Introduction

Main publication
(abstract)

Overview of the results

Comments and responses
[you are here]
List [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

De Melo Jorge Barbosa H. (2006) Interactive comment on "Biotic pump of atmospheric moisture as driver of the hydrological cycle on land" by A. M. Makarieva and V. G. Gorshkov. Hydrology and Earth System Sciences Discussions, 3, S1466-S1469. www.cosis.net/copernicus/EGU/hessd/3/S1466/hessd-3-S1466.pdf
The comment discusses the main biotic pump publication.
Informal title of the comment:
Still about the physics

Makarieva A.M., Gorshkov V.G. (2006) Interactive comment on "Biotic pump of atmospheric moisture as driver of the hydrological cycle on land" by A. M. Makarieva and V. G. Gorshkov. Hydrology and Earth System Sciences Discussions, 3, S1492-S1498. www.cosis.net/copernicus/EGU/hessd/3/S1492/hessd-3-S1492.pdf

In response to the second set of comments by H. de Melo Jorge Barbosa:

The author's reply has cleared some of my doubts, but I still have some important points which I would like to discuss.

It does not seem to me that the latent heat release during condensation is considered anywhere in the calculation. I think this is important since it warms the air, increasing the local pressure, and thus decreases the total pressure gradients and the effect of the evaporative force.

Condensation and evaporation are reversible processes. If at a given temperature in a closed system "liquid water" - "saturated water vapor" there occurred condensation, the released latent heat would warm the system. This would immediately lead to additional evaporation accompanied by the corresponding drop of temperature to its initial value, so that on average the net flux of molecules from the gaseous to liquid state (condensation) is zero. Therefore, condensation can only occur when the system's temperature is externally lowered, for example, when the system loses heat or moves to a colder environment. In accordance with the Le Chatelier principle, the release of latent heat can only decrease the drop of temperature that initiated condensation, but it cannot in principle make the air warmer than it had been before its temperature started to drop.

In the stationary case (and namely this case is considered in our paper, p. 2638) the release of latent heat due to condensation of the upwelling water vapor is taken into account in the stationary value of the observed lapse rate of air temperature G_ob, which would have been larger if the atmosphere had been dry. Expression for the evaporative force (14), as can be seen from formulae (10)-(12), includes both the latent heat of evaporization and the value of G_ob.

It should be noted that, as was discussed in our reply to Dr. Sherman (p. S1133, lines 20-29), condensation and latent heat release do not directly generate dynamic processes in the atmosphere. Introduction of the evaporative force provides clue to the so far unresolved fundamental physical question of the atmospheric circulation theory (Lorenz, 1967), namely by means of which physical processes solar energy, the apparent driver of atmospheric processes, is converted into the kinetic energy of moving air masses. The energy of solar radiation spent on the evaporation increases the amount of water vapor in the atmosphere; this enhances the evaporative force, which accelerates air masses to the observed velocities. The dynamic energy of moving air masses is transformed to heat due to friction. The energy conversion process is completed with the release of latent heat in the course of condensation of the upwelling water vapor. Latent heat is transformed into thermal energy of air molecules and ultimately leaves into space in the form of thermal radiation. As is well-known, the power of the flux of latent heat significantly exceeds the dynamic power of air circulation.

2) Precipitation before RH=100%

With the author's theory, precipitation occurs due to excessive moisture in the atmosphere. However, it is well known that condensation occurs at relative humidity well below 100% because all different kinds of aerosols act as cloud condensation nuclei. Therefore, well before the vertical profile is fully saturated, there will be precipitation, which reduces specific humidity, pressure gradients and lapse rate, etc...

Relative humidity is defined with the reference to the saturated water vapor concentration above a plane surface of pure water. Saturated water vapor concentration above the curved surface of pure water droplets is higher than the reference concentration. Saturated concentration of water vapor above various solutions (e.g., NaCl) is lower than the reference concentration. Due to this fact the observed values of relative humidity corresponding to the initiation of condensation in the atmosphere can be somewhat lower or higher than 100%. If, on average, condensation occurred "well before" 100%, as stated in the comment, for example, at 80%, which is the mean relative humidity at the surface, we would have seen precipitation originating just at the surface rather than at a certain height in the atmosphere. This is apparently not the case and condensation occurs at RH close 100%. In any case, condensation of water vapor at RH < 100% would have only enhanced the evaporative force, as it would lead to even further compression of the vertical distribution of atmospheric water vapor.

An atmospheric model does not use only one parameterization for describing convection.

Usually there are:

- Deep convection parameterization, which occurs, as the authors say, in unstable atmospheric columns (G > 6.5 K/km)

- Shallow convection parameterization. This takes into account vertical motions of moisture and energy without precipitation and happens for G < 6.5 K/km, as the evaporative forced does.

- Large scale condensation, which removes excessive water (if any, after deep and shallow cumulus parameterizations) from the column.

I would like to know what is the difference between a shallow cumulus parameterization and the authors evaporative force theory, as they both do the same: vertical transport of moisture and heat well before deep convection develops.

The principal difference between the established tradition of describing convection and the evaporative force theory, as referred to in the comment, lies in the following. In the conventional consideration the force that generates air movements is lacking. Instead, one operates with the notion of instability leading to turbulence; properties of turbulent eddies are further considered from the statistical viewpoint by considering various order moments of the studied variables. As mentioned in the paragraph on the Archimedes force in our paper (p. 2637), when averaged over a spatial scale linked to the vertical scale of the convective region, in the employed parameterization schemes there are no force-induced movements of air masses (see, e.g., Moeng and Wingaard, 1989). Since there is no dynamic driver of air motions, the major quantitative parameters and functional dependencies, including those on the subgrid scale, have to be borrowed directly from, or fitted to, observations, with little or no theoretical clues as to why they are such as they are and what physical processes and fundamental physical characteristics of the Earth's environment could determine their values. Reflecting this situation, it is widely admitted that the modern representation of atmospheric convection in GCMs is a parameterization, not a theory.

The proposed physical approach based on the evaporative force aims at describing air motions based on the fundamental physical principles. In our paper we have introduced the evaporative force acting in the troposphere and described its large-scale implications, including the continental biotic pump. The planetary boundary layer, where the processes of shallow cumulus convection develop, has its well-known peculiarities. They are manifested, in particular, in the diurnally variable vertical lapse rates that are often much larger in absolute magnitude than the mean tropospheric lapse rate. In section 3.4 "Water preservation by closed canopies" we discussed several issues regarding the implications of the evaporative force mechanism at the surface layer, see also (Gorshkov and Makarieva, 2006; Makarieva, Gorshkov and Li, 2006, in press). In our response to Dr. Dovgaluk we showed how the regional mean Bowen ratio can be quantified using this approach; the problem of hurricane formation, also outlined in our paper, awaits further studies. Generally, we believe that the application of the evaporative force mechanism to every atmospheric problem, including the description of shallow convection, can be expected to yield meaningful results, as would do consideration of an important process previously unaccounted for.

I think that the 10% change in the temperature and the other 10% change in the value of cannot be neglected. For instance, a 20% change in the author's estimates in sections 3.2 and 3.3 (either making f_ev larger or smaller) would bring their estimated values out of agreement with observations.

As can be deduced from the paper's text, we do not neglect the change of tropospheric temperature. Formula (8) that describes the decrease of equilibrium pressure with height does take into account the change of temperature with height. The approximate formula for isothermal atmosphere, that is given in the text (p. 2634, line 11) and lacks a number, is not used in any derivations. As we explained in our previous response, this formula is retained in the text for the explanatory sake, as it helps to visualize, by comparison with the exact formula (8), that an account of the temperature drop makes pressure decrease with height more rapidly than in the exponent describing the isothermal atmosphere. Regarding the value of , its value in the troposphere (h ~ 8 km) changes by only 5%. Moreover, Eq. (10), which is the basis for estimating the fundamental parameter , is exact with respect to temperature- related changes of , as noted also by Dr. de Melo Jorge Barbosa in his first comment.

After the authors answer to my first comments, I agree that the upward motion of the water vapor molecules will make the whole moist air moves upward. However, this means that the nitrogen and oxygen will become into a state where their partial pressure at the surface are less than the weight of their masses. Therefore, as soon as the moist air start to move upwards, there will appear a partial pressure gradient pointing downwards. This difference of concentration will induce diffusion of unbalanced gases N₂ and O₂ from top to bottom, and the final equilibrium situation should be such that, for all constituents, partial pressures are in equilibrium with gravity at all heights. The downward diffusion of the initially hydrostatically balanced gases should reduce the evaporative force.

Question to the authors: what would be the velocity of the downward diffusion of N₂ and O₂ in the moist atmosphere case? Is it really much slower than the upward H₂O velocity that it can be neglected?

It is rightly noted in the comment that the action of the evaporative force making moist air rise should change the vertical distribution of the other atmospheric gases as well. This question has been already quantitatively tackled in this discussion, see the responses to Dr. Sherman (pp. S1131-S1132) and to Dr. Dovgaluk (pp. S1179-S1180).

Briefly, the evaporative force acting on moist air as a whole make air parcels rise; when they expand, the relative amount of various dry air constituents does not change, yielding a constant mixing ratio of the dry air (p. 2643). When dry air has a constant mixing ratio and hence constant molar mass of M = 29 g mol^-1 and a single scale height h = RT/(Mg), all dry air gases appear out of hydrostatic equilibrium. Gases with M_i < M (such as N₂) appear to be vertically compressed as compared with their equilibrium distribution with a scale height h_i = RT/(M_ig) > h; these gases diffuse upward. Gases with M_i > M (as O₂, CO₂ etc.) are "overstretched" compared with their equilibrium distributions; these gases diffuse downward.

As we have shown, osmotic forces acting on dry air gases, that can be introduced in a manner similar to the evaporative force, precisely compensate each other in case of height-independent M (response to Dr. Sherman, pp. S1131-S1132), yielding hydrostatic equilibrium of dry air as a whole. Moreover, diffusional fluxes caused by the non-equilibrium distribution of dry air gases are all significantly smaller by their absolute value compared to the upward dynamic flux of moist air induced by the evaporative force (response to Dr. Dovgaluk, pp. S1179-S1180), which therefore appears to be the dominant vertical process.

- Specific comments on the authors answers to my first comment -

The authors give an example of two different gases in two different compartments. I believe that in this case there is will be motion (contrary to what was stated by the authors) since different gases will have different molecular masses. Initially, the center of mass (CM) will not be at the center of the box, but closer to the heavier side. At equilibrium, the CM will be exactly at the center of the box. In fact, doing this experiment in vacuum, one would see the box oscillate with decreasing amplitudes before reaching its final position. If done on a workbench, the workbench would exert a force on the box that would be responsible for changing the position of the CM.

The authors say: "we do not state anywhere in the paper that it is the same molecules that evaporate into the atmosphere that immediately condense"... but on page 2636, lines 1 to 3, it is written:

Since you say: "We do not base our consideration on the Archimedes force and, hence, we do not need to prove this statement." I think you should take "Archimedes paragraph" out of the paper.

Discussing the box horizontally divided into two compartments filled with different gases at equal pressure. While it is said in the comment that the center of mass will move to the center of the box, it will only do so in the case when the gases have different molar masses. If these are different gases having equal molar masses (e.g., isomers), the center of mass will not move anywhere being in the center of the box from the very beginning. Generally, speaking of motion in the atmosphere one usually does not mean motion of the center of mass, but motions of air masses, i.e. winds. No wind will occur in the box where two compartments are filled with different gases at equal pressure. Moreover, in the course of the diffusion process the vertical position of the center of mass does not change. Thus, if the box is much heavier than the gases within it (as is Earth with respect to the atmosphere), the impulse of the moving center of mass will be absorbed by the box and no motion of whatever object will be observed, even if the gases have different molar masses.

Regarding the water that evaporates and undergoes condensation "immediately at a microscopic distance above the surface" (p. 2636). We now see that this is a language issue. As follows from both comments of Dr. de Melo Jorge Barbosa, this phrase can be interpreted as implying that those very molecules that evaporate immediately condense. (Note that due to the quantum principle of identity of molecules, it is impossible to tell for the condensing molecules whether they are the same molecules that have just evaporated, or they are those having resided in the atmosphere for some time, unless the gas and the liquid have different isotopic composition).

To our knowledge, the word "immediately" has at least two meanings, a temporal - "at once" and a spatial - "just there", see, e.g., the Princeton University vocabulary (wordnet.princeton.edu), which gives an example of the usage like "he passed immediately behind her". It is in this second sense that we used this word, emphasizing that condensation occurs at a microscopic distance of the order of one path length from the liquid surface, right there and not further up in the atmosphere.

Gorshkov, V.G., Makarieva, A.M.: Biotic pump of atmospheric moisture, its links to global atmospheric circulation and implications for conservation of the terrestrial water cycle, Preprint No. 2655, Petersburg Nuclear Physics Institute, Gatchina, Russia, 2006.

Lorenz, E. N.: The nature and theory of the general circulation of the atmosphere, World Meteorological Organization, Geneva, 1967.

Makarieva, A.M., Gorshkov, V.G., and Li, B.-L.: Conservation of water cycle on land via restoration of natural closed-canopy forests: Implications for regional landscape planning, Ecological Research, 21, 897-906, 2006.

Moeng, C.-H. and Wyngaard, J. C.: Evaluation of turbulent transport and dissipation closures in second-order modelling, J. Atm. Sci., 46, 2311-2330, 1989.