Chris C

John B, the global temperature was not 3 degrees warmer than present day until millions of years ago...you're looking at Vostok, not the globe. More importantly, rates of change generally matter much more than absolute numbers. The global warming situation today could hit 3 C in a century as opposed to glacial-interglacial timescales which take thousands of years. There's also a lot of differences in the background extinction rate due to humans and the amount of human settlement, coastal infrastructure, and the necessity to maintain the status quo. As far as societal and ecological impacts, there's really no good past analog.

SkepticLance's stance that models (or anything for that matter) should be treated with a healthy skepticism is not a bizarre notion, and it's hard to believe that many people would disagree. There is a very detailed treatment of the strengths and limitations of GCM's here, and at least the summary should be read by those with a strong care of the issues http://www.climatescience.gov/Library/sap/sap3-1/final-report/ They are not currently very good at the task of simulating sub-grid scale processes such as cloud microphysics, etc. Regional precipitation changes are not robust among ensemble members. Aerosol indirect effects and other things are still heavily paramaterized or not included at all (despite how realistic those paramaterizations may be) but since we're not up to the task yet of building a full-blown model with no empiricism whatsoever, care needs to be taken in long-term future projections. Care is usually taken in primary documents as to how uncertainty in models, assumptions, observations, etc are likely to effect the overall conclusion(s)--if at all. We do not yet have adequate knowledge of ice sheet physics to make highly confident predictions concerning the magnitude and rate of ice loss (though they are certainly better than best guesses). OTOH, I have a lot of confidence in basic responses to changes in external forcings such as CO2, solar irradiance, etc...the "basic physical responses" to warming such as decreases in the pole-to-equator temperature gradient, landmasses heating up faster than oceans, etc are treated well in models. Responses which are robust among all models and different modelling groups (even with changes in initial conditions) such as "more CO2 gives you warming" are probably robust and it is unlikely that uncertainty in cloud physics or ocean dynamics are going to take away from that conclusion, especially ones with a solid foundation in theory and have reliable paleoclimatic constraints. It is very difficult to explain on a blog/forum what kind of results are well known (and if the model result is likely good enough for policy-decisions) and what kind of stuff is still treated very much inadequately. You'll need to read authoritative documents such as the one above, IPCC, etc to see the recommendations of scientists and a more thorough treatment of individual sub-topics. Also, many things which are treated inadequately may still be constrained enough (by theory or observations) to not limit confident projections or hindcasts of other climatic variables. The notion in the swindle video or other venues for instance that "if one of the thousands of variables is incorrect, then the whole thing is worthless" is clearly not correct, and whether they are "useful" or not depends on what climatic variable you are talking about and the context around it. C http://www.chriscolose.wordpress.com

Thanks for the input on my post...I might re-work it though. It was suggested in the comments and elsewhere that I make it more similar to the original crackpot index, just reworded a bit to make it applicable to climate change (see the evolution version for an example). Pioneer-- I pose a question for you. If an arsonist is on trial for setting a fire in the woods, would you declare him innocent on the basis that fires naturally happened before? Something tells me the prosecution is not pulling his name out of a hat; you need to read my posts at http://chriscolose.wordpress.com/2007/12/18/the-scientific-basis-for-anthropogenic-climate-change/ And by the way, the proxies *are* calibrated to the instrumental target just so that they will be comparable.

Hi John, I'm doing fine first of all, just so we're all on the same page with shorthand LW= longwave=IR SW=incoming solar (shortwave) The dominant effect of changed water vapor is indeed in its longwave trapping effect. There is a much smaller, but still significant effect on solar energy...actually there's some influence at higher latitudes where reflected solar radiation is sent back up and water vapor can absorb in SW areas as well. This is discussed in a recent paper here for instance http://ams.allenpress.com/perlserv/?request=get-abstract&doi=10.1175%2F2007JCLI2110.1 The paper also states, "...the increased water vapor acts to increase the net incoming solar radiation (i.e., it results in increased solar absorption). The magnitudes are roughly a factor of 5-10 smaller than those computed for the longwave part of the spectrum, except at high latitudes." So actually it's more of a positive-positive feedback. The only substantial negative feedback is the planck response (i.e., that a hotter planet emits more radiation). There is also a small but negative lapse rate feedback, but that only cancels out a small portion of WV. Yes, Venus has much more CO2 than Earth, but it's temperature differential between an atmosphere and no-atmsophere case is ridiculously large!! Again we're talking about something like below freezing to something like double the boiling point. For a planet that receives much less sunlight than Earth (due to its cloud cover, so when I said below-feezing in a no atmosphere case, I was still leaving the albedo the same) the role of CO2 highlights how much it matters to make an atmosphere more opaque to infrared. What's more, Earth is relatively sensitive to external changes because of the presence of water vapor on a planet situation where it changes phase so easy, so for that reason doubling CO2 matters quite a bit, even as the absolute values are relatively small. Concerning your graph--After the ice core record, especially when you go tens of millions of years back, there are substantial uncertanties in both temperature and CO2. If you look at the IPCC AR4 chart (in the paleoclimate chapter in the section on deep-time climate) the CO2 uncertainty ranges for instance can be more than 1000 ppmv, and you need to be careful at interpreting these temperatures-- such deep time proxies may reflect bottom water conditions on the shelf and not open oceanic SSTs, further on the dataset has only a very poor temporal resolution for many intervals. I've not looked at the papers that it credits, but I have a hard time believing a graph like that was published in a refereed article. The numbers are far too concise, and there are no error ranges, which makes such a graph essentially useless for this kind of comparison. I'd recommend a couple of papers https://wesfiles.wesleyan.edu/home/droyer/web/KurschnerCommentary(2008).pdf https://wesfiles.wesleyan.edu/home/droyer/web/PhanCO2(GCA).pdf

npts, and others Let's bear a few issues in mind: most of you are discussing the surface energy balance, not the top-of-atmsophere energy balance. Greenhouse warming does not rely on greehouse gas increases being able to directly increase the downward IR. It's possible to increase CO2 and not get a big change in downward IR if the lower atmosphere is already emitting like a blackbody at its temperature, and any change in downward IR may not really warm the surface (depending on evaporation, sensible heat, etc). Actually the downward IR will increase moreso because the atmosphere is warmer, not because of the direct contribution of more GHG's. And clearly energy does flow from a colder atmosphere to a warmer surface, but the net flux is always from the surface to the colder atmosphere, so there is no violation of the thermodynamics. More GHG's primarily affect what’s going on higher up in the atmosphere, which warms the whole troposphere (which is meshed together by convection so it tends to warm and cool as a unit). Adding GHG's creates a situation where the bulk of emission from the planet to space comes from higher altitudes where it is colder. Because it is colder, it emits radiation weaker than it was before, so the now the planet is less efficient at losing heat, but the incoming heat is held steady. Because the net energy flux into the planet is now greater than zero, the planet warms. From this angle, you can see that GHG mixing into higher altitudes as well as the fact that temperature drops with altitude is going to be a requirement for an enhanced greenhouse effect (actually there would be no GHG effect in an isothermal atmosphere). The atmospheric warming then warms the surface through increasing all the heat fluxes which couple the surface to the air, not just the radiative ones. //" I have always thought (maybe wrongly) that warming followed CO2 concentrations pretty closely with the increasing temps lagging some time behind increasing CO2."// Depends on when you're talking about. If you widen your scope to the geologic timescale as a whole there really is no "lagging" issue...we see higher temperatures when CO2 concentrations are higher for the most part. In fact most of geologic time had more CO2, higher temperatures, less ice, etc. During the PETM we see a big spike in CO2, and a big spike in temperature. You have a similar situation on Venus: the planet should be below freezing if it were for no CO2, but it's the hottest in the solar system. During the last million years when we see transitions between glacials and interglacials, CO2 doesn't seem to matter much for forcing the climate change. That's related to orbital changes. But there are carbon feedbacks related to changing biosphere, changing ocean temperatures, etc, and the changes in CO2 acted as a positive feedback to amplify the orbitally-paced warming. Right now, the CO2 is going up too fast, and is much higher in concentration to be a feedback. C

Atoms want their outer shells full.

Chris wonders where the error bars are on temperature and CO2 levels on this graph?? Given that there are errors on scales of 1000-3000 ppmv when you go back tens to hundreds of millions of years (see graph on pg. 441- http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter6.pdf ) , and paleotemperatures have their own uncertainty (further on the dataset has only a very poor temporal resolution for many intervals), it's hard to take this graph (and what it is trying to imply) seriously. Trying to get information from such a graph for purposes of human-released CO2 impacts on temperature is strange enough. You need to be very careful about high-frequency versus low-frequency changes, and as others have mentioned, discuss causes...not just the detection process. The currently accepted idea seems to be that CO2 is the most important "knob" for changes in climate over geologic time.

I don't know if either of us are trained professionally in climate science (I certainly am not, I'm a student here though) but you can find "debate" in a lot of topics over the internet, from evolution to climate change to the shape of the Earth. The discussion in this forum is not indicative of the scientific "consensus" or lack of. There is a lot of "real" debate amongst credible scientists on many specifics such as how hurricanes might respond, how ecosystems will respond to a warmer climate, how much warming it will take to melt Greenland (and how fast), how soon we might cross various "tipping points," why strengthening convection enhances ocean heat uptake in one region while weakening convection enhances it in another, etc. But the broad brush details like "more CO2 will cause warming" or "massive releases of greenhouse gases and deforestation are influencing climate" are not seriously questioned, and these details are in fact "settled." The real consensus on this can be found in the academic literature, but also every major scientific organization has accepted these conclusions. See for example http://nationalacademies.org/onpi/06072005.pdf http://royalsociety.org/displaypagedoc.asp?id=13619 http://gristmill.grist.org/story/2006/11/13/221250/49

It may seem "well duh" which isn't the point, but rather that the models successfully capture these responses. Concerning AGW vs. "natural cycles," there are very few examples of "fingerprints" associated with ghg warming that you don't get with solar warming, stratospheric cooling being one. Night/winter temperatures going up faster than day/summer temperatures is another, and that's observed. You might expect increased downward infrared radiation, which seems to be consistent (though this isn't exactly why the planet warms with more ghg's) with observations. The changes in tropopause height also seem to be consistent with externally forced ghg's. The ability to simulate anthropogenic + natural, but not natural variations with models helps, and the simple underlying physics and paleoclimatic evidence (what ought to happen in theory) all helps. A good read on detection and attribution (which also goes over the surface/atmosphere warming and models observations) is at https://e-reports-ext.llnl.gov/pdf/315840.pdf See http://www.agu.org/pubs/crossref/2007/2007GL029875.shtml

My apologies on the quote w/o credit. AGW theory does not "require" extra tropical heating; this is a principle of moist convection that is expected from any cause. In fact, the greenhouse effect requires a temperature decline with altitude, so if observations showed less atmospheric warming than models show, then that would mean climate sensitivity was *underestimated* because of less negative feedback from the lapse rate response. Not overly reassuring that the models could be wrong. The models are in fact not right with the sea ice decline either, but not on the conservative end (Rahmstorf 2007). The weather noise over the interval looked at is very important, and because of that, the spread of the runs due to weather is important. An ensemble is a collection of simulations run by one model, for one experiment, and is good for a set of possible trajectories for some forcing x. It's fine if you use data, but make sure you understand that there is noise, and not just the forcing from some external perturbation. It's like "global warming stopped in 1998 while ghg's have increased." Silly. Because there is an incorrect treatment of the model uncertainty, and no discussion on the data choices (the original paper would have been nice, rather than being lost in blog comments), it seems that the other analysis which show no difference between models and observations are worth looking at, no? ................................... Now I don't have a scientific background, however in my time here I've found that those who do, espouse (and have to repeat it endlessly to posters in the pseudoscience subforum) a simple step process to check a theory. 1. Formulate theory. (Check.) -- Based on a centuries+ work of radiative physics, fluid dynamics, etc 2. Make testable predictions. (Check.) --Like the temperature rise over the last century, strat cooling, increased ocean heat content, sea level rise, polar amplification, increased NH response, trop warming a bit more than surface, ice declines, species response, etc 3. Compare to observations. (Che......Oops.) --Like the temperature rise over the last century, strat cooling, increased ocean heat content, sea level rise, polar amplification, increased NH response, trop warming a bit more than surface, ice declines, species response, etc. Now models aren't perfect, but when comparing them to observations, maybe a correct analysis would be good. Right now, the main and "broad" predictions are being borne out.

You'll have to ask Gavin what exactly he meant, but the technical literature is generally intended for audiences that understand the science at a high level. Find me a peer-reviewed document that discusses how freezing water decreases the entropy of the system. Maybe you will, but it is such a simple concept, it is probably assumed knowledge and can be found in a textbook. Now a lot of this could be in very, very, old papers (I'm sure the information gets out somehow) but I'm not sure if Gavin is talking about this. This seems to be a distraction though. John Christy's statements in the "swindle video" and in various quotes I can find in op-eds across the internet have already undermined his credibility. There is plenty of indefensible nonsense that comes out of him (which is hardly an "opinion"). If you have serious credentials, you can't just go throwing sloppy statements around and expect people to take you seriously, regardless if you do it in the technical literature, a blog, or channel 4. The error in Douglass et al is that their estimate of the uncertainty in the model projections is the uncertainty in the determination of the mean of the model projections rather than the spread. It's like saying the mean of rolling a dice 100 times is 3.5 +/- 0.1 and claiming that one throw of 2 is a mismatch. Using older data without justification, and looking at time periods of high internal variability doesn't help either.

Just ask yourself "where does the energy go?" to allay confusion to all of your questions. Or ask yourself why El Ninos are different in the last 30 years, then say, over the entire Holocene. I would not expect a climate change with more frequent internal oscillations, I'd expect a more variable climate (it is a statistical difference concerning the trends and the amplitude of variation about the mean). So if you remove El Nino effects over the last century, you don't lose any of the warming. Concerning your last line, if you could hold the incoming solar radiation, albedo, and outgoing longwave radiation constant on centennial timescales, then climate will be stable over that time. I would say that the climate generally is in equilibrium (or at least close enough) on decades to centuries, but new equilibriums will be reached when you externally force something on the system, which is bound to happen. The reason climate does change over those scales is because you can change the eccentricity of the Earth's orbit, tilt it up and down (more sunlight at the poles=lower albedo), change greenhouse gas concentrations and aerosols/dust, etc., long-term variations in solar output

My guess is a few introductory courses in climate? It appears Christy and friends could use that as well.

There’s a decadal vacillation superimposed on an overall warming trend. This isn't incredibly difficult to understand. Over long term scales (climate) you need to change the global mean radiatively at the top of the atmosphere (change solar in, change albedo, change the efficiency of infrared out). There’s a lot of cold water down in the deep ocean, and under certain internal conditions when you can bring it to the surface you can offset some GHG warming, but eventually that's going to heat up, unless you don't believe in thermodynamics. Likewise during an El Nino you're clearly "mining" some heat out of the ocean and putting it into the atmosphere, and that tends to create a spike for a year if you look at a plot of global T vs. time, but there is a small net tropical mean warming associated with most El Nino events. But the 0.1-0.2 K spike is superimposed on a longer-term, and much larger ~0.8 K trend, which is being externally forced. The circulation anomaly in an El Nino year is very different from that for a warmer climate. It's the net SW and LW radiation that change the global climate in equilibrium, but the climate system doesn't need to be an equilibrium on very short timescales. There is year to year variability all the time, and the ocean and atmosphere are always exchanging heat back and forth on time scales which can be more than a year. And changes in ocean circulation can cause changes in atmospheric circulation, which cause changes in clouds and water vapor, which change SW and LW radiation. When you increase greenhouse concentrations you create a situation where the planet takes in more radiation on net than it can get rid of at the top of atmosphere, and it will heat up. Without the CO2 increase you’d get warm,cold,warm cold, etc. with no long term trend.

The PDO doesn't have any evident signal on the global mean. If you want to explain with internal fluctuations how the heat content of the ocean has increased (globally) over the last century, or the direction of the radiative imbalance at the top of the atmosphere, you'll make some interesting contributions to thermodynamics. This is beside the point, but I also don't see the correlation you're seeing...shouldn't 1940 be just as warm as today if that is all? Any internal fluctutions you're seeing are superimposed on a long-term trend from greenhouse gases. This is just century old physics, I'm really not sure what you're arguing here. Concerning this 1998 stuff, if you were to run a best fit line using any temperature analysis (GISS, HadCRUT, etc) there is warming since then. But, it doesn't really matter if you seen warming or cooling, as any signal over such a short time interval is contaminated by noise.