EDUCATIONAL SOFTWARE FOR ENGINEERING UNDERGADUATE

COURSES: SPREADSHEET FOR THE COMPUTATION OF ATMOSPHERIC

POLLUTANT FATE
André Luiz de Lima Reda1, Carlos Alberto de Moya Figueira Netto2, Larissa Xavier de Melo3

Abstract  Air pollution has been a problem since the as Environmental Sciences, Introduction to Environmental
industrial revolution, in the late nineteenth century – Engineering, Environmental Impacts or other related ones
particularly in terms of the atmospheric degradation due to [7]. Students at Escola de Engenharia Mauá (Mauá School of
effluent release by industrial chimneys. Pollutants are Engineering) can practice this matter by mathematically
transported in the atmosphere subject to translation, simulating cases of air pollutant dispersion as it leaves a
convection, diffusion and chemical reactions – in this case, chimney or stack. Programmed spreadsheets are prepared to
they may undergo either increase in concentration, or perform this simulation and provide answering patterns
consumption. The mathematical spreadsheet presented here useful to check the results of student attempts to solve the
has been applied in undergraduate Engineering courses, problem by hand. Such expert spreadsheets accept users to
helping students practice the numerical operations used to change the entering data or adjust internal equations such as
simulate the fate of suspended or dissolved pollutants to reflect situations and assumptions different from those
discharged by chimneys on a moving atmosphere. The assumed in the dispersion problem originally conceived.
spreadsheet considers the transformations of a Gaussian The following sessions detail this spreadsheet, an
plume of a conservative pollutant undergoing translation and education tool useful for activities either in, or extra-class.
dispersion, while following a horizontal translation axis The next session contains a brief theory review and an
under particular atmospheric conditions. It has been explanation of the computation methodology applied to the
successfully used by the authors to teach “Introduction to pollution phenomenon simulated in the proposed case.
Environmental Engineering” – a subject taught in the basic
years of undergraduate Engineering courses. THEORY REVIEW AND COMPUTATION METHOD
Gas or smoke discharged from a chimney may find different
Index Terms  Air pollution, pollutant dispersion, smoke
and variable atmospheric conditions; thus, forecasting their
diffusion, Gaussian plume.
fate is a difficult task. The aspects to consider in the
INTRODUCTION mathematical simulation of their propagation, after leaving a
steady source, encompass the following: meteorological
Air pollution has caused growing worries since the Industrial conditions in the region, physical and chemical phenomena to
Revolution made it generalised over industrial urban cities. which the specific pollutant is subject and eventual changes
However, even before that, as humans started to territorially in the chimney height and outlet geometry. It is also important
concentrate and burn vegetal or fossil fuel for survival, or to to know meteorological aspects such as air turbulence and
produce handcrafting, the problem arose of understanding the stability (or instability, as one may prefer); air speed projected
dispersion of atmospheric effluents (aerosols, gases or in each horizontal direction; vertical air velocity; humidity;
suspended particles) from chimneys and stacks – though at a temperature, and topography around the source and along the
scale much smaller than that seen nowadays at industrial path to be followed by the plume.
plants. A history of air pollution is presented in [3] (based on According to [5], atmospheric pollutant transport may
the case of London), and may surprise readers by showing that reach different distances at several length scales. The scale
illnesses and death among humans caused by this sort of classification proposed there may be adapted as follows:
pollution date back to the Middle Age. a) Microscale: distances of the order of a few kilometres – for
A solution to disperse air pollutants away from the low example, propagation of a pollutant plume while it can still be
altitudes where most humans live (the soil level) or, in some identified by association to its source;
cases, from the tops of buildings, has been the construction b) Mesoscale: distances of the order of a few hundred
of chimneys, or perhaps increasingly taller stacks from kilometres – for instance, plumes boosted by winds between
whose outlets pollutants are discharged into the atmosphere. mountains (e.g., a mistral wind), valley winds, or orographic
In order to teach this simple method for estimating the air streams;
impacts caused by atmospheric pollutant discharges, some c) Macroscale: distances of the order of thousands of
Engineering courses may include this issue in subjects such kilometres – for example, by high altitude or low altitude air
1
André Luiz de Lima Reda, Universidade Presbiteriana Mackenzie, Escola de Engenharia, Rua da Consolação, 930, Prédio 6, 01.302-907, São Paulo, SP,
Brazil; Escola de Engenharia Mauá, IMT, São Caetano do Sul, SP, anderreda@maua.br
2
Carlos Alberto de Moya Figueira Netto, Escola de Engenharia Mauá, IMT, São Caetano do Sul, SP, carlos.moya@maua.br
3
Larissa Xavier de Melo, Escola de Engenharia Mauá, IMT, São Caetano do Sul, SP, larissaxavier2005@hotmail.com

DOI 10.14684/WCCSETE.2.2016.31-35
© 2016 COPEC September 04-07, 2016, Castelo Branco, PORTUGAL
II World Congress on Computer Science, Engineering and Technology Education
31
streams – either between continents, at a synoptic scale, or perpendicular to the axis – vertical and the horizontal. This
across a continent. Gaussian analogy supposes that, for a given position x at the
Each of these scales of propagation length relates to a longitudinal axis, the form how concentration decays along
different order of transport-time magnitude – thus, I may take one of the orthogonal axes is proportional to the form how
either hours or weeks for the plume to reach its final probability decays for a Gaussian variable whose axis is
destination. This modelling exercise focuses on the dispersion represented by that same orthogonal axis. Thus, the Gaussian
of a continuous Gaussian plume, being restricted to model equations may be adapted for estimating pollutant
microscale simulation. The text below discusses theory and concentration at different distances from the longitudinal axis.
concepts of interest to solve a typical problem in this area. This physic-statistic analogy is summarised below and may be
The case analysed is the dispersion of a continuous plume studied in detail in [1] [6] or [9] – the latter, a practical,
released by an outlet at the upper end of an industrial stack didactical approach for undergraduates.
whose height is H’, measured from the ground (considered Initially, one may consider an observation point at a
horizontal), subject to a horizontal air stream (as a vertical distance z-H from the longitudinal axis of the plume,
simplification of reality). Different levels of atmospheric for a given x distance measured on this axis – see the z-H
turbulence may be supposed, and associated to a ‘dispersion vertical axis shown at the right side of Figure 1, on which a
coefficient’ – as explained further here. Figure 1 illustrates the Gaussian distribution is represented for the z-H variable up-
simple case in which all points exposed to the effect of plume and downwards from that longitudinal axis. The Gaussian
propagation (“observation point”) are so near to the pollution plume analogy supposes that pollutant concentration at an
source (i.e., the upper end of the stack), that plume observation point at a certain distance from the longitudinal x
propagation could not yet have reached the ground. Still in axis, measured along the z-H axis, is proportional to the
Figure 1, the vertical axis z starts at the soil surface and the estimated probability of pollutant occurring at that same
other vertical axis, z-H, departs from the horizontal plume point. By following this analogy, this probability may be
axis, x. The latter starts at a distance of H-H' above the stack estimated employing the Gaussian distribution, considering
top. Thus, after the plume is discharged, it goes rising up to a that its standard deviation in the vertical direction grows as x
level where its density equals that of the atmospheric air (i.e., grows. This standard deviation is represented in Figure 1 by
where it reaches a neutral situation in adiabatic terms) – see the vertical distance z, measured on the z-H axis vertical
[2]). At this level, it reaches stability and stops rising – i.e., it (originated where it crosses the x axis), both up- and
goes on propagating only along the longitudinal axis. downward in that axis, given the symmetry of the Gaussian
(or “normal”) probabilistic distribution. By the Gaussian
plume analogy, the z vertical distance is linearly related to
the plume “vertical dispersion coefficient”. The nomograph
of Figure 2 yields the value of dispersion coefficient z as a
function of distance x, what allows for the estimation of the
reduction in pollutant concentrations as the plume propagates
along the x axis. This analogy also allows one to estimate
concentrations at vertical positions other than z , above or
below the x axis, just by considering proportionality to the
value of the Gaussian probability density function the
required vertical position for a given x.
Pollutant dispersion occurs also in the lateral direction, as
the plume propagates longitudinally, at a similar way as it
happens in the vertical direction. On the right side of Figure
1, there is a y axis shown in perspective, with its origin where
it crosses the longitudinal, x axis. As a function of the y axis,
it is possible to see another graph showing a Gaussian
distribution that represents pollutant concentrations decaying
FIGURE. 1 laterally, to both sides along the x axis. By the Gaussian plume
SCHEMATIC OF POLLUTANT PLUME DISPERSION ALONG A HORIZONTAL analogy, that graph suggests that the lateral reduction in
AXIS FOR A LOCAL ATMOSPHERE UNDERGOING HORIZONTAL MOVEMENT. pollutant concentration is proportional to the reduction in the
The scheme of Figure 1 also serves to represent the density of the probability of pollutant occurring as y grows,
“Gaussian plume analogy”, in which a conservative and the standard deviation shown along that y axis stands,
atmospheric pollutant (gas, gas mixture or smoke) leaves a respectively, for the dispersion coefficient z, a function of x.
continuous source at a constant discharge rate, along a Given the symmetry borne by the Gaussian model, such
horizontal mass axis, x. As the pollutant moves away from the deviation could be represented to both sides, from y=0. The
source, it undergoes dispersion in the two directions

© 2016 COPEC September 04-07, 2016, Castelo Branco, PORTUGAL

II World Congress on Computer Science, Engineering and Technology Education
32
nomograph in Figure 3 yields the value of y as a function of
x and atmospheric stability conditions.

FIGURE. 2
LATERAL DISPERSION COEFFICIENT, y , A FUNCTION OF THE PROPAGATION DISTANCE FROM THE SOURCE (ADAPTED FROM [4]).
The mathematical equation to estimate pollutant U is the average longitudinal wind velocity in the x direction,
concentration at any observation point inside the ‘cone’ considered as the advective pollutant velocity in that direction
occupied by the Gaussian plume centred in the x axis (and (m/s);
departing from the stack upper end) if it did not yet reach z is the plume vertical dispersion coefficient (or “vertical
ground level is derived from the Gaussian probability standard deviation”, in the Gaussian analogy) – a function of
distance x obtained at the corresponding nomograph (m);
distribution function (see [1]), and is shown in (1),
y is the plume lateral dispersion coefficient (or “lateral
standard deviation”, in the Gaussian analogy) – a function of
(1) distance x obtained at the corresponding nomograph (m).
However, as the lower plume boundary (or its lower
‘enveloping surface’) touches the ground, the relationship
where: shown in (1) is no longer valid – a more complex equation
Q is the pollutant emission rate (g/s) = fuel burning rate x should apply. According to [8], assuming that the soil cannot
fraction of pollutant in fuel; be considered as a sink for conservative air pollutants, it is
more realistic to adopt a mathematical model that supposes
H is the geometric stack height above ground level (m);
total ‘reflection’ of the plume as it reaches the ground. Thus,
H’ is the effective stack height = plume stable propagation
the pollutant would behave as if ‘rising from the soil surface’
altitude (at the x axis) minus ground altitude (m);
at vertical velocities opposite to those it could bear if it
x is the longitudinal distance from the plume source to the
cross section containing the analysed observation point (m); continued to propagate downward (supposing the soil surface
did not exist). In simple geometric terms, the ground would
y is the lateral distance from the analysed point and the x axis
(m); ‘reflect’ pollutant propagation as a mirror does. Then,
pollutant concentration could be estimated at any observation
z is the vertical distance from the analysed point and the x axis
(m); point inside the plume, according to [8], by (2) – where the
meaning of the symbols is the same as in (1).

© 2016 COPEC September 04-07, 2016, Castelo Branco, PORTUGAL

II World Congress on Computer Science, Engineering and Technology Education
33
 (2)

FIGURE. 3
VERTICAL DISPERSION COEFFICIENT, z , A FUNCTION OF THE PROPAGATION DISTANCE FROM THE SOURCE (ADAPTED FROM [4]).
The spreadsheet solution shown by the computer screen
PROPOSED EXERCISE FOR EDUCATIONAL USE image of Figure 4 starts by the computation of SO2 emission
The text below describes the case taken as a didactical flow, Q, which is worth the product of the coal burning rate
example of this numerical method. by the fraction of SO2 in the fuel, 0,054. This is converted to
A thermoelectric power plant discharges unfiltered coal g/s and registered on the first spreadsheet line at the top of
burning smoke from its stack top at a concentration of 5.450 Figure 4. Then, vertical and lateral dispersion coefficients are
kg/h. The plume reaches a stable propagation height at 75m obtained from the nomographs of figures 2 and 3,
above ground. The carbon dioxide concentration in that coal respectively, for distances at every 200m on the x axis. Their
is 5,4% (in terms of mass). Wind velocity at that effective values are registered in the lower table of Figure 4 and their
propagation height is 6,0m/s (consider this as the advection variation along x is shown in the upper graph of that figure.
velocity). Atmospheric dispersive conditions are moderately Equation (2) is registered at the top of Figure 4, which
unstable. Dispersion must only be computed in the y and z shows, also, all the data values fed into the spreadsheet – all
directions, since the plume discharge is continuous – meaning in their respective spreadsheet areas, in a self-explanatory
that computation of dispersion in x is unnecessary. fashion and bearing units. The spreadsheet computes the C
The, the student will estimate: and SO2 concentration values required, at each 200m from the
a) The SO2 emission rate (g/s). x axis. The curve of the second graph of Figure 4 illustrates
b) Pollutant concentration at each 200m from the source, the decay of this concentration along x.
measured over the longitudinal plume axis, such as to show
its variation along that axis until 1200m from the origin.

© 2016 COPEC September 04-07, 2016, Castelo Branco, PORTUGAL

II World Congress on Computer Science, Engineering and Technology Education
34
 FIGURE. 4
SCREEN IMAGE OF EXCEL SPREADSHEET SIMULATING THE PROPAGATION OF A GAUSSIAN PLUME OF POLLUTANT ALONG A HORIZONTAL AXIS

[5] KIELY, G. “Environmental Engineering”. New York: McGraw-Hill.

CONCLUSION ISBN 0-07-709127-2, 1996, 980p.
[6] PEAVY, H.S.; ROWE, D.R.; TCHOBANOGLOUS, G. “Environmental
This spreadsheet can estimate concentrations of any engineering”. New York: McGraw-Hill International Editions, Civil
conservative atmospheric pollutant, at any position inside the Engineering Series. 1986, xiv p, 718 p.
pollutant plume – either at its longitudinal axis or points of its [7] REDA A.L.L.; LEO, L.F.R.L.; DURO, M.A.S.; FUMIÓ, B.L.C.;
neighbourhood. Although limitations to the article text ROCHA, A.J.F. “O projeto pedagógico de curso e o papel da educação
extension hindered any inclusion of other examples of the use ambiental gerando competência em Engenharia Civil. Caso da ponte sob
of the spreadsheet, e.g, computing concentrations at risco por ação de uma enchente”. Proceed.: XLII Congresso Brasileiro
de Educação em Engenharia. Engenharia: Múltiplos saberes e atuações,
observation points outside the x axis, more versions of this Juiz de Fora, 16 to 19 sep. 2014, 12p.
software, for more complete cases, have been and are being [8] STERN, A. C.; WOHLERS, H. C.; BOUBEL, R. W.; LOWRY, W. P..
prepared – and may be provided to any interested student. “Fundamentals of air pollution”. New York: Academic Press. 1973,
492p.
REFERENCES [9] WESTERN UNIVERSITY. “Self study notes. Gaussian Plume”.
[1] BOUBEL, R. W.; FOX, D. L.; TURNER, D. B.; STERN, A. C. Ontario, CA: Western University. Electronic file. 7p. Available at:
Fundamentals of air pollution. 3.ed.. San Diego, CA: Academic Press, http://www.eng.uwo.ca/people/esavory/Gaussian%20plumes.pdf.
Inc., 1994, 574p. ISBN 0-12-118930-9. Available at: http://ac.els- Access on 01/05/2016.
cdn.com/B9780080507071500020/3-s2.0-B9780080507071500020-
main.pdf?_tid=8491fff2-8282-11e6-9456-
00000aacb35d&acdnat=1474740956_8e2f715f07b5031f540e9d933b40
b7de. Also available at: http://www.sciencedirect.com
/science/article/pii/B9780080507071500020. Access on 22/06/2016
[2] BRAGA, B.; HESPANHOL, I.; CONEJO, J.G.L.; MIERZWA, J.C.;
BARROS, M.T.L.; SPENCER, M.; PORTO, M.; NUCCI, N.;
JULIANO, S.; EIGER, S. “Introdução à engenharia ambiental”. 2.ed.
São Paulo: Pearson Prentice Hall. 2005, 318p.
[3] BRIMBLECOMBE, P. “The big smoke: a history of air pollution in
London since medieval times”. Abingdon, Oxfordshire: Routledge.
Series Routledge Revivals, reimpr.2012 (1.ed.: 1987), 204p. ISBN-13-
978-0415672030. Available at: https://books.google.com.br/books?id=
bx4wAFE8xGUC&printsec=frontcover&hl=pt-BR#v=onepage&q&f=
false. Access on 17/05/2016
[4] DAVIES, M.L. “Air resources management primer”. New York: ASCE.
1973.

© 2016 COPEC September 04-07, 2016, Castelo Branco, PORTUGAL

II World Congress on Computer Science, Engineering and Technology Education
35