May 31 2007
Basic Climate Science Refresher - I
This is an open response to Ponder the Maunder. A copy will be sent via email to the author.
Common Sense
Common sense is the collection of prejudices acquired by age eighteen.
Albert Einstein. Quoted in E T Bell Mathematics, Queen and Servant of the Sciences. 1952.
Common sense is not always right. Newton’s first law says that an object in motion will remain in motion at a constant velocity until a force is applied. Common sense tells us that an object in motion will steadily decrease its speed until it stops. Friction is what causes objects to slow down. It is not trivial to think of a frictionless world. In a frictionless world, you would not be able to walk, because it’s the friction between your shoe and the floor that allows you to propel yourself forward.
Common sense tells us that a heavier object falls faster than a lighter object. Newton’s third law of motion tells us that the acceleration of an object due to gravity is the same; a hammer and a feather will both accelerate towards the Earth at exactly the same rate.
Einstein’s theory of special relativity says that the velocity of light is a constant, and that both space (length) and time are variable. Common sense tells us that neither space nor time can vary.
CO2 Absorption
CO2 absorption spectrum is measured and theoretically computed. The results are within 3% error, which is really good. The HITRAN database uses the observed data for CO2, unless there is a good reason to not use it, as explained in The HITRAN 2004 molecular spectroscopic database. Another recent paper measuring the absorption of CO2 (there have been many over the years of varying accuracy and precision) is Near infrared spectroscopy of carbon dioxide I. 16O12C16O line positions.
CO2 Concentration
“Were you aware that the total man made CO2 increase in the atmosphere over the last 150 years is just that, one ten-thousandth of total atmosphere?”
The atmosphere is composed of approximately 78% nitrogen, 21% oxygen and 1% argon. That adds up to 100% which must mean there are no other species in the atmosphere. Of course that’s silly, the numbers above have been rounded to the nearest whole number, but it does mean that almost the entire atmosphere is composed of three molecules (in fact, three elements). Nitrogen, oxygen and argon make up 99.96% of the atmosphere. Water vapor concentration is highly variable, and can constitute as much as 4% of the atmosphere or very little - near 0%. Because of this high variability, it is difficult to place an exact number on the concentration of water vapor. Carbon dioxide is the fourth most abundant non-spatially-variable gas in the atmosphere with just 0.035%.
 |
 |
Above are two graphs of the absorptivity of the atmosphere. Absorptivity is the amount of radiation absorbed normalized by the amount of incoming radiation. Therefore, the maximum absorptivity is 1, and the minimum is 0. We can see from the second panel in the top graph that O2 and O3 are highly absorbing at short wavelengths - the ultraviolet, xray, and gamma ray wavelengths. N2O (nitrous oxide or laughing gas) has two main absorption peaks around 4.5 and 8 micrometers. H2O has the most absorption peaks scattered from 0.7 to 10 micrometers. And CO2 has two main peaks at about 2.7 and 4.3 micrometers. The bottom panel of the top figure is the same as the bottom figure in the bottom figure and they show the spectral absorptivity as measured at the surface of the Earth. The top figure identifies which which feature belongs to which molecule, and also contains a plot of the absorptivity at 11 km, which is around the tropopause. The top panel of the top figure is interesting because it gives the blackbody radiation spectrum of bodies at 5780 K (the Sun) and 255 K (the Earth). You’ll notice that the main two CO2 spectral features lie between the two blackbody curves. However, there is a small CO2 feature at 12 micrometers. This feature is a peak in the absorptivity spectrum that does not reach 1. This means that is is not “saturated”. More CO2 will cause that feature to increase until it reaches 1.
Most of the absorption in the atmosphere is not due to the main components; molecular nitrogen doesn’t even show up on the figures, and molecular oxygen (O2) is mostly at small wavelengths.
The sun emits radiation roughly as a blackbody at a temperature of 5780 K (9950 F). The peak wavelength, as shown in the figure above, for an object with this temperature is around 0.5 micrometers (what we perceive as green). This is radiated isotropically, equally in all directions. A very small fraction of the light that the Sun emits is intercepted by the Earth. We can calculate this value based upon the energy emitted by the sun, the distance from the Sun to the Earth, and the radius of the Earth. This value is approximately 1367 W/m2.
The Earth also emits as a blackbody. If we assume that the temperature of the Earth is in equilibrium (it’s not, but it’s close enough), then the incoming energy needs to equal the outgoing energy. The total energy absorbed by the Earth is 1367 W/m2 (S0) multiplied by the area it intersects - πR2. But we need to account for the albedo (or reflectivity) of the Earth. So the actual energy absorbed equal S0(1-A)πR2.
(No Ratings Yet)
Loading ...Related Posts:
18 Responses to “Basic Climate Science Refresher - I”
To reduce spam, comments are automatically closed 30 days after the last comment. If you would like to comment on any closed thread, please use the contact form at the top of this page.
Hi Nathan,
Thanks for the invite to visit your site and rudely or otherwise rip your blog apart. LOLOL
I am just finishing up an article that will be published shortly then have finals all next week. Then I will be away for a few weeks visiting family.
I have breezed through your post and I do indeed see debatable material and will be happy to respond when I have time. It may be in parts as time allows. In the mean time I will send your link around and see if anyone wants to take you up on reviewing your material.
Kristen
[Ed: All ad hominem attacks have been removed from this post, which makes it a lot shorter than what it originally was. If you wish to converse in a civil way that's fine, but your reply was unacceptable. Please see my non-existent comments policy which says that I can say whatever I like on my blog, but commenters are to remain civil or their replies will be edited or tossed in the trash.]
Let us look at your statistical analysis of SOI. You compare raw data of pressure to raw data of temperature. Then you have them up side down. You do not know enough about modeling to adjust for volcanoes, bad records during WW2, time lag solar factors and etc. Climate is a rather chaotic system. You did not identify your source of global temperatures or SOI.
[Response: Actually, in the second plot just the SOI is negated. And it is only done so that it is easier to see the correlation between SOI and temperature. Modeling has absolutely nothing to do with ploting a temperature or a pressure measurement. I did source the data I used, they were linked in the post. The SOI came from UCAR, and the global temperatures came from the Hadley Center. Reread that section and follow the links to get the data. Radiation from the sun takes about 8 minutes to reach the Earth; there is no solar time lag.]
Invert your SOI and it becomes basic ENSO. You can make the proper adjustments by multiplying SOI by -14 assuming you are using SOI from NOAA [Ed: I'm not. See above and read the original post.] and comparing it to Land + Ocean Index from NOAA. When you graph it keep the monthly and smoothed (20 month smoothing is what a friend sent in a spreadsheet) data on the same graph. You will need to delete the data for about 2-3 years after each volcano; Agung 1960’s, El Chichon 1983 and Pinatubo 1991. When you look at your graph you will be able to see the lag time, it depends on the season the volcano blew.
[Response: Invert my SOI and it becomes ENSO? That makes no sense. The Southern Oscillation Index is a measurement of air pressure difference between Tahiti and Darwin and is used to tell what stage the El Niño-Southern Oscillation is in. By the way, this is exactly what I have plotted in my second figure, except I used a smoothing of 8 years and not 20. The amount of smoothing should not affect the results. Ask your friend to send you the raw SOI and temperature data, and then smooth it yourself. And in my original post I wrote, "Using a frequency filter with a cutoff that corresponds to about 25-30 year produces a similar, although not exact, plot as shown in the blue curve directly above." So I have looked at exactly what you did - the SOI with a ~20-year smoothing.]
During WW2, for obvious reasons, the temperature data is flawed. ENSO data before 1950 is from proxies so you have a data trust problem there but I think the margins of error are kind of on the low side. You might also notice that in 1910 temperature departs ENSO on the lower side. The same happens after about 1976 (on the high side) after ENSO shifts phases to positive (PDO as well) while solar is high (probably more to do with Svensmark effect than TSI but it is still cumulative). If you adjust for the solar departures, your R-value will be about 80. That is because most of the chaos in the system has been accounted for, but there is still the chaos related to weather, NADC, albedo and etc. Now consider that as all that warm water spreads out across the equatorial Pacific (El Nino) more water vapor goes into the air (greenhouse gas) and that is probably what causes temperature to change. You also need to stop using a flat anomaly line from (usually) 1960 – 1991 as your base temperature. The correct base temperature will be ocean temperature from the previous year (or as good as you get).
[Response: Most of that is gobbledygook, but taking out the mean in a time series will not affect its correlation. If you like temperature to be an actual temperature, just ad the mean Earth temperature from 1960-1991 to the numbers on the left axis. If you just use data that supports your opinions, that's called cherry-picking, and it's what you're doing here. You have a time series of SOI and temperature and you're finding where they have a high correlation and then disregaring the other data. You can't do that. If you want to compare SOI to temperature, do so, but do it for the entire data record. ENSO does have teleconnection (if you don't know what this means, look it up) with areas around the world, but they are poorly understood. At present time, I know of no mechanism that would cause global temperatures to rise because of ENSO. In fact, the opposite is probably what's occuring. As global temperatures increase, the tropical Pacific gets warmer (see Cane et al., 1997).]
I am glad that William responded to the whole molecule/ photon thing, I hate that stuff. But one thing you might consider is density changes at different altitudes and the angle of the round Earth. The whole 50 : 50 argument really goes down the drain at that point.
Response: If you don’t believe in the greenhouse effect, you’re in a league of your own. William does by the way, he just had issue with the way I explained it. I don’t believe I was wrong, just that I was simplifying it. No respectable scientist doesn’t acknowledge the greenhouse effect.
As for the temperature and CO2 changes in the ice core record, go to NOAA and look at the data by Monnin et al 2004 and Stenni et al. 2001. Notice that when temperature rises, CO2 lags hundreds of years later. But then when CO2 does start to rise, the temperature increase does not accelerate. This to me is the most compelling evidence that CO2 increases do not cause more than negligent warming. But the same goes with the cooling period between 1944 and 1976. You can calculate all you want, but when ENSO and PDO observations explain this period of cooling, I cannot accept the other arguments that are based on guesses, maybes or “my computer says so.”
[Response: You have not shown that PDO (Pacific Decadal Oscillation) observation or ENSO have anything to do with it. You have provided no mechanism and while there may be a correlation, it is weak at best.]
As for your sources of information, I read RealClimate too. I gave them a lot of fair consideration but came to the conclusion that they are not correctly described as scientists, but political activists with science degrees. Free your mind and your butt will follow. [Ed: The following comment was made at the beginning of the original post. It is moved here to keep the ad hominems at the end, where people are less likely to see them.] When I look around this site and see a generally slanted political view, and rhetoric such as “right wing nut job” or “denialist” common sense tells me that the person operating this blog is a political activist, not interested in coming to the truth in science.
Response: I am not a political activist. I’m not political. I have never voted. I don’t care who is supposedly representing me, because they do a horrible job at it. I don’t identify with any political party, but my ideas mostly align with the Libertarians. I, the scientists at RealClimate, and other scientists interested in climate change are attempting to educate not indoctrinate. People need to understand that we are not trying to push a political agenda, even though there are some “environmentalists” that are political. Science is hard to understand, but that is not a reason to dismiss it. Also, my post on “nutjobs” wasn’t specific to right-wing ones, even though that example was about a right-wing nutjob. There are many left-wing nutjobs out there as well. The only time I’ve used the term “denialists” was to point to a different site.]
[Response: Actually, in the second plot just the SOI is negated. And it is only done so that it is easier to see the correlation between SOI and temperature. Modeling has absolutely nothing to do with ploting a temperature or a pressure measurement. I did source the data I used, they were linked in the post. The SOI came from UCAR, and the global temperatures came from the Hadley Center. Reread that section and follow the links to get the data. Radiation from the sun takes about 8 minutes to reach the Earth; there is no solar time lag.]
[Ed: This is my response to Kristen's first comment. Following is her response to it, and then my response to that. Whew. Get all that?]
I went back over your post and found your data source and did indeed find that you sourced your data. But the point is unchanged. SOI and ENSO are inverted. That’s why your R-value in your second plot rose. My response “lag time solar factors” lacks a comma. It should be lag time and solar factors as separate. What lag time means is the amount of time that that it takes for temperature to change after ENSO changes. When you plot out your data you will see what I mean. As for modeling, it is another difference in terms. Plotting and modeling mean the same thing here. Some people use different language. But the point is the same, you cannot plot just one thing in the system while leaving everything else out and expect that you will have an answer. Did you do the plotting as I suggested? Fascinating stuff.
[Response: I don't understand what you mean by "SOI and ENSO are inverted." If you mean that during times of negative SOI that the tropical Pacific is in a state of El niño, that's correct. But ENSO refers to the entire oscillation, both time of warming and cooling.
I did do the plotting that you did in your report. I actually did something very similar before I wrote my original posts, as I said in my reply to your comment. My temperature data doesn't have significant cooling after the 3 major volcano incidents; I'm not sure why not, since they definitely did cool the globe. I'll need to read the actual journal articles in that link to figure out how they came up with this measure of global temperature. It's possible, though unlikely, that they already accounted for them. What's more probable is that by using a filter of ~20 years, I "smoothed out" the relatively small ~2 year cooling caused the volcanoes. But just because we both see a correlation between recent SOI and temperature doesn't mean I need to come to the same conclusions as you. I think the most probable explaination of the recent correlations between SOI (or NINO3.4 or other ENSO indices) and temperature is that the warming is causing a shift in the natural mean state of the tropical Pacific, as described in Cane et al. (1997). This has an easy to understand mechanism - the increasing global temperatures are increasing the temperatures of the tropical Pacific, which is why there is an apparent correlation between temperature and SOI. The problem with your theory, is that there is no mechanism for ENSO to affect global temperatures. When there is an el niño, there will be increased temperature in the central tropical Pacific, but how does that temperature increase affect other areas of the world such as Asia?
Modeling and plotting are generally not used as synonyms, but looking back at your first comment, it does make more sense that way. The way the words model and modelling are generally used can be found here "a scientific model is simply an idea that allows us to create explanations of how we think some part of the world works."]
As was explained in the previous post, you and I are using different terms. But ENSO and SOI are inverted. As for the smoothing, I was using 20 months, not 20 years. I am the one that provided the data, the friend did the spreadsheet work. But if you do the spreadsheet work yourself, make sure to use the NOAA data so we are on the same page.
[Response: I misunderstood. I thought you were using 20 year smoothing, not 20 month. I'll need to recompute that before commenting further.]
[Response: Most of that is gobbledygook, but taking out the mean in a time series will not affect its correlation. If you like temperature to be an actual temperature, just ad the mean Earth temperature from 1960-1991 to the numbers on the left axis. If you just use data that supports your opinions, that’s called cherry-picking, and it’s what you’re doing here. You have a time series of SOI and temperature and you’re finding where they have a high correlation and then disregaring the other data. You can’t do that. If you want to compare SOI to temperature, do so, but do it for the entire data record. ENSO does have teleconnection (if you don’t know what this means, look it up) with areas around the world, but they are poorly understood. At present time, I know of no mechanism that would cause global temperatures to rise because of ENSO. In fact, the opposite is probably what’s occuring. As global temperatures increase, the tropical Pacific gets warmer (see Cane et al., 1997).]
So what part are you saying is gobbledygook? Accounting for volcanoes or bad temperature records because the records were bombed during the war? [Ed: Ad hominem removed again.] Didn’t you know that volcanoes affect climate and the length of those effects depend on the season they erupt? And you are going to include years where the data is poor because the data was bombed or used for toilet paper by the soldiers that invaded those cities during the war? And you are going to use a temperature baseline from 20 years before rather than the previous year? There’s a difference between cherry picking and eliminating the chaos in the system. Did you even take the time to look at the data as I suggested?
[Response: Poor data is better than no data. There are other ways of accounting for poor data besides throwing it out. As I said before, where you choose as a temperature average doesn't matter.
And teleconnections, they should call them can’t telleconnections because they can’t tell you what the connections are. But if you look at the teleconnections map, the warming in the Pacific Northwest and Alaska during ENSO events is caused by PDO. The warming in the Northeast is caused by weather patterns bringing increased water vapor from the mid equatorial pacific, something you can see for yourself on an IR satellite image during an El Nino. As for the changes over Africa, I haven’t gotten that far yet. Try looking at these things before arguing further. If you need help finding data I’ll look it up for you.
[Response: ENSO and PDO are two different things happening on two different timescales. ENSO happens about every 7 years, and PDO every ~20 years.
As for the opposite “probably occurring” plot out the data and tell me what you see. No probably’s here, it’s a definite.
[Response: 'Definite' is probably a word you want to stay away from when talking about ENSO. There's a lot that's not known about that particular phenomenon.
I am glad that William responded to the whole molecule/ photon thing, I hate that stuff. But one thing you might consider is density changes at different altitudes and the angle of the round Earth. The whole 50 : 50 argument really goes down the drain at that point.
Response: If you don’t believe in the greenhouse effect, you’re in a league of your own. William does by the way, he just had issue with the way I explained it. I don’t believe I was wrong, just that I was simplifying it. No respectable scientist doesn’t acknowledge the greenhouse effect.
Greenhouse effect from? CO2? Water vapor? Didn’t I say something about greenhouse effect from water vapor? ENSO modulates water vapor in the atmosphere, different ocean temperatures modulate water vapor in the atmosphere. The fact that CO2 Greenhouse Theory predicts more warming in the upper troposphere than is actually occurring should tell you something. The greenhouse gas that causes the abundance of the warming is water vapor.
[Response: I have no idea where you're getting your upper troposphere temperature data, but if it's from MSU measurements (particularly UAH), then you should be aware that there are problems with them.
There is a simple map on NOAA’s El Nino Theme page, you can see the affects of ENSO events there for different times of the year. I am going by memory here due to time but I do not think they show an impact on Asia.
Response: I couldn’t find exactly what you’re talking about, but it doesn’t really matter. There are several indices used to monitor ENSO. They all have advantages and disadvantages. [Added later]This is the page refered to. I always thought of that as the ENSO TAO page, but I guess it does say ‘Theme Page’ on it.
Once again, the physical mechanism would be water vapor.
Response: Water vapor has a residence time in the troposphere of about 10 days. “Due to the short residence time of water vapour in the atmosphere the horizontal (zonal) distances between its sources and sinks remain near 2000 km.” (Hense et al., 2005) The circumference of the Earth is about 40,000 km.
I am surprised that you did not know about volcanoes and how much they impact climate. Look into it a little more, it’s the sulfur gases in the stratosphere that cause most of the changes and to a small extent the ash and soot. Whenever you look at temperature, you have to factor in volcanoes. Simple smoothing will not do.
Anyways, I have to run, it’s been fun.
Response: I do know that volcanoes have an effect on climate. It’s just that their effects are relatively short term. As I said before, I was using ~20 year smoothing, which will smooth out any effects due to volcanoes, which roughly last about 2 years. Your smoothing of 20 months allows those effects to still be present in your plots.
[Response: I misunderstood. I thought you were using 20 year smoothing, not 20 month. I’ll need to recompute that before commenting further.]
Cool! That’s the spirit, do the same with the rest of it, and if you have questions, just ask.
But I have a study group and will be gone til tonight.
[Ed: I combined two of Kristen's comments into this one.]
As for the temperature and CO2 changes in the ice core record, go to NOAA and look at the data by Monnin et al 2004 and Stenni et al. 2001. Notice that when temperature rises, CO2 lags hundreds of years later. But then when CO2 does start to rise, the temperature increase does not accelerate. This to me is the most compelling evidence that CO2 increases do not cause more than negligent warming. But the same goes with the cooling period between 1944 and 1976. You can calculate all you want, but when ENSO and PDO observations explain this period of cooling, I cannot accept the other arguments that are based on guesses, maybes or “my computer says so.”
[Response: You have not shown that PDO (Pacific Decadal Oscillation) observation or ENSO have anything to do with it. You have provided no mechanism and while there may be a correlation, it is weak at best.]
When you plot out the data you will see what I mean.
Response: I doubt it, but we’ll see.
Just a simple request, when you reference journal articles, could you provide a link to the abstract or at least mention which journal it’s in. That would save time trying to find which particular article you’re talking about, as most scientists will produce more than one paper per year. Thanks.
Regarding my comment above about MSU temperature trends, see Using Ancillary Zero Trend Levels as a Means to Elucidate Microwave Sounding Unit Derived Tropospheric Temperature Trends Methods. I’m not actually sure if this is public information. I can view it. If you can’t, let me know and I’ll try to get permission from the author to send you a copy.
[Response: I have no idea where you’re getting your upper troposphere temperature data, but if it’s from MSU measurements (particularly UAH), then you should be aware that there are problems with them.
The problems with the MSU data are well known and argued about. Regardless if you use the Spencer data or those that have other observations, it does not take away from tha fact that the upper troposphere is not heating the way greenhouse theory predicts.
[Response: I'm not entirely sure what you mean when you say that greenhouse theory doesn't predict upper tropospheric temperatures correctly. Greenhouse gases interact with radiation. In the troposphere, even the upper troposphere, heat transport is almost exclusively via convection, not radiation. I couldn't find a paper about the apparent temperature difference between observations and theory. Even if I did, which you could provide if you have one available, I do know that there is a level in the atmosphere above which there will be cooling due to greenhouse gases, which includes everything above the tropopause and the upper troposphere. And temperatures have been decreasing in the theromosphere ( Lastovicka et al.,2006), which is predicted by greenhouse theory. I was mostly concerned where you were getting the data; MSU, radiosondes, LIDAR, other satellites such as GPS, etc. It appears you were using the MSU data.]
Response: Water vapor has a residence time in the troposphere of about 10 days. “Due to the short residence time of water vapour in the atmosphere the horizontal (zonal) distances between its sources and sinks remain near 2000 km.” (Hense et al., 2005) The circumference of the Earth is about 40,000 km.
I could only pull up the abstract. But this is a number that I have heard of before. It does not matter, you are talking about a process that is constantly occurring and is modulated by the changes in ocean temps and ENSO to a great extent. You may have seen the literature that talks about how dry glaciations were. Or that the warmer the air the more water vapor. I had a professor warn me about using the term “warm air holds more water” I have not looked into it that much but they do not seem to agree on the term, but I think you get the point.
[Response: Saturation vapor pressure, see here and here. But just because the saturation vapor pressure increases does not mean that the actual vapor pressure (the amount of vapor in the air) increases. And it for sure doesn't mean that water vapor will have a longer residence time.]
Response: I do know that volcanoes have an effect on climate. It’s just that their effects are relatively short term. As I said before, I was using ~20 year smoothing, which will smooth out any effects due to volcanoes, which roughly last about 2 years. Your smoothing of 20 months allows those effects to still be present in your plots.
Pretty much. Although you might look up the effects of Agung in the 60’s, I recall that that one put a lot of SO gas into the stratosphere for more than 2 years. That would actually be a help if you can find it, I will be working on my plot for it in the future but it is still a bit away.
[Response: I couldn't actually find anything substantial on Agung in the literature. I probably missed it. I'll admit it was just a brief search. I found a lot on Pinatubo, but that which I found still said the radiative effects of eruptions last about 2 years.
[Response: Saturation vapor pressure, see here and here. But just because the saturation vapor pressure increases does not mean that the actual vapor pressure (the amount of vapor in the air) increases. And it for sure doesn’t mean that water vapor will have a longer residence time.]
LOL Nathan, did you even include heat in all that? Are you saying that heat has nothing to do with water vapor? The heat increasing the amount of water vapor is the idea.
[Response: Poor data is better than no data. There are other ways of accounting for poor data besides throwing it out. As I said before, where you choose as a temperature average doesn’t matter.
Feel free to show me how you would fix the problems with the WW2 data and I'll be glad to look at it. From everything I have read so far, they can't and the temperature data from WW2 is very unreliable.
[Response: Even if I did, which you could provide if you have one available, I do know that there is a level in the atmosphere above which there will be cooling due to greenhouse gases, which includes everything above the tropopause and the upper troposphere. And temperatures have been decreasing in the theromosphere ( Lastovicka et al.,2006), which is predicted by greenhouse theory. I was mostly concerned where you were getting the data; MSU, radiosondes, LIDAR, other satellites such as GPS, etc. It appears you were using the MSU data.]
Research Spencer and Christi on this one. In fact, I think Spencer has it right on his MSU page. I cannot access it for some reason. That reminds me, I have to ask some one to get the MSU data for me, thanks.
We’ve put a nice new tool online at http://www.spectralcalc.com
Here you can compute the transmittance spectra of CO2 or any other common atmospheric gas. The computations are done with state of the art line-by-line radiative transfer codes and molecular absorption databases. This should help sort things out.
Reopening comments on this because it’s been linked to fairly often lately.