Unforced Variations: July 2012

Have at it.

561 comments on this post.
  1. Unsettled Scientist:

    Sphaerica (Bob), actually I think the third link which you blow off has your answer: “For carbon dioxide the main 15-um band is saturated over quite short distances. Hence the upwelling radiation reaching the lower stratosphere originates from the cold upper troposphere. When the CO2 concentration is increased, the increase in absorbed radiation is quite small and the effect of the increased emission dominates, leading to a cooling at all heights in the stratosphere.”

    There is another similar quote from a different paper there: “In the stratosphere, however, temperature increases with altitude and as a result the cooling to space is larger than the absorption from layers below. This is the fundamental reason for the CO2 induced cooling.”

    If you want to dig into and get a deeper understanding, I would recommend reading the papers from which those quotes were taken (PDFs linked below), where you can read their words and probably dig into the math a bit.

    Trace-Gas Greenhouse Effect and Global Warming, Ramanathan, Royal Swedish Academy of Sciences (1998)

    Stratospheric Temperature Trends: Observations and Model Simulations, Ramaswamy et al, Review of Geophysics (2001)

    Hope this helps.

  2. Paul S:

    Obviously the sea rose much faster from 20,000 years to about 8,000 years ago than it is presently rising – but there has been many meters of melting in the pipeline for 20,000 years.

    After all, we are in an interglacial – isn’t this to be expected?

    A few papers have looked at this question. There were a couple of studies which used archaeological evidence to establish early Roman-period sea level relative to today. Lambeck 2004 looked at Roman fish tanks, which were placed at sea level so their present relative position can be used to deduce the level at the time of building (around 0AD). Sivan 2004 dealt with differential heights at which wells were built in Israel during the first millennium AD. From that they could infer changes to the height of the water table and therefore sea level. Both papers suggested a level roughly the same as it is now.

    Another paper (on which two RC contributors and one regular commenter coauthored), Kemp et al. 2011, studied salt-marsh sediment data to track sea level changes back to two thousand years ago. Their results suggested a level at 0AD ~20cm below that in the late-nineteenth Century. This paper also handily graphs other historic sea level studies, including the two mentioned above.

    So, the evidence suggests sea level two thousand years ago was no more than +/-30cm different from the late nineteenth Century. This indicates a negligable, and perhaps even negative, trend across the period and therefore that pipeline melting from the deglaciation period is effectively non-existent at this point.

    This picture is consistent with the long-term data and ice sheet models, which suggest a cessation of significant sea level rise 3-4 thousand years ago.

  3. Unsettled Scientist:

    Some day I’ll get good at doing these links… failed on the second PDF link there, so here it is raw: http://www.gfdl.noaa.gov/bibliography/related_files/vr0101.pdf

  4. Thomas Lee Elifritz:

    If all ice sheets melted, I think that is only 80 total meters in additional sea level rise. So your 70 meters is assuming almost all ice melting.

    I don’t see that happening over time frame of hundreds or even thousands of years – but only over millions of years).

    I guess you missed the transition to the Holocene. That took place in a few thousand years primarily via slow orbital forcing with relatively minor carbon dioxide feedbacks, and quite a bit more than 70 meters of sea level rise occurred during that period.. a very serious observational fail RickA.

  5. Thomas Lee Elifritz:

    Second, East Antarctica just won’t melt – raise the temperature 11C and it is still below freezing their year-round.

    Sorry, I missed this. The oceans control the air temperature down there. The oceans are composed of fluids with a tremendous heat capacity. Once the ocean turn over, it’s game over for East Antarctica. This is happening far faster than you are even willing to consider. I am not kidding you.

    This is an extremely dangerous problem that is now fixed for the next several thousand years, even if we stop burning carbon right now. The more carbon we burn from now on, it becomes even more dangerous much faster.

    I am frankly astonished on how deeply even the scientific community is about this very serious problem that is unfolding even as we speak. But then again, people know they only have a few decades left anyways. This is something that should have been taken care of several decades ago already but condensed matter physics just wasn’t up to the task then. It is now.

  6. t_p_hamilton:

    RickA sez:”In addition, Greenland is bowl shaped, is it not, so a bunch of ice would stay in the bowl or turn into a lake.”

    There is a couple of holes in your idea: http://membrane.com/sidd/greenrockturn.gif

  7. Kevin McKinney:

    #450–RickA–“Second, East Antarctica just won’t melt – raise the temperature 11C and it is still below freezing their year-round.”

    Except that that’s wrong. It’s wrong because there is dynamic ice loss–accelerated flow into the warming ocean. It’s wrong because if the Arctic goes ice-free, not to mention the WAIS, then how likely is a limit of 11C locally around East Antarctica?

    And BTW, Greenland may be roughly bowl-shaped, but the bowl is pierced–it drains to the sea.

    I’m wondering just how naive you really are? No offense, but surely such things are not excessively subtle to imagine.

  8. RickA:

    Thank you all for the information.

    I knew Greenland was bowl shaped, but did not know it would all drain to the sea.

    I looked at the topo map you linked to, and it seems that some of Greenland is below sea level.

    However, I could not estimate how much.

    Would not the bowl, hold back some and maybe a bunch of ice, preventing it from flowing into the sea?

    Won’t a bunch of melt form lakes on the surface of the ice and stay?

    I guess, in my naivety, that is seems that it is not as straigtforward as start melting and all of Greenland ice goes into the ocean.

    At least not for a very long time.

    As for East Antarctica – of course if the temperature gets warmer than 11C, that could make a difference.

    However, I have read papers which indicate that even in the middle of the last interglacial, that it was only 6C warmer (on average) at the poles – so is an 11C increase (on average) really remotely possible? Even over thousands of years? Maybe on the edge of the WAIS I could see that, but in the middle of the EAIS?

    Anyway – thanks for the comments and information.

  9. Tokodave:

    And by the time all the ice has melted into a “giant” (http://en.wikipedia.org/wiki/File:Topographic_map_of_Greenland_bedrock.jpg) lake in Greenland’s bowl shaped interior sea levels would rise somewhere close to 7 meters….and that’s just Greenland’s ice. Hey, maybe Antarctica is bowl shaped too??? http://en.wikipedia.org/wiki/File:AntarcticaRockSurface.jpg OK, not so much, so there’s another 61 meters if it all melted…

  10. Radge Havers:

    Getting a little more clinical about what we’ve no doubt already noted:
    New study shows correlation between so-called “skeptics” and tendency towards “conspiracist ideation”

    “The researchers offered no simple solution on how to counter the trend of climate change denial. ‘Conspiracist ideation is, by definition,’ they write, ‘difficult to correct because any evidence contrary to the conspiracy is itself considered evidence of its existence.'”

  11. Ric Merritt:

    Re inappropriate comments:

    Thomas Lee Elifritz, #433, mentions another commenter’s religion (known, purported, or whatever) in what is clearly intended as a sneering manner.

    This is over the line (and I speak as someone who does not profess the religion mentioned).

    The moderators should expunge the reference. I dislike reading blogs when I see this stuff.

  12. Thomas Lee Elifritz:

    Denialism of the science being present here is a religion, Ric. What DanH is engaged in here is far more insidious and dangerous in my humble opinion, considering the stakes, and I consider religion to be an extremely dangerous thing already. But I apologize – as you are right, bringing up IP addresses here is inappropriate. If DanH is offended, he can always sue me for libel.

  13. sidd:

    very pretty paper by all the usual suspects on GPS measurements of seasonal bedrock rebound under GRIS

  14. Patrick 027:

    Re RickA @ 458 and earlier – I’m no expert on ice flow dynamics but here’s a few thoughts:

    Would not the bowl, hold back some and maybe a bunch of ice, preventing it from flowing into the sea?

    Yes it would hold back some temporarily. However, if the depth of the remaining ice were not much greater than the depth of the basin (below sea level), the buoyancy could allow water to force it’s way under the ice. Of course, at that point the ice is floating so it would only add to sea level by it’s lack of salt (thus it’s melt will still take up a tiny bit more volume). It is the volume of ice above sea level by a height of ([density of liquid/density of ice – 1] * depth of basin from sea level) that is the concern w/ respect to sea level rise – Except for isostatic rebound, which will raise the basin with removal of extra weight – the plastic portion of this is long-term but there is an elastic component which acts fast.

    If the ice shrinks to the point that it is within the basin in contact with the ocean, then you get a long boundary where calving can occur.

    Won’t a bunch of melt form lakes on the surface of the ice and stay? – depends. Liquid water can sometimes find drainage pathways through the ice if not on top of it. And you need concavities on the ice surface; for a given ice topography, there’s a limit to the volume those could hold. The liquid water will absorb solar radiation (more than would the ice).

    I guess, in my naivety, that is seems that it is not as straigtforward as start melting and all of Greenland ice goes into the ocean. I’m not sure that’s what people meant to say. However there is some warming still ‘in the pipeline’ (there is a remaining radiative disequilibrium) and people globally have not even started to reduce CO2 emissions, etc. Also, it may need to get colder to restart an ice sheet than to merely preserve it (ice surface elevation helps keep it cold). Also, even if snowfall increases (as I’m guessing it will over EAIS), the warmer temperatures at the boundaries or higher level of the boundary between growth and ablation will thin the edges – with melt and possibly calving – and steeper slopes (and basal lubrication with meltwater, I’d guess) overall will tend to increase the flow rate from accumulation to ablation regions. It’s not like the Greenland ice sheet is now set to fall completely apart within a decade after one time with nearly all the surface near or above freezing, but it’s going to keep shrinking.

    At least not for a very long time.

    However, I have read papers which indicate that even in the middle of the last interglacial, that it was only 6C warmer (on average) at the poles – so is an 11C increase (on average) really remotely possible? Even over thousands of years? Maybe on the edge of the WAIS I could see that, but in the middle of the EAIS?

    The last interglacial was only a fraction of that warmer than it has been recently in this interglacial; if the global average temp went up by 6 C than why couldn’t EAIS warm up a lot more? How much coal is there left? Too much. Setting aside slower-acting (non-Charney) positive feedbacks.

  15. Patrick 027:

    431 Mike

    (PS Re 436 Unsettled Scientist – aside from physicists, I’d expect better of an engineer anyway)

    On the appropriateness of global average surface (or surface air) temperature as a metric for climate

    Climate is not even approximately a one-dimensional thing of course (except perhaps on a very simple planet, made out of silver or recieving the same stellar radiant intensity from all directions … no atmosphere, not rotating? … etc.). An equilibrium climate state is perhaps well-described by a strange attractor in some n-space, with extremely large n, or approximately so with smaller n and a probabilistic strange attractor with trajectories bluring into each other. Given climatic states may encompass the seasons and diurnal cycles, etc, there will be both forced and unforced (internal) variability – the strangeness of this attractor tending to reflect the later (although some internal variability is almost periodic (QBO)), except if over such long time periods that the chaos of multibody orbits and obliquities comes into play. Using not just the (forced cycles of) the (time-weighted) centroid of the attractor but even the (forced cycles of) standard deviation, still misses a lot (although for some purposes that lot may not be necessary).

    However, there are boundary conditions (externally-applied forcings) which determine the shape, size, and location of this attractor and all its intricacies.

    (PS over short time scales, or for purposes of understanding how things work, atmospheric CO2 can be treated as such a forcing, as can thick ice sheets. Over longer time periods, it is the climate-independent (part of) sources and sinks of CO2 that are the forcings, which determine atmospheric CO2 in combination with climate-depended effects (feedbacks). The distinction between forcing and feedback may shift with the timescale, as perhaps may be the number of dimensions used to describe climate.)

    If we leave some of these BCs fixed (coriolis effect, the mass and composition of the (approximately/mostly) inert and optically-irrelevant portion of the atmosphere, the mass of the Earth, the shapes of ocean basins, the topography of continents, the thermal conductivity and expansion of substances, etc, the mechanical stirring of the tides) and specifically adjust those forcings that are energy inputs and sinks:

    Solar/stellar (shortwave – or SW) heating – or more correctly, incident solar/stellar radiation, as albedo is partly climate dependent, to an extent depending on time-scale;

    the longwave properties of the atmosphere (Greenhouse effect) (in terms of atmopsheric composition and spectra of substances, etc.) and surface, and the LW darkness of space.)

    the geothermal, tidal-dissipation, anthropogenic direct heating, energy of meteors – all of those are, on the (for most of geologic history of the Earth (not Jupiter) – very small in comparison with the former and often can be ignored – as can (for the Earth, I’d guess most planets?, setting aside forcing/feedback distinctions) the absorption and emission of non-thermal radiation and the ‘evaporative cooling’ of H-escape to space, etc.

    – then we can expect that changes in climate will involve changes in heat fluxes; moreover, since most – nearly all – of the energy flux in and out of the climate system is radiative and the outgoing is almost entirely thermal radiation emitted as a function of temperature, we can expect that the temperature response will be of some importance in determining how the equilibrium climate changes in response to a forced change in energy flux. (Even non-radiative and radiative transports and transfers (thanks, Unsettled Scientist) of energy within the climate system depend on temperature – substances have their adiabatic lapse rates which affects convection, conduction requires a temperature gradient, etc.)

    In particular, we can leave obliquity and eccentricity and the alignments of those alone, we can leave the seasonal and latitudinal distribution of incident solar radiation at TOA (top of atmosphere) alone and deal with global average radiative forcing. (The climate response to orbital forcings depends on regional/seasonal effects, but those effects can have a global average feedback, which acts like a global average radiative forcing.)

    In the global average, at equilibrium, vertical fluxes must be balanced. Imbalances caused by a change in forcing will result in convergence or divergence of energy fluxes, which may go into latent heating, but can also change the temperature; the amount of heat required for a given temperature change depends on heat capacity, BUT LW fluxes will generally be part of how balance is restored and these require temperature changes – the equilibrium climate requires an equilibrium temperature and thus can be described by it.

    Yes, radiation depends on temperature in a nonlinear way, so the arithmetic mean isn’t exactly what the response depends on. However, for example, on Earth the difference between the global annual average temperature and the fourth root of the global annual average of T^4 is quite small (I think it might be ~ 1 K, roughly – google Trenberth Kiehl Fasullo for more (I hope I spelled those right, it’s been awhile)). And aside from that, if the variation in surface temperature horizontally and over the course of a year or a few ENSO fluctuations changes, or for that matter the vertical variation (temperature profile), or the alignment with spatially-variable optically-important matter (clouds, H2O vapor … also surface types) changes, this change will tend to have some particular shape and size (the climate change) and it changes the relationship between global average surface temperature and global average OLR (outgoing longwave radiation) – or global average LW fluxes at any level (like at the tropopause) – which means it acts like a feedback (example: lapse rate feedback, due to the effect of temperature on moist adiabatic lapse rates), just like water vapor, clouds, and sea ice.

    (The changes in spatial and temporal temperature variations may be different for different forcings (with different vertical and horizontal and temporal forcing structure), and the effects of this on circulation and feedbacks well as directly on radiation will give different forcings different ‘efficacies’ (climate sensitivities relative to some standard forcing agent’s effect) – however, the effects of spatial and temperorally varying feedbacks may dominate some of the temperature variation changes if the forcings are not too different.)

    So when the climate changes due to a forced change in (global average) energy inflow or outflow, global average surface temperature will tend to change; other dimensions will also change, and those are very important (precipitation distribution in space and time, in particular) – but they change along with the global average surface temperature, not independently of it (remember that internal variability is included in the description of climate).

    If there are multiple climate equilibria for a given set of BCs, then

    – if on the timescale considered the strange attractors of the equilibria interconnect then they are all part of a larger climate state; see above.

    If the climate stays on one until sufficiently perturbed, however, then each equilibrium state may be regarded as seperate – and it’s possible each may have global average surface temperature as a useful metric.

  16. Craig Nazor:

    RickA – I am no scientist. However, my fascination and curiosity about climate change leads me to believe that you are unaware of certain facts.

    Ice weighs a lot. The massive amount of ice now on Greenland pushes the land elevation down. When that ice melts, the land will rise. This is called isostatic rebound, as mentioned in an earlier comment. Little, if any, of Greenland will be left below sea level once the weight of the ice is gone.

    On top of the fact that it gets relatively little solar radiation and it is a landmass, the Antarctic is unusually cold right now because it is somewhat thermally insulated from the rest of the earth’s surface (it is far colder than the Arctic). This is at least partially due to the thermohaline ocean current that now completely encircles Antarctica. Since the Antarctic is right now colder than it would be if surface air and water mixed optimally, it will eventually heat up more rapidly than the rest of the planet, particularly if there is any change in the thermohaline current. This is a pretty complicated topic, and there are others on this list that could fill in the details far better than I.

    The Antarctic has been completely ice-free in the not-to-distant past. What would keep this from happening again, if CO2 keeps rising unabated?

    ReCaptcha: “many upities.” I’ll say!

  17. Brian Dodge:

    I wonder if Mike the engineer realizes that the argument that “more heat is going into creating water vapor than the scientists (“scientists”) realize” means that the dynamic sensitivity is lower but the equilibrium sensitivity is higher, since more water vapor results in a greater amplification. This might also mean that the lower rate of rise seen since ~98 would be accompanied by more intense thunderstorms and other “rainfall events” – and larger, more frequent, damaging floods plus other storm events, e.g. tornadoes.
    It might be educational to find out what insurance companies have seen.

  18. Patrick 027:

    re my last comment – should’ve bolded this phrase too:
    the equilibrium climate requires an equilibrium temperature and thus can be described by it.


    Re Sphaerica (Bob)

    You might find this useful

    It doesn’t seperate the OLR between stratospheric and other components, but at least you can look at the downward LW flux at the tropopause. There isn’t a single globally-representive setting, though, that I know of.

    For a global average OLR spectrum, there is a CO2 valley, and it widens with a doubling of CO2, reducing OLR by an area proportional to valley depth and widenning. A seperate increase in OLR is related to the spike at the center of the valley associated with the peak in CO2 effective emmitting level reaching upward toward or into the warmer part of the stratosphere (associated with UV absorption by ozone, of course). Since H2O optical thickness is nearly zero above the tropopause within the immediate vicinity of the CO2 band, at the tropopause, the effective CO2 valley in *net flux* goes almost or all the way to zero at the bottom (because it’s saturated or close to saturated there), while the flux absent CO2 is equal to the OLR value at TOA – so the depth of the valley in net flux is greater than at TOA and for the same widenning, there is that much more forcing. Even without the ozone layer’s heating effect, unless the temperature got near enough to 0 below the skin layer and CO2 were sufficiently unsaturated at the tropopause, there should still be some stratospheric cooling (at least before the surface+troposphere warms, and setting aside SW feedbacks).

    Although not to the point you wanted, and maybe you already know this, but’s it’s interesting to consider four cases:

    grey gas in LW, no direct solar heating above the tropopause (set tropopause at surface for pure radiative equilibrium with no direct solar heating of atmosphere): temperature increases with height up to TOA, approaching skin temperature Tskin = TOLR * (1/2)^(1/4)); TOLR is blackbody temperature corresponding to the equilibrium OLR flux per unit area.
    In the optically-thin limit, except for the troposphere the whole atmosphere is at Tskin.

    Increasing LW opacity at first reduces the heat source for the skin layer, causing transient cooling, but at equilibrium the warming merely decreases to zero going up to TOA.

    grey gas, direct solar heating above tropopause
    – PS without any LW emissivity in the atmosphere, temperature must increase upward from the surface (assuming no mechanical stirring); initially adding a grey gas causes an OLR increase, but the tropopause (surface in this case) forcing is positive; the stratosphere cools

    (PS in general: initial cooling = instantaneous tropopause forcing – instantaneous TOA forcing; some fraction of that cooling is realized as an OLR reduction and some is realized as a change in tropopause level forcing (although with a sufficiently optically thick stratosphere, the lower stratosphere could warm and the tropopause level forcing could actually increase with stratospheric adjustment – cooling could be confined to an upper portion, depending).

    In the skin layer, the greater the optical thickness relative to solar heating, the more the equilibrium temperature is pulled down toward the skin temperature absent solar heating. But the skin layer must get thinner with increasing opacity.

    3. non-grey gas, no direct solar heating above tropopause

    Increasing opacity where it is already largest displaces equilibrium OLR away from the band that determines the skin temperature, thus cooling it; what happens below the skin layer – well that depends, I guess. There could be some rebound after climate reaches equilibrium but some cooling would remain.

    The CO2 valley in OLR widens so there are additions of opacity where it is not so large (at least above where water vapor may be abundant and in clear skies or above low clouds, at least); but it still leaves parts nearly transparent in the stratosphere, so equilibrium OLR will be displaced and reduced in the CO2 band of the stratosphere (what about the water vapor bands in the stratosphere – well, H2O increasing in the troposphere so…) – unless positive SW feedbacks increase total OLR so much this overwhelms that effect.

    Skin temperature Tskin only equals TOLR * (1/2)^(1/4) if determined at one part of the spectrum or some specifically-weighted set of bands; if determined at other parts, it could range from close to TOLR (shorter wavelengths) to near TOLR/2 (longer wavelengths). So you could get cooling or warming by shifting bands or adding bands in different places.

    Graphing the lapse rate over optical thickness in terms of the Planck function – it can be concave at some wavelengths and convex at other wavelengths; concavity on the scale of moderate optical thickness will lead to warming and convex curvature will lead to cooling, etc;

    combine 2 and 3.

  19. Tokodave:

    Brian at 467…the link didn’t work…try this one: http://www.guardian.co.uk/commentisfree/cifamerica/2011/jun/28/climate-change-climate-change-scepticism

  20. wili:

    Did anyone else see this article? (apologies if it has already been discussed and I missed it):

    “UV Dosage Levels in Summer: Increased Risk of Ozone Loss from Convectively Injected Water Vapor”


    Looks like a major new dynamic that I, at least, hadn’t heard about before.

  21. sidd:

    Mr. Craig Nazor wrote on the 27th July, 2012 at 8:21 PM:
    “..the thermohaline ocean current that now completely encircles Antarctica.”

    Antarctic Circumpolar Current is wind driven ?


  22. Patrick 027:

    I had posted a comment several months ago about stratospheric cooling that was better organized and probably clearer; I can’t find it, but in that comment I refered to this

  23. dallas:


    A little more up to date discussion of ozone depletion variation.

  24. Kevin McKinney:

    For those who may not yet have read Michael Mann’s book on the “hockey stick,” you can get the (fairly extensive) gist here:


    On another topic–wili, thanks for an interesting link. It seems that CO2 and ozone depletion continue to intertwine in fascinating and sometimes unexpected ways…

  25. Brian Dodge:

    “Antarctic Circumpolar Current is wind driven ?” Yup.

    Sinking cold air over the Pole slides down to the edge of the continent, and Coriolis effects cause an Antarctic Circumpolar Vortex. This west to east wind couples to the ocean surface creating ripples, which increase the friction and coupling between the ocean surface and the wind, which increases the wave height, and bingo! the Roaring Forties. Because the Antarctic Peninsula concentrates this surface flow where it enters the South Atlantic, enhanced Eckmann transport contributes to the AMOC; IMHO, this forcing will maintain the AMOC in some form even if when there is catastrophic failure of the Greenland Ice Sheet – it won’t shut down, but it will weaken and move somewhere less favorable for clement weather in the UK & norther Europe.

  26. Kevin McKinney:

    Off-topic–but I’m hoping for a little help with a literature search. I’m looking for anything that sheds light on the *effects* of an ice-free Arctic. Certainly we can reasonably predict some things, like further increases in Arctic coastal erosion, or polar bear population crashes, without straining too hard.

    But what does the literature say about things like, say, general atmospheric circulation changes, or marine ecological changes? Any leads for me out there?

  27. Steve Fish:

    27th Patrick. Your Stoat reference seems to say that the stratosphere doesn’t actually cool, it just doesn’t warm as much as the troposphere does from the CO2 greenhouse effect. Steve

  28. dhogaza:

    Steve Fish:

    Your Stoat reference seems to say that the stratosphere doesn’t actually cool,

    Directly from the last sentence of Stoat’s explanation:

    hence it cools. Please turn off your timer.

  29. Patrick 027:

    Re 477 Steve Fish – I don’t think it says that, or it’s not supposed to (not about the CO2 effect, anyway).

    This would be true for a grey gas (optical thickness is constant over the LW part of the spectrum) in the absence of direct solar heating (in which case, the lapse rate is positive in the stratosphere, just smaller above the tropopause than below it) – adding more grey gas, OLR is temporarily reduced – the upward LW flux from the solar-heated troposphere-surface (as a unit) is reduced and at least some portion of the upper atmosphere including the skin layer cools in response. However, as heat accumulates and warming occurs below, equilibrium is restored when OLR returns to its prior value (assuming that was in equilibrium before) (setting aside any SW feedbacks that would change equilibrium OLR); the skin temperature must warm in response and return to what it was. The temperature at TOA returns to what it was, and the temperature profile is vertically compressed against TOA – it warms everywhere below TOA but less near TOA.

    With direct solar heating sufficiently close (optically) to TOA, adding more grey gas opacity increases the ability of that volume of air to emit radiation, so it cools – if within the skin layer, this continues with asymptotic approach to the skin temperature (what it would be absent solar heating).

    However, adding more grey gas opacity changes where this can happen – as the solar-heated layer becomes more deeply embedded in the optical thickness of the atmosphere, more photons emitted from there fail to reach space directly; the photon mean free path length gets shorter, requiring a larger temperature gradient to sustain the same net flux, etc – it warms.

    (Also, as a solar-heated layer gains optical thickness, emission within that layer keeps going up but the portion that escapes that layer can only approach a blackbody value for that layer’s temperature(s) (and then the temperature of the top or bottom part, etc.))

    Adding opacity (PS of the emission-absorption at LTE kind – scattering is too complicated to discuss right now (not enough time)) within just one band, … (equilibrium OLR is displaced to other parts of the spectrum, allowing skin temperature to be reduced at equilibrium…



  30. Patrick 027:

    Re last part of that comment – I was going to respond to Sidd; didn’t get to it yet.

  31. Patrick 027:

    (PS grey gas no solar heating above tropopause: if the stratosphere were optically thick then at first only the upper part would cool with the reduced OLR; down below, each layer recieves less radiation from below but more from above. If the heat capacity of the troposphere+surface were relatively small then perhaps cooling couldn’t propagated down into the stratosphere much before the surface+troposphere warmed, etc.)

    … Adding opacity to one band reduces OLR in that band; the skin layer cools, cooling may reach down farther, etc. Warming of the surface and troposphere increases OLR, including in bands where opacity was not added; thus OLR ends up higher than before at some parts of the spectrum – if solar heating doesn’t increase, OLR must be reduced where opacity was added, so at least some of the upper atmospheric cooling must remain – Especially if this band is at longer wavelengths, where emission is less sensitive to temperature.

    Widenning a band: well, consider adding another band and then having reach the same opacity as the first – then, the OLR will be even greater outside these two bands, but the temperature below will be warmer than it was at the same opacity in a narrower band, so the skin layer and stratosphere in general should get warmer with band widenning (at least in the case of a rectangularly-shaped absorption spectrum).

    Given a strongly-absorbing band, Adding bands with weaker opacity can tend to make parts of the stratosphere warmer, especially if this opacity is concentrated in the stratosphere (absorbing greater fluxes from lower down at the surface and in the lower troposphere, as the LW ozone band does) – especially if the band is added at shorter wavelengths, where emission falls a lot for a small temperature decline. Heating at other LW bands is like solar heating to the extent that it can provide more potential cooling response to an increase in opacity in a more opaque band under some conditions.

    With a grey gas, in radiative equilibrium, the net LW cooling over the whole LW portion of the spectrum balances SW heating, but over the spectrum LW cooling may vary and can switch signs…

  32. Patrick 027:

    With a grey gas or wavelength-dependent opacity , in radiative equilibrium, the net LW cooling over the whole LW portion of the spectrum balances SW heating, but over the spectrum LW cooling may vary and can switch signs…

  33. Patrick 027:

    Re 471 sidd

    Currents can be wind-influenced and thermohaline – wind directs warm currents into places where they cool and get saltier from evaporation, etc.

    My understanding is that parts of the Southern Hemisphere, in particular in a ring around Antarctica and maybe neighboring places, will have some delay in warming because of wind-driven upwelling of water which hasn’t been at the surface in a while.

  34. deconvoluter:

    Re # 431

    Averaging these atmospheric temperatures is generating a meaningless number.

    [My italics]. I think you meant useless . That depends what you want to do with the answer. Remember that this refers to the output of a calculation. At this stage, you will have finished with the basic physics. If on the other hand, you intend to use the average as the input in some sort of model, that would be quite a different matter; then you would have to be more careful. Only then would you have to substitute the answer into one of the basic equations and the errors could build up.

    For example a proper estimate of the total IR emitted by the surface would involve weighting the temperature with the local emissivity (assuming that the T^4 is first linearised) before taking the average. If this step was omitted and you performed a back of the envelope calculation the result might be good enough for an introduction in Wikipedia, but it might not be the best that you could do with your data.

    Suppose instead you just want to compare a simulation on a computer with the real world. There is no reason whatsoever why you should not compare the average surface temperatures. If they turned out to be consistent then you have some reason to trust the simulation. You could just as well compare the mean of any function of the temperature. It would be a question of numerical analysis not physics. I don’t see why an engineer should not appreciate that.

    Correct me if I am wrong.

    the climate ‘scientists‘ like to incorrectly use temperature as a metric for heat energy

    I think that proposition is rendered tautological by the use of quotes. Perhaps you are underestimating RC?

  35. Patrick 027:

    Re my 468 re my 465 re 431

    the equilibrium climate requires an equilibrium temperature and thus can be described by it.

    not described in full or even much – a better word would be identified, like a label – someone says warming from 288 to 291 K and people familiar enough with the subject know what that means for precipitation, sea level, etc….

    the equilibrium climate requires an equilibrium temperature and thus can be identified by it; temperature change is a key measure of climate change

  36. Patrick 027:

    (In a hurry, typing fast, sorry for broken grammar whatever)

    Re 483 re 471 – positive feedback – greater temp contrast between unwarmed upwelling water and warmed water affects circulation drives winds that cause upwelling

    Re my 472 regarding stratospheric cooling – FOUNT IT!


    Correction/clari fication: (end of 2.)
    if o ptical thickness in increased in an atmospheric window (assuming surface emissivity is high enough in that window, to generalize this more), this tends to warm the skin layer by intercepting OLR that is originating in warmer places.

    – except skin layer is affected most by strongest bands – skin layer is thin enough to be skin layer in strongest bands, then much weaker bands barely matter to the skin layer at least directly

    but weaker bands allow heating of thicker layers such as when stronger bands are nearly saturated (net LW flux tends to zero in sufficiently strong bands; photon mean free path can traverse larger temperature variations in weaker bands)- may indirectly affect skin layer temp, although the signal may have trouble propagating upward through the strong band to do so, could leak out through weaker band…?

  37. Patrick 027:


    emission and absorption per unit volume (in x,y,z or x,y,p) increase in proportion with emission cross section density; emission with Planck function, absorption with a m b i e n t intensity – important for thinner layers. If thin layers are stacked with some solar/other heated and some not, increasing optical thickness tends to even out equilibrium temp by increasing emission out of thin warm layers, etc. – up to point when layers are themselves moderately optically thick… (pertains to skin layer in particular)

    net LW flux ~ d(Planck function)/d(optical thickness) – averaging over thinner temperature variations which don’t emit much; peaks when photon mean free path (or mean emission-absorption displacement if scattering occurs) is some fraction of wavelength of temp variation…

    LW heating ~ d^2(Planck function)/d(optical thickness)^2, averaging over thinner variations – a temp max or min not required; changes in slope can be sufficient …

    try linearly superimposing different sin/cos functions (qualitatively makes sense, but nonlinearities here)

    … but curvature in Planck function necessary for zero net flux at relatively sharp boundaries in optical cross section density – see my “September 28, 2011, 9:30 pm” comment here

  38. Patrick 027:

    … but curvature in Planck function necessary for zero net flux (di/con)vergence at relatively sharp boundaries in optical cross section density – see my “September 28, 2011, 9:30 pm” comment here

  39. John E. Pearson:

    Richard Muller in the New York Times

    CALL me a converted skeptic. Three years ago I identified problems in previous climate studies that, in my mind, threw doubt on the very existence of global warming. Last year, following an intensive research effort involving a dozen scientists, I concluded that global warming was real and that the prior estimates of the rate of warming were correct. I’m now going a step further: Humans are almost entirely the cause.


  40. Patrick 027:

    Re my 488 and prior:

    that’s specifically when the optical thickness per unit temperature change has a relatively sharp change. The complexity arises because the flux up or down comes from intensities (weighted by cosine of zenith angle) at different directions contribute. Relative to optical depth, the surface is a good conductor and space is also basically (effectively as a heat sink for OLR) isothermal. And … see link above.

  41. Patrick 027:

    Re John E. Pearson – the Richard Muller oped – (for some reason I feel skeptical about the research – not the conclusion, just the quality of the research. Has it passed peer review yet?) After ‘turning’ the page, I see he may still have some catching up to do. But it’s progress.

  42. Chris Dudley:

    John (#490),

    A lot of bogus claims in that article. Perhaps it is to be expected from a guy who is paid to be wrong for as long as possible.

    [Response:Way, way over the top.–Jim]

  43. Hank Roberts:


    Many youngsters nowadays — most of you born after 1950 or so — never heard this stuff growing up, where you grew up. Listen.

  44. Kevin McKinney:

    #493–For one, it’s certainly not true that cooling elsewhere in the world more than compensated for the US heatwave–global lower trop from UAH was a toasty-enough .37C, while NCDC had June as the 4th-warmest ever, with this map:


    I’m guessing that Dr. Muller’s comment in this regard was not exactly data-driven.

    But I have to whole-heartedly endorse his conclusion:

    “I hope that the Berkeley Earth analysis will help settle the scientific debate regarding global warming and its human causes. Then comes the difficult part: agreeing across the political and diplomatic spectrum about what can and should be done.”

  45. Pete Dunkelberg:

    CO2 & carbon cycle news from the sane world:

    Lead author, Dr Jean-Baptiste Sallée from British Antarctic Survey says, “The Southern Ocean is a large window by which the atmosphere connects to the interior of the ocean below. Until now we didn’t know exactly the physical processes of how carbon ends up being stored deep in the ocean. It’s the combination of winds, currents and eddies that create these carbon-capturing pathways drawing waters down into the deep ocean from the ocean surface.

    Meanwhile I’m surprised that “unforced variations” has barely noticed the BEST development which is rather important in the other part of the world.

  46. Hank Roberts:

    > Science is that narrow realm of knowledge that, in principle,
    > is universally accepted. — Muller

    That does rather narrow the field, doesn’t it?

    I’d guess congratulations to Robert Rohde would be deserved.

  47. sidd:

    Prof. Pelto has posted a graph from Dr. Fettweiss at the Arctic Sea Ice blog.

    The second panel is awesome (in an older sense of the word,) this may be a teraton year for GRIS mass waste


  48. SteveF:

    Hail the new king of climate science! Hilarious piece by Roger Sr:


  49. Patrick 027:

    Re my 487

    net LW flux ~ d(Planck function)/d(optical thickness)

    net LW heating in K/time = d(net flux)/d(heat capacity)
    = d(net flux)/d(optical thickness) * d(optical thickness)/d(heat capacity)
    ~ d^2(Planck function)/d(optical thickness)^2 * optical cross section/(mass*cp)

  50. Jan Galkowski:

    The BEST group has, after a media splash, released their latest findings, at http://berkeleyearth.org/results-summary/ and the draft paper at http://berkeleyearth.org/pdf/results-paper-july-8.pdf. Note their temperature averaging process paper is http://berkeleyearth.org/pdf/methods-paper-with-appendix-may-14.pdf although I look forward to the full publication because the PDF’s equations are messed up. (Don’t they use LaTeX?) It seems it’s an adaptation of kriging, which is just BLU Estimation. (See Christensen, PLANE ANSWERS TO COMPLEX QUESTIONS,1987, or Glover, Jenkins, Doney, MODELING METHODS FOR MARINE SCIENCE, 2011, or Schabenberger and Gotway, STATISTICAL METHODS FOR SPATIAL DATA ANALYSIS, 2005, or Banerjee, Carlin, Gelfand, HIERARCHICAL MODELING AND ANALYSIS FOR SPATIAL DATA, 2004.) Apparently hasn’t (?, don’t know) been applied to temperature data, but, surely, it’s used in oceanographic work, e.g., SSTs (Glover, Jenkins, Doney). After I get equations, I’m interested in how they ascertained uncertainty in their estimates. This BLUE assumes Gaussian errors, but that’s not really necessary these days, as confidence intervals can be obtained through dependent data bootstraps.