Understanding the Role of Latent Heat Energy Transport in Climate Change

[Andre]

Meanwhile, I don't think that things get even more complicated when looking at the evaporation map again.

A closer look would reveal that the evaporation around the equator is less than in the areas around the both tropics. (+/-20 degrees lattitude) where the evaporation tends to be at maximum. Also evaporation rates towards the polar areas seem to keep up high values despite much cooler sea surface temperatures.

Evaporation rate is defined by the Penman equation, however that doesn't help us much because we are dealing with variation in temperature, windspeed and relative humidity here as main factors governing the evaporation rate and those only...

...impact the values of m, g, cp, ρ, and δe.

However the reduced evaporation at the equator despite higher sea surface temperatures can be explained by a reduction in windspeed (doldrums) and increased relative humidity accumulated during the trade wind phase (leg #4) in the Hadley cell.

The relatively high evaporation rates in the colder lattitudes can best be explained by higher wind speed (roaring forties). And this appears to make sense as we learn from blowing a hot spoon of soup. It is especially the mechanical mixing of the lower level atmosphere under the turbulence of higher winds, while calm winds remain mostly sort of laminair in stable atmospheric conditions, preventing the moisture to mix in the higher layers.

Therefore a reduction of the convection rate in the hadley cell, would slow down the trade wind speed and this would decrease the evaporation rate, which as we have seen, deals with a lot of watts per square meters. Moreover, remember, in the increased greenhouse effect setting we needed an increase in the absolute humidity to account for the 'maintaining relative humidity' assumption to generate the positive feedback in climate sensitivity

Hence things seem certainly a bit more complex, enough perhaps to obscure the judgement about the prevailance of positive or negative feedback. This would make the different measured outcomes of negative feedback in the multiply quoted Karner and Lindzen studies maybe more acceptable.

 

[vanesch]

Therefore a reduction of the convection rate in the hadley cell, would slow down the trade wind speed and this would decrease the evaporation rate, which as we have seen, deals with a lot of watts per square meters. Moreover, remember, in the increased greenhouse effect setting we needed an increase in the absolute humidity to account for the 'maintaining relative humidity' assumption to generate the positive feedback in climate sensitivity

I think that if you really want to know the real humidity, that you need to do an entire climate modeling. The "keeping constant relative humidity" is simply a very coarse idea, that you keep things "all else equal" in a way, and that the drive for higher absolute humidity (because constant relative humidity) is the higher temperature of the water at the surface.

Again, don't confuse "evaporation rate" with "humidity". If you don't blow over your hot soup, then the evaporation rate is probably lower, but the humidity above the soup is higher. If you blow over the soup, the evaporation rate is higher, but (or rather BECAUSE) the humidity in the air flow is lower.

So if the trade winds slow down, humidity might even rise stronger than "keeping relative humidity constant", all together with less evaporation. Or not. If you really want to know, and go beyond the educated guess of a simple model like this, you need to do all the modeling, in a detailed way, including winds, ocean currents, and all that.

 

[Andre]

.
So if the trade winds slow down, humidity might even rise stronger than "keeping relative humidity constant", all together with less evaporation..

That would be really hard to quantify because as said, you loose the mixing with higher levels when the wind decreases. Sure at sea level under no wind condition the relative humidity can rise easy to 100% but at -say- 10 meters altitude the humidity will be a lot lower, and that's where the greenhouse effect should have been enhanced.

Light winds can be laminair, with little mixing with higher layers under the stable conditions in the trade wind area, again, still limiting evaporation as well as limiting humidity where it counts for the greenhouse effect. Only when the wind strenghen and more pronounced waves are formed, the mechanical turbulence can cause water vapor mixing to 'higher'-lower layers, say some 100s of meters. Pilots of low flying gadgets like hang gliders balloons, sail plane can tell about mechanical turbulence for hours.

So yes reduced wind can increase humidity, but only at levels directly at the ground and under stable atmospheric conditions. But not at higher levels where it counts more for greenhouse effect.

Yes modelling. It would seem logical that the relationship between the pace of the hadley cell structure, evaporation and radiative effects are incorporated in the Global Climate Models, but I don't see evidence of that so far.

 

[Skyhunter]

So yes reduced wind can increase humidity, but only at levels directly at the ground and under stable atmospheric conditions. But not at higher levels where it counts more for greenhouse effect.

Do you have a citation to support that hypothesis?

If there is no wind, or light wind, Then it would seem to me that convection would carry the water vapor to higher altitudes as the surface warmed. The surface flux would have to be high to maintain 100% relative humidity, which would lead to strong surface warming, especially considering the strong LW absorption by the saturated air.

 

[Andre]

Do you have a citation to support that hypothesis?

If there is no wind, or light wind, Then it would seem to me that convection would carry the water vapor to higher altitudes as the surface warmed. The surface flux would have to be high to maintain 100% relative humidity, which would lead to strong surface warming, especially considering the strong LW absorption by the saturated air.

You missed the second condition: stability:

Light winds can be laminair, with little mixing with higher layers under the stable conditions in the trade wind area

meaning no convection, meaning not a lot of mixing.

 

[Skyhunter]

You missed the second condition: stability:



meaning no convection, meaning not a lot of mixing.

But that is not representative of the conditions under which WV saturation would occur. If there is enough warming at the ocean surface to maintain 100% relative humidity there must be convection to maintain the adiabat.

 

[Andre]

That's irrelevant

We're talking about the transition of leg #3 to leg #4. In the subsidence phase of leg #3 the descending (very dry) air heats up dry adiabatically, such a process with colder air below is usually suppressing convection.

The warm (desert) air that get's in contact with the ocean in leg #4 cools due to evaporation, therefore the lower levels are cooler than the environment lapse rate making the vertical temperature conditions stable. Hence no convection, only mechanical turbulence.

 

[granpa]

Do you have a citation to support that hypothesis?

If there is no wind, or light wind, Then it would seem to me that convection would carry the water vapor to higher altitudes as the surface warmed. The surface flux would have to be high to maintain 100% relative humidity, which would lead to strong surface warming, especially considering the strong LW absorption by the saturated air.

his point seems pretty simple and clear to me. higher winds = more turbulence = more mixing with higher levels, meaning the lower 1000-2000 meters of the troposphere (marine layer?)

 

[Skyhunter]

his point seems pretty simple and clear to me. higher winds = more turbulence = more mixing with higher levels, meaning the lower 1000-2000 meters of the troposphere (marine layer?)

What he said was:

Sure at sea level under no wind condition the relative humidity can rise easy to 100% but at -say- 10 meters altitude the humidity will be a lot lower,

10 meters is not 1000-2000, so I fail to see how you can interpret it as such.

I agree that higher winds = more turbulence = more mixing. But his point about no convection above 10 meters is erroneous because it is not a stable condition. With 100% humidity the surface flux will cause the air to warm and rise.

 

[Skyhunter]

The warm (desert) air that get's in contact with the ocean in leg #4 cools due to evaporation, therefore the lower levels are cooler than the environment lapse rate making the vertical temperature conditions stable. Hence no convection, only mechanical turbulence.

So where does the desert air make contact with the ocean?

Most of the air in contact with the ocean does not come from a desert.

It seems to me that you are describing an extremely rare condition that would have little effect on global climate.

 

[granpa]

he said under 'no wind' conditions. he was going to extremes to to make his point more obvious.

also I think 'desert air' means any air in the descending leg of the cycle. (where all the worlds deserts are located)

 

[Skyhunter]

he said under 'no wind' conditions. he was going to extremes to to make his point more obvious.

also I think 'desert air' means any air in the descending leg of the cycle. (where all the worlds deserts are located)

Dry air, not desert air would be the accurate term. And he was not exaggerating to make a point. He specifically used 10 meters because all LW radiation from the surface that can be absorbed by GHG is absorbed in the first 10 meters, provided there is water vapor present. That is why he said there would be no added WV to the layers where it would have a greater effect.

He is attempting to redefine the adiabat using exceptions that may or may not exist in nature.

 

[granpa]

http://en.wikipedia.org/wiki/Planetary_boundary_layer

 

[Andre]

This is what I mean with humidity of the lower 10 meters under stable no wind conditions:

http://www.bbc.co.uk/bristol/content/images/2008/02/11/fog_shallow_john_rawlings_470x258_2.jpg

 

[Skyhunter]

http://en.wikipedia.org/wiki/Planetary_boundary_layer

The stably stratified boundary layer is a night time phenomenon except in the higher latitudes where the surface is colder than the air, even during the day. Andre is proposing this mechanism for the tropics using the penman equation which includes solar forcing, the penman equation is not applicable for this phenomenon in the tropics since this condition will not exist during the day over the tropical oceans.

The reason that there is less evaporation over the ocean at the equator is evident in the Hadley cell diagram. The circulation is from high latitudes toward the equator at the surface, in the tropics, where most of the evaporation takes place. (4), by the time it gets there it is already near saturation. This nearly saturated air rises once again (1). The WV either condenses around CCN releasing it latent heat, remains as a vapor, or precipitate out. Both clouds and vapor are carried to higher latitudes (2a), (2b). Finally the air cools and falls back to the surface (3), to begin the cycle again..

During the night time there is less convection as the surface cools relative to the air, but come morning, solar energy will once more drive convection regardless of the relative humidity.

Low clouds (fog), which are liquid not vapor will inhibit evaporation from the surface as they absorb and reflect SW radiation during the day.Day and night they will trap LW in significant frequencies from surface emission. Net forcing for clouds can be positive or negative depending on numerous factors.

What Andre is proposing is a well understood and phenomenon, and well accounted for in the wet lapse rate. There will be plenty of water vapor in the layers above the first ten meters to once again with CO2 trap most of the outward LW flux from the first 10 meters of atmosphere, and without convection and 100% humidity, you would almost have an actual greenhouse with a glass ceiling. The Hadley cell diagram shows that this is not the case.

 

[Andre]

The Hadley cell is indeed accounted for but I cannot see that the changes in the Hadley cell are accounted for with increase of greenhouse effects, especially what the required changes in evaporation rate are for maintaining relative humidity to explain positive water vapor feedback. Also, how the required energy is provided to attain those evaporation rates.

All I see is the assumption that relative humidity is assumed to be more or less constant in climate sensitivity calculation. Period. I attempted to demonstrate here that it would be pretty hard to maintain that assumption, given the substantial energy required to increase evaporation rates. If there is more evaporation required, which Vanesch challenged, then this process would act as a negative feedback, which I have not seen to be accounted for.

 

[granpa]

Ah. now I understand. you're not from San Francisco are you? over here we call the 2000 meter (or is it feet?) thick marine layer 'fog'.

 

[granpa]

http://en.wikipedia.org/wiki/Planetary_boundary_layer

The PBL depth varies broadly. At a given wind speed, e.g. 8 m/s, and so at a given rate of the turbulence production, a PBL in wintertime Arctic could be as shallow as 50 m, a nocturnal PBL in mid-latitudes could be typically 300 m in thickness, and a tropical PBL in the trade-wind zone could grow to its full theoretical depth of 2000 m.

 

[Skyhunter]

The Hadley cell is indeed accounted for but I cannot see that the changes in the Hadley cell are accounted for with increase of greenhouse effects, especially what the required changes in evaporation rate are for maintaining relative humidity to explain positive water vapor feedback. Also, how the required energy is provided to attain those evaporation rates.

All I see is the assumption that relative humidity is assumed to be more or less constant in climate sensitivity calculation. Period. I attempted to demonstrate here that it would be pretty hard to maintain that assumption, given the substantial energy required to increase evaporation rates. If there is more evaporation required, which Vanesch challenged, then this process would act as a negative feedback, which I have not seen to be accounted for.

When you raise the temperature you also reduce the heat of evaporation. So your calculation for energy required to vaporize water needs to be constantly adjusted downward as SST increases.

 

[Andre]

Very true,

roughly in the order of magnitude of one promille per degrees, I'd say, judging from that slope at the left hand. But the point indeed is with countering effects, quantititive judgement is difficult.

 

[granpa]

http://en.wikipedia.org/wiki/Sunlight
Hence the average incoming solar radiation, taking into account the angle at which the rays strike and that at anyone moment half the planet does not receive any solar radiation, is one-fourth the solar constant (approximately 342 W/m²)

from:
http://en.wikipedia.org/wiki/File:Solar_Spectrum.png

 

[granpa]

what effect might increased temp and increased water vapor have on the frequency of severe storms? if they become much more common then that could have a drastic effect on the average pressure at the equator (air pressure under a severe storm being much less than under a regular thunderstorm) and therefore greatly increase the rate of convection

 

[mheslep]

Maybe this could explain what Lindzen and Choi 2009 observed:
Lindzen, R. S., and Y.-S. Choi (2009), On the determination of climate feedbacks from ERBE data, Geophys. Res. Lett., doi:10.1029/2009GL039628, in press.

The paper doesn't appear in that list of links. Is GRL the actual publisher?

 

[vanesch]

When you raise the temperature you also reduce the heat of evaporation. So your calculation for energy required to vaporize water needs to be constantly adjusted downward as SST increases.

That's in principle true, but we are talking about small temperature variations over which we can take the latent heat of evaporation rather constant in a first approximation I would think.

The real point is that there is no *required* amount of power to establish a given *humidity* (which was, if I understood correctly, the initial claim of the OP). You can, in principle, have any humidity in relationship with any power, as convection will transport exactly the amount of heat it has to, with the given air mixture and will hence be able to transport a low amount of power or a high amount of power, and this by adjusting the "looping speed" of the convection cell.

What will determine the amount of humidity has hence not much to do with the forcing just by itself, but rather with the temperature, the winds, the mixing of air masses and all that. Humidity and mixing and so on will also determine the final amount of evaporation.

If you want to find out how much all this changes, I don't think there's any other option but a very detailed modeling of all these processes. If you're too lazy to do so, then a reasonable assumption (which might turn out not to be correct, but only after thorough modeling) is to keep relative humidity constant (which would imply similar mixings as today between 'wet' air and 'dry air' in similar proportions, with 'wet air' completely saturated).

In a way, Andre is right of course that as convection will adapt, this will change winds, convection patterns, and so on, and this will have an influence on humidity and evaporation - only, it is not clear how much and in what sense without a deeper investigation. There could be more or less humidity. But it will never be because 'power is lacking'.

 

[Andre]

But it is an assumption, right? maintaining relative humidity constant with increased greenhouse effect, without any visible justification in the decription of all the processes as we attempted here. So one could argue about it being a reasonable assumption seeing the considerable energy required for excess evaporation and because the observations don't appear to confirm the assumptions.


For mheslep, the former link to Lindzen and Choi was to the article in press. It is now:

http://www.agu.org/pubs/crossref/2009/2009GL039628.shtml

 

[vanesch]

But it is an assumption, right? maintaining relative humidity constant with increased greenhouse effect, without any visible justification in the decription of all the processes as we attempted here. So one could argue about it being a reasonable assumption seeing the considerable energy required for excess evaporation and because the observations don't appear to confirm the assumptions.

You seem to stick to this "considerable energy required" and "excess evaporation" which I argued all the time, doesn't have anything to do with it.

With lower hadley cell turnover velocities, you can have higher humidities together with less evaporation and hence less required energy for evaporation. However, with higher horizontal mixing you might get lower or higher relative humidities and higher rates of evaporation. As the hadley cell turnover velocity will adapt to the required heat transport, there will never be "lack of energy for evaporation": if any lack is there, the hadley cell turnover will slow down.

You could get higher relative humidities, or lower relative humidities, depending on how mixing of dry and wet air will change as compared to now, and this without putting any burden on "energy required" or even on "evaporation required" and the only way to really find out is to do a very detailed simulation.

I'm not saying that there is any proof that relative humidities will remain constant, just that it is a very sensible assumption if you don't know any better, much more so than assuming that absolute humidities will remain constant, as if "nothing else changes" (although it probably will), the humidities are mainly given by the ocean (and "wet" land) temperatures AND how they mix together with "dry air". If the mixing is done in the same way, and the temperature rises, then there's no reason not to assume that the relative humidities will remain the same (one will mix together 100% wet air with dry air in the same proportions, and the "100% wet air" is determined by the temperature of the water reservoir with which it was in contact). This is NOT fixing any "evaporation rate", as this evaporation rate is determined also by the VELOCITY of the turnover, and this velocity is something that will adapt to whatever heat that has to be transported in order to get the lapse rate close to the adiabatic lapse rate. As this velocity can hence change, the evaporation rate can also change, with given humidities.

In fact, it would be rather difficult to keep "absolute" humidities constant, while having warmer water. Warmer water will normally lead to higher absolute humidities because the evaporation drive is greater in that case.

 

[Andre]

I am merely sticking to the requirement that assumptions would need some validation while Lindzen and Choi (2009) demonstrate that there is no such a validation.

Edit: and while we are at that, maybe that http://www.agu.org/pubs/crossref/2008/2008JD010064.shtml also hints into the same direction:

Rondanelli, R., and R. S. Lindzen (2008), Observed variations in convective precipitation fraction and stratiform area with sea surface temperature, J. Geophys. Res., 113, D16119, doi:10.1029/2008JD010064.

...We find that the fraction of convective precipitation increases with Sea Surface Temperature (SST) at a rate of about 6 to 12%/K and the area of stratiform rainfall normalized by total precipitation decreases with SST at rates between −5 and −28%/K. These relations are observed to hold for different regions over the tropical oceans and also for different periods of time. Correlations are robust to outliers and to undersampled precipitation regions. ...

 

[vanesch]

Andre,

can you indicate what this is supposed to mean in the context of what we are discussing here (namely, whether there is not enough power to raise absolute humidity to levels that keep relative humidity constant) ?

What does this observation of precipitation have to do with what is supposed to happen when we increase CO2 levels on humidity ?

Maybe it is explained in the article but I only have access to the abstract of that paper.

 

[mheslep]

Thanks for the Lindzen/Choi pointer.

I am merely sticking to the requirement that assumptions would need some validation while Lindzen and Choi (2009) demonstrate that there is no such a validation.

I don't follow how that paper relates to this discussion - it addresses radiation (claims mistaken assumptions about radiation). Even if all the models are wrong per Lindzen/Choi, so what per this discussion? Edit: I see Vanesch beat me to the question.

 

[Andre]

Both questions are not the same, but Lindzen and Choi are the core matter where, with the thread trying to hint what they may be seeing, with the big question being the dominance of positive or negative feedback on climate sensitivity.

The leading idea is that positive feedback prevails in climate sensitivity, one of the main factors being enhanced greenhouse effect of more water vapor at higher temperature, under the premisse that relative humidity remains constant.

Here I have tried to argue that maintaining relative humidity may require a lot of additional energy to enhance evaporation to sustain it, with the hadley cell as mechanism to show the open -air conditioner- function, transport of (latent heat) energy aloft for easier out radiation. This would act as a big negative feedback element, consuming the energy that was thought to go into heating. I wondered if the models did account for this energy as I see no evidence for that in the model descriptions, only that relative humidity is kept constant.

But there are many factors modifying and influencing effects, therefore I suggested to look at the end result. What are the measured effects of feedback? Positive or negative feedback? That answer is given by Lindzen and Choi 2009.

I'll attempt to get into the second publication tomorrow.

 

[mheslep]

Alright, but I don't think it is logical to point from your relative humidity discussion to Lindzen Choi as a basis. Rather, if Lindzen Choi are correct then perhaps some other negative feedbacks are at work (and unidentified by them), even so they are simply suggesting: go look for them.

 

[vanesch]

Both questions are not the same, but Lindzen and Choi are the core matter where, with the thread trying to hint what they may be seeing, with the big question being the dominance of positive or negative feedback on climate sensitivity.

The leading idea is that positive feedback prevails in climate sensitivity, one of the main factors being enhanced greenhouse effect of more water vapor at higher temperature, under the premisse that relative humidity remains constant.

What I don't understand in the (other) Lindzen article is that they try to establish feedback effects on a time scale of the order of a year (that's what they do when they analyse the correlations over time periods of this order in time) I thought the eventual positive feedbacks were over several decades, so I don't see how their very short term correlations can say anything about climatological trends. At most they say that climate models are bad weather forcasters over 6 months or a year.
In as much as their comparison with climate models is correct, they do establish that *those* climate models don't make correct predictions on that short term. But honestly, I'm not really surprised at that, as climate models are supposed to model climate, which is an average over several decades, no ?

Here I have tried to argue that maintaining relative humidity may require a lot of additional energy to enhance evaporation to sustain it, with the hadley cell as mechanism to show the open -air conditioner- function, transport of (latent heat) energy aloft for easier out radiation. This would act as a big negative feedback element, consuming the energy that was thought to go into heating.

I don't think people "think" that this is going into heating in the first place. It is what I tried to argue here all along: if you take it that convection tries to establish the adiabat, and you take it that that adiabat is established by convection you take already this cooling into account. I'm NOT saying that this convection, and that this evaporation is not giving rise to a serious cooling, I'm saying that you already include it when putting down the adiabat. It is clear to everybody that convection is a powerful heat transport mechanism - it is the principal mechanism of heat transport in the lower troposphere.

The whole question is what will happen to humidity, and there I think if you really want to know, you need to do very detailed modeling, but that the best "first guess" is indeed keeping relative humidity constant. As the adiabat already takes into account all the heat loss by evaporation and all that, there won't be any problem with "not enough power to evaporate the water" by assuming constant relative humidity, because this is already included.

I wondered if the models did account for this energy as I see no evidence for that in the model descriptions, only that relative humidity is kept constant.

Because the adiabatic lapse rate *already* takes this into account!

 

[Skyhunter]

Two things.

1. Emission is the primary feedback.
2. Evaporation is only half of the water cycle.

When it gets warmer there is an instantaneous increase in emission, negative feedback.

Unless precipitation increases there is no need for a dramatic increase in evaporation to maintain relative humidity.

 

[Andre]

In a way, Andre is right of course that as convection will adapt, this will change winds, convection patterns, and so on, and this will have an influence on humidity and evaporation - only, it is not clear how much and in what sense without a deeper investigation. There could be more or less humidity. But it will never be because 'power is lacking'.

bumbing because I stumbled upon http://www.sciencemag.org/cgi/content/abstract/317/5835/233

...However, the climate models predict that global precipitation will increase at a much slower rate of 1 to 3% per kelvin.

Maybe because indeed consideration was given to the additional energy required to evaporate additional water

... the observations suggest that precipitation and total atmospheric water have increased at about the same rate over the past two decades...

Since precipitation removes water from the atmosphere, it seems that evaporation has to increase to the same rate/flux to maintain dynamic equilibrium. It seems that this was exactly the dispute between Vanesch and me.

And as I was trying to demonstrate, the increase in evaporation requires more energy than increased greenhouse effect from increased CO2 levels could bring, this effect would work as a dominant negative feedback.

It also suggests that a part of the required increase in precipitation would require another energy source for evaporation.