Editorial

Winds and Rain: The Role of the Biotic Pump

By Peter Bunyard

In 2007, Anastassia Makarieva and Victor Gorshkov correct explanation for the conundrum of the western
came up with the Biotic Pump Theory (Makarieva A. a., reaches of the Amazon Basin, such as that of Colombia,
2007). The two scientists, mathematician/physicists at receiving as much rainfall, if not more, than at the Atlantic
the Petersburg Nuclear Physics Institute, St Petersburg, Coast 2,500 km distant.
Russia, described the Biotic Pump Theory (BPT) in terms
For them, the Trade Winds (vientos alisios), flowing
of the fundamental laws of physics pertaining to the
between tropical Africa and converging at the Intertro-
phase change when water evaporates and then condenses
pical Convergent Zone (ITCZ) over the Amazon Basin,
through the process of cloud-forming, as humid air rises
were drawn in by the high rate of evapotranspiration and
in the troposphere and chills with altitude. The degree of
then condensation over the closed-canopy rainforest.
water vapour saturation follows the Clausius-Clapeyron They based their conclusion on the evapotranspiration
equation for ideal gases, taking into account temperature rate per square metre of the rainforest, with its multi-
(Kelvin), relative humidity and barometric pressure. levels of leaves, being an order of magnitude greater
In essence, Makarieva and Goshkov took exception to than the rate of evaporation over the ocean at the same
the standard climatological idea that the Hadley Cell latitude. Makarieva and Gorshkov describe the force
air mass circulation between Africa and South America generated by the partial pressure change as the evapora-
was caused primarily by latitudinal differences in tempe- tive force Fe in hPa/km and they postulate that a gradient
rature between the tropics and the equatorial zone of is created between the Fe over the ocean and that over
the rainforest, acting to draw in the air from the ocean to
the Amazon Basin. They theorized instead that air mass
the continent (Makarieva A. M., 2013).
changes, brought about through condensation and preci-
pitation, would lead to atmospheric pressure changes If the Biotic Pump theory is correct, the consequences
such as to draw air upwards to the cloud-forming region, of widespread deforestation in the Amazon Basin will be
more or less at an altitude of 2.5 km and beyond. They considerably more devastating in terms of the hydrolo-
then proposed that the air which replaced that moving gical cycle and reduction in precipitation compared with
upwards would come from surface air flowing over the a scenario where the air mass circulation, including the
ocean, bringing humid air to the continent interior. They Hadley Cell and Walker circulation, results primarily from
suggested that the biotic pump theory provided the latitudinal differences in air temperature.

Recibido / Received: 02 de Febrero del 2018  Aprobado / Aproved: 03 de Marzo de 2018

Tipo de artículo / Type of paper: Investigación Científica y Tecnológica terminada.
Afiliación Institucional de los autores / Institutional Affiliation of authors: Cambridge University
Autor para comunicaciones / Author communications: [email protected]
El autor declara que no tiene conflicto de interés.
By Peter Bunyard

Based on measurements that the average distance of fall- flow of humid air from the tropical Atlantic ocean and
out of a molecule of water vapour, as it evaporates and over the rainforest will continue, irrespective of whether
then precipitates, is some 600 km, Makarieva and Gors- the land is forested or not. In general, climate models
hkov calculated that the central and western regions of predict that the mid and western regions of the Amazon
the Amazon Basin would desertify. That claim of deser- Basin would convert to savannah. On the basis of the
tification, with rainfall reduced over the Colombian BPT, deforestation of the Amazon Basin would remove
Amazon to less than 1 per cent of today’s value, is to be the Fe gradient and, therefore, would reduce the flow of
contrasted with the projection of only a 12 to 15 per cent the Trade Winds. Such a reduction in the flow of humid
reduction in rainfall according to those climate models, air from the ocean would result in the forming of desert
primarily temperature-based, which determine that the rather than savannah.

Figure 1. Deforestation impact according to the biotic pump theory. Makarieva & Gorshkov (Makarieva A. M., 2007) use
a length, l, of 600 km, as the distance over which a molecule of water will evaporate and fall-out through precipitation.

PX = P0 exp −  x 
The formula,  l  describes the reduction in precipitation as the distance increases between the
coast and inland.

2500
Intact forest

2000
Conventional models
Precipitation P mm/yr

Impact of deforestation
1500
Deforestation and failure
of biotic pump
1000
Manaus 13% 5.4% intact coastal forest
coastal deforestation
500
Colombian Amazon
0.625%
0
0 1000 2000 3000 4000

Distance Km (x) from Ocean

Makarieva and colleagues then cite, as evidence for the In essence the Biotic Pump Theory maintains that:
BPT, the precipitation history of those river basins that
1. The rainforest provides the moisture for cloud
are forested compared with those that lack forest (see forming by means of a high rate of evapotranspira-
Fig. 2 below). They find that river basins such as those of tion (ET);
the Amazon, Congo and Yenisei show no decline in rain-
2. The rising air results in condensation and precipitation;
fall and even a slight increase with distance inland. That
finding is in sharp contrast to river basins without forest: Leading to
they show an exponential decline in rainfall with distance 3. Monsoon rain;
from the coast, just as the formula in Fig. 1 indicates. For
4. An abrupt reduction in local pressure when clouds form;
example, the West African river basin shown in Fig. 2 has
a precipitation history of 1600 mm at the Atlantic coast. 5. The vertical motion of air from the forest canopy upwards;
Some 1,700 km away from the coast, the annual precipi- 6. The drawing in of horizontal surface air to replace
tation has reduced to just over 20mm (Fig.2). the air flowing vertically upwards.

6 ¦   Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15

 Winds and Rain: The Role of the Biotic Pump

Figure 2. The left-hand diagram is for forested river basins, with the X-axis showing distance, x, from the coast and the
Y-axis showing an exponential annual precipitation rate. The annual precipitation is seen to rise with distance from the
coast. The right-hand diagram shows river basins without forest. In all instances, the annual precipitation rate declines
exponentially with distance from the coast.

6400 3200

3200 Amazon
1600
Congo
1600 800
P, mm por año

P, mm por año
800 400
Yenisey N América, 40º N
400 200
N. Australia
200 100 NE China, 42º N
Argentina, 45º S
100 Argentina, 31º S
50

50 25
W Africa

0 500 1000 1500 2000 2500 3000 0 500 1000 1500 2000 2500 3000

Kilometres desde el océano (x) Kilometres (x) desde el océano

Derivado de A.M. Makarieva & V.G. Gorshkov, Hydrol. Earth Syst. Sci., 11, 1013-1033, 2007 Derivado de A.M. Makarieva & V.G. Gorshkov, Hydrol. Earth Syst. Sci., 11, 1013-1033, 2007

By means of the biotic pump, moist trade winds Ocean, the Amazon Basin functions on a far grander
are pulled in and feed moisture to the forest. The scale than the Chocó (Salati, 1987). Marengo showed
forest recycles more than 50 per cent of that mois- that evapotranspiration rates were sufficiently high over
ture by means of evapotranspiration. the rainforest to increase the volume of the Amazonian
atmospheric river in the form of the South American
Aside from the precipitation data for river basins, Low Level Jet Stream as it approached the western
what further evidence do we have for the biotic reaches of the Amazon Basin and was then deflected
pump? A recent pan-tropical study of rainfall and land- both upwards and southwards on encountering the
cover, as indicated by the leaf area index (LAI), led Spracklen Andes (Marengo, 2006).
and his colleagues to the conclusion that satellite-derived
rainfall measurements could be positively correlated with On reviewing daily meteorological date from
the degree to which model-derived air trajectories were La Selva Biological Station in Costa Rica, Peter
exposed to forest cover (Spracklen, 2012). Bunyard found a significant correlation of an
association between surface air flow and abso-
Poveda and colleagues provide evidence that when the lute humidity changes during daylight hours. The
low level jet streams pass over forested regions preci- absolute humidity changes in hectopascals were
pitation levels stay high and constant, whereas over suggestive of changes in the rate of evapotrans-
regions which lack forest, precipitation levels decline piration. In reviewing the data on a daily basis
exponentially, just as the BPT suggests should happen. between October 2013 and April 2014, Bunyard
Poveda and his colleagues look at the low level Chocó jet discovered an average of approximately 10 pulses
and comment that the change in direction of the Pacific of absolute humidity during 12 hours of daylight
Austral Trade Winds from Easterlies to Westerlies just which coincided with an equivalent number of
over the Equator at 4° N, may owe their abrupt switch to pulses in surface airflow, the wind direction being
the unsurpassed degree of evapotranspiration and subse- from the Atlantic Ocean. In general, the airflow
quent condensation over the Chocó in Colombia (Poveda pulses came within 30 minutes of the pulses in
G. a., 2000) (Poveda G. L., 2014). surface humidity, the lag in time suggesting the
With more than 380,000 cubic metres per second of time necessary for the humidified air at the surface
water vapour brought in from the Tropical Atlantic to flow upwards and reach condensation altitudes.

Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15   ¦ 7

By Peter Bunyard

Figure 3. Data from La Selva, Costa Rica, October 21st, 2012. The airflow changes are delayed by 30 minutes. The X-
axis shows the hours of the day.

2 4,00

1,5 3,00

1 2,00

Change in absolute humidity, hPa

Change in airflow, m/s

0,5 1,00

0 0,00

6 8 10 12 14 16 18
-0,5 -1,00

-1 -2,00

-1,5 -3,00

-2 -4,00

Windspeed plus 30 min rate change fE Rate Change

Figure 4. La Seva, Costa Rica, hydrological data July 14th, 2011.

Comparison between absolute humidity & wind

1,5 4,00

3,00
1

2,00

Change in absolute humidity, hPa

0,5
Wind change m/s

1,00

0 0,00

0 2 4 6 8 10 12 14 16 18 20 22 24
-1,00
Hour
-0,5

-2,00

-1
-3,00

-1,5 -4,00

Wind speed change, m/s Partial Pressure change hPa

The table gives the average monthly rates of peak change of hours of 6:00am and 19:30pm. For instance, from one day’s
the evaporative/condensation force, fE, (hPa/km), of winds- results as seen in Fig. 3 and Fig. 4, we obtain some 9 to 11
peed (m/s) and of temperature (°C) between the daylight peaks for both variables, namely fE and windspeed.

8 ¦   Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15

 Winds and Rain: The Role of the Biotic Pump

Data from La Selva, Costa Rica

Evaporative/condensation force
Date Windspeed (m/s) Temperature (°C)
(hPa/km)
October 2013 10.06 9.94 9.55
November 2013 10.47 9.87 9.9
December 2013 10.42 10.68 9.74
January 2014 10.35 10.52 9.58
February 2014 10.14 10.29 9.68
March 2014 10.29 10.29 9.84
April 2014 10.39 10.25 9.71

The meteorological data from La Selva, Costa Rica, equatorial Tropics and consist of a double sinusoidal
also provide insight into the daily gross changes to the wave, with their peaks and troughs following the same
barometric pressure. Such changes are typical of the pattern throughout the year.

Figure 5. Data from La Selva, Costa Rica, showing the barometric pressure wave from March to May, 2012
and from May to June, 2012. The peaks and troughs occur at the same time over 24 hours (X-axis). The si-
mulated curve is derived from the formula below (Plamen Netchev & Peter Bunyard), with coefficients, P1, P2.

The coefficients enable the simulated curve to follow closely to the actual data.
nt nt
Bar pressure p = p0 + ( p1 − p2 cos ) * sin
12 12
La Selva
1012

1010

1008

1006

1004

1002
0 4 8 12 16 20 24

Average La Selva data March to May 2012 Simulated Bar with p1, p2 Simulated Bar with p4, p5

La Selva averaged Bar, May to June 2012 Simulated Bar for May to June 2012

As stated before the barometric pressure wave is a at the time of the September equinox, when the daylight
phenomenon of the equatorial Tropics with 12 hours length was close to 12 hours, followed a similar pattern
of daylight and 12 of night. From observing data from a to that obtained throughout the year for La Selva, Costa
boreal forest region in Finland, Jokioinen, Latitude 60° Rica., A similar, less clear form of the barometric pattern
48’ 50.44’’, Bunyard found that the barometric pressure could be seen in the March equinox data for Jokioinen.

Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15   ¦ 9

By Peter Bunyard

Figure 6. The barometric pressure wave for Jokioinen, Finland averaged over 10 years during the September and
March equinoxes. The pattern is similar to that obtained throughout the year for La Selva, Costa Rica.

September 1961 - 1970 March 1971 - 1980

1013,2 1012,4

1012,2

1013
1012

1011,8
1012,8

1011,6

1012,6 1011,4
0 5 10 15 20 0 5 10 15 20

Figure 7. The data from Jokioenen, showing the barometric wave over 24 hours.

June 1961 - 1970 December 1961 - 1970

1010,5 1011,6

1011,4
1010

1011,2

1009,5
1011

1019 1010,8
0 5 10 15 20 0 5 10 15 20

Could the curves, with their sinusoidal waves, be asso- 1999 and December 2000, measured the rate of conden-
ciated with the solar diurnal cycle and, later in the day, with sation over the Southern Tropical Pacific (17.5 – 20S)
the hydrological cycle? The pressure wave of Jokioinen in and obtained the curve as shown in Fig.8 ( Wood, 2002).
June, when the daytime length is at its maximum, shows A comparison of the ‘liquid water path’ derived from the
a hint of the first part of the wave to be contrasted with a atmosphere over the Pacific Ocean with the simulated
drawn-out second wave. During the ‘shortest day’ at the cosine curve of the barometric pressure wave shows an
time of the December solstice, the second wave does not extraordinary coincidence, as in Fig. 8. The indication is
appear, whereas the first wave does. that changes in the partial pressure of water vapour are
responsible for significant changes to the pressure wave
Some answer to that question may be obtained from data during daylight hours. Those changes, indicating that
derived from the tropical Pacific Ocean. Wood, Bretherton condensation and cloud-forming are taking place, are of a
and Hartmann, over a two-year period between January sufficient magnitude to affect airflow (Bunyard P. P., 2014).

10 ¦   Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15

 Winds and Rain: The Role of the Biotic Pump

Figure 8. : Atmospheric liquid water path as determined over 24 hours in the Tropical Pacific reveals similarities with
the cosine component derived from the simulated barometric wave for La Selva. Source: Peter Bunyard, 2011; Wood,
Bretherton, & Hartmann, 2002.

[82.5-85W, 17.5-20S, Jan 1999-Dec 2000 Cosine component of Bar Wave

150 1,5

100 0,5

0
1 4 9 14 19 24
50 -0,5

TMI -[+error]
-1
Sinusoidal fit

-1,5
0 5 10 15 20

time of day [hr local]

means of a double layer of cooling coils, resulted in

Experimental Evidence the condensation per second of a few grams of water
of the Biotic Pump vapour. A Gill ultrasonic 2D anemometer measured any
airflow, including its directionality. By applying ideal
To gain further evidence of the association between gas physics to the air in different parts of the structure
the condensation of atmospheric water vapour and so as to measure changes in specific (absolute) humi-
airflow, Bunyard designed and carried out experiments dity, correlations between the rate of condensation
in a special chamber in which some 20 cubic metres and airflow can be determined (Bunyard P. P., 2017),
of atmospheric air were enclosed. Refrigeration, by (Bunyard P. P., 2019).

Figure 9. The experimental chamber for measuring absolute humidity changes and airflow. The lean-to is the laboratory.

Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15   ¦ 11

By Peter Bunyard

When water vapour condenses through cooling the air, the implosion of air surrounding the locus of condensa-
a measurable wind can be detected which circulates tion. That energy is some 1000 times greater than that
around the structure. The correlation between the rate of air density changes, the latter being insufficient by an
of condensation and the unidirectional wind flow is order of magnitude to account for the measured airflow.
highly significant. By use of basic physics, (Bunyard P. P., The following physical equations account for the energy
2019), the kinetic energy for the airflow is derived from derived from condensation:

∆Pa 3
J= m = 1000∆Tv and ∆Tv = 0.621∆qT
∆t

Where J ( joules) is equal to the partial pressure change heat capacity of dry air at constant pressure is 1,000
(pascals) over time by volume and is the reduction J kg-1K-1. Meanwhile, is the absolute humidity change
in temperature by volume as the surrounding air per volume and is the temperature of the surroun-
expands into the space vacated by water vapour. The ding air.

Figure 10. Experiment June 27th, 2016. The graph shows 4 refrigeration cycles. The left-hand axis shows the partial
pressure change in water vapour in watt.seconds during the refrigeration cycle and the right-hand axis shows the
anemometer readings. The directionality is clockwise and therefore down from the cooling coils.

Figure 11. Experiment 30th January, 2016. The left-hand axis shows the partial pressure change as a parcel of air is
cooled. The right-hand axis shows the airflow. During the four refrigeration cycles the airflow is directed upwards,
against gravity and therefore requires additional work for the air to move.

12 ¦   Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15

 Winds and Rain: The Role of the Biotic Pump

Climate models currently do not account for partial pres- fore approximately equal and opposite to one-twentieth
sure changes from water vapour condensation being a of latent heat energy) translates to some 5 Ws for every
prime cause of air mass circulation, thus raising the possi- square metre of forested land, as if occurring throughout
bility that the exclusion of the biotic pump from such 24 hours.
models brings about a gross underestimate of the conse-
quences of widespread deforestation (Fig.1). Whereas 5 Ws would give airflows of 2.85 ms-1, daytime
pulses of absolute humidity from ET (some 10 pulses
In addition, the pulsing of the absolute humidity encoun- during daylight hours) could give airflows of 7 ms-1 (Trade
tered in the meteorological data from La Selva Biological Winds) and require a condensation rate of 30 pascals per
Station in Costa Rica (Figs. 3 & 4) opens the possibility second per cubic metre (30 Ws). The 5 Ws is the average
that the rainforest synchronizes the opening and closing over 24 hours and, on the assumption that transpiration
of its leaf stomata at particular times of the day. The occurred in pulses during daylight hours, the 30 pascals
above experiments show clearly that the airflow velocity per second of condensation and partial pressure change
depends on the rate of condensation. Consequently, it per cubic metre would be readily obtained.
would be advantageous on evolutionary grounds, were
some synchronization of evapotranspiration to take place A preliminary experiment with an isolated hydrangea
so as to increase the rate of cloud-forming over the forest. plant (Hortensia) indicates that the absolute humidity
measured at the surface of separate leaves on different
The evapotranspiration (ET) over the 5 million square stems appears to be synchronized in contrast to a control.
kilometres of the Brazilian Amazon amounts to an Further experimentation is required on rainforest species
average of 1,370 mm per year (Maeda, 2017). The kinetic to see whether such synchronization of surface humidity
energy associated with the implosion as water vapour is a shared property both on individual specimens and
from evapotranspiration condenses into clouds (there- more widely.

Figure 12. Experiment on Hydrangea, 5th July, 2019. Tests on separate leaves on separate stems of a Hydrangea
potted plant compared to the control.

Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15   ¦ 13

By Peter Bunyard

Figure 13. The biotic pump in action. Source: Peter Bunyard, diagram by Andrew Ayres

In conclusion, empirical evidence suggests that the rimental evidence. DYNA, 84(202), 92-101. http://
biotic pump is no longer an abstract theory, but is a prin- doi.org/10.15446/dyna.v84n202.61253
ciple which needs urgently to be accounted for in those
[9] Bunyard, P. P. (2019). Further experimental evidence
climate models which pertain to rainforest regions such
that condensation is a major cause of airflow.
as in the Amazon Basin. Were deforestation to continue
DYNA, 86(209), 56-63. http://doi.org/10.15446/
at its current 2019 pace in the different countries of the
dyna.v86n209.73288
Amazon, in particular Brazil, but including Colombia,
the consequences for South America could be devasta- [10] Maeda, E. E. (2017). Evapotranspiration seasonality
ting. Bogotá, for instance, obtains its fresh water via the across the Amazon Basin. Earth Syst. Dynam., 8,
upper moorland páramos and, in turn, they depend on 439-454. doi: 10.5194/esd-8-439-2017
the air mass circulation over the Amazon to deliver rain.
Were that circulation to collapse because of the abrupt [11] Makarieva, A. a. (2007). Biotic pump of atmospheric
reduction in the biotic pump brought on by deforesta- moisture as driver of the hydrological cycle on land.
tion, countries such as Colombia would face a dramatic . Hydrology and Earth System Sciences (11), 1013-
shortage of fresh water. 1033. doi: 10.5194/hess-11, 2007
[12] Makarieva, A. M. (2013). Where do winds come
References from?. A new theory on how water vapor conden-
sation influences atmospheric pressure and
[7] Bunyard, P. P. (2014). How the Biotic Pump links dynamics. Atmospheric Chemistry and Physics, 13,
the hydrological cycle and the rainforest to 1039-1056. doi: 10.5194/acp-13-1039-2013
climate: Is it for real? How can we prove it? Bogotá,
Cundinamarca, Colombia: University of Sergio [13] Marengo, J. A. (2006). On the hydrological cycle of
Arboleda. http://repository.usergioarboleda.edu. the Amazon Basin: A historical review and current
co/bitstream/handle/11232/397/How%20the%20 state-of-the-art. Revista Brasileira de Meteoro-
biotic%20pump.pdf?sequence=2 logia., 21(3), 1-19.
[8] Bunyard, P. P. (2017). Condensation and partial [14] Poveda, G. a. (2000). On the existence of Lloro
pressure change as a major cause of airflow: expe- (the rainiest locality on Earth): Enhanced ocean-
land-atmosphere interaction by a low-level jet.

14 ¦   Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15

 Winds and Rain: The Role of the Biotic Pump

Geophysical Research Letters, 27(11), 1675-1678. [17] Spracklen, D. V. (2012). Observations of increased
doi:10.1029/1999GL006091, 2000. tropical rainfall preceded by air passage over forests.
Nature, 489, 282-285. doi:10.1038/Nature11390,
[15] Poveda, G. L. (2014). Seasonal precipitation 2012
patterns along pathways of South American low-
level jets and aerial rivers. Water Resour. Res., 50, [18] Wood, R. B. (2002). Diurnal cycle of liquid water
98–118. doi:10.1002/2013WR014087, 2014. path over the subtropical and tropical oceans.
Geophysical Research Letters, 29(23), 7-1 - 7-4. doi:
[16] Salati, E. (1987). The forest and the hydrological 10.1029/2002GL015371
cycle. In R. Dickinson, Geophysiology of Amazonia.
(pp. 273-296). New York: Wiley and Sons.

Autor

Peter Bunyard

Peter Bunyard received his MA honours in Natural Sciences from Cambridge University in 1964 and in
the same year an MA from Harvard University for his research in insect physiology which culminated
in the publication of RNA Metabolism in Pupae of the Oak Silkworm, Antheraea pernyi: the effects of
diapause, development and injury. Proceedings of the National Academy of Sciences. Vol 52, No 6, pp
1572-l580, Dec 1964. He was a founding editor of The Ecologist), 1969 to 2005 and has a number of
books to his name, including The Politics of Self-Sufficiency, OUP, 1980, Gaia in Action: Science of the
Living World, Editor, Floris Books (Edinburgh), 1996, Breakdown of Climate: Human Choice or Global
Disaster? Floris Books, 1999, Extreme Weather: The Cataclysmic Effects of Climate Change, published
December, 2008(Barnes & Noble), Caos Climático: Amenaza a la Vida en el Planeta Tierra. Editorial
Educar, Bogotá, Colombia, August 2010. Segunda impresión, August 2011, and, recognised by Colciencias
as original research, How the Biotic Pump links the hydrological cycle and the rainforest to climate: Is
it for real? How can we prove it? Instituto de Estudios y Servicios Ambientales – IDEASA, Escuela de
Ciencias Exactas e Ingeniería, Universidad Sergio Arboleda, Bogotá, Colombia, Edición: febrero de 2014.
He was elected a Fellow of the Linnean Society of London in 1995.

Revista de Tecnología ¦ Journal of Technology ¦ Volumen 17 ¦ Número 1 ¦ Págs. 5-15   ¦ 15