Recent Comments
Prev 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 Next
Comments 128051 to 128100:
-
MattJ at 17:48 PM on 19 June 2009CO2 lags temperature
An argument I was hoping to find the refutation for here is: "CO2 has been much higher than now in teh past, and species flourished; why wouldn't they do the same now?" My understanding has been that because of greatre climate sensitivity to ocean current changes, a little CO2 now has a much bigger change on the climate than back in the days (for example), of Pangaea, when having only one continent created very different ocean currents. But I was unable to confirm the correctness of my understanding from this skeptical science site. -
thingadonta at 10:52 AM on 19 June 2009The CO2/Temperature correlation over the 20th Century
I have read Usoskin 2005 and various others you refer to, and your 'Its the sun' page, and very few of the papers refer to longer term lag effects-most 'look' for a mechanism-over the last 35 years or so, and rarely consider lag effects after a SUSTAINED peak. Also, your graph at the bottom of 'Its the sun' page uses T with respect to the mean from 1960-1990? on its Y axis,-which would be EXPECTED to show anomalies/a rise to the mean, after ~1980, if there was a lag effect. So the 'anomalous T' axis you use is irrelevant to this argument, and is misleading with respect to lag effects, if anything- it supports it!. Have another look at your summaries of papers on 'its the sun page'-very few of these address lag effects. Most look for a 'mechanism' to sustain a recent warming trend, which they wont easily find if the warming is largely a longer-term, lag effect (or if tipping points in solar effects exist-the favourite of human-warmists). I have looked through a few of these papers, but not all (some wont open in the links);there is little proper investigation of lag effects from a sustained high peak in Solanki 2004,Solanki 2003, Usoskin 2005. HOWEVER Haigh 2003 does look at longer term time lags with respect to sustained peaks,and actually partially supports my sustained high peak argument, to quote: "chemical and dynamical processes in the middle atmosphere may act to amplify the solar impact", even though she doesnt think solar variables can explain recent warming since 1980. Usoskin 2005 simly states, 'we didn't look at it', and 'it must have another cause'. Thus the paper does not address lag time effects ONCE A SUSTAINED HIGH PEAK HAS BEEN REACHED. The 10 year time lag referred to by Usoskin 2005 only refers to short period peaks and troughs, and doesnt negate the last 30-40 years of warming, since other peaks in the past were not sustained at these high levels. Also, how does one know that a 'tipping point' (the darling of global warmists), hasnt been reached with regard to the sustained solar peak since about ~1970-80, where reduced magnetic field effects and cosmic rays really only become influential once the sun has reached a certain sustained level of activity?? (If this argument is weak, note that it is one of the KEY concepts often used by human-warmists with regard to C02). All the papers you refer to do not disprove the key argument, that the sustained high level of solar activity can't have time-lagged a T rise since 1980, especially if you use raw troposphere temperature (which has shown little rise since 1979), rather than anomalies to the 1960-1990 mean, which distorts the effects of time-lags(eg your graph at the bottom of the page on 'Its the sun' page). Also, the paper of Usokin 2005 does pick up trends on longer time scales, but says they are weak: ie 'centennial' and 'intracentennial'. So, I'm afraid you have still not properly addressed lag effects after a sustained high peak, (and NOT in relation to the 1960-1990 mean, which would be EXPECTED to show a marked rise in 'anomalous T' if there was lag effects after 1980). -
canbanjo at 06:56 AM on 19 June 2009The CO2/Temperature correlation over the 20th Century
Re point 1 and the Usoskin 2005, I understand there is greatest correlation between temperature and solar when the temp lags solar by ten years. But the graph also shows 40 year lag periods, albeit with a weaker correlation. Could you please explain why this rules out solar activity accounting for the last 30 years temperature rises? I am just an interested layman btw. Many thanks.Response: Can you clarify where you see the 40 year lag periods? Usoskin 2005 has the correlation peaking at a 10 year lag: -
chris at 03:08 AM on 19 June 2009The correlation between CO2 and temperature
re #19 Unsupported assertions on dodgy web sites are rarely helpful HS and you've again dumped several on this thread. The author of the web page you linked to is hugely misinformed about several things, including this in reference to the lifetime of enhanced atmospheric CO2 levels:"Instantly on a climatological scale would be a few years or decades at most. It would depend on the half-life of CO2 in the atmosphere, which is a disputed topic. Many processes, including photosynthesis and dissolution into bodies of water, act to reduce CO2."
However the evidence is pretty clear on the expectation that greatly raised atmospheric CO2 levels remain raised by a very long time indeed (100’s to 1000’s of years, with a long tail lasting 10’s of 1000’s of years and more). Some of the evidence for this is given below. It’s worth pointing out that the odd notion that “photosynthesis” will “act to reduce CO2” is a meaningless statement, especially without consideration of timescales and mechanisms. In the real world photosynthesis has very little effect on nett atmospheric CO2 levels since draw down of atmospheric CO2 by photosynthesis is largely balanced by oxidative organic decay and the return of CO2 to the atmosphere. There is only so much scope for enhanced nett primary production of flora (and the sequestration of CO2 by plant fossilation only occurs on the hundreds of thousands to million years timescale). Unfortunately the situation is worsened in our current world with rainforest degradation and the expectation that primary production (nett plant growth) will decrease in a warming world. Some of these points are discussed here http://www.skepticalscience.com/Is-Antarctic-ice-melting-or-growing.html (see posts #13/14). Likewise the rate at which oceans sequester enhanced levels of atmospheric CO2 is slow (100’s of years timescale) [1]. Additionally the ability of the oceans to absorb is reduced as enhanced atmospheric CO2 “drives” CO2 into the oceans, and already the evidence indicates that the expected tendency for the upper oceans to saturate with respect to CO2 uptake is already occurring [2,3]. A review of some of the evidence that informs our understanding of the lifetime of enhanced atmospheric CO2 was published late last year [1], and indicates that sequestering of CO2 occurs on multiple timescales, the most rapid of which (ocean uptake) takes many hundreds of years, with very long tails of recovery lasting 1000’s and 10’s of 1000’s of years. This is consistent with evidence from analysis of carbonates in sediments that are used to determine CO2 levels in the deep past, that indicate that the massive increases in atmospheric CO2 associated with extinctions at the Triassic-Jurassic boundary [4,6] or the Paleo-Eocene Thermal Maximum [5,6] (for example) take many 10’s of 1000’s of years to return to pre-extinction event levels. I haven’t spent very much time looking up this stuff, and no doubt one could come up with more and better examples of the evidence that informs our understanding of the lifetimes of greatly enhanced CO2 levels. [1] Archer D and Brovkin V (2008) The millennial atmospheric lifetime of anthropogenic CO2 Climatic Change 90, 283-297 Abstract: The notion is pervasive in the climate science community and in the public at large that the climate impacts of fossil fuel CO2 release will only persist for a few centuries. This conclusion has no basis in theory or models of the atmosphere/ocean carbon cycle, which we review here. The largest fraction of the CO2 recovery will take place on time scales of centuries, as CO2 invades the ocean, but a significant fraction of the fossil fuel CO2, ranging in published models in the literature from 20-60%, remains airborne for a thousand years or longer. Ultimate recovery takes place on time scales of hundreds of thousands of years, a geologic longevity typically associated in public perceptions with nuclear waste. The glacial/interglacial climate cycles demonstrate that ice sheets and sea level respond dramatically to millennial-timescale changes in climate forcing. There are also potential positive feedbacks in the carbon cycle, including methane hydrates in the ocean, and peat frozen in permafrost, that are most sensitive to the long tail of the fossil fuel CO2 in the atmosphere. [2] Le Quere C et al (2007) "Saturation of the Southern Ocean CO2 sink due to recent climate change" Science 316, 1735-1738. [3] Schuster U et al (2007) “A variable and decreasing sink for atmospheric CO2 in the North Atlantic” J. Geophys. Res. Oceans 112, art # C11006 [4] B. van de Schootbrugge et al (2008) Carbon cycle perturbation and stabilization in the wake of the Triassic-Jurassic boundary mass-extinction event Geochem. Geophys. Geosys. 9 Q04028 [5] G. R. Dickens et al. (1997) A blast of gas in the latest Paleocene: Simulating first-order effects of massive dissociation of oceanic methane hydrate Geology 25, 259-262. [6] A. S. Cohen et al. (2007) The Late Palaeocene–Early Eocene and Toarcian (Early Jurassic) carbon isotope excursions: a comparison of their time scales, associated environmental changes, causes and consequences J. Geological Soc.164, 1093-1108 -
Alexandre at 00:24 AM on 19 June 2009The CO2/Temperature correlation over the 20th Century
Great post, John. You´re website is a great source to finding AGW-related scientific references. -
Stig Mikalsen at 19:52 PM on 18 June 2009The CO2/Temperature correlation over the 20th Century
John, thanks for the post. I just want to add a point about the "mid-century-cooling". In this post by Tamino, John Mashey perhaps has a useful comment (dated May 24) about global sulfur dioxide emissions. The data can be found here: These emissions approximately tripled between 1945 - 1975, and then fell back with the Clean Air Acts. Mashey finds it plausible "that a 52Mt increase in yearly emissions could generate a ~ .4C NH dip from ~1945-1975", and thereby helped "masking" the effect of CO2 in this period. I don't know whether NASAs model of 2002/2004 accurately reflects this or not. Just a thought. Or perhaps the "mid-century-cooling" correlated also with natural changes in some of the longer ocean circulations... (PDO?) To thingadonta, I would guess that your lag time effect of 20-30 years for solar irradiance should be visible in the later historic record, which I believe it isn't. -
thingadonta at 17:59 PM on 18 June 2009The CO2/Temperature correlation over the 20th Century
You pick and choose where you want to apply CO2 effects,and where you want to intergrate other effects. It's circular reasoning. What about lag time effects with regard to the solar irradiance?. No mention at all. You haven't addressed lag-time effects with your sun-temperature disconnect theory from about 1980 (eg in other posts). Tying in effects on clouds, or the earths magnetic field (which has been declining naturally for the last few hundred years-which would amplify any warming-an important point), and you could explain a global emperature peak around the 2000s. For example: -When is the hottest part of the day?-around 3 hours AFTER the sun reaches its peak (~30% of the entire day-warming trend). -Which is the hottest part of the year?-about 6 weeks AFTER the longest day of the year (~25% of the annual warming trend). If one increases temperature to a pot on a stove, and then flattens or decreases the temperature, the potwill show a small lag time effect, then will CONTINUE to heat for a period afterwards, and then slowly decline. How long does it take the earth system to absorb lag time heat, 1 year? 5 years?, 20 years?, 50 years 100 years?, Does anybody know?? You claim that in the last 35 years C02 has been dominant, but you do not discuss lag time heat effects from the sun, which you say has disconnected since 1980, without discussing lag time effects. If 25-30% heat lag effects are a guide, then adding 25-30% of lag time to the suns warming trend, from a peak around 1980 fits perfectly with a peak around 2000s. To pre-empt a reply based on the lag time only, don't forget, there may be OTHER factors that enhance lag time, such as the very slow lag-time heating of the ocean, the earth's magnetic field reduction, and clouds etc etc. Please address? It is one of the biggest arguemnts of skepics, but I don find your site addressing it much anywhereResponse: I recommend you read Usoskin 2005 which looks at both the correlation between solar activity and climate and the time lag between long term changes in solar activity and global temperature. They compare solar and climate data and find the correlation is highest when there's a 10 year lag. It also finds the correlation breaks down when the modern global warming trend begins in the mid-70's. I also look at Usoskin 2005 on the "It's the Sun" page and touch on the time lag issue at the bottom of the page. -
thingadonta at 10:17 AM on 18 June 2009This just in - the sun affects climate
You haven't addressed lag-time effects with your sun-temperature disconnect theory from about 1980. Tying in effects on clouds, or the earths magnetic field (which has been declining naturally for the last few hundred years-which would amplify any warming-an important point), and you could explain a global emperature peak around 2002. For example: -When is the hottest part of the day?-around 3 hours AFTER the sun reaches its peak (~30% of the entire day-warming trend). -Which is the hottest part of the year?-about 6 weeks AFTER the longest day of the year (~25% of the annual warming trend). If 25-30% heat lag effects are a guide, then adding 25-30% of lag time to the suns warming trend, from a peak around 1980 fits perfectly with a peak around 2002. To pre-empt a reply based on the lag time only, don't forget, there may be OTHER factors that enhance lag time, such as the very slow lag-time heating of the ocean, the earth's magnetic field reduction, and clouds etc etc. Please address? It is one of the biggest arguemnts of skepics, but I don find your site addressing it much anywhere.Response: The issue of climate time lag is addressed in the post coincidentally titled Climate Time Lag. -
HealthySkeptic at 12:43 PM on 17 June 2009The correlation between CO2 and temperature
re #16 >> I assume that when he says that if we stop emitting CO2 the level would "instantly return to pre-industrial levels." it is just a typo and was not what he intended to say. John, I queried the author personally on this and this is his response'- "Instantly on a climatological scale would be a few years or decades at most. It would depend on the half-life of CO2 in the atmosphere, which is a disputed topic. Many processes, including photosynthesis and dissolution into bodies of water, act to reduce CO2." HS. -
HealthySkeptic at 14:28 PM on 16 June 2009The correlation between CO2 and temperature
re #16 John, With regard to saturation of the CO2 bands, do we know what percentage of energy, in the wavelengths that CO2 can absorb, is already being absorbed? I have seen figures as high as 99%. Some have even quoted that satellites equipped with infrared spectrometers have confirmed that there is already little or no infrared energy from the earth's surface escaping the earth's atmosphere in the CO2 absorption bands. Do any of these claims have any substance? If not, do we know at what atmospheric CO2 concentration the CO2 bands WILL become completely saturated? Regards, HS. -
Andrew K at 11:14 AM on 15 June 2009The correlation between CO2 and temperature
John Cross -- I suppose my point is purely emperical. The current non-warming is explained away as the result of recent weather noise masking the long term warming trend, but a statistical test using for example the d* method of Santer (which using data up to the end of 2007 proved the above explaination) now shows that the lack or warming is more than expected from known (ie previously observed) weather noise. To counter this statistical failure it has been pointed out that there have been decade or so long non-warming (or even cooling) periods within the long term warming trend as shown in Figure 2 above, but previous non-warming (or cooling) periods have followed large volcanoes whilst the current period doesn't. In summary the current period cannot be explained away either as statistical noise or as a simple occurance of known natural events that will soon reverse. So what is causing the non-warming when 'all the consesus science' says there should be (note: some are now claiming it is the sun although a few years ago we were told that solar changes and effects were all known and included in the models)? Either the measurements are wrong or there is a problem with 'the consesus science' -
gruntled at 13:58 PM on 14 June 2009CO2 measurements are suspect
I was looking for some information on the reliability of ice cores. A common argument I'm hearing lately is that chemical effect in ice cores breaks down CO2, making them unreliable proxies. -
Patrick 027 at 11:21 AM on 11 June 2009Arctic sea ice melt - natural or man-made?
"This is why this is a political hot potato and needs to be approached with unassailable science that everyone can accept, and they can not do that. " There is no such thing - unassailable science that everyone can accept. In recent weeks I've witnessed a case of a person who cannot accept simple arithmetic. We need good science. We already have a lot. We shouldn't stop trying to get more, but we needn't wait to use what we have in appropriate ways. However good the science is, there will be people who cannot accept it. I don't really care anymore what they think - I hope those of us living in reality are a strong enough majority, and those who stubbornly bury their heads in the sand while feigning self-righteous protest - they will just have to live with the decisions the rest of us make. Should I be so optimistic as to expect a room of adults to agree that the sun is a star? -
John Cross at 22:39 PM on 10 June 2009The correlation between CO2 and temperature
HealthySkeptic: Thanks for the link. I will note that the article (I don't believe it is a paper) is interesting but misses a couple of important points. As simple physics shows, even if the CO2 bands were totally saturated, adding more would still cause an increase in temperature. In addition, when the author calculates his projection he assumes that all warming takes place instantaneously. This is of course not true - warming takes place over numerous time scales. I assume that when he says that if we stop emitting CO2 the level would "instantly return to pre-industrial levels." it is just a typo and was not what he intended to say. So, by all means read it, but as with everything, read it with a skeptic's eye. Regards, John -
Gord at 16:47 PM on 10 June 2009It's the sun
Patrick - Re: your Post #511 You remind me of "Baghdad Ali". Although, your logic may be just a tad inferior to his. Don't understand?...It's OK....everybody else will. -
HealthySkeptic at 14:06 PM on 10 June 2009The correlation between CO2 and temperature
JWC, This paper may answer some of your questions regarding the relative effect of increasing CO2 concentrations on global temperatures. http://brneurosci.org/co2.html -
HealthySkeptic at 12:34 PM on 10 June 2009The correlation between CO2 and temperature
Now, if we can just get that sunscreen up into the upper atmosphere.... -
Patrick 027 at 04:35 AM on 10 June 2009It's the sun
sure, Gord, and why is anybody talking about the graviational pull of the moon and sun when the tides can be so easily explained by waves and currents? Your posts are UTTERLY STUPID. But I hope most people reading this do not need me to point it out. -
Craig Allen at 03:40 AM on 10 June 2009The correlation between CO2 and temperature
JWC: This analogy may help - The mass of the sunscreen that you need to smeer on your arm to stop almost all the UV light getting through is tiny compared to the mass of the atmosphere sitting above you which failed to stop it (there are about ten metric tons of atmosphere per square meter of surface I think). A trace amount of a substance can have a big effect if that substance is very effective at what it does. The amount of cyanide that it would take to kill you vs. the total volume of air that you breath is another example. -
John Cross at 00:14 AM on 10 June 2009The correlation between CO2 and temperature
JWC: just because something is trace does not mean that it has a small effect. The lethal dose of botulism is about 5 parts per billion. Now, if you are asking about the actual physics of the interaction of longwave radiation with the CO2 molecule, that is a different question - but one I am sure we could look at if you wish. Regards, John -
JWC at 22:06 PM on 9 June 2009The correlation between CO2 and temperature
If CO2 is such a trace gas, currently 380 PPM of the atmosphere, someone needs to tell me, an inquiring layperson, how an incremental increase in that amount can possibly have much affect on temperature increase. It doesn't make much sense, to me, and, though I've asked numerous times elsewhere, no one seems to be able to explain it to me. I was referred to this website by someone I was disagreeing with about AGW. A site to debunk the debunkers? Those skeptics must certainly be gaining some ground. Many AGW "alarmists" seem to think "the debate is over." Seems to me, it would be very hard to do further unbiased research if that's the attitude taken by these (I believe misguided) followers of King Albert of Gore. Sorry; I just call it like I see it. -
Gord at 21:24 PM on 9 June 2009It's the sun
The STUPIDITY of AGW. ---- Trenberth's Energy Budget Incoming Solar Radiation = 342 w/m^2 Solar Radiation Absorbed by atmosphere = 67 w/m^2 ------------------- (342 - 67) Leaves 275 w/m^2 available. Reflected by Clouds etc. = 77 w/m^2 Reflected by Surface = 30 w/m^2 Total due to reflection = 107 w/m^2 The percentage of reflected energy is 107/275 = 0.389 or 38.9%. Leaves 168 w/m^2 absorbed by the Surface of the Earth. 168 w/m^2 and an emissivity of 1, gives a temperature of 233.31K or -39.69 deg C. -------------------- Now what happens if the reflected energy was decreased by 1% to 37.9%? 0.379 X 275 = 104.23 w/m^2 so an additional (107 - 104.23 = 2.77 w/m^2) is available to heat the Earth. 168 + 2.77 = 170.77 w/m^2 is now absorbed by the Surface of the Earth. 170.77 w/m^2 and an emissivity of 1, gives a temperature of 234.26K or -38.74 deg C. -------------- The Earth just warmed by (39.69 - 38.74) 0.95 deg C !! That's just due to a ONE PERCENT change in reflected energy!! ----------------- Why the Hell is anybody talking about CO2, positive feed-back loops, carbon taxes etc. to explain something so easily explained? Especially since the AGW'ers admit that their "computer models" can't and don't handle CLOUDS well and the SUN is the ONLY ENERGY SOURCE! ----------------- AGW is UTTER STUPIDITY no matter how you look at it! -
HealthySkeptic at 14:09 PM on 9 June 2009This just in - the sun affects climate
Then, there's this... New Paper Demonstrates Anthropogenic Contribution to Global Warming Overestimated, Solar Contribution Underestimated A new paper has been published in GRL by Scafetta and Willson entitled: ‘ACRIM-gap and TSI trend issue resolved using a surface magnetic flux TSI proxy model’ The Abstract states: “The ACRIM-gap (1989.5-1991.75) continuity dilemma for satellite TSI observations is resolved by bridging the satellite TSI monitoring gap between ACRIM1 and ACRIM2 results with TSI derived from Krivova et al.’s (2007) proxy model based on variations of the surface distribution of solar magnetic flux. ‘Mixed’ versions of ACRIM and PMOD TSI composites are constructed with their composites’ original values except for the ACRIM gap, where Krivova modeled TSI is used to connect ACRIM1 and ACRIM2 results. Both ‘mixed’ composites demonstrate a significant TSI increase of 0.033%/decade between the solar activity minima of 1986 and 1996, comparable to the 0.037% found in the ACRIM composite. The finding supports the contention of Willson (1997) that the ERBS/ERBE results are flawed by uncorrected degradation during the ACRIM gap and refutes the Nimbus7/ERB ACRIM gap adjustment Fröhlich and Lean (1998) employed in constructing the PMOD.” The authors state in their conclusions that: “This finding has evident repercussions for climate change and solar physics. Increasing TSI between 1980 and 2000 could have contributed significantly to global warming during the last three decades [Scafetta and West, 2007, 2008]. Current climate models [Intergovernmental Panel on Climate Change, 2007] have assumed that the TSI did not vary significantly during the last 30 years and have therefore underestimated the solar contribution and overestimated the anthropogenic contribution to global warming.” Scafetta N., R. C. Willson (2009), ACRIM-gap and TSI trend issue resolved using a surface magnetic flux TSI proxy model, Geophys. Res. Lett., 36, L05701, doi:10.1029/2008GL036307.Response:The ACRIM vs PMOD debate is a question over which composite of satellite TSI data is more accurate. ACRIM shows a slight warming trend. PMOD shows a slight cooling trend. Even using the ACRIM data, the warming trend is so slight, it is insufficient to explain the steep warming over the past 35 years. However, various independent measurements of solar activity all show closer agreement to the PMOD reconstruction which indicates the sun has been showing a cooling trend over the last few decades. This is explained in detail at Determining the long term solar trend.
UPDATE: I read the Scafetta paper which claims the ACRIM data (which shows slight warming) is more accurate than the PMOD data (slight cooling) because it shows closer agreement with the TSI reconstruction by Krivova and Solanki. I found this curious as Krivova 2007 itself compares its results to PMOD and finds close agreement. So I contacted Sami Solanki, one of the authors of the Krivova model. He replied (very promptly, a very approachable scientist) that they had actually written a paper responding to Scafetta's claims that is currently awaiting publication. I will keep everyone posted when the paper is published.
Incidentally, Krivova's TSI reconstruction is the data I use in Figure 1 above.UPDATE 2 Nov 2009: The response from Solanki and Krivova has been published. I summarise it's findings in ACRIM vs PMOD, the rematch.
-
Patrick 027 at 10:10 AM on 9 June 2009It's the sun
"I think that type of reflecting surface has inverted square pyramids" ... Oh, maybe that's just bicycle reflectors - but you get the idea. -------------- And now the moment very few people have been waiting for: What is the spectral (monochromatic) Ibb? For N = 1: Where the blackbody flux per unit area = π*Ibb = sigma * T^4 where sigma (also known as BC above) =2 * π^5 * k^4 /(15 * c^2 * h^3) and Where c, h, and k, and sigma are: (Where there are two values, the second value is from a physics textbook (or calculated from physics textbook values), and the first value is from Wikipedia ( http://en.wikipedia.org/wiki/Boltzmann's_constant , http://en.wikipedia.org/wiki/Avogadro_constant , http://en.wikipedia.org/wiki/Stefan%E2%80%93Boltzmann_constant ): c = 2.99792458 E+8 m/s h = 6.62606896 E-34 J*s h = 6.626075 E-34 J*s k = 1.3806504 E-23 J/K k = 1.380658 E-23 J/K sigma = 5.670400 E-8 W/(m2 K4) sigma = 5.670511 E-8 W/(m2 K4) Ibb(T,L) = |d[Ibb(T)]/dL| = ( 2*h*c^2 / L^5 ) / ( exp[h*c/(L*k*T)] - 1 ) Ibb(T,v) = |d[Ibb(T)]/dv| = ( 2*h*v^3 / c^2 ) / ( exp[h*v/(k*T)] - 1 ) where: c = L * v v = c/L |dv| = c / L^2 * |dL| L is the wavelength in a vacuum, v is the frequency. The L of maximum Ibb(T,L) is equal to (2897 microns/K )/ T (Wien's displacement law, from class notes but can be found elsewhere.) The L of maximum Ibb(T,L) is 10 microns at T = 289.7 K. -
Patrick 027 at 10:10 AM on 9 June 2009It's the sun
"I think that type of reflecting surface has inverted square pyramids" ... Oh, maybe that's just bicycle reflectors - but you get the idea. -------------- And now the moment very few people have been waiting for: What is the spectral (monochromatic) Ibb? For N = 1: Where the blackbody flux per unit area = π*Ibb = sigma * T^4 where sigma (also known as BC above) =2 * π^5 * k^4 /(15 * c^2 * h^3) and Where c, h, and k, and sigma are: (Where there are two values, the second value is from a physics textbook (or calculated from physics textbook values), and the first value is from Wikipedia ( http://en.wikipedia.org/wiki/Boltzmann's_constant , http://en.wikipedia.org/wiki/Avogadro_constant , http://en.wikipedia.org/wiki/Stefan%E2%80%93Boltzmann_constant ): c = 2.99792458 E+8 m/s h = 6.62606896 E-34 J*s h = 6.626075 E-34 J*s k = 1.3806504 E-23 J/K k = 1.380658 E-23 J/K sigma = 5.670400 E-8 W/(m2 K4) sigma = 5.670511 E-8 W/(m2 K4) Ibb(T,L) = |d[Ibb(T)]/dL| = ( 2*h*c^2 / L^5 ) / ( exp[h*c/(L*k*T)] - 1 ) Ibb(T,v) = |d[Ibb(T)]/dv| = ( 2*h*v^3 / c^2 ) / ( exp[h*v/(k*T)] - 1 ) where: c = L * v v = c/L |dv| = c / L^2 * |dL| L is the wavelength in a vacuum, v is the frequency. The L of maximum Ibb(T,L) is equal to (2897 microns/K )/ T (Wien's displacement law, from class notes but can be found elsewhere.) The L of maximum Ibb(T,L) is 10 microns at T = 289.7 K. -
Patrick 027 at 04:42 AM on 9 June 2009It's the sun
LW radiation: "If this is specular reflection (As might be expected for relatively calm water), then, except in an inversion with sufficient opacity, the reflected radiation will be absorbed over a shorter distance in the air" ... as opposed to Lambertian reflection. -
Patrick 027 at 04:41 AM on 9 June 2009It's the sun
"I have actually noticed in lawn grass in sunny conditions that the grass appears brighter just around the shadow of my head - this means that there is a concentration of reflected radiation going back near the direction from which it came - similar to the reflecting surfaces used for traffic signs. I noticed a similar phenomenon in gravel." But the process that produces this effect is different for the traffic signs (and the effect itself will probably be a little different). I think that type of reflecting surface has inverted square pyramids, where each flat triangular surface is at a 45 deg angle to the plane of the larger surface, and each flat triangular surface has specular reflection. In contrast, what may be happening with the grass and the gravel is that the individual surfaces have nearly Lambertian reflection - they will look at bright from any direction - but the surfaces themselves are angled differently and thus have more or less sunlight per unit area to reflect. When looking toward the shadow of one's own head, one would see more of the surfaces that are nearly normal to the direct solar rays. ---- Sometimes the brightest and deepest colors can be seen when looking at the diffuse transmission through leaves and petals. Highly recommended viewing. -
Patrick 027 at 04:27 AM on 9 June 2009It's the sun
Additional notes and observations of scattering of solar radiation: When there is a hole through which direct solar rays pass, scattering along the path of the beam allows the beam to glow with scattered radiation - it can be seen from outside itself. This can be observed in a dusty room with sunlight coming through a window. A shadow cast through the air can also be seen from outside itself by the same mechanism. Variations in direct solar ray intensity can be seen to the extent that they have optical thickness along the line of sight and there is not too much optical thickness along the line of sight between the viewer and variation being observed. These variations are called crepescular rays and can be seen when the direct sun is blocked by a layer of clouds with holes, or there are patches of clouds casting shadows, or when the sun is behind a cloud with an irregular edge. One particularly interesting case is the shadow cast be a long thin straight contrail (the cloud left by a jet when conditions allow). Such a contrail casts a shadow that is a thin planar slice through the air; along lines of sight nearly parallel to this shadow, a dark streak can be seen through the sky; it will be darkest to viewers within the shadow. But the shadow will not be observed along lines of sight in most other directions because the shadow is thin in those directions. Clear air atmospheric scattering is generally stronger for shorter wavelengths. This is of course why the midday clear sky is generally blue (it can be white near the horizon - possible contributors to that: less blue light reaches to air near the surface, while there is nonzero scattering of other wavelengths, and lines of sight near the horizon pass through a greater thickness of lower-level air as well as the total atmosphere; some aerosols with different scattering properties may also be abundant in the lowest level air). White objects near the horizon may appear slightly red because blue light is scattered out of the line of sight. However, some blue light is scattered into the line of sight from an overhead sun and/or overhead thin clouds. Thick clouds overhead can reduce this effect, so that distant clouds lit by the sun will appear more red. Of course, the extinction of blue light and the nonzero scattering of other wavelengths explain the colors of sunset and sunrise. The overhead clear sky still appears blue, and the sky near the horizon can appear blue after sunset and before sunrise - this is because of the curvature of the Earth; the solid/liquid Earth casts a shadow on it's own atmosphere, but the edge of the shadow is at some finite height within the atmosphere at dawn and dusk, and when reaching the Earth nearly horizontally, sunlight travels through a smaller mass of air when it goes by higher above the ground, so there is plenty of blue light to scatter. --------- Specular reflection (as in a mirror) can be observed for some smooth surfaces, such as calm water. Wavy water still has locally specular reflection but images will break up and be distorted. Diffuse reflection takes an incident beam of light and reflects it over a range of directions. It is a form of scattering. Lambertian reflection is diffuse reflection in which the reflected radiation is isotropic. Reflected radiation can be a mix of Lambertian and specular or nearly specular (images would appear fuzzy), or more complex. I have actually noticed in lawn grass in sunny conditions that the grass appears brighter just around the shadow of my head - this means that there is a concentration of reflected radiation going back near the direction from which it came - similar to the reflecting surfaces used for traffic signs. I noticed a similar phenomenon in gravel. Have you ever noticed crepescular rays when looking down into murky water? When the surface has a high albedo, the sky can appear brighter than otherwise because it can scatter back to the surface some of the radiation reflected from the surface. I have seen a picture of a blue light on the base of clouds over a patch of shallower water (which appears bright blue because there is less absorption of sunlight through the thinner water layer and there is reflection from the underlying surface) - perhaps an atoll. (Clear) water is generally blue because, while nearly transparent (to visible wavelengths) in thin layers, it does absorb light, and more strongly at longer visible wavelengths than for blue light. Of course, looking near the horizon, one sees less into the water and more the reflection of the sky off the water. When there are waves, one can see into the water best on the side of the wave closest to normal to the line of sight, while mainly a reflection of the sky will be seen from the other side of the wave or where the line of sight just grazes the water surface. PS for LW radiation: Then nonzero LW albedo of the surface: If this is specular reflection (As might be expected for relatively calm water), then, except in an inversion with sufficient opacity, the reflected radiation will be absorbed over a shorter distance in the air because a greater portion of it will come from angles near the horizon where the LW glow fo the air will (except for a low level inversion with sufficient opacity) generally appear brightest near the horizon. -
Patrick 027 at 03:46 AM on 9 June 2009It's the sun
A condition where the phase of the wave could matter (?) is when there is significant nonlinearity - In nonlinear optics, passage of very high amplitude electromagnetic waves through a medium can alter the optical properties of the medium, allowing photon-photon interaction. When evaluting I# at specific values of P, the P at one location is not necessarily the same P everywhere, and when finding the emission distribution, there could be multiple P values at the same location corresponding to different paths that photons took between locations and the different changes to P along the way. P is obviously relative to the orientation of processes that depend on P. ---------- For radiative energy transfer in the atmosphere, there are some useful approximations (for at least Earthly conditions): 1. Assume local thermodynamic equilibrium, at least below some height level (that is at least above the tropopause). Ignore non-thermal emission. Assume emission cross section = absorption cross section. 2. For wavelengths shorter than about 4 microns (SW radiation), assume the only emission is from the sun. 3. For wavelengths longer than about 4 microns (LW radiation), assume there is no solar contribution (a bit less accurate than assumption 2, but still a good approximation). 4. Aside from absorption and emission, assume N = 1 within the atmosphere on a macroscopic scale (larger than scattering mechanism scales), and ignore gravitational lensing, so that radiation propagates in straight lines within the atmosphere and space, except for scattering and reflection. 5. Also ignore gravitational redshift, and the blueshifting and redshifting of solar energy at sunrise and sunset, etc. 6. Below some level that is above the vast majority of atmospheric emission and absorption at most wavelengths, ignore the curvature of the Earth and atmosphere, so that radiation propagating in straight lines is assumed to have a constant angle relative to the local vertical, and the total horizontal area around the globe at any height can be assumed constant over vertical distance. 7. Except at the land and ocean surface, assume zero scattering and zero reflection for LW radiation within the atmsophere. (Note that even when there is some significant single-scatter albedo, the multiple scatterings required (when that is the case) for radiation to scatter back from, for example, a cloud, can result in very low albedo due to absorption between each scattering.) 8. For some purposes, scattering and reflection at the surface can also be neglected for LW radiation. -
Patrick 027 at 03:17 AM on 9 June 2009It's the sun
In so far as the differential equation following a path locally in the direction x over distance dx, for I# in the direction x: At specific v and P, per unit spectrum and polarization dv and dP (P could have more than one dimension, actually): d(I#) = IL#(G) * Lcsv * dx + Ibb#(T) * ecsv * dx - I# * (acsv+scsv) * dx + Is# * scsv * dx - I# * R + Ir# * R The terms on the right hand side (if this were written out in one line) 1. non-thermal emission into the path (such as fluorescence), where Lcsv is a cross section density for that process and IL#(G) is a function of the the energy available for such a process and the nature of such a process. 2. thermal emission into the path, where ecsv is the emission cross section density and Ibb#(T) is the blackbody intensity. 3. absorption and scattering out of the path, where acsv and scsv are the absorption and scattering cross section densities, which sum to give the extinction cross section density. acsv = ecsv at local thermodynamic equilibrium (it could be possible to define them as equal even when not at local thermodynamic equilibrium since a non-thermal emission term is also included). 4. scattering into the path, where scsv is the same scattering cross section density, but Is# depends on I# in all directions at that location (including backwards along the same path) and the type of scattering. 5. specular reflection out of the path, where R is the reflectivity of any interface encountered within dx. 6. specular reflection into the path, where Ir# depends on the I# going backward along the path taken by reflection out of the path. Inclusion of reflection at a discontinous interface (relative to the scale of the wavelength) in the differential equation works so long as dx is small enough that the other terms are very small. -
NewYorkJ at 01:46 AM on 9 June 2009This just in - the sun affects climate
Contrarians have a tendency to not only recylce old arguments, but in this case to recycle old arguments and then distort them to support a political agenda. This is related to the same strawman that claims climate scientists think only CO2 affects climate. In NASA's 2007 temperature summary, Hansen notes: "This cyclic solar variability yields a climate forcing change of about 0.3 W/m2 between solar maxima and solar minima. (Although solar irradiance of an area perpendicular to the solar beam is about 1366 W/m2, the absorption of solar energy averaged over day and night and the Earth's surface is about 240 W/m2.) Several analyses have extracted empirical global temperature variations of amplitude about 0.1°C associated with the 10-11 year solar cycle, a magnitude consistent with climate model simulations, but this signal is difficult to disentangle from other causes of global temperature change, including unforced chaotic fluctuations. " http://data.giss.nasa.gov/gistemp/2007/ -
Dan Pangburn at 01:29 AM on 9 June 2009Models are unreliable
Climate Scientists have adopted the word 'feedback' but apply it completely differently from the way engineers had already successfully used it for decades. Correct use of 'feedback' combined with paleo temperature data proves that added atmospheric carbon dioxide has no significant effect on average global temperature. Any activity to curtail the atmospheric carbon dioxide level puts freedom and prosperity at risk. -
Gord at 18:24 PM on 8 June 2009It's the sun
"The forces of American colonialism began to drop containers that produce a sound explosion, a very huge sound. I remind you that they said that their strategy is based on shock and awe. Those failed ones manufactured a type of container that has an explosive substance, which they drop. They cause a very huge explosion in terms of sound, as if the universe was shaken. After a while, you go out and you don't find anything. You find some nails, screws, pieces of metal, but the important thing here is the sound. Those failed ones think that through the huge sound explosion, people would be shocked and consequently would collapse and be defeated. What happened? The contrary." ....a "Baghdad Ali" quote. Sounds a lot like Patrick? -
Patrick 027 at 14:49 PM on 8 June 2009It's the sun
________________________ MORE ABOUT REFRACTION: COMPLEX N: Let the complex index of refraction be N. Previously I used n to represent the real component of the index of refraction. It would be better to refer to that as nr. (From class notes): Imaginary component of the index of refraction ni: The absorption coefficient (equal to the absorption cross section per unit volume) = 4 * pi * ni / L# where L# is the wavelength in a vacuum of radiation with the same frequency v. ---- The real and imaginary components of the index of refraction, N = nr + i*ni , do not vary independently of each other over v or L#. nr and ni are related by the Kramer-Kronig relationships. ---- My understanding is that, When (magnetic) permeability is not different from a vacuum, the complex dielectric coefficient is equal to the square of the complex index of refraction. ------------------------ **** IMPORTANT CLARIFICATION/CORRECTION **** The statement that I / nr^2 was conserved in the absence of absorption, reflection, or scattering is true at least in so far as the group velocity and phase propagation are in the same direction. However, my intent was that the change in I over a distance dx along a ray path would then be given by letting I# = I / nr^2, and then the differential formula would be: d(I#(Q,v,P)) = Ibb#(v,T) * ecsv - I#(Q,v,P) * (acsv + scsv) * dx + Iscat * dx where Ibb#(v,T) = Ibb(v,T) * nr^2 is the blackbody intensity for the index of refraction (given per unit frequency dv and per unit polarization dP) and: ecsv, acsv, and scsv are the emission, absorption, and scattering cross sections per unit volume, Iscat * dx is the radiation scattered into the path from other directions, and any reflection (removing and/or adding to I along the path) at an interface within dx is included in the scattering terms. And of course, ecsv = acsv if in local thermodynamic equilibrium. --- Such a relationship is constructed with I# = I/ nr^2 is based on Snell's law where nr*sin(q) = constant (not quite true, actually (???) - see below) is the relationship that determines q as a function of N, where N varies only in one direction z (planes of constant N are normal to the z direction) and q is the angle from the z direction. I# = I / nr^2 is derived from Snell's law, by determining that the solid angle dw that encompasses a group of rays expands or compresses with variation in the index of refraction, specifically so that dw is proportional to 1/nr^2. **** HOWEVER: When N is complex, Snell's law actually still uses the complex N, not just it's real component. I haven't entirely figured out what that means for q, though I have the impression that nr*sin(q) = constant should be at least approximately true. Snell's law itself is based on the requirement that the phase surfaces of incident and tranmitted waves line up at an interface, and that the phase speed is inversely proportional to N when N = nr. What is the phase speed when N has a nonzero imaginary component? And then there is also the complexity of what happens if group velocity is not in the same (or exact opposite) direction as the wave vector (the wave vector is normal to phase planes and thus is in the direction of phase propagation). -------- So for the time being, let I# = I / n^2, where n is whatever N-related value that works in that relationship and also: d(I#(Q,v,P)) = Ibb#(v,T) * ecsv - I#(Q,v,P) * (acsv + scsv) * dx + Iscat * dx At least when N = nr and N is not a function of direction Q, n = N = nr. ________________________ REFLECTION AND EVANESCENT WAVES: When the entirety of wave amplitude and wave energy is reflected by an interface (as in total internal reflection), some of the energy actually penetrates across the interface. The portion of the wave across the interface that is associated with the reflected portion of the wave is called an evanescent wave. The energy and amplitude of the evanescent wave decay exponentially away from the interface into the region where the reflected wave cannot propagate (in the time average, there is no group velocity component normal to the interface within the region of the evanescent wave). The evanescent wave is mathematically required for continuity of the electric and magnetic fields, and for the time-integrated divergence of the energy flux to be zero over a wave cycle when there is a constant incident energy flux (and no absorption or emission). If there is another interface, beyond which the wave could propagate, then waves energy can emanate from that interface to the extent that the evanescent wave penetrates to it. This flux of energy must pull energy from the evanescent wave, which results in reduced reflection from the first interface. This is how wave energy can 'leak' through a barrier. The same general concept applies to mechanical waves, including fluid mechanical waves in the atmosphere and ocean, and to quantum mechanical waves - electrons tunnel through barriers via evanescent waves. Absorption within the region of the evanescent wave allows some net energy flow across the interface and thus reduces the reflectivity, so that even when a nonzero 'transmissivity' is not allowed except for tunneling, the reflectivity can be less than 1. I'm not sure but I think absorption on the side of a transmitted wave might also reduce reflectivity even when trasmission is allowed. (??) ________________________ I think I actually have the equations necessary to find some answers to questions just raised (if unable to find them elsewhere), but it will take time and it isn't pertinent to the matter here (I've already gone off on these tangents far enough). ________________________ GROUP VELOCITY: Group Velocity in geometric space specifically is equal to the gradient of the angular frequency (omega = 2*pi*v) in the corresponding wave vector space (a wave vector is a vector with components of wave numer; wave number is equal to 2*pi / wavelength measured in the correpsonding dimension; the wavelength in the direction of phase propagation is equal to 2*pi divided by the magnitude of the wave vector). Where the wavevector = [k,l,m], where k,l,and m are the wave numbers in the x, y, and z directions Angular frequency = omega group velocity in x,y,z space = [ d(omega)/dk , d(omega)/dl , d(omega)/dm ] phase speeds cx, cy, cz in directions x,y,z cx = omega/k cy = omega/l cz = omega/m IN the direction of phase propagation, the phase speed c and wavelength L are related as: c = v*L -------- IT IS POSSIBLE for some materials, at some values of v, to have a real component of the index of refraction less than 1. This (tends to or approximately??) corresponds to a phase speed that is greater than the speed of light in a vaccuum. This does not violate special relativity; the group velocity is the direction and speed of energy flux and information transport, and the group velocity will not be larger than the speed of light in a vaccuum. ________________________ Relativity: Effects are gravitational lensing and gravitational redshift, and effects related to relevant motion (of the Earth and sun, for example. One effect of relative motion is that small dust particles orbiting the sun tend to fall in toward the sun over time (their semimajor axes shrink) because as they orbit, they recieve radiation more from their leading side because of their motion, so radiation pressure tends to slow them down. Of course, radiation pressure also pushes them out - on the other hand, any stable orbit would be such that the outward radiation pressure would only partly cancel the gravitational acceleration, thus causing a slower but stable orbit in the absence of the radiative pressure torque on the orbit from orbital velocity. Larger objects of a given density have more mass per unit surface area and are less affected by radiation pressure. ________________________ Another way of looking at magnification by atmospheric refraction (considering mainly N = nr cases): For flat surfaces, leaving some vertical level moves upward across falling n values, total internal reflection keeps a portion of rays trapped; the other rays spread out to fill the full hemisphere solid angle, reducing the intensity and, if the radiation was initially isotropic, keeping the total upward flux per unit horizontal area proportional to n^2 at each level (so long as n only either decreases or remains constant with height). But for concentric spherical surfaces (with N decreasing outward to N = 1), the cone of acceptance defined for flat interfaces is narrower than the cone of rays that is able to escape upward to any given level, because as the rays curve over and downward, the interfaces - the locally defined horizontal surfaces - curve downward. Thus, the height to which any ray can reach is raised, and a greater solid angle of rays escapes all the way out to N = 1. This means a greater total upward flux per unit horizontal area reaches to any given height and to N = 1. But the intensity of the radiation still falls by the same amount as it passes to lower N (being proportional to N^2); so the greater flux requires a greater solid angle - hence, the underlying surfaces that are the source of the fluxes occupy a greater solid angle from any given viewing point - they appear larger - hence they are magnified. Some rays will still be trapped, so the magnification must be less than N^2. ---------- From some class notes: N of the air, in the visible part of the spectrum, at STP (standard temperature and pressure) is about 1.0003. Near sea level conditions are broadly similar to STP. If N ~= 1.0003, then N^2 ~= 1.0006. In that case, the effect on I would be a 0.06 % increase relative to I in a vaccuum. The magnification of the surface as seen from space cannot be more than that (and is also limited by the vertical extent of different N values). The magnification of higher layers will tend to be less because N will tend to decrease with height toward 1. ------------------ A particular important point about the climatic effects of increasing surface area and changes in N with height is that they would not be a source of significant positive or negative feedback (at least for Earthly conditions). As the greenhouse effect increases, the distribution of radiative cooling to space does shift upward. But a doubling of optical thickness per unit mass path would only shift the distribution within the atmosphere of transmissivity to space upward by about 5 km, give or take ~ 1 km (it is less at heights and locations where the temperature is colder, more where warmer). Doubling CO2 would have that effect only over the wavelengths in which it dominates (covering roughly 30 % of the total radiant power involved), and not quite even that, since at most wavelengths, emission cross sections are smaller with increasing height, at least for the troposphere and maybe lower stratosphere, and also, there would be some overlaps with clouds. Also, the tropopause level forcing would be due mostly to expansion in the wavelength interval in which CO2 significantly blocks radiation from the surface, water vapor, and clouds, given the shape of the CO2 absorption spectrum and that the central part of the CO2 band is saturated at the tropopause level with regards to further radiative forcing. The water vapor feedback is less than a doubling of water vapor, and water vapor density decreases much faster with height than air density within the troposphere. The height of the troposphere will shift upward to lower pressure as the temperature pattern shifts, so the highest cloud tops will be higher, but not by much in terms of affecting the emitting surface area. An upward shift in the LW radiative cooling distribution of about 5 km would result in about a 0.1 K cooling; it seems that a vertical shift less than 5 km likely causes a warming of about 3 +/- 1 K, 20 to 40 times larger (PS the vertical shift includes all LW radiative feedbacks; if there were no positive LW feedbacks, the warming would be reduced, but so would the vertical shift. If SW feedbacks were excluded, then maybe the warming would be at the lower end of the range given, I think). What would truly be required to result in a 0.1 K cooling by this process is if all LW radiative cooling that occured within the troposphere and at the surface were confined to be below 5 km from the tropopause initially (or else, to have the tropopause level rise to accomodate the shift?), and then to have the whole distribution raised 5 km, so that there would then be no LW cooling below 5 km from the surface. This would actually cause warming of roughly 30 K, given a lapse rate of 6 K/km. Thus the cooling by area increasing with height would be roughly just 1/3 % of the warming. There is yet one other way to vertically shift the LW cooling distribution: Thermal expansion. The atmospheric mass in the troposphere would only expand about 1 % in response to a 3 K warming, which would only contribute a 0.002 K cooling for an initial 10 km effective level for LW cooling. (This mass expansion is a seperate issue from the increasing height of the tropopause mentioned in the previous paragraph - the later is a rise in the tropopause relative to the distribution of mass - essentially a transfer of mass from the stratosphere to the troposphere. This is actually an expected result of global warming in general, although nothing on the order of 5 km so far as I know). --------- Feedbacks involving changes in the index of refraction of the air, and forced changes in the index of refraction of the air, can also be expected to be insignificant. ________________________ MORE ON THE 'I CAN SEE YOU AS MUCH AS YOU CAN SEE ME' PRINCIPLE (THAT IS REQUIRED if THE SECOND LAW OF THERMODYNAMICS IS TO APPLY): SPECULAR REFLECTION: For simplicity of illustration, consider an interface between materials, where on side A, N = NA, and on side B, N = NB, and in both cases, let the imaginary component be zero. Let N be invariant over direction within each material. Identify directions Q by two angles: Q = (q,h). q is the angle from the normal (perpendicular) to the interface (specifically, measured from the direction leaving the interface on the side that q is taken), and h is the angle around the normal to the surface; h going from 0 to 360 deg at constant q traces a circle in a plane parallel to the interface; rather than reversing the direction h is measured across the interface to keep the same overall coordinate system (right-handed or left-handed) for each side A and B, measure h on each side in the same sense (clockwise or counterclockwise) as viewed from just one side. Consider four ray paths that approach this interface. Two rays, 1 and 2, are incident from side B with directions Q1 and Q2, respectively. Two other rays, 3 and 4, are incident from side B with directions Q3 and Q4. Q1 = (qA,h0) Q2 = (qA,-h0) Q3 = (qB,-h0) Q4 = (qB,h0) So rays 1 and 2 have the same q = qA, rays 3 and 4 have the same q = qB, and 1 and 4 have the same h = h0, while 2 and 3 have the same h that is the opposite of the h of the other two rays. Let NA*sin(qA) = NB*sin(qB). A portion of each rays is reflected (denoted r) and a portion is transmitted (denoted t). For example, from the incident ray 1, the reflected ray is 1r and the transmitted ray is 1t. Notice: If all incident rays intersect the interface at the same point, then: ray 1t goes backwards along the path of ray 3 ray 3t goes backwards along the path of ray 1 ray 1r goes backwards along the path of ray 2 ray 2r goes backwards along the path of ray 1 ray 2t goes backwards along the path of ray 4 ray 4t goes backwards along the path of ray 2 ray 3r goes backwards along the path of ray 4 ray 4r goes backwards along the path of ray 3 etc. The incident rays have spectral (monochromatic) Intensities I# = I/N^2 of I1, I2, I3, I4. Along paths 1, 2, 3, and 4, the I/N^2 toward the interface are: I1 I2 I3 I4 IF the reflectivity from side A is RA and the reflectivity from side B is RB then: Along paths 1, 2, 3, and 4, the I/N^2 away from the interface are: RA*I2 + (1-RB)*I3 RA*I1 + (1-RB)*I4 RB*I4 + (1-RA)*I1 RB*I3 + (1-RA)*I2 Suppose each ray is emanating from a blackbody, and each blackbody has the same temperature. In that case, the net intensity (forwards - backwards) = 0 along each path so that there is no net heat transfer, assuming the second law of thermodynamics holds for the consequences of reflection and refraction. RA*I2 + (1-RB)*I3 - I1 = 0 RA*I1 + (1-RB)*I4 - I2 = 0 RB*I4 + (1-RA)*I1 - I3 = 0 RB*I3 + (1-RA)*I2 - I4 = 0 Also, I1 = I2 = I3 = I4. Then: RA + (1-RB) = 1 RA + (1-RB) = 1 RB + (1-RA) = 1 RB + (1-RA) = 1 Each relationship yields the same conclusion: RA = RB = R. Reflectivity is the same for any two rays approaching the same interface from opposite sides in which each of their transmitted rays goes backwards along the other incident ray. Reflectivity can vary with polarization P, so this only applies if either the incident rays are completely unpolarized or polarized specifically to fit some variation of N over P (as in perfect blackbody radiation), or if the intensity is evaluated for each polarization seperately. The formulas (one for parallel and one for perpendicular polarizations) for reflectivity (Fresnel relations) as a function of NA/NB and qA or qB (each q determines the other) give the same R for RA and RB for each polarization. The Fresnel relations are not determined by the second law of thermodynamics; they are determined (at least in part) with the constraint that the electric and magnetic fields only vary continuously in space; they each have only one value at any one location and time, including on the interface. I'm not sure if it is necessary for deriving the relations, but the Fresnel relations would also have to fit with the conservation of radiant energy (except for allowed sources and sinks such as emission and absorption). An interesting further point: Along ray 1, the I/N^2 coming back from the interface = R * I2 + (1-R) * I3 What happens to the radiation going toward the interface along I1 ? R * I1 is reflected backward along ray path 2, and (1-R) * I1 is transmitted backward along ray path 3. In other words, the distribution of where the radiation going forward along a path reaches matches the distribution of the sources of the radiation that goes backwards along the same path (before weighting by the strength of each source). ------ There should be a similar pattern of behavior for scattering. That is, if I is isotropic, then the I scattered out of a path must equal the I scattered into the path over the same unit of path dx. Otherwise, anisotropy in I# could spontaneously arise, which would break the second law of thermodynamics (clever use of such spontaneous anisotropy could run a perpetual motion machine). Of course, a scattering cross section per unit volume, scsv, can vary over direction, but it should (except for macroscopic scatterers where absorption cross section on one side can be matched with scattering cross section on the opposite side) be the same for a pair of opposing directions along a ray path. More generally: What I expect is that the distribution over directions of radiant intensity that is scatterd out of a ray going in the x direction over a distance dx should fit the distribution of the source directions of radiant intensity that is scattered into the ray in the same dx to go in the negative x direction, before weighting by anisotropy of the source direction intensities. When the same type of scattering has the same scattering cross section in all directions (and the scattering distribution has 0-fold rotational symmetry about the direction from which I is scattered), I think this can be shown to be true: Consider a fraction of radiation that is scattered by an angle q' into the solid angle dw in the direction Q2, from an initial direction Q1 and solid angle dw. The same angle of scattering applies to radiation that is scattered from the direction Q2 into the direction Q1. Thus the same fraction of radiation is scattered into Q2 from Q1 as is scattered from Q1 into Q2. ________ CONCLUSION (in the expectation that more complex forms of scattering and reflection and dependence on complex N values that vary over direction, and where group velocity is the direction of propagation of intensity, while phase propagation may be in a different direction - that these situations still fit the general concepts illustrated above): At a given frequency v, and if necessary, polarization P (could be a function of position along paths): At a reference point O along a path in the direction Q (which can bend as refraction requires), with distance forward along the path given by x: Considering the forward intensity I#f in the solid angle dw, the distribution of where I#f goes is proportional to the derivative of the transmission with respect to x. It can be visualized by considered a distribution of distance x over dw, where each cross section per unit area normal to the path over the unit distance dx (including reflectivity at any interface within dx) occupies a fraction of dw equal to itself multiplied by the transmission over x. The solid angle dw is the sum of many fractions of dw that are each filled by cross section at a different distance. The fractions are the fractions of I#f that are intercepted at the corresponding distances. They are also the fractions of the total visible cross section from point O at the corresponding distances. This is essentially a distribution of dw over x, that describes the distribution over x of what is visible along x from point O and of how much visibility of I#f at O exists at x. d(dw)/dx can be integrated along x to give a visible cross section per unit area of 1. I#f(O) * d(dw)/dx is the interception at x per unit dx of whatever I#f passes through O. Integration of d(I#b(x)) * d(dw)/dx gives the total backward intensity I#b(O) that reaches O; it sums the proximate sources of I#b(O) over the fractions of dw that correspond to a distribution over x. This can be extended farther. Suppose we are interested in the absorption of I#f. For all the fractions of dw that are scattering or reflection cross sections per unit area, the fate of those fractions of I#f can be traced farther, through successive scatterings and reflections, until every last bit is absorbed. This will be a distribution of dw that may extend outside of the path and over other paths, perhaps over some volume. It is the distribution of the absorption of I#f(O). Assuming local thermodynamic equilibrium within each unit volume, It is also the distribution of the emission of I#b(O) - multiplying the distribution density by I#bb(T) and integrating over the distribution gives the I#b(O) value, where I#bb(T) is the blackbody intensity (normalized relative to refraction) as a function of T, however it varies over space. I#f(O) also has an emission distribution that can be found, tracing back in the opposite direction from point O. The net I# at point O in the forward direction is I#(O) = I#f(O) - I#b(O), and will be in some way proportional to the difference in temperature T (specifically, the difference in the weighted average of Ibb#(T) over the emission distribution) between the emission distribution of I#f(O) and the emission distribution of I#b(O). Assuming the forward direction is from higher to lower temperature within the emission distributions, if the two distributions extend over a larger range in T (at a given overall average T), then the net radiant transfer I#(O) will be larger; if the two distributions extend over a smaller range in T (at a given overall average T), then the net radiant transfer I#(O) will be smaller. I#(O) can be changed by either changing the emission distributions or the temperature distributions. Increasing overlap of the distributions will tend to reduce I#(O) (the distribution of emission for either I#b(O) and/or I#f(O) can wrap around the point O when there is sufficient scattering and/or reflection). EXAMPLES: ----------- PURE EMISSION AND ABSORPTION (loss to space is 'absorption' by space for climatological purposes): The emission distributions of each of I#f(O) and I#b(O) are entirely along the ray path, with zero overlap. Each distribution density decays exponentially from point O in optical thickness coordinates; the distributions are more concentrated near point O in (x,y,z) space when there is greater cross section per unit (x,y,z) volume. If there is an emitting/absorbing surface (approximately infinite optical thickness per unit distance), that the ray path intersects, increasing the opacity of the intervening distance between the surface and O reduces the portion of the distribution that is on the surface, pulling it into the path between the surface and O while concentrating the distribution within the path toward point O. For a given temperature variation along the path, if the temperature is increasing in one direction, increasing the cross section per unit volume decreases the net I#(O). IF the temperature fluctuates sinusoidally about a constant average, then increasing the cross section per unit volume may have little effect on net I#(O), until the emission distribution becomes concentrated into a small number of wavelengths; if the temperature is antisymmetric about O, the maximum net I#(O) will occur when the emission distribution is mostly within 1 to 1/2 wavelength, so the nearest temperature maximum and minimum dominate the emission distribution; beyond that point, further concentration of the emission distribution reduces I#(O). ----------- PURE SCATTERING within a volume between perfect blackbody surfaces at temperatures T1 and T2 (with 100 % emissivity and absorptivity): When the scattering cross section density is zero, net I#(O) is the difference between Ibb#(T1) and Ibb#(T2), only except for paths that are parallel to the surfaces, or with variations in N, never intersect both surfaces. The emission distributions will only be at the surfaces and not in the intervening space. The effect of scattering will be to redirect radiation so that a portion of the emission distribution for I#(O) from one side of O will come from the other side of O; this reduces the net I#(O) by mixing some of both Ibb#(T1) and Ibb#(T2) into both I#f(O) and I#b(O). Without multiple scattering, forward scattering makes no difference for direction Q that is everwhere normal to the surfaces (assuming constant N everywhere). Increasing either the scattering cross section density, the deflection angles of forward scattering, the portion of single scatter backscattering, or the angle from the normal perpendicular the surfaces, will increase the portion of the emission distribution for I#(O) from one side of O that is shifted to the other side of O. Interestingly, scattering could introduce some net I# into paths that do not intersect both surfaces, such as those that intersect the same surface twice due to total internal reflection. But increasing scattering will eventually reduce I# for even those cases. Partially reflecting surfaces between the emitting surfaces will have the same general effect as scattering; the emission distributions will not fill a volume with specular reflection alone, but they will spread out over a branching network of paths that penetrates space both back and forth. **** Having an emitting surface that have emissivity and absorptivity less than 100% would be analogous to placing either some partially reflective surfaces and/or a layer of high scattering cross section density immediately in front of the surface so that none of the points O considered would be between that layer and the emitting surface. ------------------------- In the case of perfect transparency between surfaces with reflectivities R1 and R2, the I#f from surface 1 to surface 2 along any path that intersects both surfaces would be (PS this assumes that R1 and R2 are not direction dependent, or happen to be constant for all the directions that occur when tracing back and forth along a given path at a point O; at least in the case of no directional dependence, this applies to both specular and diffuse reflection): (1-R1) * Ibb#(T1) + R1*(1-R2)*Ibb#(T2) + R1*R2*(1-R1)*Ibb#(T1) + ... = SUM(j=0 to infinity)[ Ibb#(T1) * (1-R1)*(R1*R2)^j + Ibb#(T2) * R1*(1-R2)*(R1*R2)^j ] --- And the I#b would be: SUM(j=0 to infinity)[ Ibb#(T2) * (1-R2)*(R1*R2)^j + Ibb#(T1) * R2*(1-R1)*(R1*R2)^j ] --- So the net I# would be: SUM(j=0 to infinity)[ Ibb#(T1) * (1-R1)*(R1*R2)^j + Ibb#(T2) * R1*(1-R2)*(R1*R2)^j ] - SUM(j=0 to infinity)[ Ibb#(T1) * R2*(1-R1)*(R1*R2)^j + Ibb#(T2) * (1-R2)*(R1*R2)^j ] = SUM(j=0 to infinity)[ Ibb#(T1) * (1-R1)*(1-R2)*(R1*R2)^j - Ibb#(T2) * (1-R1)*(1-R2)*(R1*R2)^j ] = [ Ibb#(T1) - Ibb#(T2) ] * SUM(j=0 to infinity)[(1-R1)*(1-R2)*(R1*R2)^j ] = [ Ibb#(T1) - Ibb#(T2) ] * (1-R1)*(1-R2) / [1-(R1*R2)] ------- When R2 = 0, I# = [ Ibb#(T1) - Ibb#(T2) ] * (1-R1) --- When R2 = R1 = R, I# = [ Ibb#(T1) - Ibb#(T2) ] * (1-R)*(1-R) / [(1-R)*(1+R)] = [ Ibb#(T1) - Ibb#(T2) ] * (1-R)/(1+R) ------- Increasing either R1 or R2 decreases I#. ------------------------- MIX OF SCATTERING/REFLECTION AND ABSORPTION/EMISSION (between two opaque surfaces with nonzero emissivities and absorptivities): ----------- Weak scattering and reflection relative to absorption/emission: Adding scattering and reflection concentrate the emission distributions closer to O, pulling them off of and away from any opaque surfaces, and spread them out from the ray path into a volume of space, and can cause some overlap of the two distributions. Specular reflection will not cause the emission distributions to fill a volume but will have other effects broadly similar to scattering. For the same scattering cross section density, single scatter dominated by forward scattering with small deflections has the weakest effect. Most of the photons may only scatter or reflect once if the emission/absorption cross section density is enough relative to scattering/reflection cross section density. ----------- Weak emission/absorption relative to scattering/reflection: Adding emission/absorption cross section density pulls the emission distributions off of the opaque surfaces; that portion which is lifted off the surfaces is concentrated near the point O. Adding more emission/absorption cross section pulls more of the emission distributions off the opaque surfaces and increases the concentration near O. With sufficient scattering relative to emission, multiple scattering will tend to diffuse an intensely anisotropic I# into nearly isotropic I#; this will make the emission distributions for both I#f(O) and I#b(O) into nearly spherical regions that are both centered near O, so that there will be great overlap of the two distributions, reducing the net I#(O). With less emission/absorption cross section density, the spheres expand; with zero emission/absorption within space between opaque surfaces, the distributions are left on the surfaces (both distributions nearly evenly divided among surfaces if scattering and/or reflection is sufficient). ----------- Varying proportions of emission/absorption cross section and scattering/reflection cross section: All cross sections tend to restrict the emission distributions, but within the larger distribution, pockets of higher densities of emission cross section will have a greater density of the emission distribution. Pockets of sufficiently high densities of scattering and reflection cross sections may reflect and deflect the emission distribution around themselves. ------------------------ -
Sam Spade at 11:10 AM on 7 June 2009Glaciers are growing
........SKEPTIC SOURCES My focus above was on increasing glaciers that were referred to by skeptics. I got them from an endless skeptic bibliography...that's split up into a hundred subdivisions. z4.invisionfree.com/Popular_Technology/ index.php?showtopic=2050 GW skepticism being what it is, the counter argument was often also there, in the same article. In addition, there'd be some "google words", on which to continue. The other skeptic source I know about is also subdivided; and includes some text, with a literature review. appinsys.com/globalwarming/ A neophyte believer blogger's first source would likely be a skeptical argument website, like skeptical science.com. Is appinsys the first source for a neophyte skeptic blogger? It would be useful to know where the other blogger is getting his information. For example, I've always been amazed at skeptics' ability to produce weather related fatality statistics. When I've never seen any. Until... There's no hope for the cynical skeptic, who'll cherrypick anything anywhere. And the gullible skeptic can probably be reached only on his own entry level conspiracies. But I think the more open-minded ('skeptical') skeptic would eventually see through this disjointed collection of counter arguments...that don't add up to a realistic whole. Are there any other useful skeptic sources? -
chris at 07:40 AM on 7 June 2009The correlation between CO2 and temperature
This is stating the obvious possibly, but it’s worth supplementing the top article with a more explicit explanation of why we don’t expect to observe a correlation between CO2 (levels) and (surface) temperature in the short term. If we take the middle of the range of climate sensitivities consistent with empirical evidence, of around 3 oC of surface warming for each doubling of atmospheric CO2 concentration, it’s straightforward to calculate the change in surface temperature at equilibrium resulting from the yearly increase in CO2. This is about 0.023 oC for current rates of increase of CO2 (e.g. 382 ppm to 384 ppm). Does this mean that the Earth’s temperature should rise by 0.023 oC each year? Obviously no, for two reasons: (i) this is the temperature rise at equilibrium. In fact the current temperature rises are occurring largely in response to the raised CO2 levels of the previous decades. If the atmospheric CO2 levels were to drop by 2 ppm per year during the next 5 years or rise by 2 ppm per year during the next 5 years, the effects on surface temperature wouldn’t be expected to be very different. In each case the temperature progression will be a response to a raised radiative forcing arising from enhanced [CO2] near 385 (+/- 10) ppm. There is no expectation of a correlation between changes in [CO2] and changes in surface temperature on this sort of timescale. (ii) The dominant contributions to year on year variations in surface temperature are intrinsic variations in the climate system. The effects of El Nino’s, La Nina’s, the solar cycle, volcanoes etc. can result in year on year changes of up to around 0.2 oC. Obviously these effects are likely to outweigh the small contribution from enhanced [CO2] forcing. Since the internal variations are expected to result in “noise” while the contribution to the surface temperature from raised [CO2] are persistent, the latter should rise out of the noise as a trend on the longer term. A temperature rise of the order of 0.2 oC per decade with internal variation of the order of 0.1-0.2 oC per year, should be apparent as a trend averaged on the decadal time scale. (iii) If [CO2] rises result in raised surface temperature, where should we observe correlations? We should expect to observe good correlations between the atmospheric [CO2] and the equilibrium surface temperature. This is apparent in the paleorecords sampled throughout the last 500 million years. If we have a good handle on the timescales of the surface temperature response to enhanced radiative forcing (i.e. the progression of the surface temperature to raised CO2 levels that result from the various inertial elements of the climate system), we should expect to observe a good relationship between the observed temporal progression of the surface temperature and modelled temperature, as long as we recognise that the stochastic variations will result in unpredictable (and therefore non-reproducible in forecast) “noise”. -
Quietman at 06:35 AM on 7 June 2009Arctic sea ice melt - natural or man-made?
Re: "We have reason now to actually impose a price signal on emissions (or some other policy) to shift the economy towards greater demand of and investment in clean energy and energy efficiency as well as other emissions-reducing pathways. I do agree that there is room for debate about how large that price signal should be," And there is where the problem resides. In countries where the civilians are disarmed it may work but it will not worked in truly free countries. This is why the gun grabbers have been trying to impose gun laws that allow them to disarm as many of the populace they can. And they already know that our military will not forcibly disarm Americans (they asked the troops if they would be willing to exactly this about 10 years and they plain out refused). This is a built-in check to keep our government from imposing laws against the will of the people and subvert the constitution. So they have to tread softly on such issues and try to get the populace to actually believe that this is for our good and not making headway. If they force the issue it will get very ugly very quick. This is why this is a political hot potato and needs to be approached with unassailable science that everyone can accept, and they can not do that. -
roverdc at 00:48 AM on 7 June 2009Models are unreliable
A common argument heard is "scientists can't even predict the weather next week, how can they predict the climate years from now". This betrays a misunderstanding of the difference between weather, which is chaotic and unpredictable and climate which is weather averaged out over time. While you can't predict with certainty whether a coin will land heads or tails, you can predict the statistical results of a large number of coin tosses. Or expressing that in weather terms, you can't predict the exact route a storm will take but the average temperature and precipitation will result the same for the region over a period of time. Using this argument is admitting that there is no such thing as climate science , only climate statistics. It is also a dodgy use of statistics in that the toss of a coin is fundamentally a result of a physical law with a random element of how hard the spin is induced. If you have any doubts on this score try a simple rig which drops a coin horizontally to catch its edge on a bar placed in its path. It is possible to get about 90% selection either way by moving the bar if the coin lands on a relatively bounce free surface. I was told that it is actually possible to get better than 99% with a fancier rig but never tried to do so as I only needed 75% for my free lunch. Can anyone point to the references for the IPCC peer reviewed projection of the effect of cloud cover changes in both type and coverage as a result of temperature and the more erratic weather patterns they claim will occur? -
John Cross at 21:33 PM on 6 June 2009The correlation between CO2 and temperature
Andrew K. I don't follow your logic. When you say that 4 cooling periods follow volcanoes, you imply that they are caused by volcanoes. For a cooling period of say 8 years to be caused by volcanoes, we would need the cooling to be somewhat minimal at the early stage and then increase towards the end. Keep in mind to cause a cooling period you need to suppress the end of the period more than the start. From what I understand about a volcano the SO2 has a residency time of about 2 years in the stratosphere. So from what I see a volcano at the start of a cooling period would actually tend to mask a natural cooling trend. Do you see something wrong with that logic? Regards, John -
shawnhet at 01:38 AM on 6 June 2009What causes short term changes in ocean heat?
I wonder if Willis 2004 is only addressing the correlation of the ENSO with ocean heat over the most recent period. Personally, I think that if the correlation goes in opposite directions at different times(as it seems to in the 70s as opposed to recent events, it is premature to ascribe the more recent cooling to ENSO shift. Further, I am not convinced that claiming strong La Ninas lead to less ocean heat actually makes physical sense. Assuming a constant input of energy into the oceans, warmer sea surface temperatures should, it seems to me, lead to more emission of heat from the ocean(and hence, less ocean heat gain). However, claiming that La Ninas cause less ocean heat to be stored is a situation where there is both less emission of OH and less storage of it. Cheers, :) -
MattJ at 01:19 AM on 6 June 2009Scientists can't even predict weather
Please update this to use more recent RSS data, as described in http://www.topix.net/forum/news/global-warming/TBSJVGR5M9Q6A7B8H/post87 -
Patrick 027 at 14:27 PM on 5 June 2009It's the sun
(continued)... ... but for the flux per unit area (integrated over solid angle), the fraction that is trapped by total internal reflection remains the same. ------------- CORRECTION: following ray paths without scattering, absorption and emission, or reflection, over variations in n: it is I / n^2 that is conserved. Not I * n^2 . ------------- Considering the case where there is no reflection, scattering, or emission and absorption along a path over varying n, the conservation of I / n^2 implies that blackbody radiation Ibb(v,T) must be proportional to n^2, so that observed blackbody radiation at any given n, such as n=1 (vaccuum, approximately in air), has the same intensity regardless of the n at the location of emission. When following ray paths, the direction of the path obviously can bend as n varies, so the direction Q is a function of location (with location itself being a function of direction from another given location - outside of some simplified conditions, calculating location and direction requires integration of directional changes along the path from an initial location.) Q in general has two components, which can be the angles in spherical coordinates; the q used earlier is the angle from the top pole (the colatitude: latitude - 90 deg), which is all that is necessary if conditions do not vary in planes perpendicular to the axis of the coordinate system; more generally, the angle around the axis (analogous to longitude) could also be specified. n can depend on direction in materials, in which case n is a function of Q. I would guess it could also be a function of polarization P (is that what produces the double images through calcite?: note to self - look up birefringence). Many people are aware that in materials it is often a function of frequency v (why we have rainbows). Of course, n = c_vacuum / c_material, the ratio of the phase speed in a vacuum to the phase speed in the material (in the direction being considered) ... BUT when n is a function of frequency v, the group velocity - the direction and speed at which energy is carried by wave amplitude propagation - can be different from the phase direction and speed. Presumably, they will still be in the same direction (or directly opposite each other, as in the case of certain metamaterials with a negative index of refraction) when n does not vary with direction; I wonder what happens if n varies in direction; then perhaps group velocity might be at some angle to phase propagation - in which case, the ray path to follow when evaluating changes in I would be along the path defined by group velocity, ... and the n for which I / n^2 is conserved in the absence of reflection, scattering, and absorption and emission, is (??????). (PS it is actually common to deal with fluid mechanical waves in the atmosphere (and, I presume, the ocean) (gravity waves, inertio-gravity waves, Rossby waves, Rossby-gravity waves, Kelvin waves) for which the group velocity and phase propagation, relative to the air (as it moves or doesn't), are perpendicular to each other.) PS n is the real component of the index of refraction; the index of refraction can actually be a complex value, with an imaginary component that is related to the absorption cross section per unit volume (the absorption coefficient). ------------ Fortunately, radiation transfer through the atmosphere is little affected by variations in the real component of the index of refraction of the bulk air (obviously it figures into the scattering properties of cloud droplets and aerosols). The most significant effects I am aware of is the twinkling of starlight and the shifting of the sunset and sunrise times. (Also, gravitational redshifting is not a big effect for radiation into and out of the Earth; I won't bother going into the relativistic effects on radiation here.) But it is interesting to consider what qualitative effect it would have, as well as the spherical geometry of the Earth (the layers of the atmosphere have slightly greater area than the surface, and so can emit the same total power at a slightly lower temperature than otherwise.) The effect of n slightly greater than 1 increases the length of the day as seen from the surface. This implies that, aside from atmospheric albedo contributions, the surface of the Earth intercepts a greater amount of sunlight than would actually pass through it's cross sectional area. Indeed, the implication of the index of refraction of the atmosphere being slightly greater than space is that the Earth would appear slightly bigger from space than it actually is - In general, any given spherical surface below the 'top' of the atmosphere will be magnified by the n greater than 1 of the air above it - but obviously, cannot appear any bigger than the spherical boundary of the layers of air that are magnifying it. This happens because, when looking at the edge of the Earth, the line of sight intersects surfaces of constant n within the atmosphere around the Earth at a glancing angle and is bent toward the Earth's surface; the visible edge of the Earth corresponds with the lines of sight that bend only enough to reach the surface nearly horizontally, and those lines of sight will reach the surface somewhat behind the front half of the Earth (as defined by viewing position). (This description applies to the view from an infinite distance; obviously, without refraction, much less than half the Earth would be visible from nearby, but refraction would extend the area visible by some amount.) If n were higher for solar radiation than terrestrial radiation, then it would increase the solar heating of the Earth and any given layer of atmosphere more than it would increase the LW radiative cooling to space from the same vertical levels. I don't know if that is the case, but it is a very small effect that can be ignored for most purposes anyway. Most of the radiation to space comes from within the troposphere, and a majority of direct solar heating occurs at the surface (or within the surface material, as in the ocean). For a very quick analysis to put some likely bound on the effect of this, suppose the radiation to space came entirely from a height of 16 km above the surface (that's within the stratosphere outside of the tropics, and close to the tropopause within the tropics), while all solar heating were at the surface. In that case, the effective surface area emitting to space would be about 2*(16/6371) ~= 0.5 % larger than the surface that is heated by the sun, which would allow the effective emitting temperature to be (0.5 % /4) = 0.125 % smaller than that required to radiate to space enough to balance solar heating with the same area as the surface. That would be about 2.55 / 8 ~= 0.32 K cooler. Not very much compared to a 33 K greenhouse effect overall. So both the increase in area with height and the refraction by the air can be ignored for climatologically-relavent radiative fluxes. Also note that the refraction would tend to counteract the effect of greater area with height, since it would magnify areas beneath greater amounts of atmosphere more than higher level areas. ------------ -
Andrew K at 14:23 PM on 5 June 2009The correlation between CO2 and temperature
Prehaps this chart can help http://globalwarmingquestions.googlepages.com/trendvolc.jpg Since 1960 there have been 5 cooling periods the first 4 followed volcanoes -- the last (this century) doesn't. So what is the mechanism? If it is just noise then we can test statisitically -- is the noise greater than we would expect from previous data or not. When we do the simple CO2 v temp or the more complex IPCC multi-model mean v temp or even the very simple 2 deg/century v temp all currently fail at 95% confidence. So if it is not noise and not volcanoes what is it? BTW it's not solar cycle as Gavin Schimdt said in 2005 at Real Climate (and since not recanted) that even a variation of 0.1 deg C per solar cycle is an over estimate.Response: There's not one single driver of climate so I don't think it's a case of looking for a single smoking gun. Two contributors to the current cooling phase would be the solar minimum (whether it be 0.1C cycle or slightly less) and the strong La Nina conditions in the Pacific Ocean. I explore this further on the global cooling page. If you're interested in the physical mechanism by which La Nina cools global temperatures, I recommend reading the previous post on what causes short term changes in ocean heat. -
Dan Pangburn at 05:26 AM on 5 June 2009It's the sun
I'm OK with radiated. I don't know which is less confusing to most people. -
Patrick 027 at 05:20 AM on 5 June 2009It's the sun
In some cases, it may be necessary to consider I not just along specific directions and at specific wavelengths, but also by polarization. At a given direction q and wavelength L, the intensity I(q,L) is the sum of contributions of I from each polarization P: I(q,L,P). Scattering and emission/absorption cross sections could be functions of polarizations in cases where the particles involved lack spherical symmetry and are not oriented at random. Within the atmosphere, ice crystals can have prefered orientations, and rain drops flatten slightly as they fall (though I wouldn't think the later has much significance overall). Reflection where there is a relatively sharp change in the index of refraction is also a function of polarization. ------- Contributions to the intensity I(q,L,P) could also be seperated by phase - I don't know if there would be much point in this except in finding the brightness temperature of laser light that corresponds with the entropy; perfect blackbody radiation, as well as being isotropic across all q, is evenly distributed over polarizations P and phases (it is incoherent); so it would be of interest to know the temperature of a blackbody that can emit radiation with I(q,L,P) for a single phase equal to the intensity I(q,L,P) of some coherent radiation (defining brightness temperature with such specificity, any significant nonzero I that is either perfectly monochromatic (dL = 0), perfectly polarized (dP = 0), or perfectly coherent would have infinite brightness temperature and zero entropy - it would be all work and no heat). --------- Particles much smaller than a wavelength of radiation can change the overall index of refraction, but an even distribution of such particles would not cause a spatial variation in the index of refraction. When the index of refraction for the bulk medium (as opposed to just the scattering and absorption/emission agents) must be considered, then it will be more convenient to use frequency v instead of wavelength L to specify the part of spectrum considered. The monochromatic intensity is better given as the intensity per frequency interval dv because wavelength intervals dL can change. Alternatively, one could refer to L# as the wavelength the radiation of that frequency would have if in a vacuum (and then define the spectral intensity as the intensity per unit L#. Following I along a path when the real component of the index of refraction, n, changes, I in the absence of scattering, reflection, or absorption and emission, is not what is conserved along the path; it is I * n^2 that is conserved. This is because a unit of solid angle dw that envelopes a group of rays will be compressed or expanded by the square of the ratio of different n values following the refracted paths. This relates to total internal reflection: for a flat surface, there is a cone of accepantance through which some radiation can pass from high n to low n (100 % if there is a perfect antireflection coating (if and when such a thing exists); outside of that cone, there will be (in the absence of scattering and absorption) 100 % reflection. If the transition from high n to low n is gradual, the cone of acceptance applies, but instead of reflecting at a specific interface, ray paths outside the cone curve back gradually, as ray paths within the cone curve to spread out to fill the full hemisphere of solid angles (additional transition to lower n defines a narrower cone of accetance for the rays that reach over the larger variation of n without curving back or reflecting). Relative to wavelength, small-scale texterization can mimic a gradual variation of n and reduce reflection within the cone of acceptance in that way; larger-scale texterization can reduce reflection via successive intersections of a ray with the interface, so that the amount that transmits across the interface accumulates with each intersection; in that case, the cone of acceptance may not be so easily defined due to different angles, but ... -
Patrick 027 at 04:39 AM on 5 June 2009It's the sun
"Near the surface, most is re-radiated by GHGs as required for the surface" This can be a point of confusion: people throw around terms like 're-radiate' a bit too loosely. The rate of thermalization is sufficiently high in the great majority of the atmopshere by mass and by optical thickness that GHGs do not radiate by fluorescence - which is what some people seem to think (and they may get that idea from the term 're-radiate'); instead, GHGs and other greenhouse agents emit photons mainly from thermal energy, which they can get from the bulk air from molecular collisions. -
Steve L at 02:18 AM on 5 June 2009The correlation between CO2 and temperature
Another problem for the senator is probably the use of graphs scaled by others, such as the ones you show. The increase in CO2 from 2002-2008 looks like a lot when the y-axis is 370-400 ppm, but the increase is about 12 ppm. That's an increase of about 3% or less than 1% per year. It's not the increase in any particular year that's having an effect so much as the accumulation over time. It took an increase of about 20-25% over pre-industrial (and a lot of time, as this post points out) before the signal started to really stand out from the noise. From this perspective, the lack of signal from 2002-2008 of a 3% increase in atmospheric CO2 in Figure 1 doesn't impress. I have no refs from attribution studies for my comments -- just common sense and the figures above. Hopefully common sense isn't misleading here. -
Dan Pangburn at 02:06 AM on 5 June 2009It's the sun
The other K&T updated values referred to are the 78 of incoming absorbed by the atmosphere, the 17 thermals and 80 latent. The 78 seems high. Reduction of this increases thermalization by the same amount and decreases back radiation by the same amount. Reducing the 78 to 10 increases thermalization by 68 to 105 and reduces back radiation from the atmosphere to 237. That which gets from average clouds to ground remains at 13. -
John Cross at 01:50 AM on 5 June 2009The correlation between CO2 and temperature
Opps, of course it should be El Chichon. At least I am consistent in my errors. John -
John Cross at 01:47 AM on 5 June 2009The correlation between CO2 and temperature
John: Just to emphasize my point to Andrew, since El Chicon (the 1982 volcano) takes place early in your second series (1981 to 1989) it would actually cause a cooler reading than you would expect in the early part of your data which is not there in the latter part. So in fact it would tend to bias the result warm. Best, John
Prev 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 Next