What Is Wrong With Dynamical Detrainment

First uploaded on 2023/02/10
Last updated on 2025/04/23
Copyright(C)2022-2025 jos <jos@kaleidoscheme.com> All rights reserved.

[NOTE]
If you are uninitiated for Dynamical Detrainment at all, please see the page of What is Dynamical Detrainment.

[NOTE]
This article is based on our interpretation for the Dynamical Detrainment (DD). The interpretation might be old, since we have not updated our understanding on DD due to lack of research environment. However, unless there have been very drastic changes in the DD constitution, the following statements would be still correct.

We are simply advocating Radiatively Driven Circulation (RDC) as a transporting mechanism around cumulus, and, of course, we have neither intention at all of attacking other people nor of confusing the research community. Unfortunately, however, RDC, which is based on the physics of the mean field of the atmosphere outside the cumulus clouds, and Dynamical Detrainment (DD) which is based on the physics of the disturbance within the cumulus clouds, are fundamentally different and mutually exclusive mechanisms. If the RDC is correct, it alone can be responsible for transport around the cumulus, and if any transport such as DD other than the RDC occurs, it can even be harmful, breaking the equilibrium state of the atmosphere which RDC would maintain.

The reasons for our belief that the RDC is right can be found throughout this website. Here we summarize what is wrong with DD to make people aware of the dangers of DD. The contents of linked items are detailed in the linked pages, while the contents of other items are scattered throughout the website. Before continuing your DD research, try to negate each of the following items. If you cannot deny even one of them, you may need to reconsider your current approach.

• DD Violates The Second Law of Thermodynamics
  

  Cumulus clouds are merely domains where mixing of the lower and upper atmospheres occurs. The motion of air in a cumulus cloud is a mass of fine vortices, and the physical scale of each individual vortex is too small to create an outflow of large physical scale from the cumulus cloud. The effectiveness of small-scale flows on which DD is based is questionable.

  We can assure you that continuing DD study will never give you an answer. The crucial reason is that DD violates the second law of thermodynamics. Some people say that the second law of thermodynamics is a trivial one that cannot be positively used. That is incorrect. It is a premium lifesaver that protects us from falling into the trap of believing in physical phenomena that seem to be true but never happen. The first step in considering any physical phenomenon is to sift it against the second law of thermodynamics to see if it is possible. DD researchers are putting the convention of meteorology above this fundamental law of physics. The second law of thermodynamics is exactly what is to be dedicated to them today.

  [NOTE]
  Dynamical entrainment is physically natural because it is part of mixing. It does not violate the second law of thermodynamics. On the other hand, dynamical detrainment is physically impossible, it violates the second law of thermodynamics. Dynamical detrainment just sounds like a play on words: If ENtrainment happens, then DEtrainment should happen. We must determine in advance whether or not a phenomenon is possible in the light of fundamental physical laws.



• Transitions of Kinetic Energy Spectrum
  

  The violation of the second law of thermodynamics by the DD method is fundamental, but it is difficult to convince people who are based on their intuition. Therefore, for those people, we show that the DD method is impossible in principle by presenting the transitions of the kinetic energy spectrum in a visual and easy to understand way.



• Anvil Is Not Evidence For DD
  

  Anvil might be the original source of the DD idea, but looking closely at the anvil photos, it seems not necessarily the physical basis for DD. The validity of chimney-type convective motion, with dynamical outflow from the top of it, is questionable.



• DD Has Methodology, But No Principle
  

  DD research is mainly focused on reproducing the outflow from cumulus domain in the current atmosphere using an ensemble of chimney-type convection. On the other hand, the underlying principles of DD have not been established yet and are still a subject of discussion. Therefore, even if DD can reproduce the current atmospheric phenomena, it is not certain that it can explain the response to global warming. The approach of methodology preceding principle is only accepted for weather forecasting. Applying the meteorological method in the same way to climatological phenomena is questionable.



• DD Is Likely To Result A Too Dried Atmosphere
  

  Since the altitudes of DD outflows are defined as the altitudes at which the air parcels lose buoyancy, they tend to be concentrated near the tropopause in general. As the amounts of water vapor in air parcels reaching this altitude are very small, when such air parcels flow out and occupy the clear-sky region outside the cumulus clouds, the total amount of water vapor in the atmosphere becomes very small. This is not consistent with the condition of the current atmosphere. (Of course, we know DD researchers have been trying to reproduce a humid atmosphere.) In addition, during warming, more activated cumulus convection forces the altitude of the tropopause even higher, so even drier air will flow out and occupy the clear-sky region. The effect of reduced water vapor, a greenhouse gas, works to suppress warming when it occurs. This is negative warming feedback due to water vapor, which is also contrary to the current warming acceleration.
  Ultimately, DD will significantly underestimate the water vapor feedback to global warming. (See References for details.)



• DD Breaks The Natural Radiative-Convective Equilibrium.
  

  The equilibrium state of the atmosphere does not change significantly each time a cumulus cloud develops. (Changes in equilibrium, i.e., climatic changes such as warming, occur over long periods of time that alter the thermodynamical properties of the atmosphere.) Cumulus clouds repeatedly develop within the same equilibrium state. This is because the motion of cumulus clouds is merely the mixing of air within the cumulus domain. However, in DD, the flow associated with the cumulus is represented by an ensemble of chimney-type circulations that extend to the outside of the cumulus based on the physical attributes within (or near) the cumulus domain. Such a flow affects the clear sky outside the cumulus in an undesirable way. When a DD occurs, the heat or mass balance in the clear-sky region will always be broken if we treat the DD flow effect strictly. Atmospheric models including DD that claim to be able to maintain equilibrium are either simply ignoring this undesirable effect on equilibrium, or introducing even more unnatural flows to mask it.

  The DD method mainly studies the flow mechanism from the inside of the cumulus to the exit of the cumulus flank. However, it is the flow after the exit that is difficult. If they assume that the dynamically driven flow within the cumulus is free to exit without regard to the thermodynamic and continuity structure outside the cumulus, then the outflow must break the thermodynamic and continuity equilibrium conditions outside the cumulus.

  In order for the cumulus outflow assumed by DD to maintain atmospheric equilibrium, it is necessary to consider thermodynamical balance and mass continuity in the clear-sky region outside the cumulus, as RDC does. However, once the physics of the clear-sky region is taken into account, it is clear that only RDC can do so in a consistent manner. Furthermore, once RDC is introduced, it alone can provide the necessary and sufficient transport of mass, heat, and water vapor from the inside to the outside of the cumulus. Therefore, DD is no longer necessary, or more specifically, it should not be there at all.

  There are indeed special types of disturbant motion allowed even in the troposphere where thermodynamical equilibrium is realized and mass continuity is ensured. These are wave phenomena such as sound waves and internal gravity waves. However, these waves are regarded to be linear and each air parcel in motion is only oscillating in situ; they are not possible to transport mass/heat/water vapor. When you want to consider general disturbant phenomena with transport other than linear waves, you must correctly treat their thermodynamical effects and their effects on mass continuity as departures from the equilibrium state in the atmosphere.

  Ultimately, DD attempts to focus only on the disturbance component of the transport, but completely ignores its effect on the mean field. Such treatment was effective in weather forecasting, where only the disturbance component (precipitation, strong wind) was important, but is a major problem in climate prediction, where the mean component is what is important. And even more surprisingly, the mean component RDC alone is sufficient to provide required transport.

  (See the main article for details.)



• DD Assumes Convection-Dominant Atmosphere, But Radiation Appears to be Dominant in the Real Atmosphere
  

  It is likely to think of the flow associated with the visible cumulus clouds as providing transport around them, which leads to the DD concept. This concept assumes that convection dominates the transport process. According to the results from our simple atmosphere models (as well as in the real atmosphere), however, the troposphere is in radiative-convective equilibrium, and rather than seeing convective motion as dominant, radiative process appears to be more dominant.

  (See

  • Two Types of Circulation in the Atmosphere
  • Convection-Dominant or Radiation-Dominant
  for details.)



• Some DD Methods Assume an Analogy to Mid/High-Latitude Cloud Dynamics
  

  Some figures describing the DD method may assume baroclinicity by the analogy to mid/high latitude cloud dynamics. However, it is at low latitudes, where the barotropic atmosphere covers half of the globe surface, that the outflow from cumulus clouds becomes climatologically important. And others may assume a cumulus ensemble to explain the multi-altitude outflows from cumulus clouds to humidify the surrounding atmosphere. However, the cumulus clouds at low latitudes are usually solitary, and a single cumulus cloud alone must account for the humidifying multi-altitude outflows.

  Thus, it is a mistake to apply the DD method, which is based on mid/high-latitude atmospheric dynamics, to the transport around cumulus clouds in the low-latitude atmosphere. This tendency may be unavoidable, considering that many climate researchers come from the field of mid/high-latitude meteorology, but it should be clearly recognized that the atmospheric properties and the cumulus cloud types are different between mid/high-latitude and low-latitude atmospheres.

  On the other hand, the RDC mechanism is independent of baroclinicity or cumulus ensemble, and can explain multi-altitude outflows from a single solitary cumulus cloud in a barotropic atmosphere at low latitudes. It is general enough not to be restricted to low latitudes, but can also be expected to work for outflows from mid/high-latitude disturbances.

• Absence of Bénard Cells and Presence of Gravity Waves
  

  If transport by convective dynamics were dominant in the atmosphere, it would not be surprising that structures such as Bénard cells appear in the atmosphere. Although we can see the Bénard cells in the Convective Boundary Layer (CBL) at the bottom of the troposphere, where convection is dominant, not a single Bénard cell is observed around an isolated cumulus cloud in a clear sky area in the middle troposphere. This means that, contrary to intuition, convection in the troposphere is unlikely to create a closed large circulation except in the CBL. Instead, cumulus convection evolves into complex overlapping vortices that eventually emit kinetic energy in the form of gravity waves, which is commonly observed. This is due to the thermal and density stratifications of the atmosphere, the short duration of cumulus convection, and the long time required for radiative cooling. This typical inefficiency of convective circulation is also seen in the results of our simple dynamical model DCM.

DD is intuitively easy to accept and appears plausible at first glance. When viewed from a different perspective, however, DD seems not only improbable, but a process with no physical basis at all. To us, it looks as if they are trying to build a perpetual motion machine. (Yes, DD and the perpetual motion machine are exactly the same in ignoring the second-law of thermodynamics.) The past half-century might have been paid for people's intuitions and preconceptions.

As a matter of fact, soon after computers became capable of handling strongly nonlinear phenomena in fluids, high-resolution simulations of the motion of a moist thermal plume assuming a cumulus cloud were also performed. We have seen some of the reports. They commonly showed only a mixing process, i.e., a large plume taking in external air from the surroundings, breaking up into fine vortices as it rises, and finally filling the whole domain with the fine vortices. There were no outflows from the plume as in the so-called dynamical detrainment. Researchers believe that there must be some other dynamical process that is missing from these simple simulations, and they continue to try to add various dynamical processes that they call "dynamical detrainment". However, from RDC's point of view, such a dynamical process does not exist. In other words, the initial simulations for the moist thermal plumes were sufficient to express the essence of cumulus motion (namely, mixing) in terms of dynamics. What they lacked was a correct assessment of thermal balance and mass continuity outside the cumulus; the radiative cooling occurring in the atmosphere, the subsidence flows that balance it in thermal equilibrium, and the horizontal flows that compensate for the mass divergence in the subsidence flows. Adding complex processes just makes the model more difficult to understand and black-boxes it. The only way to contribute to the understanding of a phenomenon is to describe it with the correct minimum of detail necessary.

In the case of analyzing the results of our very simple radiative-convective cloud-resolving model, DCM, we initially assumed that DD was correct and tried to explain the cumulus outflow on that basis. However, we could not find any significant relationship between the outflow from the cumulus and the physical attributes of the disturbances in the cumulus. Not at all. We wasted more than half a year on this DD study. Once we turned our attention to the mean structure of the atmosphere outside the cumulus, however, we found that the outflow from the cumulus had a very clear relationship with the thermodynamical equilibrium and mass continuity of the atmosphere outside the cumulus, and we were able to extract the RDC. Our modelling study and the modelling studies currently being conducted around the world should be essentially the same. They differ only in size: the complexity of the model, the computing environment in which it is run, and the number of people involved in the project. The research period is also much longer, more than half a century. It seems to us that the climate research community is now repeating on a much larger research scale what we did initially on a smaller research scale. If that is the case, then there is nothing to be gained by continuing DD research. We believe we have a duty to suggest this to DD researchers. However, in our experience, when we suggest RDC to DDs, they usually laugh at first and then become furious. This is because RDC is so far removed from DD, RDC is the opposite of DD. Before getting emotional, however, we would like you to calmly judge which is physically correct. If the DD researchers cannot make a calm decision, we would like to ask for advice from fluid dynamics and thermodynamics people who have not been directly involved in climate research. They have no preconceived notions of meteorology.

We are doing our duty by presenting this website about the dangers of DD and about its alternative, RDC. Although only brief and verbal descriptions are presented on this website (except for the main article), we are ready to produce more rigorous reports showing the DD error as soon as we have the research environment in place. As for the actual application of RDC parameterization to the practical climate prediction models, however, there is nothing we can do on our own, because we have neither a practical climate model nor a running environment for it. We only hope that someone in the research institutes that actually run the numerical models for climate prediction will be interested in RDC.

(At this time point, the only work remaining in the RDC parameterization is writing code to divide the entire area into partitions that RDCs cover according to the distribution of radiative cooling rates in the horizontal plane for each altitude, and parameter tuning that specifies how the RDC in each partition should be limited when there is insufficient mass flux available in the supplying source [cumulus] domain.)

Of course, we are always willing to abandon the RDC scheme if someone points out a flaw in it that we find acceptable. Please give the RDC scheme serious consideration. The worst thing you can do is ignore our advice and stick with existing methods that have not worked.

If we are lost in a deep cave and someone says he sees a light on a side path, it is imperative that we first make sure that the light is not from the outside world before continuing on our original path. Or, when a rower sees from the end of the rowers' seats a large waterfall ahead of the boat which is going at full speed, it is his duty to warn the crew and suggest an alternative route if possible. The first thing the crew have to do before kicking him down into the water is to investigate carefully which route of the two is right. Especially if there are 8 billion passengers on board.