RealClimate logo

On a Weakening of the Walker Circulation

by Ray Pierrehumbert and Rasmus Benestad

Second article of our 3-part series on atmospheric circulation and global warming

In Part I we outlined some general features of the tropical circulation, and discussed ways in which increases in anthropogenic greenhouse gases might affect El Niño. Now we take up the question of how global warming might affect the quasi-steady east-west overturning circulation known as the Walker Circulation. The Walker circulation affects convection and precipitation patterns, the easterly Trade Winds, oceanic upwelling and ocean biological productivity; hence, changes in this circulation can have far-reaching consequences. It also provides the background state against which El Niño events take place, and so changes in the Walker circulation should form an intrinsic part of thinking about how global warming will affect El Niño. In a paper that recently appeared in Nature, Vecchi, Soden, Wittenberg, Held, Leetmaa and Harrison present intriguing new results which suggest that there has already been a weakening of the Walker circulation in the past century, and that the observed changes are consistent with those expected as a response to increases in anthropogenic greenhouse gases. The discussion in Vecchi et al. also raises some very interesting issues regarding the way the hydrological cycle might change in a warming world.

1. The main result of the paper

Vecchi et al. compared the observed trend in the Walker circulation between 1861 and 1992 to that yielded by simulations from the GFDL CM2 general circulation model, run with and without anthropogenic forcing. The comparison was done on the basis of surface pressure, because the humble barometer is a simple instrument which, with low technology, can nonetheless yield very accurate results; hence there are good long term instrumental records from the logs of intrepid tropical mariners. Instrumentation for accurately measuring winds only came along later, to say nothing of instrumentation for monitoring convection patterns or the subtle circulation aloft. Surface pressure provides a good proxy for the Walker circulation because, near the Equator, winds tend to flow from regions of high pressure to regions of low pressure, under the acccelerating action of pressure gradient forces. In the Walker circulation, the low pressure is in the West Pacific and the high pressure is in the East Pacific. This gradient strengthens the Easterly Trade Winds to the east of the rising branch (above the low pressure cell) and counters the Easterly Trades on the west side of the low pressure cell, weakening them or even turning them into westerlies. (See the Walker cell sketch in Part I ) .

The following figure shows maps of the observed and modeled pressure changes between 1861 and 1992. The observations (upper left panel) show an increase of pressure in the Western Pacific and a decrease of pressure in the Eastern Pacific, indicating a weakening of the east-west pressure gradient associated with the Walker circulation. The model simulations (upper right panel) driven by all known climate forcings over the period in question show a very similar pattern of weakening. The bottom two panels demonstrate that this weakening is due entirely to the anthropogenic forcings — greenhouse gas increases offset by sulfate aerosol effects. The simulations shown are the mean of an ensemble of five simulations of the period starting with slightly different initial conditions.

Sea level pressure maps from the Vecchi et al. paper

Most press reports summarized this result as a "weakening of the Trade Winds" in response to global warming. As a description, that’s not too bad, given that the indicated trend in the Walker circulation does indeed lead to a weakening of the Trades over most of the Pacific. However, the Trade Winds are primarily caused by the Hadley circulation, and are only modulated by the Walker circulation, so it is more precise to think of this result as indicating a change in strength of the Walker circulation.

2. Circulations and the hydrological cycle in a warming world

One of the things that makes the findings of Vecchi et al. especially interesting is that they are consistent with some rather robust theoretical arguments linking the strength of circulations to certain aspects of changes in precipitation and water vapor. These arguments are explained in more detail in Held and Soden 2006 (preprint available here). The argument begins by noting that the Clausius-Clapeyron equation, predicts a strong increase of boundary layer water vapor content with temperature (about 7% per degree of warming); the increase of low level water vapor with temperature is not controversial, since oceanic boundary layers are in contact with their moisture source and stay rather near saturation. Observations tend to support the expected increase (e.g. Wentz and Schabel Nature 403, January 2000. Some of the observational references given by Vecchi et al. in support of the increase actually deal more with the extratropics than the tropics, but the general principle is not seriously in question.) One might then expect that there would be a precipitation increase in proportion to the increase in water vapor content. However, it has been known since the earliest general circulation simulations by Manabe that as the Earth warms in response to increasing CO2, the precipitation increases much more slowly than Clausius-Clapeyron would suggest — typically only 2-3% per degree of warming. Because latent heat release in the course of precipitation must be balanced in the global mean by infrared radiative cooling of the troposphere (over time scales at which the atmosphere is approximately in equilibrium), it is sometimes argued that radiative constraints limit the rate at which precipitation can increase in response to increasing CO2. This argument is stated, for example, in Allen and Ingram.and repeated in Vecchi et al. The argument isn’t actually as firm a constraint as generally believed, since the infrared radiative cooling of the atmosphere is affected by the temperature difference between air and the underlying surface, which can adjust to accommodate any amount of evaporation Nature wants to dump into the atmosphere (as shown in Pierrehumbert 1999 ("Subtropical water vapor…" available here)). This is why single-column radiative convective models can show stronger increases of precipitation with temperature, even approaching the Clausius-Clapeyron limit. However, the relatively weak increase in precipitation with temperature seen in general circulation models is robust across models, suggesting that with suitable additional conditions the argument given in Allen and Ingram can be made to work.

Taking the slow increase of precipitation with temperature as a given, the more rapid increase of boundary layer humidity implies that the rate of transport of moisture from the boundary layer to higher levels where it rains out must go down. One way to do this is to decrease the strength of large scale circulations like the Walker and Hadley circulations.

If this argument seems obscure, here’s an analogy that may prove helpful. Suppose you live on a tropical island where water must be brought in buckets to your hut by the local authorities. The size of the buckets (which are always full) stands in for the boundary layer water vapor content. You need the water because you raise pigs, which are very temperature sensitive; to keep them comfortable you need to throw a certain amount of water over them each day. The amount needed per day is determined by the temperature, via aspects of pig physiology we need not go into. The amount of water per day you need to dump on your pigs stands in for precipitation. The standard bucket is one gallon, and at a normal temperature you need to throw four gallons per day over your pigs, meaning you have to rouse yourself four times per day and go out to empty your buckets. Now (parbleu!) global warming strikes, and it is two degrees warmer. By the physiology of pigs, you now need to dump eight gallons of water per day over your pigs. However the government has gone overboard and passed the Clausius-Clapeyron law, which mandates that each pig farmer now gets four-gallon buckets when the temperature gets two degrees warmer. That means that you now only need to get up twice a day (i.e. half as often) to throw water over your pigs. The rate at which you have to go dump water over the pigs (oh happy pigs!), which also equals the rate at which the local authorities must come refill your buckets, is analogous to the strength of the moisture-transporting atmospheric circulation.

Note that the above argument only shows that the rate of moisture exchange between the boundary layer and the free troposphere should decrease. This does not prove that the large scale circulation itself must decrease, for the moisture exchange consists of a small scale convective mass flux as well as a portion due to the large scale circulation. It is only the combination of the two that must become more sluggish. Even if the net rate of moisture exchange were to remain fixed, the Walker circulation could still become stronger or weaker, if the circulation reorganized itself to put less or more of the exchange in the form of small scale convective motions. The factors governing this partitioning remain to be elucidated.

3. Caveats and other viewpoints

The biggest caveat leaps out at the reader upon examining the upper panel of Figure 3 in Vecchi et al. In this figure it is evident that the observed trend in pressure gradient is almost entirely due to a precipitous drop in the late 1970s , which persisted through most of the 1990s. This shift coincides with an apparent 1976-77 climate shift in the character of ENSO, the attribution of which to global warming has been much debated. It’s not evident why the smooth trend in 20th century climate forcing should give rise to such an abrupt shift, and indeed the individual members of the model ensemble do not show a clearly analogous shift. Comparison of individual model runs with the observations is further complicated by the large decadal variability of the simulations. It remains disconcerting that the whole trend appears to rest on a meager handful of anomalous individual El Niño events. On the other hand, there is no reason to believe that the Walker circulation should change smoothly as a function of climate forcings; perhaps the potential for change builds up over many years, and manifests itself all of a sudden, in the fashion of an avalanche. Backing up this speculation with equations is a challenge for the future. Another decade or two of data will greatly clarify the situation.

Reconciling the picture in Vecchi et al with other analyses of climate change in the late 20th century also poses some difficulties. In particular Cane et al (Cane, M. A. et al. (1997). "Twentieth-Century Sea Surface Temperature Trends." Science 275: 957-960) have suggested that the upwelling of cold water in the Eastern Pacific provides a kind of thermostat which keeps the Eastern waters from warming as much as the Western warm pool waters. (Their result must be treated with some caution, since it doesn’t enforce the top of atmosphere balance, and should disappear in the long term after the water tapped for upwelling begins to warm; still the idea has a lot of merit in the transient warming situation we are now in.). This suggests an intensification of east-west sea surface temperature gradients, which ordinarily ought to yield a strengthening in the Walker circulation. In fact, Cane et al (1997) argue that the tendency toward increased SST gradient is precisely what is seen if one uses a robust trend analysis to decrease sensitivity of the trend analysis to outliers such as the very large 1982/1983 El Nino event (this event, and the equally large 1997/1998 El Nino event, greatly influence the estimate of a weakening trend of the Walker circulation in Vecchi et al). Hoerling and Kumar (Hoerling, M. and A. Kumar (2003). "The Perfect Ocean for Drought." Science, 299: 691-694 ) find suggestions of a similar pattern to that argued by Cane et al. Then too, it should be remembered that Vecchi et al present results only for one model, whereas there is evidence that the anthropogenic changes in tropical circulation can be model-dependent (Collins et al (2001), Climate Dynamics, 17: 61-81).

So, this paper probably shouldn’t be seen as a smoking gun for global warming, nor should confirmation of the results in the paper be seen as a crucial test for global warming theory in general. For the most part, the somewhat speculative nature of the results are in the nature of the data itself, and will only be resolved by another decade of observations. Nonetheless, Vecchi et al provides an important harbinger of what may be in store for the Tropics, reminding us once more that there is more to climate than just temperature and precipitation. We will be watching future trends in the Walker circulation with bated breath.

55 Responses to “On a Weakening of the Walker Circulation”

  1. 1
    Ian K says:

    Are we supposed to be able to click on the charts to enlarge them? Doesn’t work for me!

    [Response: No, I thought the chart was pretty legible so I didn’t bother with that. I just went in and made the plot a little bigger. –raypierre]

  2. 2
    Michael Tobis says:

    In the 1st paragraph of section 2, last sentence begins

    “However, the relatively weak *decrease* in precipitation with temperature seen in general circulation models…”

    Should this read “relatively weak *increase*” ?

    Thanks for a nice presentation. The porcine model was memorable.


    [Response: You’re right. It’s now fixed. –raypierre]

  3. 3
    Isaac Held says:

    Ray and Rasmus:

    Minghua Zhang and Hua Song, SUNY Stony Brook, have a nice complementary paper coming out on the same subject in GRL very soon. Minghua has made a preprint available at

    There is one point that I think worth emphasizing for starters, before thinking about the counterintuitive reduction in strength of tropical convection. It is often stated, as a cause for concern, that the “strength of the hydrological cycle” will increase as the Earth warms. If what we mean by this term is the global mean precipitation (or evaporation), then models agree that this response is going to be quite weak, as you say. The majority of the AR4 models generate less than 2% increase per degree K global mean warming. This is just a model result, of course, but one that the models happen to agree on, more or less. On this basis, an increase in the “strength of the hydrological cycle” is not something that should rank high on our list of concerns, I suspect.

    Global mean precipitation is of less interest than global mean temperature in general because we expect temperatures to warm nearly everywhere while we expect precipitation to increase in some regions and decrease in others. Fractional changes in local precipitation are expected to be larger than those in the global mean. Here is a simple picture that I think is worth keeping in mind:

    The atmosphere is continually transporting water horizontally from relatively arid regions (the subtropics in particular) to relatively wet regions (especially subpolar latitudes as well as tropical convergence zones and areas of monsoonal rains). Now suppose that atmospheric circulations don’t change at all and that humidity in the lower troposphere increases, following Clausius-Clapeyron. The atmospheric flux of water vapor and the associated convergence and divergence increase in amplitude. So wet region tend to get wetter and dry region drier. One shouldn’t take this picture too literally as a prediction for any particular region, but it help one understand why the signal in rainfall will be positive in some region and negative in others, with the pattern tending to increase gradients.

    To confuse matters, this increased horizontal transport is at times what is meant when one see reference to “an increase in the strength of the hydrological cycle”. Unlike the relatively small change in global mean precipitation, increased horizontal transport and increasing rainfall gradients are very much a cause for concern.

  4. 4
    Ian K says:

    Ray, could you answer some relevant basic questions before I try to digest this article? There are certain basics of the absorption of heat by the ocean which are unclear to me and which seem relevant here. A skeptic’s website avers that, paradoxically, water is not warmed by infrared radiation because it is such a good absorber(!) ie this radiation is completely absorbed by a top skin layer. This leads to re-radiation and evaporation rather than heating of the bulk of the water! Perhaps he has a point, although commonsense suggests to me that if I have two, open-top, insulated, pans of water, in darkened rooms of, say, 10 and 20 C, then that in the warmer room will warm up further and faster.

    A corollary of what he is saying is that the sea is basically heated by visible light absorption rather than by heat radiation and is pretty impervious to global warming!

    [Response: This line of argument is total nonsense, and your intuition about what happens to a can of water in a warmer vs. colder room is correct. While it is true that infrared is absorbed in a thin skin at the top of the water, even if the water were completely quiescent this would still lead to the skin layer heating up until emission (plus evaporation and all the other terms we include in the surface budget) equalled the energy input. By the way, if the fluid were quiescent, the solar heating which penetrates the water would cause the water to boil (or close to it) since the heat that enters would have a hard time getting out by diffusion, which is slow. However, the top 50-100 m of ocean is well stirred by turbulence, so energy dumped into the top skin of the ocean gets mixed downward quite rapidly

    An addendum to the above remarks is that the skeptics’ site you refer to is probably trying to imply that the greenhouse effect couldn’t increase at all, if infrared can’t heat the ocean. Aside from the fact that infrared can heat the ocean, the argument is misconceived on another level. As explained in our article “A Busy Week for Water Vapor,” the anthropogenic increases in greenhouse effect do not heat the surface primarily by directly increasing downward infrared radiation at the surface. Primarily, they affect what’s going on higher up in the atmosphere, which warms the whole troposphere (which is yoked together by convection so it tends to warm and cool as a unit). 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. The downward radiation to the surface increases mainly because the low level air temperature increases, not because of the direct effect of increased CO2 in the air. –raypierre ]

    A further point where I need clarification is that, in Part I, you seemed to be suggesting that the West Pacific warm pool develops due to the trade winds blowing surface water in that direction. Could other mechanisms contribute to the warm pool, such as differences in cloud cover? I have also read in a reputable book that one of the reasons the warm pool is higher is merely due to the fact that warm water occupies more room, ie for hydrostatic equilibrium to apply the relatively deep warm pool should be higher than the cooler water to the East. How significant is that effect?

    Thanks for the article. I am sorry if I am not yet up to speed on this. Feel free to edit or reduce the above if it is too off topic. You can even eliminate it completely or answer such issues at a later time if you want. The larger graphs are also appreciated

    [Response: It’s not so much that the warmer water is pushed to the warm pool, as that the wind pattern thickens the warm layer, making it harder for upwelling to tap into the colder deeper waters. The contribution to sea surface height by thermal expansion is significant, but doesn’t play a very big role in determining the temperature of the warm pool. In many ocean models, the free surface effects are neglected and replaced by a rigid lid, yet they still get a warm pool. There are other mechanisms affecting the temperature of the warm pool, however, like variations in evaporation rate. Clouds could have an effect on temperature, but in the present climate they don’t do much, because the cloud greenhouse effect just about cancels out the cloud shading (cooling) effect. The clouds affect precipitation and the strength of the Walker circulation, and may indirectly affect the warm pool configuration via that influence. –raypierre]

  5. 5
    C. W. Magee says:

    The cool Eastern Pacific is sunnier and drier than the warm Western Pacific. The Galapagos islands are a desert, while New Guinea, Indonesia, and the Solomon Islands are rainforest.
    If you look at pictures of Earth from space that show South America (e.g. the Gallileo flyby pics), the Amazon is generally covered in clouds, while the Atacama and eastern Pacific are perfectly clear.

    To what extent does the disequilibrium between atmospheric warming and oceanic warming suppress rainfall? I’m just a chemist who only understood half your post, but it seems obvious that if you raise atmospheric temperatures faster than you raise oceanic temperatures, relative humidity will fall. So rapid warming ought to be drier than gradual warming, and rapid cooling ought to be wetter than slow cooling.

    Also, if radiative limits are preventing tropical precipitation, wouldn’t that just increase the height of the convection cell, if it can’t radiate heat as efficiently? I seem to recall last year’s Atlantic hurricane season reported a lot of unusually cold cloud top temperatures, even in the “smaller” storms. After all, if a -50C cloud top can’t radiate due to CO2 absorption, what choice does it have other than to keep rising until adiabatic cooling drops it out of the CO2 absoption band?

    Finally, if temperature only slightly increases rainfall, then why does paleoclimactic data indicate more widespread rainforests in pre-glacial cenozoic?

    p.s. Your Clausius-Clapeyron link is dead. Here’s a good link to the hard-rock version:
    Just don’t ask me how this applies to the atmosphere…

    [Response: I tried our own Clausius-Clapeyron link just now (in Safari) and it seems to work fine. Clausius-Clapeyron applies to water vapor in the atmosphere just as it would apply to any other phase transition. For any given temperature, it tells you how high you can make the partial pressure of water vapor before the vapor starts to condense into liquid or ice. The specific humidity (i.e. “concentration” of water) is proportional to the ratio of partial pressure of water vapor to total atmospheric pressure. As it gets warmer, a kilogram of air can hold more grams of water before condensation starts to occur.

    Regarding changes in rainforest area, keep in mind Isaac’s point that local precipitation can go up a lot while global precipitation is increasing just a bit. Further, precipitation over land is a small fraction of the total, so there’s a lot of room for changes in precip there without altering the result on the global mean. Additionally, while some areas in the warmer parts of the cenozoic were more forested (e.g. tundra turns to forest), I wouldn’t describe too much of this as “rain forest,” and I haven’t seen firm data that the actual rainforest area increased. There were a lot of dry New Mexico type land areas, and the dinos seemed to have liked these. The steamy Cretaceous murals you see in the Field Museum and similar places are not really faithful to paleoclimate reconstructions. I’ll comment a bit later on your question regarding the effect of disequilibrium. –raypierre]

    [Response:A few more responses concerning the interesting issues you raise. First, concerning disequilibrium, it has potential effects, but they are more subtle that the one you suggested. Raising the atmospheric temperature doesn’t generally lower the relative humidity, since the boundary layer is in good contact with the moisture source at the ocean surface, and can easily maintain near-saturation. The disequilibrium referred to comes from the fact that the ocean has a lot of thermal inertia and takes a long time to warm up, whereas the atmosphere has a short response time and quickly comes into equilibrium with any given ocean temperature, corresponding to the current amount of greenhouse gases. The implication is that when you dump more GHG in the atmosphere but don’t give the ocean time to warm up, then the atmosphere needs to warm up until the sum of the energy lost to space and the energy lost to the ocean surface comes back into balance. This warming is less than it will ultimately be, because the cool ocean surface holds back the warming — allowing more energy loss out the bottom than will ultimately be the case. I think this disequilbrium mostly shows up in an increase in the air-sea temperature difference, but lots of other aspects of the surface energy budget will be affected as well, and all of these can affect the amount of evaporation. Held and Soden’s paper (linked in the article) do document some changes in the moisture transport that differ between equilibrium and transient runs, but these are affected by circulation changes and it’s not clear that one can get at them through simple thermodynamic arguments. This is a very good topic for future research.

    Now, as to what the precipitation limitation does to convection, first note that I don’t entirely buy the argument for the inevitability of the radiative constraint, but in any event let’s think about what happens when you warm the atmosphere but don’t let it precipitate more. The main constraint (and I’d claim it is a diagnostic one without a whole lot of predictive value) is that somehow or other the increase in infrared radiative cooling of the troposphere must be balanced by the increase in latent heat release (precipitation), since that is the dominant balance in the tropics where most of the rain takes place. The problem is that there are a whole passel of ways this balance can be maintained. Increasing the height of the convection cell doesn’t generally help the planet radiate away heat, since the higher the tropospause (loosely the height of convection) goes, the colder it gets, inhibiting radiation. If you were in a situation where there was initially more precipitation than radiative cooling could handle, then the atmosphere could just warm up until the radiative cooling increased — though then you’d have to worry about how much the warming affects precipitation, etc. In some sense, the possibility of convection (which is responsible for a lot of the rainfall) is maintained by radiative cooling, which allows the atmosphere to get cool enough relative to the surface to allow convection. If there isn’t enough cooling, then convection can’t be maintained, and the convection and rainfall can go down. Still, it’s not easy to sort this out by words alone; you need equations. It’s interesting enough that I’m thinking of putting together a simple paper contrasting the way all this works in radiative convection models (no large scale dynamics) vs. models with a simple Walker circulation. –raypierre]

  6. 6

    Re 4 “A corollary of what he is saying is that the sea is basically heated by visible light absorption rather than by heat radiation and is pretty impervious to global warming!”

    Visible light causes evaporation (and so cooling!) It is the heat radiation which is warming the oceans and causing the global intensification of the hurricanes.

    Kirchhoff’s Law states that for heat radiation, absorption equals emission at equilibrium. Thus if the downward radiation of heat to the Earth’s surface increases, then the ocean temperature will rise until it is emitting radiation at the same intensity. This is a slow process because the layer of warm water that needs to be heated is about 100 meters thick.

    [Response: It’s not true that its only the heat radiation that causes warming. The energy inputs to the surface are from absorption of solar radiation and absorption of downward infrared radiation. In equilibrium these would be balanced by upward transfer of infrared radiation emitted by the surface, by sensible heat flux (warm air carried upward) and by latent heat flux (i.e. evaporation — moisture carried upward). While the energy input allows more evaporation to occur, it’s confusing the matter to say that this “cools” the surface, since the evaporation is just helping to carry away the heat put in by the other terms. The picture I gave neglects the effect of ocean dynamics — cooling by upwelled water entering the mixed layer and warming by imported warm water from the side. –raypierre]

  7. 7
    Ileana Blade says:

    Thanks for the paper commentary and the comparison with Cane et al paper. I agree the caveat about the trend is pretty obvious. It seems kind of a stretch for the authors to describe the 1976 drop as an “amplification in the trend”, when in fact the trend seems to be close to zero before then. Also, the trend seems to have decreased in recent years, and again that is referred to as a “disappearance” in the amplification of the trend …

    On another note, I was wondering how valid it is to assess the significance of an observed trend by comparing it to “natural” trends in unforced GCM simulations, particularly when the GCM’s internal variability is not that good. Would that trend be significant “per se” if robust trend analysis was applied?

    “it should be remembered that Vecchi et al present results only for one model”

    Actually, Vecchi et al. do use more than one GCM and state that the GFDL results are matched in “most” of the remaining IPCC GCM models (unfortunately, the supplementary Nature jpeg figures did not work for me). The GFDL model does happen to show the largest trend (and yet, as you point out, the SLP gradient in recent years does not exactly mimic the observations …).

    ps. The link to the upcoming Zhang and Song paper given (in Isaac’s comment) is incorrect. Should be:

    [Response: Thanks for the additional information. I had somehow overlooked that the supplementary material included information about the other AR4 models, in Supplementary Figure 4] However, looking at that figure, it’s not particularly clear that most of the AR4 models show a trend toward weakening. In fact there are only a very few for which the trend lies outside the confidence limits for the period under study. The figure doesn’t alter our statement that models differ greatly in tropical response. One particular thing to look at is the extent to which the different models exhibit a Clement/Cane type limitation in the Eastern Pacific. Clouds can also affect the strength of the Walker circulation. By the way, in the copy of Supplementary Figure 4 I downloaded, the identification of the individual models for each bar in the graph seems to be missing. Does it show up for others, or is the information perhaps found somewhere else? –raypierre]

  8. 8
    pat neuman says:

    Seasonal changes in precipitation need special attention in the mid latitudes because of the importance to the growing season and the portion of precipitation which runs off. At the Langdon experimental climate station (northeast ND), the sum of May and June precipitation increased 2.3 inches (linear regression 1906-2005) while annual precipitation increased only 1.3 inches. The elevation of Devils Lake increased 26 ft from 1993 to current. For more, see:

  9. 9
    Alexander Harvey says:

    Re #4

    Water is a good absorber of IR and a poor absorber of visible light.

    But I think it is a poor reflector in both cases.

    The albedo of sea water is low ~.1 (please check). Most of the visible and IR are absorbed but the visible penetrates deeper.

    I think that some confusion may exist about emission following absorption of IR (well for me at least). As I understand it, once the energy has been absorbed, there is a race between conversion to heat (by molecular collisions) and spontaneous emission from the excited state. For H2O in a very low pressure gas the collision rates are low and emission wins. At atmospheric pressures and as a liquid conversion to heat by collision wins.

    In the second case the emission rate is mostly that due to the temperature of the water (or water vapour). That is, it radiates because it is hot not because it is excited. In that case it does not matter how the water is heated but simply the temperature of the layer of surface water down to a few multiples of the inverse of the IR absorption coefficient which is I think varies from around a few cm to less than 1mm with increasing wavelength.

    I think my contribution has taken so long that it has been overtaken by others but I will post anyway.

    [Response: Actually, there isn’t much difference between the absorption/emission process for vapor in the atmosphere and liquid water. Both are in thermodynamic equilibrium at each individual wavelength, much as Kirchoff describes, and there’s not much business about excitation and delayed emission going on (as there is in phosphorescence and luminescence). The main difference between the liquid ocean and the water vapor in the atmosphere is that condensed water is a much better infrared absorber across the spectrum than water vapor (due to collisional effects), and that the entire atmosphere contains only about as much water as a few centimeters of ocean, so absorption is spread out over a bigger distance in the atmosphere. –raypierre]

  10. 10
    Tom Fiddaman says:

    This is OT, but the Al Gore’s Movie thread is closed. The WSJ published an editorial on the movie, which trots out the usual claptrap (CO2 saves us from impending ice age) as well as a new 4th level of skepticism: “we’re all dead in the long run”:

    Warmed Over
    May 31, 2006; Page A13

    …a valid service is performed in satisfying the eternal human appetite for gloom and doom (and no virgins were sacrificed), distracting people from the reality of life, which is that we all are doomed, while the universe, the earth and all that environmentalists hold dear will go remorselessly on and on without us.

    In a million years, the time it takes the earth to sneeze, the planet will likely be shorn of any conspicuous sign we were ever here, let alone careless with our CO2, dioxins, etc. Talk about an inconvenient truth.

    How much more securing, in a way, to believe we are ruining the planet than the planet just does not care about us, and will run rampant with life long after we are dust. And how pleasant to be able to transmute our fury over our fate into incoherent feelings of self-heroism against our present “enemies.” …

    [Response: Dear Gore fans: I have re-opened the discussion on the Gore movie. I closed it prematurely, neglecting the likelihood that there would be a flurry of new interest once the movie opened in more theaters. –raypierre]

  11. 11


    Is this not in contradiction with the findings of Wielicki and Chen, that the Hadley and Walker circulations actually increased in strength over the past decade(s)?

    From the abstract of Chen e.a.:

    Similar analyses of upper tropospheric humidity, cloud amount, surface air temperature, and vertical velocity confirm that these flux changes are associated with a decadal-time-scale strengthening of the tropical Hadley and Walker circulations. Equatorial convective regions have intensified in upward motion and moistened, while both the equatorial and subtropical subsidence regions have become drier and less cloudy.

    The model may be right over the full 131 year period, but in this case doesn’t reflect natural cycles including El Nino and longer cycles (as is the case for ocean warming, where models – significantly – don’t reflect any cycle with a length between 10-100 years).

    [Response: If you look at the time series, the rise seen in the Chen and Wielicki papers is clearly there in the SLP, but is a reflection of short time variability (and comes from a very low base). -gavin]

  12. 12
    Isaac Held says:

    Ray (I can’t seem to call you Raypierre) and Rasmus:

    [Response: No problem. They only turned me into “raypierre” at RC to distinguish me from Ray Bradley. It’s pretty easy to tell which of us is who by context. –ray(pierre)]

    It is quite common to see the claim that the atmosphere will become more energetic as it warms, without any attempt at justification, as if this were self-evident. It is not self-evident. I liked the pig model, but here is an alternative:

    The fans at Wrigley field pile out of the stadium after another Cub loss and try to get on the El to get home. They file into the cars and are eventually deposited at their stops at this predetermined rate. But ominous signs have appeared of a trend towards increasing weight of the average Cub fan. What effect will this have on the frequency of the El and the rate of circulation of fans out of the stadium and to their homes?

    Flash forward 50 years. Still no World Series championships, but the size of the average Cub fan has increased by 10%! The frequency of the trains has increased a bit, but has not kept up with this unfortunate weight gain. Indeed, the latter was simply ignored during the political infighting that determined how much money (not much) went into financing upgrades to the El. At first, out of habit, the same number of fans try to cram into each car, but the pressure is too great and some are pushed out. Eventually the system equilibrates, but with a slower rate of circulation of Cub fans through the El and to their homes.

    [Response:Very nice analogy! It’s much more believable than the pigs model, which (I have to admit) was a bit contrived from the standpoint of actual animal husbandry. –raypierre]

  13. 13
    Jeffrey Davis says:

    re 12: Another Pig Substitution

    The Bulls make money and The Bears make money, but The Cubs are led to slaughter.

  14. 14
    Ian K says:

    Thanks Ray and all (especially Alexander’s query because I had a similar uncertainty about the interplay of absorption line energy and thermal energy.) Is there a not too mathematical book, which explains these basics so that we don’t have to bother you with such stuff? Thanks also Ray for your earlier discussion of the Philipona paper which helped me with understanding the importance of top-of-atmosphere balances and was also instructive on how important exchanges between scientists can be!

    I am still, however, not clear what is the mechanism for warming of still water under still air when subjected to only infrared radiation.(In real life I understand that mixing is the main agent of deeper warming in the ocean due to winds, currents, etc.) Only the top skin of water heats up and therefore lower warming must be by diffusion, or are convection cells within the water inevitable?

    Another point: could the slowdown in moisture exchange with temperature increase be due to the fact that there are still only the same numbers of nuclei for rain drops to form around? Or don’t the models include this feature?

    Final point: how does the slowdown postulated gibe with hurricane intensification?

    [Response: There’s no need to think about an example as unrealistic and contrived as completely still water. The same issues that are bugging you (or your skeptical acquaintance) are perfectly well illustrated, indeed better illustrated, by the heating of bare, dry rock from solar radiation and from infrared coming down from the atmosphere. Rock doesn’t move during the short time scales needed for the surface to come into equilibrium, so the only vertical heat transport is by diffusion. At the same time, rock is very opaque to both infrared and solar radiation; the radiative heating is deposited only in the top few microns of the rock. To determine the temperature of that top few microns you use the fact that (by an application of Kirchoff’s law) the infrared emission from the surface comes from the same few microns in which the infrared is absorbed). Hence, the rate of infrared energy loss from the surface is determined by the temperature of those top few microns — making the idealization that the only loss mechanism is infrared, the top few microns will nearly instantaneously heat up until the temperature is high enough for the infrared radiation to balance the absorbed solar radiation. Now the slow diffusion processes come into play: heat diffuses from the skin layer downward, and over a long period of time, the entire body of rock becomes the same as the surface temperature. Once the heated layer becomes more than a few centimeters thick, the heat loss of the skin layer due to downward conduction of heat by diffusion stops having any significant effect on the surface temperature, since rock is such a good insulator that the heat flux by conduction in rock is tiny compared to the heat loss by infrared radiation out the top. In this way, after a relatively short time, you can determine the surface temperature by energy balance without knowing much about the details of what is going on deeper down. On the real Earth, temperature actually increases with depth rather than relaxing to the global mean annual mean surface temperature, because of diffusion of heat upward from the deep hot layers. This diffusion has essentially no effect on surface temperature, because the heat flux is tiny compared to the other terms in the surface budget (solar, infrared and turbulent heat fluxes).

    On the matter of the role of condensation nuclei, a few general circulation models do have some crude representation of nucleation microphysics in their convection or cloud schemes, but it certainly isn’t the key factor in the weak increase of precipitation with temperature, which is seen in all GCM’s including those with very basic representations of convection. As of the previous IPCC report (the TAR) it was not possible to find any clear influence of parameterized microphysics on the general circulation, but that might have been because of inadequate vertical resolution or the crude manner in which microphysics was parameterized. I’m not aware that the situation has gotten any clearer since, but I haven’t read through this section of the AR4 IPCC report (the current one) yet. I could see nucleation issues affecting the precipitation efficiency, since they determine (in part) how much rain falls near the convective tower vs. how much falls through drier air farther away, evaporating on the way down. Figuring how this would influence the global mean precipitation rate is well beyond what I can do in my head.

    As for your question about hurricanes, the argument given for the global mean hydrological cycle doesn’t apply to the hurricane because the global mean argument assumes an equilibrium between radiative cooling and latent heat release. That applies (nearly) for the atmosphere as a whole, which is in equilibrium on time scales of around a month, but it doesn’t apply to a local precipitation system like a hurricane or even a thunderstorm. These phenomena are far out of equilibrium, the hurricane living by transeferring energy stored in anomalously warm ocean water into the atmosphere. A gentle global increase in precipitation with warming can be made up by localized systems becoming more intense while vast other areas get drier and more sluggish. As Isaac says, global mean precipitation is a less useful summary statistic than global mean temperature, if you are interested in what life will be like in a doubled CO2 world. –raypierre]

  15. 15
    Alexander Harvey says:

    I was going to add a little more to my post but having read Ian’s (#14) I think that it would be better to use this opportunity to second his thoughts.

    For some of us, it would be wonderful, to have some of the basics explained and explored in a general but non-trivial way. It seems to me that the basic science is often not obvious and sometimes counter-intuitive.

    It is quite easy to get caught out and find oneself believing nonsense because it seems just so reasonable.

    This I think is true of IR absorption/emission. I won’t get into this here because it is not material to this thread. Which is also one of Ian’s points. We have questions that do not always hang well on specific climatic topic.

    I have considered trying to put together my thoughts and making them available on the web but I am not an academic and can not pass myself of as one. Material needs to be reviewed, corrections made and commentaries added or it is just so much noise.

    I have to say that I am distressed by the “apparent” state of the science, in that, it feels like more heat than light is being produced. It seems that you are operating in a noisy environment which is not at all helpful for the trickle down of authoritative information.

    If there any other sites, like this one, that seek to aid the process then I would be grateful for a link.

    My personal preference is for a better understanding of the hard (as in non-controversial) aspects of the science. You may think that these are so obvious, and so firmly put to bed, that there is little need for further elaboration, I would say that I know of no resource on the Internet that adequately provides a solid primer in the basics that also stretches the reader. By solid I mean, grounded in scientific principles not analogies. “Greenhouse” is a case in question.

    I could supply a list of topics that perplex me, if anyone is interested, but I feel that there should be a better forum for what would be, in effect, an essay on my ignorance.

    Many thanks for this website.

    [Response: For a not-too mathematical introduction, but one which still focuses on the basic physics in a quantitative way, I recommend Dave Archer’s new book, coming out soon from Blackwell. I think that a draft is still available on David’s web site, though you’ll probably want to buy a real copy once it’s out. That gives you some idea of the basic nature of infrared effects of greenhouse gases, but it still won’t address some of the subtler points concerning radiative transfer and the effect of band saturation — except insofar as band saturation is used in the explanation of why the radiative effect of CO2 is roughly logarithmic. In my own book (coming out eventually from Cambridge), I go into these things at a more mathematical level — less dense than Goody and Yung, but probably still more technical than you are looking for. The fallacy that band saturation precludes an anthropogenic greenhouse effect is tied in with the fallacy that the greehouse effect works through the direct effect of CO2 on downward infrared. I addressed that in my post “A Busy Week for Water Vapor.” There are other flavors of the band-saturation fallacy floating around, but they don’t get very wide play and have taken in rather few people. In writing books, its always a question whether one should tell people what’s wrong with certain arguments that few people have either heard of or subscribe to. That could just lead to more confusion. Most of us prefer to concentrate on explaining the correct view of things in as clear a way as we can. It would be useful to me if you posted a brief list of basic things which ought to be explained but for which there isn’t any readily accesible explanation for the scientifically literate layperson. –raypierre]

  16. 16
    George A. Gonzalez says:

    I brought up this point in an earlier discussion on how the heating of the planet is predicted to dramatically slow the ability of the ocean to absorb carbon dioxide. The result would be substantial increases in atmospheric carbon dioxide, and the likelihood of run away warming. With this prediction, and the trend toward the slowing down of the jet stream, is it time to consider/advocate geoengineering as a response to global warming? Is this our only chance at survival as a species? The evidence already indicates that too much warming is already built into the global biosphere in order to expect that emissions abatement would avert a global warming crisis. Not that abatement is on order in the near future. By geoengineering I am referring to such schemes as screens in the earth’s orbit to regulate the amount of sunlight reaching the biosphere, or the injecting of aerosols into the stratosphere to deflect sunlight out to space.

  17. 17
    Ike Solem says:


    Even if we’ve had some scientific training, these issues are very complex! Here is my attempt at understanding just one of the paragraphs in this well-written article:

    “Because latent heat release in the course of precipitation must be balanced in the global mean by infrared radiative cooling of the troposphere (over time scales at which the atmosphere is approximately in equilibrium), it is sometimes argued that radiative constraints limit the rate at which precipitation can increase in response to increasing CO2. This argument is stated, for example, in Allen and Ingram.and repeated in Vecchi et al.”

    So, as water condenses around nuclei in the atmosphere, heat is released as the vapor->liquid transition occurs. This is accessible to one’s personal experience – a warm burst as the first raindrops hit.

    The radiative balance over equilibrium timescales – the heat released by raindrop formation will locally warm the atmosphere, but it takes time for the atmospheric circulation to average this out. I beleive this relates to the points that Isaac Held (#3) raises with regard to local vs global precipitation patterns.

    Then, the radiative balance limits the rate at which precipitation can increase – a question of rates, which is always going to be more complicated then a question of quantities. We could include ‘on a global basis, not on a local basis’. So a local spike in precipitation releases a lot of heat – but as the heat increases, this negatively affects the vapor->water transition (precipitation, or raindrop formation), since warm air holds more water then cool air – and so the limit on precipitation vis-a-vis the radiative balance of the atmosphere appears.

    I hope I’ve worked my way through that statement correctly – it reminds me of the introduction to a Science or Nature article – condensed information (It helps to know that such articles are severely restricted in terms of length, so once sentence can carry a large amount of information).

    I think if you work through the whole article in this fashion, you will be envious of those pigs and their cold water baths…

  18. 18
    Grant says:

    Re: #16

    I’m very skeptical of the wisdom of geoengineering. We’ve already screwed up the planet; I doubt that it’s wise for use to attempt to alter it.

    I’m particularly concerned about the danger of injecting aerosols into the atmosphere. The possible unforseen consequences could easily make the “cure” worse than the disease.

  19. 19
    Ian K says:

    Thanks again Ray. Your example of rock heating is informative. However I still have a problem with water. I can visualise evaporation fighting diffusion and winning easily. I have no way of knowing who wins that fight;)

    How to convey basic physics relevant to climate change?. I had a quick look at the draft of your book a while ago and have just had a look at David’s. I don’t think they (yet) cover the issues at the level that suits me.

    Saturation and/or overlapping of absorption lines of CO2 and water. Spencer Weart’s (History of) The Carbon Dioxide Greenhouse Effect ( includes interesting snippets on this. I would have thought that the “drama” of the historical, intellectual obstacles to the recognition of our warming the planet would appeal to students and the public.

    I have found this book useful: “Is the Temperature Rising? The uncertain science of global warming”, (my version written in 1998), published by Princeton University Press and authored by S George Philander. Although it tackles global warming somewhat sideways it is a lucid, non-mathematical primer on weather and climate.
    As an aside: it has a graph (page 51), which shows a jagged curve representing the spectrum of radiation from the Earth as seen from a satellite. Over this is superimposed a set of smooth curves of ideal blackbody radiation, labeled with temperatures. The line just above the true spectrum is labeled 27C, (the temperature of warm and humid Guam, over which the satellite was sailing).

    What I found interesting is that the spectrum, although lowered somewhat by water vapour absorption, shows a significant dip due to CO2. This goes down past the -48C line and the author says “we can infer that carbon dioxide absorbs primarily at elevations where the temperature is approximately -50C”. This graph brings home to me the significance of CO2 and how it affects the top-of-atmosphere radiation budget. I have seen nothing so informative elsewhere.

    As I am trying to illustrate, information on more technical aspects of global warming is scattered. Global warming deniers however, as you have said, are obsessed by water vapour and good information is needed to show them how they are fooling themselves.

    If I were to give RealClimate a wishlist, I would like part of your site to be permanently open so that viewers could post technical questions. Other viewers could then post explanations and you guys need only intervene if the explanations were inadequate. This could be divided into categories, eg radiation, water vapour, oceans, wind, ice sheets, glaciers, model issues, etc, etc.

    It is rather hard to find such info by searching the ever-increasing number of threads on your site and it would aid in keeping threads shorter.

  20. 20
    Grant says:

    Re: #19

    I’ll second Ian K’s request for a section on RealClimate for readers to post questions — technical or otherwise. We definitely have enough scientists who are regulars here to handle most of the workload; a bit of monitoring by the moderators would still be required (a lot of us are scientists but not climate scientists).

    This might be an important resource to persuade the general public of the reality about global warming (climate crisis!). Before any of you protest that you’re “too busy” or have insufficient resources to shoulder yet another burden, I’ll remind you: we’re talking about the future of the planet and all its life.

    I’ll also mention that while RC is my relied-upon resource for climate information, I often hesitate to send others here because both the posts and the discussion can get rather technical. As for me, that’s exactly what I want. But for my somewhat skeptical in-laws …

  21. 21
    Hank Roberts says:

    Spaceships in fiction have been using refrigeration lasers to cool themselves for years. Here’s someone propsing something like that for the planet.

    Hmmm, if other species do this to themselves, we ought to be looking for extrasolar beacons in the frequencies these people propose to use, ones useful for getting rid of trapped heat on a planetary scale.

    “… The Earth Cooler/spl trade/ radiator has been designed to transmit electromagnetic energy in the infrared range for spectral bands (or windows) that naturally occur in the atmosphere and are essentially transparent to the traveling energy. Upwards of 140W/m/sup 2/ can be transmitted unattenuated through the atmosphere into deep space utilizing a simple panel at 290 K. A one-dimensional model is used to predict the temperature difference that exists between two sections of a uniquely designed panel surface that radiates thermal energy into space. The model includes all the meteorological parameters necessary to provide meaningful information for electromagnetic waves propagating through the atmosphere, and shows that a temperature difference of up to 4/spl deg/C can be produced depending on the moisture content of the surrounding air. The theoretical performance of the model coincides well with data collected from several prototypes. The paper discusses many of the technical issues involved with the propagation of infrared thermal energy through the atmosphere. The presentation of the physical laws of nature that govern the phenomenon provides a better understanding to those on both sides of the global warming issue.”

    [Response: Geoengineering involves much bigger technical challenges than carbon sequestration which in turn involves much bigger challenges than simply reducing emissions. As sombody already noted, both of the former involve much more intrusive big government and more fragile international agreements than the latter, which can be done largely through market forces. I don’t want to see this thread get hung up on geoengineering but the device Hank describes offers some nice opportunities for thinking about infrared radiative transfer and the greenhouse effect, so let’s all give it a go. –raypierre]

  22. 22
    George A. Gonzalez says:

    Re: #18

    I find it difficult to believe that there is a greater risk than global warming — which currently does not look like a risk but a certainty.

  23. 23
    George A. Gonzalez says:

    In response to Ray’s comment on carbon sequestration through market based programs: there is no evidence that market schemes will abate carbon dioxide emissions internationally. Moreover, it is questionable whether market-based approaches have ever worked to reduce airborne emissions. Additionally, the U.S. is not even on board with this rather meek and dubious approach to abating climate change emissions. Finally, carbon sequestration itself is highly perilous

  24. 24
    Ian K says:

    I have rummaged about the IEEE site, because the article might be useful for learning but it seems it is available only to subscribers. Surely this device is not really geoengineering? I have thought of using reflective plastic sheeting on my flat metal roof. This would cool the house in summer and also reflect nearly ALL the solar energy straight back to space. Seems a no-brainer to me compared to the black-tiled roofs that are now popular. And while I am at it, what about mandating that asphalt uses stones, fillers etc of high albedo? I have never seen these issues discussed. Surely they are worth a carbon credit or two.

    [Response: You are talking about something a lot simpler: increasing the visible wavelength albedo of your roof by making it whiter. That works because the atmosphere is quite transparent to visible radiation.What the IEEE abstract seems to be outlining is something more interesting and more complicated: turning broadband infrared radiation (much of which would be captured by the atmosphere) into narrowband radiation in infrared window channels that are fairly transparent to infrared. In essence it allows the surface to radiate to space as if the atmosphere were pure nitrogen, with no greenhouse effect. I’m just guessing, since I haven’t gotten my hands on the paper yet, but I’m pretty sure that’s what they’re getting at. –raypierre]

  25. 25

    There is very good information on this site. We all understand that it critical to understand what is going on. That is why we started the “Global Warming Forum.” We must be willing to share ideas and talk about the climate changes that are taking place.

  26. 26
    jhm says:

    Climate change call-in 9:30 am today on C-SPAN, with Mr. Dan Vergano from USA Today. FYI.

  27. 27
    Alexander Harvey says:

    I will give the selective radiator as described briefly in the IEEE abstract above a go. Hold on tight.

    I MUST STRESS that this is a exercise in thinking as prompted by Ray above. It is not a comment on the paper which I HAVE NOT READ.

    1) If the radiator is NOT transmitting in the atmospheric absorption bands then in those bands e(missivity) = 0 implying a(bsorptivity) = 0 and r(eflectivity) = 1.

    a = e; a + e + r = 1; (after Kirchoff).

    In the absorption bands (where there is a downward flux originating from the atmosphere) it will be a strong reflector elsewhere it would be a strong emitter which in combination would be similar in terms of upward flux to a dark grey IR radiator which is not dissimilar from many standard surfaces.

    2) A difference arises from any thermal gradient (thermal lapse) above the radiator which normally gives rise to a net flux from the radiator in the absorption bands. Acting as a mirror there is no net flux so there is a deficit to the atmosphere compared to a standard radiator. On the face of it, this is precisely the opposite of what is intended.

    3) At the effective radiative height (near the tropopause) where the atmosphere radiates in the absorption bands into space, (most/much) of the ground radiated IR in the absorption bands has been extinguished, the atmosphere is radiating IR largely due to its temperature (~220C). In order for the amount of radiated IR (in the absorption bands) to increase significantly then the temperature at this height needs to change significantly. This temperature is controlled in some complex way that gives rise to the lapse rate and it is not clear how a selective radiator, that would tend to reduce IR in the absorption bands, would achieve this.

    4) I can not see why this would be any better or as good as a radiator with the highest achievable broadband IR emissivity.

    If the surface of the earth was covered in just this sort of selective radiator and the atmosphere had zero absorptivity (no greenhouse gasses) then it would seem that we would have zero emissions in the current absorption bands (which is less than is currently the case) and the world would get hotter despite the lack of greenhouse gasses due to ground surface temperature rise (less efficient radiation) and conduction of heat to the atmosphere (no longer any IR heating) despite the lack of greenhouse gasses. Adding greenhouse gasses to this scenario would tend to increase the totally emissivity of the world and hance be a source of cooling not heating.

    Well that is what I might guess from the abstract, no guarantee that any of it is right.

  28. 28
    hunter says:

    Solar heating has been going on a long time, yet the deep oceans seem to be uniform in temperature, so the solar energy is dealt with at much shallower levels. I can’t see how ocean temps could be mined for chronologically useful information, unlike ice cores. But perhaps in the sediment layers there would be some evidence that could be extrapolated for something useful. It also seems that if an energy level was piled up over oceans receiving more heating, they would, unlike Cubs fans, find new ways to transport energy to an area of lower energy. In other words, either the current transport system will gain capacity, or new ones will develop.

  29. 29
    Eric says:

    You said: “Taking the slow increase of precipitation with temperature as a given, the more rapid increase of boundary layer humidity implies that the rate of transport of moisture from the boundary layer to higher levels where it rains out must go down.”
    Wouldn’t the rate of upward moisture transport depend primarily on the distribution of convection? My understanding is that concentrated convection produces a drier upper troposphere. How do you account for concentrated convection as you outlined in #14 with a moister upper troposphere that would allow more warming?

    [Response: It’s just a diagnostic statement. It doesn’t say how the reduction in mass exchange is achieved. It could be through a reduction of convection, and there are any number of factors that could come into play to reduce convection. It could also, on the other hand, be achieved through a weakening of the large scale circulation. –raypierre]

  30. 30
    Gabriel Vecchi says:

    Ray, nice write up on our paper. The atmospheric and oceanic changes we mention in the paper, in particular a shoaling of the zonal-mean equatorial Pacific thermocline, are consistent with the results of Mike McPhaden and Dongxiao Zhang (2002, “Slowdown of the meridional overturning circulation in the upper Pacific Ocean”, Nature, v.415, 603-608; and Zhang and McPhaden, 2006: “Decadal Variability of the shallow Pacific meridional overturning circulation: Relation to tropical sea surface temperatures in observations and climate change models”, to appear in Ocean Modelling.)

    We also have a few comments:

    1) Regarding the 1970s shift, Ray mentions that: “It’s not evident why the smooth trend in 20th century climate forcing should give rise to such an abrupt shift, and indeed the individual members of the model ensemble do not show a clearly analogous shift.” Actually, our Supp Fig. 1 shows similar shifts in ensemble members 3 & 4. (we are trying to get the supplementary material available, it should be available on Nature’s website).

    Also, for what it’s worth, since the paper came out we got access to the latest version of the Hadley Centre’s SLP product (HadSLP v2.0), which we did not analyze in the letter to Nature. This updated dataset includes more data sources than the HadSLP v1.0 and is updated to April 2006, this dataset is documented in an upcoming J. Climate manuscript (Allan, R. and T. Ansell: A new globally-complete monthly historical gridded mean sea level pressure data set (HadSLP2): 1850-2004. J. Climate, in press). The DSLPA index computed from HadSLP2 shows a much more “trend-like” reduction than the datasets shown in the manuscript, in which the 1970s shift plays a less pivotal role; though the amplitude of slope of the linear trend is consistent with the model and observations. A plot of the timeseries of DSLPA from the HadSLP2 dataset is available at:

    2) Ray also wrote; “It remains disconcerting that the whole trend appears to rest on a meager handful of anomalous individual El Nino events.” We checked this — the 150yr blended dSLP trend was still negative when we nullified the 1982-3 and/or 1997-8 events. Under the assumption that the control runs have reasonable natural variability, the influence of natural variability has been addressed, since the probability that a 150yr trend this large would appear by chance in the control run is less than 1%.

    Among our two biggest concerns are (a) the realism of the control-run variability in our model and the AR4 ensemble, and (b) the quality and spatial representativeness of the pre-1900 obs. We attempted to address “(a)” by assessing the natural variability in all of the IPCC-AR4 pre-industrial control integrations (Supplementary Figure 3), and the observed DSLP changes are inconsistent with the variability from all of the models.

    3) The issue of whether there is any inconsistency of the results of a weakening DSLP and the results of a strengthening tropical Pacific SST gradient shown in Cane et al (1997, Science) has also been raised. The question is: has there been a change in SST gradient across the Pacific that is inconsistent with the SST gradient changes in the GFDL CM2.1 model (a model that shows a decrease in SLP gradient)?

    It is worth mentioning that different observationally-based SST products indicate a different change in the E-W SST gradient in the Indo-Pacific. Over the period 1880-2005 the Kaplan SST product and HadISST both show an increase in E-W SST gradient (more “La Nina”-like state), while NOAA-ERSST shows a moderate weakening of the E-W SST gradient (more “El Nino”-like state). The discrepancy appears to arise from the different estimates the products have of the 1920s-1930s, with Kaplan/HadISST indicating a more “El Nino”-like state and NOAA-ERSST indicating a more “La Nina”-like state. We’ve put up figures of the near-equatorial SST anomalies from both Kaplan and NOAA-ERSST, and the different evolution is clear:

    The time-series of an E-W SST difference, defined analogously with our DSLP index (difference of the average (80E-160E,5S-5N) and (160W-80W, 5S-5N) ), shows the different evolution in the two products:

    The two time-series show a clearly distinct behavior. We believe that emphasis should be placed on reconciling this difference in evolution of near-equatorial SST between the products.

    It should also be noted that of the five “all forcing” ensemble members in our analysis, two showed no change in the E-W Pacific SST gradient, two showed a decrease (“El Nino”-like state) and one showed an increase (“La Nina”-like state). So the evolution of the tropical Pacific SST gradient across the various GFDL-CM2.1 ensemble-members is not inconsistent with a tendency towards an SST “La Nina”-like state as shown in the Kaplan/HadISST products. We note that none of these SST-gradient changes are significantly different from zero at p=0.05 given the model internal variability.

    4) Finally, it is noted that: “…it should be remembered that Vecchi et al. present results only for one model…” It is true that we use only one model (the GFDL CM2.1 coupled model) for the attribution part of our paper. However, for the detection part we used all of the AR4 models, and for none of them was the change in the observed DSLP index consistent with “internal variability”.

    [Response: Thanks for the additional information. I’m confused about the last statement, though. I looked at the supplementary material (see my response above to Ileana Blade, #7) and if I’m reading your figure right it looks like the GFDL model is the only one for which the individual ensemble members lie outside the confidence limits for zero trends. At least that’s what the bars on the figure seem to say. Could you clarify? –raypierre]

    [Response: Some of these arguments are simply implausible. The trends in the SLP gradient do not look they would survive a robust trend determination to eliminate the influence of the obvious late outliers. By “survive”, I don’t mean “trend is still negative”, I mean “trend is negative and statistically significant“). In fact, I doubt the trend is statistically significant if you simply stop anywhere before the late 1970s. The model simulation analysis used to bolster the result shows a systematic long-term trend. If the observations show no trend through the late 1970s, can we really conclude that the observations support the model prediction? This is unconvincing. As far as the SST trends are concerned, your arguments are even less convincing. Two of the SST products (in my view, the two more reliable products, based on the optimal interpolation procedures used) show the opposite trend you are arguing for, and one (NOAA-ERSST) shows little or no trend. That’s not a ringing endorsement. If you take the NOAA-ERSST product, while the La-Nina like trend is clearly diminished, there is still not any statistically significant El Nino-like trend, as would be required for the self-consistency of your argument. In short, an objective look at the data is equivocal at best as to the true trend in the tropical Pacific ocean-atmosphere system over the past century. It is simply unconvincing to claim otherwise. –mike]

  31. 31
    Eric says:

    Thanks. It seems like a hypothetical decrease in convection would have to be specific since convection in the west is a part of the cycle. Wouldn’t that weaken the circulation? Wouldn’t the circulation weakening change the distribution of convection (allow more hurricanes at least in the Atlantic)? All these seem to me to be negative feedback for moisture in the upper troposphere.

  32. 32
    Gabriel Vecchi says:

    Ray, I haven’t been able to get at the version of our figures, so I don’t know what supplemental Fig.4 looks like online.

    However, in Supp. Fig. 4 there should be at total of eight models (including GFDL CM2.1) that have at least one ensemble member outside the p=0.05 confidence interval for weakening. Also, about 2/3 of the individual ensemble-members (46 out of 68) from all the model runs have linear trends that indicate at least a nominal weakening – this is significantly different from what one would be expected from a Binomial distribution with a 50% probability. So the multi-model ensemble shows a significant tendency to a weakening of the Walker Circulation (based on this index).

    It is true that GFDL CM2.1 shows the most consistent and strongest responses across the various models, and I’m not sure why that is: luck, skill or deficiency. Again, based on the model estimate of internal variability of this index, we are barely past the limit of detectability right now. If one looks at the IPCC-AR4 integrations for SRESA1B, a vast majority of the models show a clear weakening of the zonal-mean zonal-stress across the Pacific through the 21st century. And a weakening of the vertical mass transport across 500hPa is a robust feature across all IPCC-AR4 models in Scenario A1B (though, how that is partitioned is – as you correctly noted – a very interesting and unclear issue).

    [Response: The figure I linked in my response to Ileana above was downloaded from It looks pretty much as you said (though note the labeling of which model goes with which bar seems to be missing, except for GFDL). In my somewhat hasty summary of the appearance of the figure, I was neglecting those few individual ensemble members that were outside the bars because if you have fifty realizations or so you’d expect few to be outside just by chance; I agree there is a tendency for a lot of the rest to show weakening, but the GFDL one is the only case where almost all of the individual ensemble members clearly show significant weakening. There is one model that has two out of three showing significant weakening, and two that show one out of two members with weakening, but I’m not sure how much credence to put on such small ensembles. I’ m content with the statement that for most of the models we’re at the limit of detectability of the trend. Time will tell. In my mind, one of the key scientific issues is whether the Clement/Cane argument for an increased SST gradient is valid, is detectable in the data, and (if valid) is present in the models. –raypierre]

  33. 33

    I’ve completed that article, in four parts, that uses a bit of algebra to demonstrate how planetary surface temperatures can be modeled. It comes out as six single-spaced pages in Word. I can provide it as a Word doc, or HTML, or PDF. Would anyone want to see it, and is there any chance of posting it somewhere on RealClimate? Naturally I would want someone competent to read it over first and advise me of any silly mistakes. Thanks.

  34. 34
    Ian K says:

    Ray your response to #24 above misses my point. What I was trying to raise was the general issue of changes in albedo, which would seem far more effective ways of altering the radiation balance of the planet than the IEEE device, no matter how ingenious. I would think a house with a reflective roof (as opposed to a fashionable black roof)in a sunny climate would affect the impact of that household. There is talk of carbon offsets: wouldn’t a household making albedo changes also be worth some level of carbon credit? Is this a factor which could also be considered, as well as emissions, at a national level? Albedo changes could be easily assessed by satellite. I have never seen albedo issues discussed except at geoengineering level with aerosols in the atmosphere, etc. Why is this so?

    [Response:Sorry, I was carried away with the cleverness of the device. I absolutely agree with you that the device seems pretty pointless compared to just increasing the visible albedo. The device (as I infer from the abstract) converts upwelling IR into a band where the atmosphere is quite transparent to IR. However, the atmosphere is already quite transparent in the visible, and the upwelling IR came from conversion of absorbed visible to begin with. So, it seems a lot simpler to just reflect the visible before it gets absorbed and converted to IR. White paint on the roof is simpler and cheaper than a fancy IR transformer. Geoengineering albedo in the urban environment makes sense, but how would you paint the oceans white? Reflecting too much sunlight would also rob plants of radiation needed for photosynthesis. –raypierre]

  35. 35
    Hank Roberts says:

    Ian, I’m guessing, probably because the total area of the atmosphere and Arctic ice, where albedo makes a large difference, is so much larger than the total area on which people can paint things with reflective paint (even if we get paint the exact “color” of the infrared bands in which the atmosphere is transparent — “Hello paint store? I need …”

    I do recall a suggestion the last time we had a local drought that people buy enough “Ping Pong” balls to cover their swimming pools to reduce evaporation. Same sort of calculations apply I think, the wetted surface area was increased and evaporation went up. Painting a lot of things to reflect heat ought to pay back in savings on air conditioning costs, without any “albedo credit” calculation.

  36. 36
    Grant says:

    Re: #34

    I agree that albedo changes might be effective. Furthermore, if this strategy turns out to have unforseen negative consequences (I’m not saying it will, just *if*), it’s easily reversible. One of my strongest concerns to the injection of aerosols into the atmosphere is that it’s not easily reversible.

  37. 37
    Blair Dowden says:

    I would like to return to the question asked by C. W. Magee in #5: “if temperature only slightly increases rainfall, then why does paleoclimactic data indicate more widespread rainforests in pre-glacial cenozoic?” I think the evidence for that is more extensive than “steamy Cretaceous murals you see in the Field Museum.”

    For example, this U.S. Geological Survey of middle Pliocene paleoclimate clearly shows that time was generally both warmer and wetter than today. See especially Table 4 where forested areas are larger and deserts smaller during that time.

    Also of interest is this study from the Oak Ridge National Laboratory. Compare this map of African vegetation today with this map of 8,000 years ago during the mid Holocene. The tropical rainforest is much larger, and the Sahara desert is covered in grassland. Given that the warming then is less than that predicted for this century, is that what we can expect for Africa’s future?

    These are only two examples that contradict the model results that presently dry areas will get drier with global warming. If models do not agree with the data, maybe they need a little more work?

    This might partly be explained by the fact that paleoclimate data is measuring a system in equilibrium, while models are predicting a climate in transition. I can see two reasons why moisture may lag temperature increases:

    1) Land warms faster than the ocean, and warmer oceans are probably the source of most of the extra moisture.

    2) Rainfall is influenced by feedback from vegetation. The fact that the Sahara was covered in grassland increased its rainfall. It will take time for the grass to establish itself.

    Those two factors may explain why a Pliocene climate may not be established right away, but I do not see support for increased desertification. I wonder what work has been done to reconcile model results with paleoclimate data?

    [Response: The period 8K years ago (sometimes referred to in a value-laden way as the “climatic optimum”) is not a good analogy for what goes on in a doubled CO2 climate. The climate change in this period is generally believed to be associated with precessional changes in the distribution of solar radiation, which primarily affect land-sea temperature contrast, and give only a regional warming, plus an enhancement of certain monsoonal circulations. The Pliocene is a more interesting case (but note the dinos I was talking about are back in the Cretaceous, so we’re talking about different climates here). I wouldn’t describe the reference you cited as supporting growth of rainforests. It’s only the dark green part of the scale that’s rain forest, and that doesn’t change notably. What you mainly see is a replacement of tundra by boreal forest, and that can be accounted for by the reduced Arctic sea ice. Still, I’ll be the first to admit that there’s a lot to be learned about the land precipitation in past climates. I don’t want to overstate the idea that dry areas get drier, etc. Precip over land accounts for a very small fraction of total precip, and so diverting a little water to the land goes a long way. Yannick Donnadieu and I have a paper coming out in EPSL in a few months where we study runoff changes throughout the Cretaceous, and find a very strong effect of the paleogeography, which can be stronger than the effect of CO2. –raypierre]

  38. 38

    Re #33 — um, anyone? Please?

  39. 39
    Grant says:

    Re: #38

    I’d be interested in taking a look. I’m not sure how to get it, since I’m loathe to put my email address here. I notice your message has a URL to your website. Could you post it there for me to download?

  40. 40
    Ian K says:

    Re your response to #34, Ray, ie “Geoengineering albedo in the urban environment makes sense, but how would you paint the oceans white?”

    Albedo changes in the desert might make sense: blindingly white artificial salt pans come to mind as they should be cheap if brackish/salty bore water is available (though you would have to choose your topography carefully!). In urban Australia ubiquitous concrete-tiled roofs are currently being spruced up with paint. Unfortunately black and other dark colours are favoured as they are also for new building. The point is: should all of these issues by part of the equation along with emissions? Why (again) isn’t it discussed? The important issue in getting change is cost-effectiveness of mitigation measures. I would think certain changes in albedo would be cheap or even free and in urban settings lead to lower emissions as well.

  41. 41
    Hank Roberts says:

    Ian (salt pans):

    Archives of Environmental Health: ‘Respiratory health effects of alkali dust in … in drinking water quality and to blowing salt and dust from the lake bed. … – 25k

    Mineral Dusts in the Southwestern US
    Owens Lake – – a world-class source of salt-bearing dusts – – Possibly the greatest or most intense human-disturbed dust source on earth is Owens Lake, … – 26k

    I’m sure roofing contractors’ suppliers know how many square yards of roof are applied or repainted — it’s basic to their business. Someone able to figure the difference doing all that in white could make?

    Point being a lot of these ideas could be checked for sanity/benefit by those advocating them.

  42. 42
    Blair Dowden says:

    Re #37: Ray, the information I have suggests the mid-Holocene warming was mostly global, with increased rainfall nearly everywhere (except western North America). Today, the Earth is nearest to the sun in its elliptical orbit in January, making northern winters milder and summers cooler. Eight thousand years ago Earth’s orbit was closest to the sun in July because of precession, as you mentioned. This would intensify the difference between northern summer and winter. I would guess summer warming would melt polar ice, leading to ice albedo feedback and global warming. I have seen data suggesting sea levels were a meter or two higher than today, supporting this idea.

    You seem to be suggesting that this slightly different warming pattern would lead to wind and ocean current changes, leading to different climate patterns. I wonder if the mid-Holocene warming patterns can be reproduced in climate models?

    I agree that Pliocene climate may be a better model for greenhouse warming, but I have not been able to find any detailed information about it.

    [Response: There’s a vast amount of modelling work on mid-Holocene climate, particularly associated with the N. African Wet-Dry cycles. There’s far too much for me to list here, but if you google “COHMAP Mid Holocene Climate Reconstruction” you’ll get a representative set of hits on both the climate reconstruction and modelling studies. The classics are by Kutzbach and Gallimore, and there are recent updates by Bette Otto-Bleisner. As I said, it’s more of a regional and seasonally concentrated climate change, and the most evident climate changes are in tropical precipitation. With regard to models, even the early ones reproduce the basic pattern of monsoon changes, but there has been continuing difficulty in getting the rain band to move far enough North. There’s a general feeling that dynamic vegetation feedback may be necessary to get this right, but it’s a subject that’s much in flux. The main changes in radiative forcing from the precessional cycle are in the latitudinal and seasonal distribution, not in the global mean, which is why the nature of the response can be expected to be different from doubling CO2. –raypierre]

  43. 43
    Emile Arens says:

    Regarding geoengineering and albedo.

    For what its worth, my modest (first) addition to a brilliant site…

    Would a retroreflecting surface have an higher albedo? There is already retroreflecting paint on roads. Alternatively you could cover your roof and walls with retroreflecting panels of glass. Send the light back to where it came from!

    See wikipedia on retroreflection.

  44. 44

    Re #39 — If you want to e-mail me, I’m at Then I can send you a copy of the article without your having to reveal your e-mail address here. Sorry I took so long to get back to you; much happening here.

  45. 45
    Hank Roberts says:

    Re albedo and roof color —Here’s a rather odd site that advertised themselves on the weblog.

    I’ll give it in human-readable form rather than a clickable link since it’s …. odd.
    –it’s interesting for the pictures showing how different roof and wall paints/colors collect heat, though it’s a bit odd in claiming this is the cause of global warming.

    Nevertheless: put a . where I wrote “dot” to make this work:

    www dot thermoguy dot com/globalwarming-solutions.html

  46. 46
    Hans Erren says:

    With regard to models, even the early ones reproduce the basic pattern of monsoon changes, but there has been continuing difficulty in getting the rain band to move far enough North. There’s a general feeling that dynamic vegetation feedback may be necessary to get this right, but it’s a subject that’s much in flux

    Wouldn’t that imply that massive rainforest removal has a significant effect on the tropical precipitation and hence cloud formation and thus the dominant cause for the observed increase in equatorial OLR?

  47. 47
    Ian K says:

    Re: 35# and 45#. Thanks for the websites Hank. My point about using salt basins was a bit whimsical and off the top of my head, though such things as quartz chips might be more benign. Perhaps I could distribute them to spell out a sponsor’s name to defray expenses ;) That thermoguy is seriously off the track. I presume he has bought himself an infrared camera and got carried away. What I am talking about, however, doesnâ��t need any fancy equipment like the IEEE article is talking about. It is just plain, boring, reflection. So we donâ��t need to â��get paint the exact “color” of the infrared bands in which the atmosphere is transparentâ��. I am talking about just mirrored or white surfaces. Most of sunlightâ��s energy that reaches us at ground level is in the visible part of the spectrum and so we can use our eyes to assess how good a job a surface will do in reflecting that part of the sunâ��s energy straight back into space. The tricky bits as I see it are: issues of glare for passersby/neighbors; working out what hours to deploy it on my roof (or may be to shade my lawn in the middle of a summerâ��s day, etc); I would have to work out sun elevations, hours of sunlight in my area, etc to determine how much visible light should be reflected; and then finally express this as a negative forcing in W/metre2 to be offset against my presumed calculated positive forcing due to my familyâ��s emissions of GHG, etc. No doubt this would be an informative exercise in some way but I think I would need some help getting started as doing science is a lot harder than talking about it!

  48. 48
    Gar Lipow says:

    Re: cool roofs. While confusing this with global warming is nonsense – it is quite true that light colored roofs tend to be cooler than dark colored roofs. In warm climates light colored or reflective roofs can reduce cooling bills substantially. It is an interesting calculation to do though in climates with hot summers and cold winters. Does the reduction in cooling costs in the summer pay for the increase in winter heating costs? I suspect the answer to that is very much a matter of local climate – that there is no one magic answer. For any place with hot summers and mild or even moderate winters though the balace clearly favors light colored roofs.

    If you live in (for example) Houston if your current roof is dark, making the next one light colored will save slightly on air conditioning, and reduce your carbon emissions a bit as well.

  49. 49
    Alan Betts says:

    I made the same argument on the slowing of the tropical mass circulation in a warmer climate, based on Betts and Ridgway (JAS1989) and the difference of the slopes of the Clausius-Clapyron and the radiative cooling

    Betts, A. K., 1998: Climate Convection feedbacks: some further issues. Climatic Change. 39, 35-38.

    [Response: Thanks for that information. I’ll keep it in mind if I write up a small paper on the precipitation-circulation-temperature relationships. I’m beginning to think a paper on that subject is necessary. I still don’t see why people keep saying that the radiative cooling is constrained to go up more slowly than Clausius Clapeyron. This is plainly not true, as can be easily seen by computing the net radiative cooling in a radiative-convective model with a consistent surface energy budget. It’s true that there are aspects of the vertical distribution of radiative cooling that can’t be controlled by adjusting the air-sea temperature difference, but I haven’t seen it demonstrated that these are crucial. –raypierre]

  50. 50
    Blair Dowden says:

    Re #49: Alan, your link does not work.