Biotic pump: Comments and questions: 6
Main page in English Russian version of current document BIOTIC REGULATION OF THE ENVIRONMENT Welcome to discussion of biotic regulation!

Nature photos News Demography Contacts Hot topic Publications
topical :
: recent
Your questions Biotic regulation
in simple words

Biotic Regulation has moved.
You are on the first web site of Biotic Regulation.
Since February 14, 2008 it is no longer updated.
Welcome to!

Hot topic: Biotic pump of atmospheric moisture



Main publication

Overview of the results

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




Anonymous (2006): Critique of spatial precipitation patterns.

The comment discusses the first biotic pump publication.

(1) It appears to me that Eq. (1) of your preprint assumes water vapor convergence to be proportional to the total flux of water vapor. Doesn't water vapor convergence in a given area depend mainly on the convection over that region?



Makarieva A.M., Gorshkov V.G. (2006): Response to anonymous critique of biotic pump regarding the spatial precipitation patterns.

Local water vapor convergence is, by definition, the input of water into the local area. In the stationary case it is equal to runoff R. Moisture is brought to any given area at a distance x from the ocean via the atmosphere, we denote this flux as F(x). While a unit atmospheric volume travels over a small unit distance dx, some part of its moisture can precipitate. Of this precipitated moisture some part evaporates back into the atmosphere during the same time while the unit atmospheric moisture is still over dx. This precipitation-evaporation cycle will causes therefore no change in F(x). However, some part of precipitated moisture will not return to the atmosphere, but will be lost to runoff. Namely this part will constitute the change of the flux F(x). Hence, simply from the law of matter conservation, the change of F(x) over unit distance dx is equal to runoff R, dF(x)/dx = R. This is the first part of our Eq. (1).

It is further clear that the amount of water locally lost to runoff cannot be greater than the amount of water brought to the local area via the atmosphere. The larger the amount of water brought to a given area via the atmosphere, the larger -- all other conditions being equal -- the runoff. In other words, as we said in the preprint, the probability that a water molecule condenses, precipitates and joins runoff, is independent of the time it spent in the flow. This is the ground for the second part of Eq. (1), namely that dF(x)/dx = -F(x)/l and (its integrated form) F(x) = F(0)exp(-x/l).

For the runoff we have R(x) = [F(0)/l]exp(-x/l) and for precipitation P = kR we have P(x) = [kF(0)/l]exp(-x/l).

The particular magnitude of runoff will depend on what is going on in the atmosphere. If there are rapid (as compared to the flow velocity) upward air motions, the moisture will condense, precipitate and be lost to runoff more quickly. This will be reflected in a smaller value of l. Conversely, when little moisture precipitates, the value of l will be larger -- in this case the moisture flows over land practically without interacting with it. Hence, what is referred to by convection in the comment is reflected in the value of l.

See also comments by Prof. Savenije and our responses.


(2) For the sake of understanding, consider the case of Amazonia vs the Brazilian Northeastern area. These are two regions very close, both edging the ocean, and both receiving a similar flux of water vapor from the ocean [F(0), in your nomenclature]. But the convection over Amazonia is much greater than that one over the Northeast, therefore it appears to me that there is much more convergence over Amazonia than over the NE. I gather from the rational in your preprint that the L constant for the NE would be much greater than that for Amazonia, because the NE vegetation evaporates much less. But prevailing winds over the NE are basically parallel and after follow into the direction of the much rainier state of Tocantins [the area on the edge of Amazonia to the east]. How can one harmonize these facts with your theory?


(2) We are not sure we have understood the above comment clearly. As we have shown in our response to your first question, the convection and precipitation processes are reflected in parameter l, not L, the latter being the length of the river basin where the biotic pump is in action. Indeed, as far as the NE vegetation evaporates less, the evaporative force and the upwelling air fluxes are weaker in this region than on the verge of Amazonia.

If there were no biotic moisture pump, then the value of l = l2 at Tocantins, where the vegetation is richer, would be smaller than it is over the NE Brazil, l = l1, l1 > l2. In this case we would have seen that F(x) would have been first changing only slowly over NE Brazil, with little precipitation and little runoff, then more rapidly over the richer vegetation and stronger convection, see the enclosed scheme. According to the formulae above, precipitation would have a peak first at the border of NEB and Amazonia (caused by change of the vegetation cover and of parameter l from l1 to l2), but then decrease even more rapidly than over the caatinga, see figure below:


Schematic representation of how flux of moisture F(x) and precipitation P(x) would change over two adjacent land regions both 1000 km in length. In the first region l = l1 = 700 km (little precipitation, little runoff, slow change), in the second region l = l2 = 100 km (large precipitation, large runoff, rapid change).

Whatever is the strength of convection, precipitation over uniform surface (with no radical changes of vegetation as in the considered example) should decrease exponentially.

The existence of forest moisture pump over the Amazon changes this pattern completely and eliminates the exponential decline altogether. Note also that the fact that the Atlantic winds reach the edge of Amazonia after having passed the arid NE Brazil is possible only due to the high cumulative power of the Amazonian biotic pump. If the situation were the reverse (a wide arid zone and a narrow forest), winds would not reach the forest, dying out at several hundred km from the coast, as is, e.g., the case at 15 oS to the south of Amazon (see Fig. 5 of Marengo (2004)).

This particular case of NE Brasil is but one of the many special cases which present no conceptual difficulties for explanation, but that cannot be all explained in a single paper.


Main page in English Russian version of current document   Modified 5 November 2006.
© 2006 V.G. Gorshkov, A.M. Makarieva