Guest Post by Willis Eschenbach
I came across a lovely photograph of a “fire devil”, also called a “fire whirl”. I liked it because the photo perfectly exemplified what is wrong with the current generation of climate models.
What is wrong with the models is that they don’t include any of the vortex-based emergent atmospheric phenomena like fire devils.
Let me start with the concept of “emergent phenomena”. Emergent phenomena are phenomena which:
- Emerge spontaneously from the background when certain thresholds are exceeded. Below the threshold there are none. Above the threshold, the number emerging can increase very rapidly.
- Have a lifetime.
- Move, adapt, and change in response to environmental conditions.
- Eventually dissipate, fade away, and die out.
- In addition, emergent phenomena generally are not naively predictable from looking at the underlying conditions.
Here’s a way to understand naive predictability when it comes to emergence. Suppose we were members of a tribe on a remote island where clouds never formed. Day after day, the sun came up and went down in a cloudless sky.
Imagine if, after generations of living like that, one day people looked up and were terrified to see a large, white, seemingly solid object had formed out of nothingness right above them! Would it fall? Would it harm them? The priests did incantations and read entrails. They prayed nothing would happen to the people. Nothing happened. The priests were given extra respect, they were clearly the cloud-masters.
Of course, once it happened over and over, day after day, people would give it a name, like “tropical thermal-generated cumulus cloud”, or maybe “cotton-ball cloud”, and go on living. Day after day. More generations pass. Life is boring again. Every morning around eleven, the first cotton-ball cloud appears. In a very short time, the sky is covered. It brings blessed cooling by blocking out the hot tropical sun. The cloud priests wax fat and multiply.
Then after centuries of such bliss, imagine if one day, one of the nice friendly cotton-ball clouds in the sky suddenly grew taller and taller, until it towered menacingly above the people. A sudden wind came up out of nowhere, and things got cooler. Then, the rain pounded down … consternation! Water from the sky! The priests claimed it was from their prayers and incantations.
Finally, just when the priests were about to finish saying their orisons to the cloud gods, there was a blinding bright white flash and then a huge sound … and of the eleven priests, four were untouched, five were burnt in strange patterns by the wrath of the sky-god, and two were dead.
The remaining living priests said they had escaped by virtue of their arcane arts, and were worshipped until their deaths as lightning priests and sons of the lightning gods.
Now, all of those phenomena, the tropical clouds, the rain, the cumulus growing into a thunderstorm, the lightning, every one is an emergent phenomenon. They all emerge spontaneously when a certain threshold is passed. They all exist for some length of time. They move and change based on environmental conditions. At the end of their lifetime they all fade, dissipate and die.
And as the priests in the story found out … none of these emergent phenomena are naively predictable from knowledge of the conditions from which they emerge.
With that as an introduction to emergent phenomena, let me return to the fire devil.
To me, it seems that far too often, climate scientists are looking for causes rather than looking at effects. This is particularly true with emergent climate phenomena like clouds and thunderstorms. Too many climate scientists ask “Why do thunderstorms form? What causes them?”.
Me, I try to avoid looking at causes for emergent phenomena. Instead, I consider their actions. I ask “What do they do? What is different from when they are born to when they die out? What is their overall effect?”
So, why are all of these emergent phenomena important to the climate and thus to the climate models?
Because all of them have the same effect—they cool the surface. Cumulus clouds cool the surface by reflecting the sun back to space. Storm-generated wind cools the surface by greatly increasing evaporation, just as a fan cools a sweating person. And tropical rain plus the entrained vertical rain-wind can leave you shivering even on a warm day.
These emergent phenomena all cool the surface because they are all generated in response to the surface being much warmer than the atmosphere. They are all heat engines, driven by what is called “delta T”, a temperature difference between the surface and the atmosphere.
The global climate models cannot model thunderstorms, much less have them spontaneously emerge from a relatively uniform background. One problem is that their gridcells are too large, much larger than thunderstorms and most other emergent phenomena that cool the surface.
So … consider the irony:
The climate models are attempting to model the surface temperature without including the very phenomena that cool the surface.
But wait, it’s worse than that. In addition, the models don’t contain one of the more common ways of moving energy aloft, the humble vortex. Not only is it common, it is incredibly efficient. Consider the photo of the fire devil again, which I’ve re-posted below.
Look at the effect of the fire devil. Rather than heating the mass of the surrounding air, and rather than mixing smoke and other byproducts with the surrounding air, a vortex functions like a pipe. It pipes the hot air and the combustion byproducts through the surrounding air with only the most minimal of mixing. Look how all the smoke is contained in the vortex, with no visible surrounding cloud of particles.
One of the most common and most important climate examples of a vortex is the tower of a thunderstorm cloud. It is a huge pipe-shaped vortex moving an incredible amount of warm air in a vertical helix from the base of the cloud to the upper troposphere. This vortex is totally contained within the cloud and is not interacting with the surrounding atmosphere either physically or radiatively.
Consider the full voyage of a bit of heat moved by a thunderstorm from the surface to the top of the troposphere. At the surface, the heat evaporates some water, cooling the surface. Then it is carried as latent heat up to the underside of the thunderstorm by the vertical circulation under the storm.
Inside the thunderstorm base, the incoming water vapor condenses, releasing the latent heat as sensible heat. This constant source of heat from condensing water vapor is what stokes first the vertical development of the thunderstorm tower vortex, and then pumps massive amounts of warm air up the vortex to the top of the troposphere.
Note that just as in the fire devil above, the air in the thunderstorm tower is NOT interacting in any way with the surrounding atmosphere.
Here’s the curious part. This means that there are escape holes in the greenhouse effect. Consider it once again from the surface upwards.
A bit of heat evaporates some water at the surface. It is now latent heat in some lifting parcel of air. Because as latent heat it doesn’t warm the air parcel, it doesn’t increase the shortwave radiation. It’s not interacting radiatively with the atmosphere.
After the latent heat is lifted up through the bottom of the thunderstorm, it condenses as sensible heat. But it is condensing inside the cloud, so once again there is only the most minimal of radiative interaction with the atmosphere.
And this isolation from the surroundings continues as the re-warmed air parcel travels up the vortex inside the thunderstorm tower. Only after the air parcel emerges from the top of the atmosphere, along with a few ice crystals, does the air parcel start interacting radiatively with the surroundings.
And of course, at that point it is far above all of those pesky greenhouse gases, and free to radiate to space.
Now, think about this a minute. There are actual physical tunnels through the greenhouse effect which let surface heat escape directly to the upper troposphere.
Through these vortex-driven vertical pipes inside thunderstorm towers, surface heat is rapidly spiraled vertically to high altitudes where it is free to radiate to space, untouched by the greenhouse effect. Estimates are that at any time there are on the order of 2,000 active thunderstorms on earth.
Now, if these escape holes for excess heat were located randomly it would be one thing. But they are not.
Instead, they form exactly where they are most effective—over a local area that is warmer than its surroundings. They preferentially cool the warmest parts of the surface.
And this is why I started by saying that the fire devil shows the problem with the current generation of climate models. The current models don’t have a couple thousand self-organizing escape tunnels for surface heat that form spontaneously and preferentially over the hottest parts of the modeled surface.
And that’s not including dust devils and waterspouts, the more pedestrian but much more numerous cousins of the giant vortex in the thunderstorm …
Finally, consider that fire devils and dust devils and cumulus clouds and thunderstorms are all driven by delta T, the difference in temperature between the surface and the atmosphere. In general the peak difference is in the mid-day and afternoon, when the surface is warm. The emergence of the escape tunnels for surface heat that we call thunderstorms typically occurs in the afternoon.
This means that the proliferation of thunderstorms once the local threshold is exceeded is a large part of what prevents the tropics from overheating. They bleed off excess energy very efficiently, removing immense amounts of energy from the hottest parts of the surface in the hottest parts of the day and moving it high in the sky. Get rid of it before it causes trouble.
(Thunderstorms can also continue into the night, in part because thunderstorms are a dual-fuel heat engine. They can run on either warm air or humid air, and they generate their own humid air from winds under the base of the storm. But I digress …)
Finally, the existence of spontaneously emerging escape tunnels for excess heat exactly where they are most needed means that temperature is NOT a simple function of forcing as the climate models assume. Instead, it implies a practical upper limit on surface temperature.
Best of a warm summer afternoon to you all, stay safe, dodge the storms, I send good thoughts to those in smoky air, high winds, pounding rain, and all the vagaries of the atmosphere.
My Usual Request: Misunderstandings start easily and can last forever. I politely request that commenters QUOTE THE EXACT WORDS YOU DISAGREE WITH, so we can all understand your objection.
My Second Request: Please do not stop after merely claiming someone is using the wrong dataset or the wrong method. I may well be wrong, as may anyone, but such observations are not meaningful until you add a link to the proper dataset or an explanation of the right method.