part for all of us is that we can put all that evil “back radiation” to work as there is well over 200 watts/m² available 24/7 coming in from our atmosphere. All we need to do is tweak the doping for solar cell absorption to cover these wavelengths and we never have to worry about power again.</sarc>
That’s an understatement. Poked around on NOAA’s SURFRAD as well and the sensor they’re using for IR is so broadband it goes all the way out to 50µm… a considerable way into FIR. It also uses Parson’s black which NOAA has had problems with in the past see Section 3 of this NOAA Tech Memo here.
Bill Illis,
If the rate of outgoing radiation is decreased but incoming remains constant, what happens to temperature?
Brandon Gates:
In the middle of the Sahara, where the GHE is the least, due to the dryness, air temperature is the highest. In the humid tropics, where the the GHE is greatest, air temp maximum is 20°F to 30°F lower than the Sahara.
Try and explain that one in terms of your “down welling” IR.
Day time highs are predominately due to Solar, and the amount of moisture available at the ground. In this case with little moisture the Sahara warms quickly to a higher temp.
Now night time cooling is the domain where DWIR helps maintain temps, but again you see that in the case of your two locations, Co2 doesn’t add much to the IR from all of that water vapor in the tropics, and when you take the water vapor away in the Sahara, there’s so little energy from Co2, temps drop like a rock.
Micro:
The principle that I was trying to illustrate is this: the regions of stronger GHE have lower maximum temps.
This was in response to Brandon’s comment wherein he implicated higher air temps to DWIR. My observation compares low GHE (Sahara) to high GHE (tropics). The lower GHE (Sahara)
with the least DWIR (theoretically), has the highest temp. Brandon’s reasoning is the sort of screwy thinking that permeates climate science.
The next time I go sun-bathing, it can have a new name – “out to catch some downwelling solar”.
Doesn’t sound quite so lazy.
Out of my comfort zone here but I wonder how accurate you expect the surface temperatures to be when contrasted with the lapse rate of one degree C per each hundred metres of altitude?
I’m guessing the accuracy at the surface is finer than the approximation with altitude.
Thanks, M. Normally I would have only reported the emissivity to one decimal … but if I did that, the difference in emissivities from the two different temperatures wouldn’t be visible, as they only differ in the third decimal point. I wanted to point out how little difference the adjustment in temperature made to the emissivity.
w.
Thanks for the response.
That makes sense – I’m sorry I missed what you were saying with that accuracy.
Told you I was out of my comfort zone.
Willis, the Stefan-Boltzmann formula applies to a blackbody, or to a grey body if emissivity is taken into account. But in either case, the emissivity must be the same for all wavelengths. This is not true of the atmosphere. The atmosphere only radiates because of the presence of radiatively active gases. These do not cover the full spectrum and so the downward longwave radiation does not match a plank distribution. The atmosphere does not behave as a blackbody.
http://scienceofdoom.files.wordpress.com/2010/04/longwave-downward-radiation-surface-evans.png
There are large gaps in the spectrum, especially in the ‘atmospheric window’ between 8 and 12 microns. The Stefan-Boltzmann formula is not valid in this context.
.
Tsk, I should have got that.
You are right, MikeB.
“The atmosphere only radiates because of the presence of radiatively active gases”
Not true as stated. Somewhere north of one half is from liquid water droplets in clouds (60% coverage) whose radiation is much closer to full spectrum, but likewise not strickly a graybody or blackbody so the Stefan-Boltzmann formula is not valid in this context either. Though radiatively active, I just wouldn’t call these droplets a gas so your statement is misleading and lacking.
Well Wayne, , you are right that water droplets in clouds will radiate but these are not normally considered to be part of the atmosphere. I don’t really want to diverge into semantics here, but atmosphere is defined as a layer of GASES surrounding a planet. In normal usage It does not include liquid water because liquid water is, ipso facto, not a gas.
Saved me a reply. The S-B formula is sort of valid, you just have to integrate. That’s what Willis was trying to do by considering height, but technically you have to integrate height against transparency and temperature, as the mean free path of IR photons varies with frequency. On the other side of things (escaping photons) you have the same problem, only even more complex as pressure broadening means that heat “leaks out along the edges” of the lines as they sharpen at lower pressures at higher heights, narrowing the absorption bands above the emission bands below.
And don’t forget clouds. Some fraction of your downwelling energy is specular reflection from clouds, and hence is blackbody radiation from the ground in the holes where the atmosphere is transparent enough to permit wavelengths to reach the clouds to be reflected and come back.
The concept of “emissivity” under these circumstances isn’t to be taken too seriously as you are looking at something that is far from being a black body — wrong spectrum, partially transparent, partially reflective (in a time varying way) and sampling from a frequency-dependent range of temperatures, not a single temperature. This is why the problem is so complex. The errors in our ability to quantitatively evaluate what is going on are larger than the the effects we are trying to predict or understand, so that any small thing that we leave out of the dynamics might suffice to make our tentative beliefs incorrect.
Consider. You have a lovely graph for a single station. How universal is the result you plot? If the station were right next to a lake, would it be the same? How about a station in the heart of downtown or the middle of a shopping mall? How does a forest station compare to one in the middle of a plowed field. How does the plowed field fare in the winter time fallow or with a cover crop, vs freshly plowed in the spring, vs covered with barley or brocolli in midsummer? How does all of that compare to a station in mid-Sahara, in the Siberian marshes, on the Tibet plateau, in central Africa, in the middle of Antarctica? Does the curve for the station change from year to year, is it fundamentally different in fall/winter/spring/summer increments, and above all, what is the standard error? What is the variance in the sample data? The curves you have are remarkably smooth, but the underlying data could not possibly be that smooth and should show considerable seasonal variation.
rgb
I’ve been measuring zenith temps with an 8u-14u IR thermometer, since the 15u-16u Co2 bands are invisible, I’ve started adding 3-4W/m^2 back into what the thermometer is reading.
I do presume that to the thermometer 8u-14u does look like a “blackbody”, and at worst it’s measuring water vapor and then whatever temp the gases are at.
At 41N, when it’s cool with low humidity, Tzenith is easily 80F colder than my concrete sidewalk, and many times it’s over 100F colder. Clouds are always warmer than anything else, low puffy clouds are only 10-20F colder than the surface, high thin clouds might be 10-20F warmer than Tzenith, but they are always far warmer than humidity, which is far warmer than dry air.
So I’ve come to see the entire surface radiating huge amounts of energy to a very cold sky 24x7x365, with a blazing torch shining part of the time. And as soon as the torch moves on the surface starts to cool, but at almost no time during the nightly cooling process is the ultimate limit from Co2, it’s from water vapor.
@ur momisugly RGB
I rarely disagree with anything you say, but I take issue with “Some fraction of your downwelling energy is specular reflection from clouds, and hence is blackbody radiation from the ground in the holes where the atmosphere is transparent enough to permit wavelengths to reach the clouds to be reflected and come back.”
Water droplets that make up the clouds are pretty close to blackbody absorbers/emitters. As such, they would absorb nearly all the IR coming up from the ground (not reflect it). They would emit *their* *own* blackbody thermal IR. So the spectrum should be at the temperature of the *clouds* not the temperature of the ground.
PS Here is a link to a site that uses MODTRAN (http://forecast.uchicago.edu/modtran.html) to give the calculated spectrum at various conditions. It is well worth exploring/bookmarking for anyone who wants to understand atmospheric IR. For example, it pretty well recreates the spectrum posted by MikeB if you look upward from the surface with mid-latitude winter conditions.
Sorry Tim, I stand corrected. I was thinking of the visible spectrum and their high albedo. I misplaced my copy of Petty’s book this summer and that’s my usual bible for checking things before I post them, as I mistrust both my own memory and my understanding.
rgb
“””””…..And don’t forget clouds. Some fraction of your downwelling energy is specular reflection from clouds, and hence is blackbody radiation from the ground in the holes where the atmosphere is transparent enough to permit wavelengths to reach the clouds to be reflected and come back……”””””
I don’t think clouds specularly reflect LWIR. Since clouds are water in liquid or solid form, they are strongly absorbing of LWIR radiation. Then they isotropically re-emit BB like radiation depending on the water temperature.
At solar wavelengths, water is transparent, but once again they only weakly reflect (about 2-3% reflectance). Mostly the droplets strongly refract the solar spectrum wavelengths, and convert the near collimated solar beam into a strongly focused beam, which then strongly diverges. Just a few drops in sequence and you get a complete wide angle scattering of the solar spectrum, which is not really specular reflection. The normal reflection coefficient for water (N = 1.333) is 2%
As for BB emission from the “atmosphere”, I thought gases could not absorb and emit thermal radiation spectra. Excuse me; that’s I thought people claim that gases don’t emit in the infrared (non GHGs)
CO_2 acts as a coupling between the atmospheric thermal reservoir and local radiation. In local thermal equilibrium (to the extent that the concept is valid in an open system conveying energy) the GHG molecules collide with the N_2 and O_2 molecules rapidly enough (as I noted) that they remain in equilibrium. The CO_2 is also coupled to the local radiation field (coarse grained over a much larger volume — mean free path order of a meter to meters). Following Kirchoff, it both absorbs radiation (instantly thermalized with the rest of the local atmosphere) and emits it (generally as energy picked up in collisions with the local atmosphere). As always, the directions of the average energy flow in the interactions is to try to establish detailed balance where local equilibrium is satisfied — the atmosphere warms if the local radiation field is warmer, cools if it is cooler, the atmosphere warms the local radiation field if it is cooler and cools it if it is warmer. This is on top of the adiabatic lapse rate, hence there is active transport of heat via radiation from lower horizontal slices to higher adjacent horizontal slices up to heights where the mean free path of in-band photons reaches “infinity” in the upward direction (becomes larger than the optical thickness of the remaining atmosphere).
Because of pressure broadening, each layer is slightly “leaky” — it emits from a slightly wider spectrum than the layer above it absorbs, so that radiation from the edges of the bands has a differential mean free path. This means that it doesn’t take all of the radiation in the surface bands the same amount of time to reach escape height, and that height itself varies with frequency. It is doubly complicated because there are multiple greenhouse gases with different absorption/emission bands. Still, the emitted in-band radiation follows a general intensity profile where the emission peaks roughly correspond to the local temperature of the emission height, see the many figures in Petty’s book or reproduced on WUWT by the article here:
http://wattsupwiththat.com/2011/03/10/visualizing-the-greenhouse-effect-emission-spectra/
which has my vote as one of the best all-time posts on WUWT. Note well that top of atmosphere spectrographs looking down and bottom of the atmosphere looking up are complements of one another, and that there is a clear relationship between the “blackbody temperature” of the emission height in-band for the primary gas(es) concerned and the spectral intensity.
To a physicist, these spectra are direct evidence of the greenhouse effect. That doesn’t make the differential radiation trapping easy to compute in a nonlinear chaotic double Navier-Stokes system evaluated on a rotating, tipped, oblate spheroid 70% covered with water and with a highly differential land surface height and character in an eccentric orbit around a rather variable star, but it leaves little doubt as to the probable average effect of increasing its atmospheric concentration.
rgb
Mi Cro — I think you are missing a lot more than 3 or 4 W/m^2. Measuring zenith at 38N here, I routinely see about -45C with an IR thermometer which is specified to cover 8-14um. However, if I make the same measurement using an IR thermometer that covers 5-20um then I get something around -21C. I have not worked out how much power is in the band from 14 to 20um but just from a visual estimate it appears to be quite a lot.
Mistake in that last post – I am seeing 7F on the 5-20um IR thermometer, and incorrectly converted that to -21C…it should be -14C.
I wonder if Miriam understood any of that?
MikeB
I think your Figure 1 is for dry air. H2O vapour is missing and would overwhelm the CO2 peak. As it says, “several greenhouse gases,” but not the most important one.
what MikeB said. The atmosphere is partly transparent in the IR, which causes the effect you are seeing.
Not to impose upon Dr. Spencer, but, although I knew vaguely that was the answer, I was hoping someone could give a black-body-for-dummies explanation of how solid-body interactions tend to smear the spectra out.
I believe many of us laymen continue to visit this site in the hope of finding such nuggets from time to time, and just sift through all the stories ridiculing the warmist dreck that issues with depressing reliability from the press and academe.
What Roy Spencer said about what MikeB said.
Willis, measuring the sky with an IR meter is more like measuring the temperature of a black-painted perforated metal sheet than a gray body. The gray body is an analogy giving an ‘average’ emissivity of what is really a ‘holey plate’. A highly emissive molecule of H2O is effectively ‘black’. But they are speckled/distributed through the air. The emissivity you calculated of about 0.6 is equivalent to saying you have a solid surface with an emissivity of 0.95 but at a lower temperature. The calculations show the total ‘back radiation’ but it is not from a homogeneous surface.
Regarding the water droplets, air has a lot of water droplets in it – they can be seen easily by illuminating them with a powerful handheld green laser pointer. Try it at night in cooling air. Water droplets are powerful emitters of IR and also reflect low incident angle radiation. The emissivity of water is approximately the same as black oil.
There is little point in discussing back radiation without also discussing the water vapour level. We can assume CO2 is evenly distributed by altitude but that is certainly not the case for water vapour. When water vapour is only 5 g per standard cubic metre, it is going to emit 10 times the IR of CO2 at 400 ppm.
Willis you say the air temperature cools quicker than the ground, but isn’t the air warmed by the ground not by the sun in which case the air should cool with the ground?
Regards kelvin
I’ll second that query
Note that the air temp correspond only to “downwelling” IR, as given in the scale of the graph, when in fact it is a flux of no particular direction. In this the jerks at NOAA bias their product.
In fact, it would be expected that air temp should correspond closely to the IR flux. No surprises here.
Also to ground temp., the ground being the source of the IR.
Yes and no. Because the atmosphere almost instantly thermalizes absorbed IR radiation, transferring the energy to the molecular KE of the surrounding air in a tiny fraction of the radiative lifetime, yes, the ground and air often/usually track one another pretty closely — there is rapid transfer when they aren’t in equilibrium.
However, it’s a dynamical system, with extremely variable heat capacity in the different parts. Cold air/warm air moving from one place to another can easily disequilibrate any given location. Given that 70% of the surface is water with a mix of latent heat and huge heat capacity and surface turbulence…
What would be really, really interesting would be to cover even one small selected patch of reality with a dense set of stations like this. For example, take Durham county (where I live in NC). Roll accept/reject random numbers to select (say) 10,000 sites all over the county (including in the middle of lakes or rivers or on the tops/sides of hills if that’s where the random gods will). Distribute specially designed stations with IR eyes that point up and down in addition to the usual thermometers, wind/rainfall-snowfall/pressure/humidity sensors on those points within (say) 10 meters (even when they aren’t conveniently located or might sample e.g. parking lots or asphalt rooftops). Monitor for a decade. In fact, since we’re going to the trouble, let’s go ahead and add a hundred blimps, similarly randomly located in 3d (including random heights, sampling full spectra facing both up and down, local temperature, wind, and so on). Or do full soundings at those sites 12x a day.
We might actually learn enough to have an idea of what “average temperature” could sensibly mean. We might learn about heat transport. We might be able to inform models, instead of just guessing and filling in whatever we think might be true.
rgb
Dr, Brown:
Excellent suggestion. For a fraction of the money squandered on miserably failed, GIGO models, real climate science could be advanced by acquiring more and better relevant data. We need more observation and fewer models, ie none at all yet, in the primitive state of our understanding.
rgbatduke
November 25, 2014 at 9:40 am
” …. What would be really, really interesting would be to cover even one small selected patch of reality with a dense set of stations like this…. (including random heights, sampling full spectra facing both up and down, local temperature, wind, and so on)…..”
===========
That would settle a lot of assumptions now being made with regards to models.
In consideration of Willis’ review suggesting that the lower region of the atmosphere may be responsible for ~ 70% of DWIR, a relatively inexpensive gathering of data from a couple of points (geographically) utilizing perhaps 4 or 5 small planes crossing over a given point at various elevations within a short period of time measuring all the parameters you stated (i.e. IR up, IR down, temp, RH/dew point, etc) could yield a valuable trove of information.
This could possibly be done using a string of balloons if the FAA would permit.
I would be surprised if this hasn’t been done.
eyesonu:
Such measurements have been made already at a number of instrumented towers scattered around the world. Miskolczy has published analysis results for, IIRC, the tower maintained by KNMI in the Netherlands. Naturally, because they lead to conclusions contrary to academic dogma, his work has been excoriated by the high priests of AGW.
Willis says: “Other than the atmosphere starting to cool a bit earlier in the day than the ground (as we’d expect from the relative masses) they match up perfectly.”
I’m not a meteorologist and i have never lived in that area, but another explanation would be fewer clouds during the afternoon. As others have pointed out, much of hte back-radiation comes from clouds, so clear skies would radiate less at the same temperature.
Perhaps the data also includes cloud cover, and you could see if is it less cloudy when the IR curve drops below the temperature curve.
Reblogged this on Centinel2012 and commented:
Good work — I was not aware of these stations!
Air temperature is not only due to radiative heating, but also conduction directly from the surface to the air, as any amateur astronomer knows. Turbulence over buildings destroys the quality of images, and they are to be avoided. It’s not like the building is venting hot air, it’s the roof and windows heating the surrounding air by contact and causing the heated air to rise. Similarly, any pilot knows heated air rises from the ground, and if it contains sufficient moisture, a cloud may form, and a big enough cloud may begin to rain. So if you want to find updrafts, go from cloud to cloud (VFR, staying underneath).
I assume the instrument measures infrared along its entire frequency? But to properly understand the post and to educate myself I must ask. What are the outer bounds of the instrument in measuring incoming IR?
I have answered my own question. From the http://www.esrl.noaa.gov/gmd/grad/surfrad/surfpage2.html website:
“Radiation measurements at SURFRAD stations cover the range of the electro- magnetic spectrum that affects the earth/atmosphere system. Global solar and its components are measured separately. Total downwelling (global) solar radiation is measured on the main platform by an upward looking broadband pyranometer. The direct component is monitored with a normal incidence pyrheliometer (or NIP) mounted on an automatic sun tracker, and the diffuse component is measured by a shaded pyranometer that rides on the solar tracker. Diffuse solar was not in the original suite of SURFRAD measurements. The shaded pyranometer was added in 1996 when a support platform with a shade arm mechanism was fitted to the trackers. A third pyranometer is mounted facing downward on a crossarm near the top of the 10-meter tower to measure solar radiation reflected from the surface. An upward looking pyrgeometer on the main platform measures long wave (thermal infrared) radiation emitted downward by clouds and other atmospheric constituents. Another pyrgeometer, mounted facing downward on the crossarm atop the tower, senses upwelling long wave radiation. These measurements of upwelling and downwelling in the solar and infrared wavebands constitute the complete surface radiation budget.”
Odd that it matches linearly. Shouldn’t it require a log scale?
Ln(1+x) = x + …..
Looks linear enough to me.
of course that’s x _x^2 / 2 + x^3 / 3 _ etc
x – (x^2)/2 + (x^3)/3 –
I’m talking about that the scale on the left for temp (C) and the scale on the right for radiation (W/m^2) are both linear scales. Assuming that P = aT^4, isn’t anyone bothered by the linear scale on the axes? What am I missing?
Over the interval in question, there’s not much difference between the linear and the T^4 variability.
w.
I think I just said what it is you are missing. And if not, then Willis just said it here too.
But in case it is of interest to anyone, I recently did a calculation of a simple model, where the BB radiation of a static T = To body is compared to the same body at the same AVERAGE Temperature To, but actually having a cyclic Temperature (sinusoidal), where T = To.(1 + k.sin (2pi t/tau))
Integrating the total radiant energy emitted during a complete cycle of the Temperature cycle:
The static case gives E = sigma. To^4 tau j/m^2
The cyclic case gives E = sigma. To^4 tau. (1 + 3k^2 + k^4 / 8)
The k^4 / 8 term is rather negligible.
For k = 0.1, the AC case gives 3% higher total black body energy emission, so such a body would cool faster, and would have a lower (not much) average Temperature, than if the Temperature was absolutely constant.
This is what Dr Svalgaard disputed, and is the reason, I don’t accept 342 W/m^2 as the TSI for planet earth. It really is 1360 whatever.
But the difference is small enough that linear scales suffice as Willis used.
(21+273)^4 / (12+273)^4 * 330 = 373.7
Graph shows about 360 @ur momisugly 21 C. Seems like a significant difference to me.
Willis
Sorry for not quoting your exact words, but I have a general enquiry.
You often comment along the lines of how much warmer it is when it is a cloudy night because clouds reradiate downwards more LWIR. How does that proposition fit in with your observation that “three-quarters of the downwelling radiation comes from the bottom hundred metres of atmosphere…”.
If three quarters of all DWLWIR comes from the bottom 100 metres of the atmosphere (which I do not challenge), then how do high level clouds at say 2,000 to 6,000 metres make a substantial difference, save other than reduce heat loss by restricting convection.
Or because as the atmosphere cools water vapor changes state to liquid on cloud condensation nuclei and radiate the latent heat of state change. There is no need for ‘reradiation’ at all the heat was up there all the time.
Ian W,
I have given much thought along the lines of your comment for quite some time.
One other thought that keeps nagging me is also the increased specific heat contained in the air that contains a likelihood of more water vapor in cloudy conditions as opposed to dry clear conditions.
I don’t have the ability to properly explain/express that as well as others here.
==========
@ur momisugly Bill Illis (November 25, 2014 at 5:58 am)
Your graphic is very interesting.
===========
Anyway, this thread is going to provide some very interesting comments and discussion.
richard verney November 25, 2014 at 6:04 am
richard, I have always assumed that it is because the downwelling IR from the clouds is absorbed and re-radiated from the lower levels … but I’m up for all suggestions.
w.
It seems to more closely track the surface, but I’ll have to scan across a cloud bottom with the Sun behind it to see if it changes or not.
Just above Dr. Brown said that IR absorbed in the atmosphere thermalizes. Absorbed and re-radiated almost never happens in the lower atmosphere, only much higher and closer to, if not at, the Top of Atmosphere. Downwelling is indeed a misnomer, as the entire atmosphere radiates in all directions at all times.
Dr. Brown also said that clouds “spectrally,” which means like a mirror (I had to look it up) reflect IR from the Ground in the Holes where the ground can radiate directly to space, these holes being in the IR spectrum from 8 to 12 microns. It is all so simple now…
The condensation of the water vapour at, say, 2000m forms clouds and releases a great deal of heat: latent heat of condensation. That heat thermalizes the surrounding….water vapour! The air has so much water vapour that is why it is condensing in the first place. It is saturated. Clouds are surrounded by water vapour, not dry air.
That water vapour, now warmed by the latent heat released by the condensing water vapour, radiates in the IR band, cooling as it does so. The cloud water droplets are themselves IR-radiative with an emissivity of about 0.98. Their behaviour in the IR band is completely different from their behaviour in the visible range. All that IR radiation goes up (to be intercepted by the clouds and sent down again) or directly down to the ground. Obviously there is leakage at the ‘edges’. The net effect is for clouds to form an insulating blanket much more real than the misnamed ‘greenhouse effect’ claimed for CO2. These water and water vapour effects overwhelm any small variations in CO2 concentration of a few hundred ppm.
“””…Dr. Brown also said that clouds “spectrally,” which means like a mirror (I had to look it up)…”””
No he did not.
He said “specularly”, which is not “spectrally”.
“richard, I have always assumed that it is because the downwelling IR from the clouds is absorbed and re-radiated from the lower levels … but I’m up for all suggestions.”
====================================================
This brings up many questions. The lower atmospheric levels GHG molecules are emitting not just LWIR from the ground I think, but conducted and convected energy as well, from the thermal dynamic equilibrium of exchanged energy through conduction, with collisional exchange of energy far more common throughout the lower troposphere, correct?
If this is correct, then is not much ? some? of the 50% of the energy sent to space from emitting GHGs, from non- GHG molecules, which if they had transferred their energy to other non-GHG molecules instead of GHG molecules, would still be in the atmosphere, thus the affect of the GHG molecule was to radiate away conducted energy, thus cooling through shortening the residence time of energy from non-GHG molecules.
????
I think I am looking for a ratio regarding how much of the energy GHG molecules radiate to space is from conducted energy, vs LWIR from the ground?
Let me further clarify my question. I think I am looking for a ratio regarding how much of the energy GHG molecules radiate to space is from conducted energy – convected energy – evaporative condensed energy vs LWIR from the ground?
(It appear logical to me that if said energy from the first three sources conducts to a GHG molecule, it has a 50 percent chance of accelerating the loss of said energy to space vs. that same energy conducting to a non GHG molecule where the chance of it escaping to space would be 0.)
I suggest you calculate dew point from relative humidity, pressure, and atmospheric temperature. Then apply the S-B calculation. I suspect the measured radiation is from the condensation/evaporation process with possibly a relatively small amount from CO2 at the same temperature. NOAA/ESRL records average hourly met and CO2 data from Pt Barrow, Mauna Loa, Samoa, and South Pole http://ds.data.jma.go.jp/gmd/wdcgg/cgi-bin/wdcgg/map_search.cgi. The annual files are in text format for easy copy and paste. These files do not contain radiation data.
I have noticed that once teh Sun sets, it cools very quickly, until rel Humidity climbs as air temps drop, this quick rate under clear skies if the limit due to water and Co2, once RelH gets to 80-85% the cooling rate drops to a fairly low rate, which continues until the Sunrises.
I think this is more important than many other factors. It is water that keeps the heat from radiating to space. When water is turn into ice the planet cools. When all the ice melts the planet warms. As has been often noted the planet appears to have a maximum temperature. That is probably because the atmosphere holds only so much water and that is that.
I think the important point is this.
From the surface (which while they keep trying to move the goal post, is what’s critical), for the ~50% of the planet under clouds, a change in Co2 has 0 influence on surface temps. For the remaining ~50% of the planet, the fast cooling rate (up to 10F/hour or more) right after the Sunset, is impacted far more by the day’s % humidity than Co2 (multiple 10’s to 100+ w/m^2 vs 4w/m^2), then once the surface water vapor starts condensing out this sets the cooling rate (to a degree/hour or less).
But the slow cooling limit is controlled by surface temps, so an increase in Co2 would at worst delay the transition from fast cooling to slow cooling for minutes.
Also, while there’s not a lot of slope on Willis’s temp graph, you can see the rate change during the early morning.
If this SURFRAD database is indeed so cumbersome, how can anyone make use of it efficiently? Quite possibly, deviant Gruberians eschew objective, rational, analysis/interpretation… so, much like AW’s original reconciliation of disparate Surface Station readings, mayhap some competent researcher of integrity could undertake to render this extraordinary 15-year resource intelligible.
Lloyd,
The files are plain ASCII text files, fixed width format, one for each day of observation for each station. IOW, quite similar to daily weather station data from NCDC. A one-line wget command downloads all the files for one station, it took an hour. While that was running, I have a boilerplate script which uses some pre-written helper functions that I configured to import the data into a database. By the time the download was finished I was ready to start the import which took another hour. So in the space of two hours I was up and running writing queries to analyze and plot ~20 years worth of data which contain up to one-minute resolution in later years (three minutes for the early years).
Fascinating stuff, really and well worth the minor effort. h/t to Willis for turning me on to it.
[Thank you for making that effort. .mod]
I have also looked at surfrad and got some information overload. I really mis the possibility to plot temperature and radiation at the same time. I wonder if it was overcast days or clear sky.
I sometimes measure the sky temperature with an infrared termometer, and clear sky is below -20C, overcast between 10 and -5C. Gives an estimate of the cloud hight.
Willis
Compliments on delving into Surfad. Your graph shows a remarkably close relationship. The difference in early afternoon may be cloud cover per your thermostat hypothesis. (Interesting comments by Bill Illis above).
Re your query on atmospheric emissivity (from the peanut gallery) I expect the quantitative value is calculable by using a quantitative Line By Line emissivity model calculated across all emitting wavelengths summed over numerous elevations and adjusted for cloud cover. Complement that with a thermodynamic model of the lapse rate pressure, temperature and atmospheric composition, or empirical model – adjusted for local humidity. See:
R. Saunders, P. Rayer, P. Brunel, A. von Engeln, N. Bormann, L. Strow, S. Hannon, S. Heilliette, Xu Liu, F. Miskolczi, Y. Han, G. Masiello, J.-L. Moncet, Gennady Uymin, V. Sherlock, and D. S. Turner A comparison of radiative transfer models for simulating Atmospheric Infrared Sounder (AIRS) radiances, Journal of Geophysical Research: Atmospheres (1984–2012) Volume 112, Issue D1, 16 January 2007
e.g., Ferenc Miskolczi’s High-resolution Atmospheric Radiative Transfer Code (HARTCODE), (Miskolczi et al., 1990) and 1D climate models, and Robert H. Essenhigh’s atmospheric lapse model.
F. Miskolczi Modeling downward Surface Longwave Flux Density for Global Change Applications and Comparison with Pyregeometer Measurements J. Atmospheric Oceanic Technology, 1994 V. 11 p 608
Note Miskolczi’s development of Tau E, the Radiative Equilibrium Flux Optical Thickness, and E sub D super A = the All Sky Long Wave Downward flux to the ground (p 32) in:
Ferenc Mark Miskolczi, The Greenhouse Effect and the Infrared Radiative Structure of the Earth’s Atmosphere, F Development in Earth Science Volume 2, 2014 http://atlatszo.hu/wp-content/uploads/2011/07/article.pdf
Note differences in clear vs cloud methodology.
On thermodynamic lapse rate, see Robert H. Essenhigh, Prediction of the Standard Atmosphere Profiles of Temperature, Pressure, and Density with Height for the Lower Atmosphere by Solution of the (S−S) Integral Equations of Transfer and Evaluation of the Potential for Profile Perturbation by Combustion Emissions, Energy Fuels, 2006, 20 (3), 1057-1067 • DOI: 10.1021/ef050276y
(Contact for Robert Essenhigh at Ohio State)
Developing a quantitative local model for downwelling radiation and surface temperature along these lines from half of Surfad would be a good quantitative modeling development with the potential for quantitative validation by the other half of Surfad.
Eratta: SURFRAD
David,
Thank you for the links above. I found the one by Ferenc Mark Miskolczi, The Greenhouse Effect and the Infrared Radiative Structure of the Earth’s Atmosphere, very interesting. I will read again when time allows.
The one linking to Robert H. Essenhigh was pay walled so only the abstract was viewed.
Essenhigh’s model was extended in the public document:
Sreekanth Kolan, Study of Energy Balance between Lower and Upper Atmosphere MSc Thesis, Ohio State 2009. See Ch. 5-7.
Thanks
So that’s my puzzle for today. Is Geiger wrong about the source of the downwelling radiation? Is the emissivity of the atmosphere really on the order of 0.6? Is something else going on?
=============
the contradiction suggest there is an error in the assumptions. the contribution of convection/conduction?
Plotting the first derivative of the curves might be informative, to see if cause and effect can be isolated.
http://web.iitd.ac.in/~prabal/gas-radiation.pdf
Link is to a short paper that has the Hottel charts in it. It can be seen that the maximum emissivity for CO2 is .2 as shown in on the charts for the temperature shown on the graph above. For the 362 W/m^2 shown on the graph presented to come from CO2 the temperature of the gas would have to be 422.7 K.
Thank you to Willis and all commenters for an interesting thread.
Water vapour v.s. water droplets: an experiment you can do at home.
Pull out your IR thermometer and set the emissivity to 0.95 or 0.98. Cheapies are fixed at 0.95.
Boil your kettle and point the IR meter sensor at the clear steam as it emerges from the kettle mouth. Get close.
Next point it at the white condensed water vapour located a centimeter further from the mouth.
Are they the same temperature?
What is the radiation source for the IR in the white condensed cloud?
Dr. Robert, as always your clear and lucid comments are much appreciated. You say:
rgbatduke November 25, 2014 at 5:39 am Edit
True dat … particularly the part about the errors.
Your usual good questions. As I said, this is a first look. Here’s a second look at the same data, but this time as a scatterplot of the entire dataset. I invite interpretations of what the different parts of the graph reveal …
The light black lines seem to run in the same sense as the S-B lines … but then there’s the black section above the line at high temperatures, and a host of other details.
In any case, my thanks to you and to all the commenters.
w.
My “seat of the pants” interpretation …
* The bottom-most data points (lowest emissivity around 0.7) correspond to clear, low-humidity days. Lack of clouds and lack of water vapor limits the back-radiation.
* The top of hte bottom band (emissivity around 0.85) corresponds to clear, high-humidity days. The extra water vapor provides a bit more back-radiation
* the upper band (highest emissivty) represents cloudy days, with lots of back-radiation from the nearly-blackbody clouds.
If I am right, then I predict that temperatures above ~ 30 C almost always have clouds, and the coldest temperatures are nearly always clear. I’ve never been to Mississippi, but I suspect these are both reasonable predictions for this area.
Two additional bits ….
1. The ranges of power agree quite well with the predictions of MODTRAN for low humidity at the low power end of things to cloud cover at the high power end of things.
2. The strokes running parallel to the colored lines would be the warming and cooling during the day with approximately constant humidity; the vertical strokes would be the rapid changes from clear to cloudy (or vice versa) at relatively constant temperature.
I am probably asking the impossible here but is there any way to put a time series here to the black lines either by color or some other mechanism. Maybe color coded by month/quarter or the holy grail by surface temps or some form of representation of both?
You are on to something big here on this thread!
Willis,
I’d love to see what the Colorado site looks like as you’ve plotted Miss. My thought is that you’ve mostly just plotted the response of water vapor, similar to the graphs you’ve done with SST’s limiting to 30C or so.
I would expect Colorado to get close to 150W/m^2 by 0C. The black shifted down to the left based on the difference in humidity between the two locations.
The black blob is the effect of low level clouds, they are the closest anything is to ground temps.
Lewindowsky and any other psuedo-sociologist will likely say that this demonstrates a deep-seated need on your part to take the:
http://theinkblot.com/
test.