Guest Post by Willis Eschenbach
At the end of my last post , I said that the climate seems to be an inherently stable system. The graphic below shows ~2,000 climate simulations run by climateprediction.net. Unlike the other modelers, whose failures end up on the cutting room floor, they’ve shown all of the runs … including the runs that ran right off of the rails.
Figure 1. Climate simulation runs from climateprediction,net.
Notice that many of the runs go badly wrong, either cooking the planet at some 8°C (14°F) hotter than at present, or spiraling down into an ice-covered unreality. To me this is a perfect example of the basic misunderstanding of how the climate works. People think that the global temperature is free to take up any temperature at all, and that if the forcing changes, the temperature must change. But it is not free to go off of the rails. Instead, the global temperature is an inherently stable system.
Now, what does it require for a natural heat engine like the global climate to be inherently stable? The general way that humans control heat engines like say an automobile engine is by controlling the “throttle”, which in an automobile is what the gas pedal connects to. The throttle decrease or increases the amount of fuel that is entering the engine. To be stable, you need some system that opens or closes the throttle based on some criterion.
In the climate system, of course, the throttle is the variable albedo of the earth. The “albedo” of an object is a number from 0.0 to 1.0 that measures the fraction of solar radiation that is reflected from the surface of the object. It’s usually given as a fraction, although I prefer it as a percentage (0% to 100%). The albedo of the earth is about 0.29, meaning 29% of the sunlight is reflected back to space.
As I showed in my last post, the albedo generally decreases with temperature … up to around 26°C or so. Above that the albedo rises rapidly. As a result, in much of the tropics when the ocean warms the albedo increases, rapidly cutting back on the incoming solar energy.
In such a system, when the earth is cooler than the equilibrium temperature, the solar input goes up, increasing the temperature. And conversely, when the earth is warmer than the equilibrium temperature, the solar input goes down, and the earth cools back to the equilibrium temperature.
In the comments to my last post, someone asked how the increase in albedo worked out on a daily basis. To answer that, I need to take a bit of a diversion.
I got interested in climate in the late nineties. Most folks I read wanted to understand why the earth’s temperature had changed over the 20th century. I had a very different question—I wanted to know why the earth’s temperature had changed so little over the 20th century (a variation of ± 0.3°C). Since the earth’s temperature is about 290 Kelvin, that’s a variation of plus or minus a tenth of a percent or so. As someone who has dealt with regulated engines, to me that was astounding long-term stability. Over the 20th century we had droughts, the clouds came and went, we had volcanoes, times of lots of hurricanes, times of few hurricanes … and the temperature went nowhere. Plus or minus a tenth of a percent.
At the time I started tackling the problem of climate stability, I was living in Fiji. At first, I spent a whole lot of time searching for the reason that there was such long-term stability. I tried to identify and understand any processes acting on multi-decadal time scales. I thought about the ebb and flow of CO2, about how the CO2 makes the rain acidic and dissolves the mountains over millennia. I thought about the purported barycentric solar cycles. I thought about the multidecadal oscillations.
No joy.
In the evenings after work I’d walk and think, think and walk. I picked up and discarded dozens of possibilities. I can’t tell you how far I walked thinking about long-term, slow compensatory systems that could keep the earth on track for a century and more.
Then one day I had a curious thought. I thought, if there were a system that kept each day within a certain temperature range, it would keep that week within that same temperature range, and it would keep that year within that temperature range, and that decade, and century, and millennium … like a fool, I’d been looking at entirely the wrong end of the time spectrum. I needed to look at minutes and hours, not decades and centuries.
This changed the entire direction of my research overnight. I started looking for processes that would regulate the temperatures on a daily basis … and since I was living in Fiji, I didn’t have far to look. I started to think that the action of the tropical cumulus clouds and in particular the thunderstorms were the real actors in the climate pageant.
I could see the daily tropical cycle unfolding most days. Clear at dawn. Then cumulus clouds form usually before noon. Thunderstorms in the afternoon, sometimes lasting into evening or night. However, I was still at a great disadvantage. I didn’t understand how the control worked. The problem was that even in the tropics you have seasons, and not every day is the same. Plus there’s day and night, it was all so complex I couldn’t see how the control was effected. I wanted some point of view where I didn’t have to deal with all of that day/night, seasons stuff.
Then one day I realized that there was a point of view which freed me from all of those problems. This was the point of view of the sun. You see, from the sun’s point of view it’s always daytime—from the sun’s point of view, there is no night. And there are no seasons—underneath the sun, it’s always eternal summer.
So to investigate the cumulus and the thunderstorms from the sun’s point of view, I used the satellite local-noon-time images from the GOES-West weather satellite. I averaged the photos over an entire year, to show the average cloudiness of the Pacific. Figure 2 shows that result:
Figure 2. Average of one year of GOES-West weather satellite images taken at satellite local noon. The Intertropical Convergence Zone is the bright band in the yellow rectangle. Local time on earth is shown by black lines on the image. Time values are shown at the bottom of the attached graph. Red line on graph is solar forcing anomaly (in watts per square meter) in the area outlined in yellow. Black line is albedo value in the area outlined in yellow.
Looking from the point of view of the sun does a very curious thing—it trades time coordinates for space coordinates. For example, in the photo above, it is always local noon at the point directly under the sun. Noon is not a time. It is the vertical line running up the middle of the picture. Sunrise is always at the left edge of the view from the sun, and the left half of the picture is the time before noon. Sunset is always at the right edge of the view from the sun, and afternoon is the right half of the picture. We’ve put spatial coordinates in place of temporal coordinates.
From this, you can see that the onset of cumulus clouds is at about 10:30. This is shown by the increase in albedo (black line at picture bottom). By 11:30 there is a fully developed cumulus field. This shift in albedo changes the reflected sunlight by about 60 W/m2. And that field of clouds persists all through the afternoon (right side of the picture above).
And most important, from the sun’s point of view I could finally understand how the albedo control is actually effected—via variations in the timing of the onset of the cumulus and thunderstorm regimes. What happens is that if the Pacific is warmer than usual, the cumulus clouds and thunderstorms shift to the left in the image above by emerging earlier in the day. This, of course, reflects more of the sunlight. And if the Pacific is cooler than usual, the clouds and thunderstorms shift to the right, emerging later in the day or not at all, and thus exposing more of the area to the stronger sunlight of the mornings. The clouds act like a reflective window screen that covers more or less of the day, depending on the temperature.
Now, from this hypothesis we can advance some testable predictions. First, albedo should be positively correlated with temperature in the tropical Pacific. This is confirmed by my previous post. Next, we should be able to detect the effect of the variations in cloud onset on the daily temperature record … which hopefully will be the subject of my next post.
Finally, while the cumulus and the thunderstorms control the throttle by regulating the amount of energy entering the system, there are a variety of other temperature regulating phenomena as well. What all of these have in common is that they are “emergent” phenomena. These are phenomena that emerge spontaneously, but only when conditions are right. In the climate system, these phenomena typically emerge only when a certain temperature threshold is surpassed.
In the tropical daytime system, once a certain temperature threshold is reached the cumulus clouds start to form. But often, the reduction in incoming sunlight is not enough to stop the daily warming. If the surface continues to warm, at some higher temperature threshold thunderstorms form. And if the surface warms even more and a third temperature threshold is surpassed, yet another phenomena will emerge—the thunderstorms will line up shoulder to shoulder in long serried rows, with canyons of clear descending air between them.
Thunderstorms are natural refrigeration cycle air-conditioning machines. They use the same familiar evaporation/condensation cycle used in your air conditioner. But they do something your air conditioner can’t do. They only form exactly when and where you need them. When there is a hot spot in the afternoon on a tropical ocean, a thunderstorm soon forms right above it and starts cooling the surface back down. Not only that, but the thunderstorm cools the surface down below the starting temperature. This can not only slow but actually reverse a warming trend.
And if there are two hot spots you get two thunderstorms, and so on … do you see why I argue against the entire concept of “climate sensitivity”? When you add additional forcing to such a system, you don’t just get additional hot spots.
You also get additional thunderstorms working their marvels of refrigerational physics, so there is little surface temperature change.
It’s one AM, big moon a few days past full. Think I’ll go back outside, I heard a fox barking outside around moonrise. Best to all, moon over your shoulder, more to come,
w.
The Perennial Request: If you disagree with someone, please have the courtesy to quote the exact, precise words that you disagree with. That way we can all understand your objection.
Further Reading:
I love thought experiments. They allow us to understand complex systems that don’t fit into the laboratory. They have been an invaluable tool in the scientific inventory for centuries. Here’s my thought experiment for today. Imagine a room. In a room dirt collects, as you might imagine. In my household…
Air Conditioning Nairobi, Refrigerating The Planet
I’ve mentioned before that a thunderstorm functions as a natural refrigeration system. I’d like to explain in a bit more detail what I mean by that. However, let me start by explaining my credentials as regards my knowledge of refrigeration. The simplest explanation of my refrigeration credentials is that I…
In a recent post, I described how the El Nino/La Nina alteration operates as a giant pump. Whenever the Pacific Ocean gets too warm across its surface, the Nino/Nina pump kicks in and removes the warm water from the Pacific, pumping it first west and thence poleward. I also wrote…
Willis, I always enjoy reading your ponder pieces. You have a great mind and ability to bring simplicity to issues, then set them in prose for all to see. The changes in warming and cooling trends that have occurred since temperature measurement began in 1659 have been triggered in both directions by the natural ocean cycles (longer term PDO and AMO, shorter term NINO and in a lesser manner and frequency NAO). The degree of change and direction depends on whether two of the oscillations are opposing, supporting or out of phase with each other. The flat spots in temperature, such as currently, last until one of the oscillations changes phase when supporting another. The temperature can then either cool or warm depending on the change. The oscillations primarily affect the winds.
JFD
LGL: You believe that GARBAGE presented as fact?
Complete NONSENSE to think that we have reliable temperature data back to the 19th century.
Even more NONSENSE to think it means ANYTHING sans the humidity tied in with it. (See: Average blood pressure.)
Willis, you focus on albedo, but I believe there are two other negative feedbacks operative in your mechanism, both testable in theory. First, the latents heatnof evaporation is released at altitude as Tstorms form percipitation. Itbhas an easier time escaping since there is less or no GHG ‘blanket’. That should show up in Ceres radiative balance for tropical grids in differences between morning and afternoon OLR. Second, the precipitation lowers specific humidity, lowering water vapor feedback, especially in the upper troposphere. Testable with morning/evening radiosonde readings, maybe also with GPS occultation if there is sufficient vertical resolution.
Backed into these because the CMIP3 and CMIP5 models get tropical humidity and precipitation wrong. Which why they model the non-existant tropical troposphere hot spot.
Magnificent post. Should be a paper.
ristvan,
“there are two other negative feedbacks operative in your mechanism, both testable in theory. First, the latents heatnof evaporation is released at altitude as Tstorms form percipitation. Itbhas an easier time escaping since there is less or no GHG ‘blanket’.”
That is actually in the models; it is called the “lapse rate feedback” and significantly reduces the water vapor feedback. It is also, I think, the cause of the predicted “hot spot” since the enhanced latent heat transfer causes warming to be greater aloft than at the surface.
“Second, the precipitation lowers specific humidity, lowering water vapor feedback, especially in the upper troposphere.”
I think that is very like Lindzen’s iris hypothesis.
“Backed into these because the CMIP3 and CMIP5 models get tropical humidity and precipitation wrong. Which why they model the non-existant tropical troposphere hot spot.”
A recent modelling study seems to indicate that including the iris hypothesis improves tropical humidity an precipitation. There is a discussion at: http://judithcurry.com/2015/05/26/modeling-lindzens-adaptive-infrared-iris/
But I don’t think that any of this is easy to test observationally since the mean effect is small compared to the natural variation.
Willis, I got hooked standing outside at night taking astro pictures and really noticed how quickly it cooled.
Averaging the surface data into long periods throws away so much useful data.
I’ve been looking at the difference between how much it warmed during the day, and how much it cools at night.
Here’s another emergent behavior as it warms during the day water evaporates, but when it cools at night it condenses excess water out, some of which ends up in the water shed. So there’s limits to daily humidity, and all of that steam that’s generated over the oceans is transported poleward, where is gets wrung out of the air.
Nice observations, Willis.
I loved your ‘dawn on the left’ and ‘sunset on the right’. I had to think about that one for about ten minutes, before I could place myself on the Sun instead of on the Earth. My normal life is dictated by timezones going the other way…. Makes a nice picture, though – a whole ‘year-day’ in one picture..
We have the same effect at 52 degrees north, but it is dictated more by radiation rather than albedo. The clouds form again at about 10am or so, but perhaps the greatest cooling effect is their transport of moisture and heat high up into the atmosphere, where I presume that more can radiate away into space, without hitting that pesky CO2 and H2O and getting re-radiated back down again.
Half pressure in the atmosphere is at 18,000 ft. So any heat transported to above this level has only half the CO2 and H2O to worry about, before escaping into space. I am fairly sure this regulates the temperature of the atmosphere – you can certainly feel the cold gusts it lets descend down to earth.
Ralph
re: “argue against the entire concept of “climate sensitivity”? ”
Yup, unfortunately even some skeptics seem to be drawn into debating values for climate sensitivity, which is essentially assuming that there is a linear relationship between CO2 and temperature, which seems a rather unjustified leap of faith. Even if past data was fit to a line for some period of time, that doesn’t mean there is a linear relationship that will usefully continue to hold in the future. On a short enough stretch any sort of curve can be appear to be linear, especially with an approximate fit to uncertain data, that doesn’t mean the relationship is actually linear and talking about it as linear can be misleading. Past history doesn’t show a direct correlation between temperature and CO2 over the long run. The whole reason for needing models rather than being able to rely on such an equation is due to the existence of interacting feedbacks (some of which may be step functions, factors only triggered at certain tipping points) and that the fact that other factors do change over time. The *only* use for talking about climate sensitivity seems to be merely for rough reality checks, e.g. comparing approximate ballpark figures to reality check things at the level of a “back of the envelope” calculation that those engaging in such things usually realize are uncertain and not to be trusted. Unfortunately of course non-skeptics seem to have trouble recognizing the degree to which an approximation can be trusted, leaping to conclusions prematurely to an unjustified level of certainty.
Thank you Willis for a profoundly simple illustration of how real critical thinking occurs. Long hours of contemplation, questioning, rearranging, along with the humility and open mindedness to question your own assumptions and biases in search of the truth, rather than validation of ego or personal theory. Bravo! You embody a skill that sooooo many people “think” they have, but clearly do not.
You caused me to think new thoughts today Willis. Thank you!
Willis: Now we’re talking, a good exposition on the necessity of looking where the sunlight actually lands versus averaging over all the places is doesn’t.
All: On a slight tangent to the topic here: The first image in the post reminds me of a basic curiosity that I haven’t seen discussed, and I’m hoping someone can point me to the answer. (I can’t fathom that it hasn’t been done yet.)
Namely, taking a black body curve for 288K and then excising the entirety of the CO2 bands, or even double it for the giggles, what is the necessary increase in temperature such that we have the same area under the curve for a black body at 288k (or whatever) and one that has the given bands excised?
It seems to me that this is a rather crucial boundary constraint on any model or theory of the thermodynamics of climate. Thanks in advance.
I’ve created a spread sheet that does something like what you’re asking.
Takes a surface temp, add some w/m^2 to it, calculates the resultant surface temp.
Reply to let me know what w/m^2 you’d like me to use.
I did take 59F=288.15K and added a full 22w/m^2 additional the surface temp was then 66.5F=292.33K
with 3.7w/m^2 = 60.29F 292.87K
Now, I show that surface temps for 30 of the last 34 years cool more at night than they warmed during the day, this is the excess loss due to a warmer climate, as 50 of the last 74 show excess cooling, which covers the the cooler period (which are slightly more warming than cooling).
So this would be the instantaneous warming of radiant IR, the ground might not warm all that much (grass, trees, etc since they are generally cooler during a warm day), so very little of the ground would warm, but we would feel it to be slightly warmer, until the sun sets and it cools again.
Humid air cools until rel humidity get above 80-90% then condensing that humidity slow nightly cool, hence why tropical nights don’t cool as fast as dry air nights.
You can feed me a net value for the Co2 notch if you don’t like what I chose.
The reply is much appreciated. I think, however, I either didn’t describe what I’m looking for or didn’t understand your response. My fault in either case, so forgive me for attempting to clarify:
1) A basic idea is that if we prevent some band of radiation from exiting a system, then we have induced an imbalance between incoming and outgoing radiation. By normally stated consequence this will cause the temperature of the object to increase until a balance is regained.
2) So if we assume, very roughly and incorrectly, that the Earth is emitting at 288.15K with a perfect black body curve, and then carve out the CO2 absorption bands in toto; we will have created such an imbalance. The temp must increase until equality is reached with the original and unmolested black body curve.
3) But as the temp increases, the curve changes, and so the area excised by the CO2 absorption bands changes. That is, we are trying to find the equality between one unmolested black body curve and another molested black body curve.
This is obviously oversimplified and thusly just wrong with respect to reality on the ground. But there are a number of sanity check sketches about the black body temperature of Earth, Earth with atmosphere, Earth with water, Earth as a grey body, and so on that are in use for just such sanity checks. Or, on occasion used rather perversely as the sole ‘target’ of higher order models.
The idea here being that if any conception starts flirting with such a boundary condition, then we are well in hand to demand a lot of work and answers shown on the subject. Doubly so for any model that uses the black body curves as priors to denote maximal warming/forcing from various values of radiative gas X concentrations.
I hope that helps illuminate my confusions for you. And thanks again for the reply.
I think I understood the first time, but I wasn’t sure what values to use, Willis provided that.
What I wrote was a excel that takes some BB temp, turn it into a W/M^2, then you can add or subtract watts, and then it turns those watt’s/M^2 back to a BB temp.
My thought was that you could add and subtract watts as needed and get the equivalent temp. But when I use Willis’s number 38W’sM^2 make bigger difference. I also have a 1 body, a 2 body, a Cooling in J/day, and surface +/- Watts/M^2 in that spreadsheet.
Unfortunately I will have to do more at another time, but if you have excel maybe you can play around and see if you can get the limits you’re looking for.
I would be interested in how you did it, if you do it.
jquip, you should take a look at the MODTRAN web site. You can set CO2 to zero and see what it does to surface temperature.
The short answer is that using 400 ppmv CO2 as a starting point, if there is NO CO2 outgoing radiation would increase by about 38 W/m2. In order to maintain radiation balance the surface would have to cool about 10°C,
On the other hand, if the CO2 doubles from its current value Modtran says the CO2 absorption will increase by 3.5 W/m2 … which it says will be offset by a temperature rise of 1.6°C.
Bear in mind with Modtran you need to set the elevation of the sensor at 17 km, and subtract the downwelling IR from the upwelling IR … because IPCC, that’s why.
w.
Appreciate the response, Willis. Granted, I’m looking for what the equivalent temperature would have to rise to to maintain the area, rather than as a cooling to maintain the balance. Sadly, I think I broke Modtran. Zeroing out everything except water vapor scale, 17km look down, I get ground temps identical for 0, 400, and 800ppm of CO2.
jquip, try reloading MODTRAN and making sure that you’re not leaving anything blank. It doesn’t like empty cells.
w.
Stable is relative but I see what you mean Willis. I prefer to look at it as an unstable system within boundaries.
Great post Willis! But it seems that this can be handled using conventional methods with feedback defined correctly as I have proposed based on changes in surface temperature, not changes in TOA forcing as IPCC and the climate models do. Your methods shows large negative feedback from increased evaporation with clear skies which moves heat to the atmosphere, some of which goes to space, and also increased albedo from more clouds.
http://edberry.com/blog/climate-clash/g90-climate-sensitivity/improved-simple-climate-sensitivity-model/
Clear at dawn. Then cumulus clouds form usually before noon. Thunderstorms in the afternoon, sometimes lasting into evening or night.
===========
Willis this is not isolated to the tropics. It happens just about everywhere there is a coastline. The land heats during the day, air rises and drags in moist sea air. Clouds form which rain over the land, cooling everything down. At night the land cools, the air descends and flows back out to the oceans to get recharged with moisture and the cycle repeats.
The big difference in the tropics is that it happens everywhere in the wet season (local summer). Over the ocean, over the land, it makes no difference. And if you live right on the equator it happens during traditional spring and fall, when the sun passes overhead, which are summer on the equator. on the equator, traditional summer and winter are winter, as the sun is no longer overhead.
ferd berple June 4, 2015 at 12:17 pm
Thanks for emphasizing that point, ferd. Thermally driven thunderstorms happen all across big parts of the US. And I saw and commented on them in England when I was there.
However, a coastline is neither necessary nor sufficient. I’ve seen the same pattern in New Mexico, a thousand miles from the coast … and on the other hand, I can see the ocean from my house but we almost never get the afternoon thunderstorm pattern.
Best regards,
w.
In Ohio we frequently get the warm day cumulus clouds, but probably don’t have either heat or moisture to get the afternoon storms in”calm” atm.
Also, when we lived in FL in the early 80’s around Orlando east and west we’d get the storms, in the 90’s they stopped but seem they might be coming back, I wondered if they just moved (north?).
And if the surface warms even more and a third temperature threshold is surpassed, yet another phenomena will emerge—
===========
Sumatra’s. The wikipoo article does not do them justice.
http://en.wikipedia.org/wiki/Sumatras
Sumatras or Sumatra Squall Lines (SSL) is a term used in Singapore and Malaysia to describe squall lines that develop over Sumatra at night usually between April and November and then steered towards the west coast of Peninsular Malaysia and Singapore by the southwesterly winds of the southwest monsoon and usually arrives during the pre dawn and early morning.[1][2][3]
http://en.wikipedia.org/wiki/Squall
Severe weather
A squall line is an organized line of thunderstorms. It is classified as a multi-cell cluster, meaning a thunderstorm complex comprising many individual updrafts. They are also called multi-cell lines
Regional terms
Argentina
Known locally as pamperos, these are characterized as strong downsloped winds that move across the pampas, eventually making it to the Atlantic Ocean.[6][verification needed]
Central America
Offshore Central America, a gully squall is characterized by strong increases of the wind forced through sharp mountain valleys on the Pacific Ocean side of the isthmus.
Cuba
A bayamo is a squall emanating from tropical thunderstorms near the Bight of Bayamo.[7]
East Indies
In the East Indies, brubu is a name for a squall[8]
Pacific Northwest (North America)
In the Pacific Northwest, a squall is a short but furious rainstorm with strong winds, often small in area and moving at high speed, especially as a maritime term. Such a strong outflow occurring in fjords and inlets is referred to by mariners as a squamish.
South Africa
Bull’s Eye Squall is a term used offshore South Africa for a squall forming in fair weather. It is named for the appearance of the small isolated cloud marking the top of the squall.[9]
Philippines (West Pacific)
In most parts of the country, squalls are called subasko and are characterized by heavy rains driven by blustery winds. Local fishermen at sea are often on the lookout for signs of impending squalls on the open water and rush to shore at its early signs.
South-East Asia
“Barat” is a term for a northwest squall in Manado Bay in Sulawesi.[9]
“Sumatra” squall is a term used in Singapore and Peninsular Malaysia for squall lines that form over the island of Sumatra and move east across the Straits of Malacca. Gusts can reach up to 28 m/s (100 km/h).[10]
For any warmunists who mistakenly think the earth’s climate is not stable, or that the earth has any remote chance of ‘runaway warming’:
http://www.geocraft.com/WVFossils/PageMill_Images/image277.gif
Willis…Great arguments in defense of your post. I have a question. What effect does the thinning of the atmosphere at higher altitudes have on cumulonimbus clouds, if any? It seems that they sort of blow up and grow larger at top.
It is the rarefied air at higher altitudes where cumulonimbus clouds release their heat and at partial vacuum that heat is….
Willis, excellent essay and discussion. Thank you very much. Thought experiments are fun and have helped me in the past.
Have you ever tried to scale your thermostat process to tornado and hurricane scale? It might be a couple of knotty thought experiments.
Take care.
A corollary is that if the onset moves all the way to the left or right, the equatorial clouds would cease to be regulatory. IOW they’re potential tipping points.
I think there’s no doubt that cloud-induced albedo variations are a part of “the thermostat”. But there’s another part of that thermostat that clouds complicate; radiant transfer of surface heat to outer space. Anyone who has ever had the bright idea of replacing power plant evaporative cooling towers with “spray ponds” has learned that the effectiveness of the spray-pond combination varies so drastically with “night cloud cover” that it renders itself unreliable for the heat rejection purpose intended.
yet another phenomena will emerge
Nitpick Alert! “Phenomena” is plural. “Phenomenon” is singular.
More importantly, thank you for the essay.
“Agendum, an agenda with a single item on it”
I always wonder why should you use an obscure Latin inflection in English, but not in all foreign words.
Just asking. I’d like people to nitpick on incorrect dual forms. (a dual is between singular and plural)
Willis, excellent thinking and point of looking on problem. I think you are right, you identified negative feedback causing upper limit in Earth temperature. I’m from Central Europe, not much sea 1000km around and I could also clearly see effect of increasing cumulus clouds during day. It is clearly function of day temperature and humidity. It can be that 26C is point of maximum temperature of ocean in current, ocean – atmosphere (density, content) setup. It is too much coincidence with 26C stable upper maximum temperature on Earth on Michaels picture above. Any increased radiation input to Earth will result only in increased cloudliness spreading from equator to higher latitudes. Final stage would be whole earth covered by clouds with very high albedo and temperature of 26C from equator to poles…
On the other side your mechanism is not providing lower limit for temperature. There can be some mechanism working on dark side of Earth where decreasing temperature coming to dew point of water will create fog which is blanketing Earth and shifting it from blackbody more toward whitebody. For sure it works in visible spectrum where invisible water vapor in air will become visible white fog reflecting white light. This means less radiated energy to space.
Earth is vulnerable to decrease of temperature, not increase. And I think main driver of temperature decrease is CO2 (surprise) but not through greenhouse effect, but through atmospheric content. CO2 is constantly sequestrated in carbonates by corals and shells. Such carbon is taken from circulation in biosphere and stored for millions years underground. So without replenishing, biosphere will suffocate itself by missing CO2. CO2 is released back only by volcanism. So that means that we are solely dependent on carbon stored in Earth millions years ago and how much of life there was. We are dependent of how much limestone is entering magma chambers under Earth crust and escaping through ocean ridges and volcanos.
So the normal state of CO2 in atmosphere is steady decrease. When it reaches around 260ppm, higher plants like trees starts to suffocate. When it is going to 180ppm only grasses can survive. And that means transformation of Earth from green lush world covered by forests to dry dusty desert and savanna. Deserts and savannas have much higher albedo than forests creating positive feedback for cooling. Glaciers are advancing to lower latitudes decreasing albedo more. Dessert dust in atmosphere is another positive feedback for cooling, increasing albedo further. It is clear from ice core samples that onset of ice ages is accompanied by increased dust content in atmosphere. So Earth becomes cloudless, dry dusty place with a lot of glaciers. Probably there is stable state which keeps balance between low albedo of open cloudless ocean and high albedo of continents and glaciers.
This is quite persistent state in which Earth is whole ice age. As Earth is cooling earth crust contracts increasing inner pressure inside of Earth. Then suddenly threshold is reached, maybe caused by Milankovitch cycle as trigger and sudden increase in volcanic activity starts release of CO2 back to interglacial levels. Higher plants spread again, decreasing albedo, lowering dust, decreasing albedo more. Hydrological cycle recovers, lowering dust spreading plants further and we are for short time back in interglacial.
Willis, we already knew you are a climate scientist. The manner in which you explain your science demonstrates that you are also a philosopher. And your much earlier post on “Here there be Dragons” proves that you are a poet as well. Made the road rise up to meet you……
“I got interested in climate in the late nineties. ”
I got interested in the 1950s and became a member of the Royal Meteorological Society as a student in 1968.
In so far as the water cycle enables albedo variations Willis is correct in that such variations in albedo do indeed make it easier for the system to remain stable than without such variations.
However, the system would remain stable even without water and its phase changes because convection alone will always adjust to remove destabilising influences.
http://www.public.asu.edu/~hhuang38/mae578_lecture_06.pdf
“However, the system would remain stable even without water and its phase changes because convection alone will always adjust to remove destabilising influences.”
I think you are right. Meridonial heat transport is a more important thermostat than changes of the albedo. But it is not so popular as the “sunshade” theory of Willis because the average global temperature goes up when you increase the meridonial heat transport. (This is caused by the nonlinearity of the SB-law). Unfortunately, nature doesn’t care about that.
Paul Berberich said:
“the average global temperature goes up when you increase the meridonial heat transport”
Only if one increases either top of atmosphere insolation or the proportion of that insolation reaching the surface.
In that case you can have a warmer surface and latitudinal shifting is needed to keep energy out equal to energy in.
Since non condensing GHGs such as CO2 affect neither TOA insolation nor albedo they will not influence meridional heat transport and nor will they affect surface temperature because the rate of convective overturning in the vertical plane adjusts instead.
My view is that latitudinal shifting deals with changes in TOA insolation or albedo variations because both are capable of warming the system but the speed of convective overturning in the vertical plane deals with changes in the radiative capability of an atmosphere.
It is necessary to carefully distinguish between the effect on the lapse rates in ascending and descending columns of air when one has, no GHGs at all, condensing GHGs such as water vapour or non condensing GHGs such as CO2.
Each of the three scenarios works out differently and I hope to have an article providing diagrams and more detail ready shortly.
Stated similarly: The earth has self regulated with the negative feedback hydrologic cycle forever. With apologies for repetition here is a version in English:
Yes Virginia, There is a Positive Feedback)
(but it augments the negative water vapor feedback in the earths climate)
Water vapor has been the primary atmospheric energy transfer and earth cooling agent since the earth first acquired a hydrosphere. It would stretch logic beyond credulity to think that it suddenly has become a warming factor in the climate energy balance equation.
There has been in the literature an assertion that although the CO2 alone greenhouse effect would only raise the earth temperature by ~1 C per century, that an additional positive feedback by thermally excited water vapor would increase this climate warming by a factor of between 1 and 6 to reach the catastrophic warming promoted in the UNIPCC Summary for Policy Makers. This letter addresses the absurdity of that assertion and hopes to refocus the point of view of the climate study community to further address quantitatively this area of atmospheric physics.
There is in the lower surface level of the boundary layer, a positive water vapor feedback which enhances the transfer of surface energy into the latent heat of vaporization of water. This enhances the cooling rate of the surface and enhances the primary heat energy transport to the level at which there can be radiation to space. It is not unusual in complex systems having outer loop negative feedback to find minor positive feedback loops within which augment the performance of important processes inside the system. These in no way detract from the overall system behavior. Water vaporization, vertical energy transport and radiation control the negative feedback response to any internal or external forcing that might drive climate temperature to change.
Stated simply; positive water vapor IR feedback exists and is limited by IR’s short, mean free path to the lowest strata of the boundary layer. Increased energy rate transfer in this strata is limited to the bulk surface energy transfer rate response but adds nothing to the total latent energy flow. Water vapor energy absorption and transport is the negative feedback factor driving the earth thermostat. Stated otherwise: Increased water vapor or carbon dioxide is not an additional radiation source since it has no internal source of energy but functions only as an incremental enhancement to energy transfer rate across the surface-atmosphere boundary. The point of view that adding water vapor and CO2 to the atmosphere will increase downward radiation and heating is backwards. They have no source of energy to radiate. Instead, looking up from the surface we see that surface IR is totally captured and thermalized by these gasses in the bottom surface layer of the atmospheric boundary layer. The vaunted 3.7 watts/m2 increase CO2 radiation to space is not relevant to its effect in this part of the atmosphere.
This discussion regarding water vaporization applies to the 90% of the earth covered by water in the form of open surface liquids and transpiring vegetation. The desert regions have their own less constrained radiation physics.
To tediously belabor the point in English:
Water vapor is of course the major ‘Greenhouse’ ie IR absorbing Gas’. By virtue of its broad spectral response as well as its dominating volumetric ratio it is dominant in the atmospheric boundary layer all the way through the troposphere to its final condensation and radiation escape altitudes. It captures IR from the surface radiation in the first several IR mean free path lengths in the first several meters above the surface. CO2 is also active in this region. By virtue of its volume density some 12 to 50 times smaller than water vapor in this region its mean free path is more than 10 times longer than water vapor. IR will thus encounter ~many attenuation lengths of water vapor absorption before encountering one absorption length of CO2. So we see that additional CO2 in the energy transfer equation will be minimum. Its effect could only serve to slightly lower the vertical level at which all IR is captured and thermalized in the boundary layer Any potential increment to the positive water vapor feedback loop slightly increases the rate but cannot increase the total of captured IR beyond that which water vapor has already done. Available IR absorption is complete in this bottom strata of the atmosphere. This short distance radiation inherent in any ‘greenhouse gas’ heat transfer thus serves only to improve the effective conductivity of air in this strata of the boundary layer. This conductivity is in series with and does not bypass the conductivity/convection resistance of the surface mass energy flow to the evaporating and radiating surface.
*ref. Complete thermalization
To recapitulate:
The total energy transfer rate is limited to the bulk surface mass conductivity/convectivity energy transfer rate supplying energy to the surface and is further limited by the cooling effect of surface water evaporating into the accumulating local relative humidity. This surface heat transfer process is thus rate enhanced by the positive water vapor feedback but self limited by these physical heat transfer rate limits. The water vapor positive feedback merely reduces the surface temperature required to effect the energy transfer from the surface. It cannot increase the energy transfer beyond that available through the surface mass.
Regardless of the balance of IR radiation capture resulting from these effects, CO2 influence on the energy transfer rate from surface to boundary layer is minimal to none. Relative to water vapor its share in capturing IR is not important but as we shall see any captured IR by any ‘greenhouse’ molecule is a desired benefit to increase the efficiency of energy transfer across the surface boundary and to the water vaporization process and thus to the ultimate water vapor energy convection transport to the radiation level in the troposphere.
Water being an IR absorption molecule it has always had its positive feedback influence on the boundary layer temperature by virtue of the complete thermalization of the captured IR., Due to spectral overlap in the 15 u band this is also true though to perhaps lesser degree than the CO2 contribution in the 2 u wide lower side band slice of unsaturated spectrum. However it plays out between conduction and radiation into the top 10 u film of surface liquid water, the end effect is still total thermalization of the IR in the lowest reaches of the boundary layer. To complete the positive feed back loop, any net thermalized air and any direct radiation of IR contributes to further vaporization of water from the surface and contributes to atmospheric convective lift of the water vapor entrained in the warm rising columns of air.
This local positive water vaporization loop through direct radiation and air thermalization does not add any energy to the system but merely lowers the resistance and thus enhances the rate at which the available energy from the surface converts its heat energy into thermalized air and latent heat of water vaporization. In other words and terms it increases the gain and therefore the rate in a surface-power limited positive feedback energy transfer loop from surface to air and water vaporization but cannot increase the total amount of energy transfer since the surface mass is the the only source of energy. This positive feedback water vapor loop is already essentially power limited (saturated) and thus can have no significant response to additional IR capture by additional CO2 gas. In feedback system terms; the loop is already driven by positive feedback to its power supply limit. If surface mass heat energy flow constraints allowed it to respond faster, it would already have evaporated more water to do so.
The surface transfer would take place without water vapor positive feedback albeit at a higher surface temperature, slower air conduction/convection limited rate. The positive feedback merely speeds up the process toward the surface energy rate limit, it generates no more total water vapor or energy but brings the surface energy transfer to its physical limit more efficiently at a lower surface temperature than would be required in a conduction-only limited heat transfer. In this respect a local delicately balanced positive feedback factor is not required to prevent runaway since the process is power limited to the rate at which the surface can transfer energy from its thermal mass to the water-air boundary. Thus positive feedback is not to be feared since it makes the surface cooling more efficient and promotes the normal hydrologic cooling cycle of the climate by enhancing latent heat capture in the water vapor at a lower surface temperature.
This provides a strong negative feedback to the surface temperature rising and to the overall climate response to either internal or external forcing, since the evaporated water with its latent heat of vaporization dominates the surface cooling effect by enhancing transfer of energy from the surface. Convection of this latent heat to the mid and upper level of the troposphere where water, both liquid and vapor dominates the total radiation spectrum to space is the major factor in the earth cooling energy balance.
Above the final condensation and freezing level the increased (doubling) CO2 content of the thin atmosphere would have a minor effect due to its low partial pressure and density and perhaps narrowing side band spectral lines. The final IR radiation temperature (~217K) of the TOA CO2 spectrum implies that its radiation to space takes place at the tropopause level where the vaunted colder temperature due to negative lapse rate is diminishing to zero. This further brings into serious question that there can be any significant CO2 ‘greenhouse effect’ anywhere in the earth climate system but that on balance additional CO2 may have at most a slight net cooling effect on the climate. Ref. IR radiation enhancement.
A note:
Perhaps this helps explain why the ocean thermal energy increase is on the order of ½ watt/m2 or less and does not reflect the predicted (1.6 watt) CO2 energy increase let alone any positive feedback energy increase effects.