1. Introduction
In south Morocco, the agricultural production is
limited by unfavorable climatic conditions, wich
penalize the production in quantity and quality.
Moreover, the presence of thrips and aphides are
responsible for significant crop damage.Therfore,the
use of very fine anti-insect proof nets has been
recognized to reduce the need for pesticide application.
They act as mechanical barriers to insects but olso
reduce the ventilation rate,and raise inside temperature
and humidity. [1-7]
The studies for the dynamic characterization of
these nets have been based on experimental studies.
Computational fluid Dynamics (CFD) have been
increasingly used to study greenhouse ventilation. The
effects of insect screens on ventilation, have been
characterized and numerically modelled,in a tunnel
greenhouse.The effect of wind speed on natural
ventilation have also been analysed in a greenhouse
using a three-dimensional and a two-dimentional CFD
simulation respectively. [8-15]
The aim of the present study, was to analyse the
distribution of temperature fields in a canarian
greenhouse installed in southern Morocco, in order to
find a structure better adapted to the climatic
conditions of our region. It is a commercial greenhouse
covered with a plastic cover, It occupies a surface of
11,250 m²and its average height is 5 m. The orientation
of the "chapels" is North-South, that is, perpendicular
to the direction of the prevailing wind. The ventilation
of the greenhouse studied is provided by seventeen
openers (0.6x125 m² each, or 1275 m2 in all) and
Covered with nets against insects type 20 x 10 (anti-
Thrips). The side walls of the greenhouse have fixed-
sized openings and are equipped with the same type of
nets.
We combined an experimental and modelling study.
The numerical climate model is based on (CFD)
simulation of sensible and latent heat exchanges
between the tomato crop and the greenhouse air, with
combination of radiative transfers at roof level. The
model was first validated by measured data and then
used to explore the details of air flow, temperature.
This CFD assisted for exploration of inside climate and
allows for a better assessment of the overall climate
and plant activity.
2. Theory
We note that the nomenclature with the explanations of
all variables can be found at the end of the paper
2.1. aerodynamic equations:
The mass, momentum, energie and concentration
equations can be represented with the following
conservation equation:.
S
xx
u
xt jj
j
j
)()(
(1)
2.2 Modeling of flow through insect screens and
plants :
The drag forces induced by insect screens and crop,
that correspond to the term
S
,is included into our
CFD study by the porous medium approach given by
the Darcy–Forchheimer equation: [16]
Study of temperature fields inside a canarian greenhouse
K. LEKOUCH, M. EL JAZOULI*, L. BOUIRDEN
Laboratory of Thermodynamics and Energy,
Faculty of Sciences, University Ibn zohrcité
Dakhla BP 8106 Agadir, MORROCO.
Key-words : Canarian Greenhouse; CFD; temperature fields ;Insect screens;
Received: May 23, 2021. Revised: March 21, 2022. Accepted: April 22, 2022. Published: May 24, 2022.
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
79
Volume 17, 2022
Abstract: The climatic factors that influence the climate inside the greenhouse are temperature, humidity, solar
radiation, wind and plant cover. Temperature is the most important factor in the growth and development of
vegetation. Thus, a well controlled temperature allows the physiological development of the plants. This study
presents an analysis and simulation of temperature fields in canarian greenhouse equipped with anti-insect nets.
The site is located in the coastal area of southern Morocco. The study is carried out during two period daytime
and night to determine cooling or heating requirements, with consideration of the long and short wavelength
balance and the radiative convective coupling. The fundamental calculation of climatic conditions is based on
CFD. The dynamic influence of the insect screens and tomato crop on airflow movement, was described ,using
the concept of the porous medium approach proposed by Darcy and Forchheimer. The coupling of convective
and radiative exchanges at the plastic roof cover is considered. This CFD study assisted for exploration of inside
climate and allows for a better assessment of the overall climate and plant activity. A good agreement was
observed between the measured and simulated values for inside temperature. The results clearly showed the
heterogeneity of the greenhouse’s internal climate, which infects agricultural production in quantity and quality.
S
=- (
2
U
K
C
U
K
f
) (2)
44.3K
×
6.19
10
(3)
f
C
=
13.2
2
1030.4
(4)
is the screen porosity deduced from the thread
dimensions. : [17]
))((
.
dWdL
WL
(5)
L = 0.788 mm and W= 0.255 mm are respectively
meshes length and width,
d = 0.28 mm, is the wire diameter.
For the low air speed (0.1-0.5
1
.
sm
), we can considers
only the second member of relation (2):
S
= -
2
U
K
Cf
(6)
The non-linear momentum loss coefficient
f
C
and the
permeability K can be deduced from equation: [15]
K
Cf
LA
I
D
C
(7)
The radiative net flux Rnet reaching each mesh ,is
partitioned into convective sensible
sens
Q
and latent
heat fluxes
lat
Q
: [15]
0 latsennet QQR
(8)
The sensible heat flux
sen
Q
was expressed with the
temperature difference between inside air and canopy :
sen
Q
=
)
)(
(
a
iv
AVpr
TT
IC
(9)
The aerodynamic resistance
a
r
was deduced from the
air speed:
5.0
)(
288.0 U
d
C
rv
p
a
(10)
The latent heat fluxes
lat
Q
was deduced from the
humidity difference.
)(
*
3
1
as
iv
AVevlat rr
ww
IllQ
(11)
The Tomato leafstomatal resistance
s
r
was deduced
from air temperature and saturation deficit values
using Boulard et al.formula:
))162910.107.6(34.0exp(11.01 max
5.237
5.7
min Dwrr i
T
T
ss
i
i
(12)
.
The solar radiation received at a height z (m) is
expressed by the following equation: : [17]
))(exp()( z
zH
IkRzR LAScgi
(13)
)()()( 21 zdRzRzRRabs
(14)
Finally the tomato crop temperature (
v
T
) can be
calculated according to the following equations:
t
ai
ev
LAVp
a
iv r
ww
LL
dz
zdR
IC
r
TT )(
)(
2
13
1
(15)
3. Materials and methods
3.1. The greenhouse
The studied greenhouse (Fig. 2) is a commercial
canarian greenhouse. The greenhouse is covered with a
plastic cover, It occupies a surface of 11,250 m²and its
average height is 5 m. The orientation of the "chapels"
is North-South, that is, perpendicular to the direction
of the prevailing wind. The ventilation is provided by
seventeen openers arranged in roof (0.6 125 m² each,
or 1275 m2 in all) and covered with protective nets
against insects of type 20 x 10 (anti-Thrips). The side
walls of the greenhouse have fixed-sized openings and
are equipped with the same type of nets.
Fig.1 : interior photo of the studied greenhouse .
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
80
Volume 17, 2022
simulated temperature profiles measured in the centre
and along the length from west to east, to two
different heights: 1 m and 4 m. It is observed that in
general, the simulated values are slightly lower than
the measured values.
Fig. 2 : Representative scheme of the greenhouse and
its ventilation system.
3.2. mesh size and boundary conditions
3.2.1 mesh size
The mesh used in this study is of the BFC type
composed of 192, 44 and 112 meshes according to the
axes respectively
x
,
y
,
z
(total 860160). The
calculation domain also includes the free space on the
windward side (30 m), the downwind sides (30 m) and
along both sides (2 x 30 m) of the greenhouse.
Fig. 3 : Mesh used to simulate the greenhouse .
3.2.2 boundary conditions
The conditions at the outside ground temperature
limits of the greenhouse correspond to the averages of
the values measured experimentally.
Thermal conditions at the walls : Generally, wall
conditions of the «WALL» type are used , however,
the side walls have been treated as porous.
4. Results and discussion
4.1. Model validation
Figures 4, 5,6and 7 represent the evolution of the
Fig.4: Evolution of daytime temperature profiles
simulated and measured at the centre of the greenhouse
at 1 m above the ground, depending on the length of
the greenhouse (cut in the centre of the greenhouse)
measured (♦) simulated (▲)
Fig. 5: Evolution of daytime temperature profiles
simulated and measured at the centre of the greenhouse at 4
m above the ground, according to the length of the
greenhouse (cut in the centre of the greenhouse)
measured (♦) simulated (▲)
303
304
305
306
307
308
020 40 60 80 100
Longueur (ouest- est) (m)
Température ( K)
302
303
304
305
306
307
308
020 40 60 80 100
Longueur (ouest- est) (m)
Température (K)
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
81
Volume 17, 2022
.
Fig. 6: simulated and measured night temperature
profiles at 1 m above ground level as a function of
greenhouse length measured (♦) simulated (▲)
Fig. 7: Evolution of the simulated and measured profiles of
the night temperature at 4 m above the ground as a function
of the length of the greenhouse
measured (♦) simulated (▲)
4.2 Detailed description of the thermal field
within the greenhouse
4.2.1 study of diurnal microclimate :
Figures 8,9 and 10 represent respectively the simulated
thermal fields in horizontal sections 1, 3 and 4 m
above ground level. It is observed that the temperature
is high inside the greenhouse and that its distribution
depends strongly on the penetration of the outside air
which significantly cools the inside air. At the
openings 4 m above the ground and above the
vegetation the temperature is almost constant, while at
the ground level it is heterogeneous. The warmest area
(311 K) is at ground level. approximately 12 m
downstream from the greenhouse entrance. This is the
most critical area of the greenhouse.
At the height of 1 m, there are two distinct zones:
Warmer areas are at the junction of the convection
cells. This temperature increase particularly at12m
downstream of the greenhouse inlet near the windward
end where interference between the incoming air
current and an internal air current blowing in the
opposite direction is observed.
In general, there is significant temperature
heterogeneity at this level. This is mainly due to the
geometry of the roof and the alternating arrangement
of the ventilation openings on the roof in high (5.5 m)
and low (5 m) position relative to the prevailing wind
direction. This arrangement has a remarkable effect on
the circulation of air within the greenhouse. the
temperature difference between inside and outside the
greenhouse is less than that observed at 1 m, This
limitation of heating is explained by the contribution of
the roof openings to the ventilation of the area above
the vegetation.
Figure 11 shows a vertical section of the air
temperature field at the centre of the greenhouse in the
direction of flow. This figure shows the very high
temperature gradient that develops in areas near the
greenhouse cover and the soil. It also shows the
existence of the 8 cold air intakes corresponding to the
roof ridge openings located in the "low" position at 5
m above the ground and the 9 hot air outlets
corresponding to the ridge openings located in the
"high" position 5.5 m above ground level .
Figure 12 shows the simulated vertical temperature
profile at the centre of the greenhouse as a function of
height. It synthesizes the previously reported
observations that the temperature is very high at the
level of the soil surface, then that it decreases with the
height, up to about 4 m. Then we observe a slight
increase in temperature which reaches its maximum at
the level of the plastic cover. These two temperature
peaks are explained by the high absorption of radiation
by the PE film and the soil surface.
Figures 13, 14 and 15 represent the temperature
profiles at 1, 3 and 4 m above the ground, respectively,
as a function of the length of the greenhouse from west
to east. It is observed that at 1 m in height, the
temperature distribution is very heterogeneous with
high values located in the areas below the hot air
outlets (309 K) and lower values in areas just below
the cold air intakes (306 K). We can also note the same
alternations at 3 and 4 m but with a strong damping of
the variations compared to those recorded at 1 m.
Figures 16 and 17 represent the temperature profiles at
1 and 4 m above the ground as a function of the width
of the greenhouse (North to South), respectively, they
highlight the existence of a great homogeneity of the
temperature distribution over the width of the
greenhouse. This phenomenon is due mainly to the
entrance of air
290
290,5
291
291,5
292
292,5
293
020 40 60 80 100
Longueur (ouest- est) (m)
Température (K)
290
290,5
291
291,5
292
292,5
293
020 40 60 80 100
Longueur (ouest-est) (m)
Température (K)
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
82
Volume 17, 2022
Fig.8:Simulated thermal field
(horizontal cut to 1 m above ground level )
Fig. 9: Simulated thermal field
(horizontal cut 3 m above ground level)
Fig. 10 : Champ thermique simulé
(coupe horizontale à 4 m au-dessus du sol)
Fig 11 : Simulated thermal field (K) in the centre of
the greenhouse
(vertical cut in direction of flow)
Fig.12: Vertical profile of Simulated Temperature at
the Center of the Greenhouse as a Function of Height
Fig. 13:Temperature profile simulated at 1m above the
ground depending on the length of the greenhouse:
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
83
Volume 17, 2022
Fig. 14: Temperature profile simulated at 3 m above
the ground depending on the length of the greenhouse
Fig. 15: Temperature profile simulated at 4 m above
the ground depending on the length of the greenhouse
Fig. 16:Simulated temperature profile 1m above the
ground depending on the width of the greenhouse
Fig. 17 : Temperature profile simulated at 4m above
the ground depending on the width of the greenhouse
4.2.2. Study of Nocturnal Microclimate
For the same geometric configuration (including
opening) of the greenhouse previously studied during
the daytime, we examined the distribution of its
microclimate during the night period. In Table 5.3, the
averages and standard deviations of the different
climatic parameters used as conditions at the initial
limits or conditions for this simulation of the nocturnal
microclimate.
Table 1: Experimental (average) measurements taken
between 2 and 5 in the morning for 3 days) and used as
boundary conditions for night microclimate simulation.
Figures 18 and 19 show the temperature field in a
vertical plane at the centre of the greenhouse in
parallel and perpendicular sections respectively to the
direction of the prevailing wind (West-East). These
two figures highlight the existence of an almost perfect
homogeneity of the internal temperature of the
greenhouse during the night.
With a slight rise in temperature in the vicinity of
the ground surface +1.5°C due to the thermal flux
imposed by the latter during the night.
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
84
Volume 17, 2022
The greenhouse roof is approximately 2°C colder than
the outside air and 1.5°C colder than the greenhouse
air. Figure 20, relating to a horizontal section of the
temperature field at 1 m above the ground, shows the
almost perfect homogeneity of the distribution
temperature within the vegetation. Beyond the height
of the canopy (Figure 21) the distribution is more
heterogeneous, with a rather small variation.
The analysis in Figure 22, which summarizes the
evolution of the temperature profile at the centre of the
greenhouse as a function of height, shows that the
maximum temperature is found near the surface of the
soil. We then observe a slight decrease in temperature
within the canopy. Above the canopy the temperature
becomes similar to that of the outside.
Figures 23, 24 and 25 represent the evolution of
temperature profiles as a function of the length of the
greenhouse, moving from west to east over three
different heights respectively: 1, 3 and 4 m. These
three figures confirm that the temperature above the
vegetation (3 and 4 m) is the same as that of the
outdoor air, with a slight decrease within the
vegetation cover.
Overall, the temperature is perfectly homogeneous for
a given height. According to figures 26 and 27, which
represent the evolution of temperature profiles as a
function of the width of the greenhouse for two
different heights (1 and 4 m). We note that this
temperature homogeneity for the same height is also
true along the North-South axis.
Fig. 18 Simulated thermal field in the centre of the
greenhouse (vertical cut in direction of flow)
Fig. 19 : Simulated thermal field (K) at the centre of
the greenhouse(vertical cut in direction perpendicular
to direction of flow)
Fig. 20 Champ thermique (K) simulé
(coupe horizontale à 1 m au-dessus du sol)
Fig. 21 : Simulated thermal field (K)
(horizontal section 4 m above ground level)
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
85
Volume 17, 2022
Fig. 22: Vertical temperature profile simulated at the
centre of the greenhouse as a function of height
Fig. 23 : Simulated temperature profile 1 m above
ground as a function of greenhouse length
Fig. 24: Simulated temperature profile 3 m above
ground as a function of greenhouse length
Fig. 25: Temperature profile simulated at 4 m above
the ground depending on the length of the greenhouse
Fig. 26 : Simulated temperature profile at 1 m above
ground as a function of greenhouse width
Fig.27 : Simulated temperature profile at 4 m above
ground based on greenhouse width
In summary, the simulation of the distribution of
thermal fields greenhouse air during the night period,
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
86
Volume 17, 2022
allows us to conclude that during the night the
temperature difference between the inside and outside
of the greenhouse is very low and that the air
temperature of the greenhouse is slightly higher within
the canopy.
5. Conclusion
Numerical simulation is a tool to characterize the
thermal and dynamic fields inside the greenhouse.
Other parameters such as culture transpiration flow
and CO2 concentration could be added to this
simulation.
Two approaches were adopted:
-An experimental approach,based on measurements
of climatic parameters (temperature) inside and outside
the greenhouse, through which we were able to collect
the input mathematical models. This approach also
allowed us to characterise the internal climate and to
understand the natural aeration in the greenhouse and
to refine the knowledge of several mechanisms
involved.
- A numerical modelling of the greenhouse
microclimate using the fluid mechanics code CFD,
which allows the prediction of the temperature fields
inside the greenhouse after the numerical solution of
the Navier-Stockes equations and the heat equation in
the considered computational domain.
In general, a significant temperature heterogeneity is
observed inside the greenhouse. This phenomenon is
mainly due to the geometry of the greenhouse and the
alternating arrangement of the ventilation openings on
the roof in relation to the direction of the prevailing
wind. This arrangement has a remarkable effect on the
air circulation within the greenhouse.
In summary, these first results give confidence in the
relevance of the numerical simulation of the
greenhouse climate and in its concrete use to improve
its climatic conditions
One of the most important implications of our study is
that farmers can use our work to predict the internal
climate changes involved in using the nets or in
arranging the crop rows. This allows them to
determine the best combination for effective protection
against insects.
Nomenclature
K medium permeability (
2
m
)
U air speed . (
1
.
sm
)
Cf
non-linear momentum loss coefficient
LA
I
leaf area index
sens
Q
convective sensible flux (W.
2
m
)
lat
Q
latent heat fluxes (W.
2
m
)
v
T
canopy temperature (K),
i
T
inside air temperatures (K),
p
C
specific heat of air at constant
pressure (
11.. KKgJ
),
AV
I
leaf area index,
a
r
leaf aerodynamic resistance (
1
.
ms
)
v
d
characteristic length of the leaf (m);
*
v
w
saturated water content of air (
1
.
kgkg
)
i
w
specific humidity of air (
1
.
kgkg
)
s
r
tomato leafs tomatal resistance (
1
.
ms
)
gi
R
global radiation inside the greenhouse (W.
2
m
)
Rge Outside radiation (W.
2
m
)
c
k
extinction coefficient of radiation,
H total height of canopy(m),
LAS
I
crop stand leaf area index (
22 .
mm
)
Greek letters
studied variable
diffusion coefficient of the quantity
dissipation of the turbulent energy.
air density (
3
.
mkg
)
dynamic viscosity
11..( mskg
)
air thermal conductivity (
11.. KmW
)
air viscosity (
12 .
sm
)
the screen porosity
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
87
Volume 17, 2022
References
[1] El jazouli et al. CFD study of airflow and
microclimate patterns inside a multispan
greenhouse.WSEAS transactions on fluid
mechanic DOI :10.37394/232013 (2021)
(102-108)
[2] Katsoulas et al Effect of vent openings and insect
screens on greenhouse ventilation. Biosystems
Engineering.(2006).01.001
[3] Hanafi, A et al. Performances of two types of
insect screens as a physical barrier against B.
tabaci and their impact on TYLCV incidence in a
greenhouse tomato in the Souss valley
ofMorocco. Mediterannean Climate.IOBC
Bulletin. Vol. 26 (10)(2003) : 39-42.
[4] Fatnassi, H et al. Optimisation of greenhouse
insecte screening with Computational Fluid
Dynamics. Biosystems Engineering 93,(3)(2006),
301- 312.
[5] Montero et al .Computational Fluid Dynamic
modelling of night-time energy fluxes in
unheatedgreenhouses.january.Acta Horticulturae
691 (2005) 403-409.
[6] Molina-Aiz et al Using computational fluid
dynamics tool to model the internal climate of an
Almerıa-typegreenhouse.ActaHorticulturae,654,
(2004) 271–278.
[7] Fatnassi et al Optimisation of greenhouse insecte
screening with Computational Fluid Dynamics.
Biosystems Engineering 93,(2006) 301- 312.
[8] Haxaire, R. . Caractérisation et Modélisation des
écoulements d'air dans une serre. Thèse de
Docteur en Sciences de l'Ingénieur de
l'Université de Nice, Sophia Antipolis.
148p.(1999)
[9] Demrati,H.etal.Natural ventilation and
microclimatic performance of a large scale
banana greenhouse. J.Agri.Engng Res 80 (3):
(2001) 291-271.
[10] Ould Khaoua et al. Modélisation de l’aération
naturelle et du microclimat des serres en verre de
grande portée sous climat tempéré océanique.
Thèse de Doctorat,Université d’Angers
France(2006).
[11] Demrati,H.et al, Natural ventilation and
microclimatic performance of a large scale
banana greenhouse. J.Agri.Engng Res 80 (3):
(2001) 291-271.
[12] Launder and Spalding, The numerical
computational of Turbulent flows. Comp.
Method App. Mech. Eng. (1974), 3, 269-289.
[13] Boulard and Wang. Experimental and numerical
study on the heterogeneity of crop transpiration
in a plastic tunnel. Computers & Electronics in
Agriculture. 34,(2002) 173–190.
[14] Bruse, Development of a numerical model for the
simulation of exchange processes between small
scale environmental design and microclimate in
urban areas.Thèse de Doctorat, University of
Bochum.(1998).
[15] Miguel et al, Airflow through porous screens:
from theory to practical considerations. Energy
and Building28 (1998) 63-69.
[16] Kittas,C.Contribution théorique et expérimentale
à l’étude du bilan d’énergie des serres.
Thèse de Doctorat Ingénieur. (1980)
[17] Kaviany, M.. Principles of heat transfer in porous
media. Springer. Verlaz, Berlin (1995).
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the Creative
Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en_US
WSEAS TRANSACTIONS on FLUID MECHANICS
DOI: 10.37394/232013.2022.17.8
K. Lekouch, M. El Jazouli, L. Bouirden
E-ISSN: 2224-347X
88
Volume 17, 2022
