|
Post by socold on Apr 8, 2009 21:20:05 GMT
For a detailed overview of physics of the atmosphere, including the kind of stuff that the models are based on there's a PDF here: geosci.uchicago.edu/~rtp1/ClimateBook/ClimateVol1.pdfIt's about 14mb. A lot of the math goes over my head but there is also some discussion in it. Chapter 2 onwards covers most of the physics for describing an atmosphere.
|
|
|
Post by tacoman25 on Apr 8, 2009 21:59:16 GMT
For a detailed overview of physics of the atmosphere, including the kind of stuff that the models are based on there's a PDF here: geosci.uchicago.edu/~rtp1/ClimateBook/ClimateVol1.pdfIt's about 14mb. A lot of the math goes over my head but there is also some discussion in it. Chapter 2 onwards covers most of the physics for describing an atmosphere. Personally, I don't think the models are based off of bad physics. I just think they are not able to take all factors into account...because the data/knowledge just isn't there yet. Physics models, as complex as they may be, basically boil down to plugging numbers into an equation. Unfortunately for them, the real world is a little bit more complicated than that. The earth is dominated by negative feedbacks that tend to keep things mostly in balance - if not, everything would have spiraled out of control a long time ago.
|
|
|
Post by tacoman25 on Apr 8, 2009 22:04:38 GMT
Ironically if something dramatic like a solar minimum occured it might just throw a spanner in the works. If for example a very deep solar minimum occured and temperatures droped 0.6C, it would be harder to then determine a high end for how much warming co2 could have caused. Ie without the co2 rise the temp drop could have been 1C rather than 0.6C. Indeed. But here's another way to look at it. If an extended period of low solar activity did actually coincide with a .6C drop, I think we could then have much more confidence in solar forcings...and then could perhaps better determine how much of the warming in recent centuries was due to solar activity - and what proxies are best for measuring that impact. Then perhaps the true impact of man-made GHGs could also be better assessed.
|
|
|
Post by jimg on Apr 8, 2009 22:28:12 GMT
Of course it could also mean that a period of energetic solar cycles v. normal (whatever that really is) was the driver of the warming trend as well.
It would also mean that the sun-earth relationship is more complex than TSI alone can explain.
|
|
|
Post by socold on Apr 8, 2009 23:33:51 GMT
Unfortunately for them, the real world is a little bit more complicated than that. The earth is dominated by negative feedbacks that tend to keep things mostly in balance - if not, everything would have spiraled out of control a long time ago. Positive feedback doesn't mean it spins out of control, it just means that a rise in temperature peaking at 1C actually peaks higher. Just 15000 years ago global temperature rose 6C and sea level rose over 300 feet. The orbital variations that are proposed to cause that and other glacial-interglacial cycles are not large enough alone to explain the rise in temperature and positive feedback is required. In fact the comment above about "the sun-earth relationship is more complex than TSI alone can explain" relies on positive feedback too. TSI is essentially all the energy we recieve from the Sun, to get more bang for buck requires amplification. The physics also currently show positive feedbacks dominate. Perhaps more details might show this is wrong, but uncertainty in the complexity could cut both ways. We could find out in 20 years time that positive feedbacks are even greater that thought now. I suspect it's more likely that it will change somewhat, but not by a significant amount. Perhaps currently thought net feedback will be found to be 20% too high or too low. That won't really make much difference. The Earth with positive feedbacks only requires forcing to not change much over time for it to remain fairly stable. There's no evidence that massive forcings have occured in the past. Orbital cycles, volcanic events, etc don't provide anything massive that if combined with positive feedback would turn us into mars or venus. So existance of positive feedback in climate does not defy the earth's existance.
|
|
|
Post by tacoman25 on Apr 9, 2009 1:10:56 GMT
Unfortunately for them, the real world is a little bit more complicated than that. The earth is dominated by negative feedbacks that tend to keep things mostly in balance - if not, everything would have spiraled out of control a long time ago. Positive feedback doesn't mean it spins out of control, it just means that a rise in temperature peaking at 1C actually peaks higher. Just 15000 years ago global temperature rose 6C and sea level rose over 300 feet. The orbital variations that are proposed to cause that and other glacial-interglacial cycles are not large enough alone to explain the rise in temperature and positive feedback is required. In fact the comment above about "the sun-earth relationship is more complex than TSI alone can explain" relies on positive feedback too. TSI is essentially all the energy we recieve from the Sun, to get more bang for buck requires amplification. The physics also currently show positive feedbacks dominate. Perhaps more details might show this is wrong, but uncertainty in the complexity could cut both ways. We could find out in 20 years time that positive feedbacks are even greater that thought now. I suspect it's more likely that it will change somewhat, but not by a significant amount. Perhaps currently thought net feedback will be found to be 20% too high or too low. That won't really make much difference. The Earth with positive feedbacks only requires forcing to not change much over time for it to remain fairly stable. There's no evidence that massive forcings have occured in the past. Orbital cycles, volcanic events, etc don't provide anything massive that if combined with positive feedback would turn us into mars or venus. So existance of positive feedback in climate does not defy the earth's existance. The warming produced by many of the climate models used by the IPCC relies heavily on positive feedbacks. So there is no reason to minimalize them either. As you say (sort of), they could be the difference between 1C warming and 5C warming...inconvenient (for some, great for others), or catastrophic. Many AGW proponents fear a chain of runaway positive feedbacks, and my point is simply that I find that highly unlikely given what we know about our natural world.
|
|
wylie
Level 3 Rank
Posts: 129
|
Post by wylie on Apr 9, 2009 1:22:54 GMT
Socold,
Thank-you for your thoughtful reply. I would welcome the kind of situation that you describe (i.e. a a 0.6C drop in temp caused by a solar minimum) in terms of the learning (But NOT in terms of the hardship for everyone including my family!).
I am not sure that I can accept your arguments on positive feedbacks though. If CO2 (for example) were to cause a large forcing at high concentrations, its concentration has been at least 10X higher in the last 50 million years. Also, negative feedbacks of considerably greater magnitude have also occured. E.g. the Yellowstone volcano. Another large geological forcing was the much larger Bering straight, combined with the absence of a land-mass at the South Pole (i.e ANtarctica). Full access to both the south and north poles by Ocean currents would (probably) have allowed the equivalent of at least another 2 or 3 Gulf-streams worth of heat to reach the poles. I do not believe that there was ANY ice-cap during this time. The albedo changes (in the positive direction) alone could very well have been suffiicient to account for the total absence of ice-caps. In fact, it could legitimately be argued that the current Quartenary Ice-Age (last 2 million years) was MOSTLY due the presence of Antarctica at the south pole and the closing of the Bering straight to deep Pacific currents.
Those positive feedbacks did not cause runaway warming and life thrived on the Earth (including our ancestors). I am NOT suggesting that you believe that the current increase in CO2 would cause a Venus-like runaway warming, I am just suggesting that there is significant evidence that there are indeed significant stabilizing factors brought about (mostly) by the presence of water and water vapor.
Keep posting,
Ian
|
|
|
Post by nautonnier on Apr 9, 2009 10:01:04 GMT
Thousands of years for temperatures to change?.... Really? Although globally that may be true, but for Northern lattitudes it may only take a decade. Northern European temperatures changed sufficiently in a ten year period to cause massive crop failures and famine. Lots of rain, lots of cold. Millions died. Read up on the Little Ice age. Not pretty. I think two things are being confused. The temperature drop to LIA and beyond could happen rapidly - the kilometer thick glaciers over Chicago and Birmingham UK will take a little longer.
|
|
|
Post by nautonnier on Apr 9, 2009 10:06:11 GMT
2002-03 had a strong El Nino, but wasn't close to the 1998 record. 2006-07 had a moderate El Nino, but again wasn't close to the 1998 record. And during the last -PDO phase, there were only 3 strong El Ninos over the course of 30 years. It would be wise to pay attention to history. 2002-2003 is 6 years ago. A strong el nino at this point would reach higher temps. And where would the ocean heat content for this 'strong El Nino' appear from? The Earth appears to be down on spare energy at the moment. El Nino would appear to work as a feed back that dumps excess OHC into the atmosphere where it then escapes into space. There needs to be that excess heat for it to occur.
|
|
|
Post by kiwistonewall on Apr 9, 2009 10:09:44 GMT
This is an excellent paper on the ir radiation given off from the atmosphere. Very interesting, particularly the fact that Ozone absorbs ir at very much the same places as CO2. arxiv.org/PS_cache/arxiv/pdf/0810/0810.2957v1.pdfIt is quite clear that the ONLY big bite in the Earths IR radiation spectrum in the peak (high power) area is caused by Ozone absorption at 9.5microns. CO2 absorbs mainly at 4.3 & 15, well off the main radiation peak of Earth's radiation to space. See fig 4. All very interesting. Ozone is a potent greenhouse gas, but acts in the upper atmosphere. ephemeris.sjaa.net/0807/f.htmlThings just are not as they have been made out to be.
|
|
|
Post by steve on Apr 9, 2009 10:32:43 GMT
Steve: Latent heat of fusion (water) ~80cal/g vaporization: 540 cal/g. Specific heat: 1 cal/g-Deg C Considerably more energy goes into evaporating water or melting ice than it does to raise its temp 1C. I would think this is non-negligible. Don't know where this is going because I'm calculating as I write: Greenland ice sheet is 2.85 million cubic metres (wikipedia). This equates to 2.85e+21 grammes approximately. By your figures it would take 10^23 joules to totally melt it. If we take the "disaster scenario", that it will melt in 1000 years, then we require 10^23/(1000*365*24*3600) = 7e+13 joules per second (Watts). The area of the earth is 5e+14 square metres. So the forcing required to provide enough energy to melt Greenland in 1000 years is: 7e+13/5e+14 = 0.14W/m^2 This is about 4% of the predicted forcing from doubling of CO2. A similar calculation shows that it would take 3.3 W/m^2 to warm the whole volume of the ocean by 1C in 100 years (though I don't think the projections show the whole ocean volume warming this amount). In short, the figures of the energy required to melt ice and warm the ocean are not out of line with the forcing projections.
|
|
|
Post by nautonnier on Apr 9, 2009 10:39:57 GMT
Statistically there is a similarity here that cannot go unaddressed. The most outlandish financial models are usually espoused from the top of a soap box in the middle of the town square and you can tell by the dress that the advocate really has no real interest in the outcome. That changes dramatically as someone gains an interest in the outcome and the first job of auditing such models is to identify what that interest might be, because once identified you can usually narrow your scope of work, deliver a good product at a reasonable price. I am truly admiring the work of Steve McIntyre, I just wish I was up to his level in statistics. The nation got sick of hucksters and created a profession to deal with it. Now that we are beginning on a path towards policy arising out of science that can fundamentally affect the well-being of entire nations it might be a good idea to figure out how to ensure accountability for using science as a cover for politics. It could prove beneficial to us all and especially to the scientific modelers who just might find themselves dangling from the end of a rope by the neck if the outcomes get nasty enough. . . .certainly would not be the first time anyway. While McIntyre does a decent job in identifying issues with papers, there are many scientists who are already doing the same to good effect. The idea that auditing will solve all the problems seems to forget that Enron was audited, Madoff was audited, Stanford was audited, AIG was audited, Leh,an...etc. etc. As I've pointed out before, many of the models are much the same models as used in weather forecasting - so their results are publicly available, and the forecast organisations are hugely competitive such that they are all working hard to produce the best model they can. Secondly, many of these model are, even if not open source or freely available for use, widely available for research use by academia. So these models are well audited by a broad scientific community. " As I've pointed out before, many of the models are much the same models as used in weather forecasting - so their results are publicly available, and the forecast organisations are hugely competitive such that they are all working hard to produce the best model they can." Steve, You should read the work of Lorenz who flagged up the effect of minor errors on chaotic models - and he was actually using weather models at the time.. All these weather forecasters if asked to give a precise result out to 18 hours would laugh at you - they can get reasonably close - like fixed point to a long float. Now take that same model and say use it to 7 days now its integer to long float - and a low Bayesian confidence level. Take those same models out months and no forecaster would trust them. So now you add in statisticians and ensemble runs and what turns into educated guess work (assumptions)- known as climatology. You cannot do iterative daily forecasts 365 times and get an accurate forecast for next year. This is basic chaos theory the start parameters must be precise or the forecast will rapidly deviate from reality.
|
|
|
Post by steve on Apr 9, 2009 11:08:07 GMT
This is an excellent paper on the ir radiation given off from the atmosphere. Very interesting, particularly the fact that Ozone absorbs ir at very much the same places as CO2. arxiv.org/PS_cache/arxiv/pdf/0810/0810.2957v1.pdfIt is quite clear that the ONLY big bite in the Earths IR radiation spectrum in the peak (high power) area is caused by Ozone absorption at 9.5microns. CO2 absorbs mainly at 4.3 & 15, well off the main radiation peak of Earth's radiation to space. See fig 4. All very interesting. It is indeed interesting and I really have to thank you for finding it, as it is the best possible illustration of how reasonable the estimates of CO2 forcing are. While 15 microns is "well off the main radiation peak" it is still well on the emission spectrum, so you can clearly see there is a significant "bite" at this point. The units are given as per steradian per micron, so I guess you have to multiply the figures on the y-axis by 2pi to get energy per square metre per micron - (there are 4pi steradians in a sphere, but we're only interested in the upward half, as the downward half is going back towards earth). I would say there was approx 24 Watt/m^2 bite in this line alone (if you calculate the area above the spectrum and below the 255K black body curve (the middle of the three green lines) and multiply by 2pi - I think it is approx 2W deep and 2microns wide - 2 * 2 * 2pi is approx 24). The projections are that a doubling of CO2 would deepen this bite from 24 to 27 or so Watts - so about 10% deeper and a bit wider. Does that seem so unreasonable?
|
|
|
Post by steve on Apr 9, 2009 11:10:08 GMT
While McIntyre does a decent job in identifying issues with papers, there are many scientists who are already doing the same to good effect. The idea that auditing will solve all the problems seems to forget that Enron was audited, Madoff was audited, Stanford was audited, AIG was audited, Leh,an...etc. etc. As I've pointed out before, many of the models are much the same models as used in weather forecasting - so their results are publicly available, and the forecast organisations are hugely competitive such that they are all working hard to produce the best model they can. Secondly, many of these model are, even if not open source or freely available for use, widely available for research use by academia. So these models are well audited by a broad scientific community. " As I've pointed out before, many of the models are much the same models as used in weather forecasting - so their results are publicly available, and the forecast organisations are hugely competitive such that they are all working hard to produce the best model they can." Steve, You should read the work of Lorenz who flagged up the effect of minor errors on chaotic models - and he was actually using weather models at the time.. All these weather forecasters if asked to give a precise result out to 18 hours would laugh at you No, they'd laugh at you for failing to read my post accurately.
|
|
|
Post by gridley on Apr 9, 2009 16:28:19 GMT
Models *are* real evidence. Steve, I'm happy to stipulate that the above statement may have been an oversimplification, but I still feel a need to address it and some of your later posts. A model, by itself, is evidence of nothing and proof of nothing. A model that correlates closely with observed behavior is limited evidence but falls well short of proof. One can use models for interpolation with some reliability. Given two sets of real data in a generally understood system it is usually possible to create a model that will accurately predict behavior between these boundary conditions. At work we use this to certify the design of a family of structures without testing every single one. If one has a very good model one can even manage small extrapolations with some confidence. The regulatory authorities, however, won't let us certify all of a family based on a single test. Why? Because they understand the limitations of computer modeling. This is theoretical computing 101, as taught at Carnegie-Mellon University and used by the commercial aviation industry. Now, in order to predict the future we are forced to extrapolate from models. And, as weather forecasts show, we can get pretty good results with short-term extrapolation from large data sets. Now again, you are correct that policy makers would be negligent if they ignored the long-term model forecasts. However risk analysis demands that they consider the probability of a problem as well as the effects of that problem when determining how much effort to put into mitigation. If a fist-sized meteor hits the cockpit of a commercial airplane, it is going to crash and a lot of people are going to die; but we don't armor the cockpits of our aircraft against such meteors (we could, but we don't), and given the odds of a fist-sized meteor hitting a flying airplane we are right (IMO) to do so. It is my opinion that the immaturity of the current generation of GCMs and the corresponding low probability that their severe projections will come to pass do not merit significant efforts at mitigation at this time. I'm not saying AGW is falsified, or that it is fundamentally flawed (although the projection onto public policies such as cap and trade CO2 systems certainly is), I'm saying we've got finite resources and AGW doesn't merit allocation of anything beyond research funding and a trickle of concurrent engineering when measured against the high-probability problems of the future or the known problems of the present. To invoke one final analogy: my opinion is that AGW is to the current status of the human race as telling you that you may have cancer right after you've been hit by a truck. Should the doctors ignore your hemorrhaging to take you in for chemotherapy? No. Should they spend some time keeping an eye out for tumors while they operate? Probably, if circumstances permit. If you survive the trauma from being hit by a truck you may still die of cancer - if you actually have it. That doesn't mean your doctors are negligent if they focus on more urgent problems in the short term.
|
|