Nativa, Sinop, v. 9, n. 5, p. 612-627, 2021.

Pesquisas Agrárias e Ambientais

DOI: https://doi.org/10.31413/nativa.v9i5.13168 ISSN: 2318-7670

Solar radiation incidence under different shading screens in tropical climate:

diurnal evolution and estimates

Daniela Roberta BORELLA1, Hercules NOGUEIRA2, Francielli Aloisio MORATELLI3,

Aline KRAESKI3, Adilson Pacheco de SOUZA1,2,3

1 Postgraduate Program in Environmental Physics, Federal University of Mato Grosso, Cuiabá, MT, Brazil.

1 Institute of Agrarian and Environmental Sciences, Federal University of Mato Grosso, Sinop, MT, Brazil.

1 Postgraduate Program in Environmental Sciences, Federal University of Mato Grosso, Sinop, MT, Brazil.

*E-mail: drborella@gmail.com

(ORCID: 0000-0003-2941-2116; 0000-0001-7056-351X; 0000-0002-0304-985X; 0000-0001-5795-9245; 0000-0003-4076-1093)

Submitted on 11/17/2021; Accepted on 12/29/2021; Published on 12/30/2021.

ABSTRACT: The use of photoselective screens improves plant productivity and quality, and it is necessary to

understand the dynamics of solar radiation transmissivity under these microenvironments to subsidize agricultural and

forestry production projects. Therefore, the objective of this work was to describe the diurnal and seasonal evolution

of incidence of global irradiance (IG), photosynthetically active irradiance (IPAR), and luminance in aboveground forest

nurseries under different shading screens. Radiometric fractions were evaluated and statistical equations were obtained

based on the external incidence. Instantaneous measurements of IG, IPAR, and luminance in the exterior and interior of

nurseries (East-West direction), solstices (12/22/2018 and 06/20/2019) and equinoxes (03/21/2019 and

09/21/2019), and local zenithal culminations (02/18/2019 and 10/20/2019) were evaluated between 7h00min and

17h00min. Estimates were evaluated and the data were grouped into two databases composed of 70% and 30%,

respectively, to generate and validate regressions for each variable. The statistical performance of regressions was

evaluated using the following statistical indicators: coefficient of determination (R2), mean square error (MSE), root

mean square error (RMSE), and Willmott index (d). IG, IPAR, and luminance presented similar dynamics of diurnal and

seasonal evolution under the shading screens for external conditions, and the transmissivity was affected by the

environmental conditions (water seasonality and solar declination) and intrinsic characteristics of the shading screens

(porosity and color). The transmission and absorption of IG, IPAR, and luminance were affected by color and porosity

of the shading screens, whereas the reflection was affected only by the color. The values of Willmott index were higher

than 0.9975 and 0.9973 for black screen and photoselective screen, respectively, and were considered as good,

indicating that the equations generated good estimates of IG, IPAR, and luminance for application in different regions.

The choice of shading screens for crops are dependent on spectral composition requirements and IPAR transmissivity

of each species.

Keywords: radiometric properties of photoselective screens; photosynthetically active radiation; irradiance; luminance;

statistical indicators.

Radiação solar incidente sob diferentes telas de sombreamentos em clima

tropical: evolução diurna e estimativas

RESUMO: O uso de telas foto-seletivas melhora a produtividade e qualidade das plantas, e é necessário compreender

a dinâmica da transmissividade da radiação solar sob esses microambientes para subsidiar projetos de produção

agrícolas e florestais. Nesse sentido, o objetivo foi descrever a evolução diurna e sazonal das irradiâncias global (IG),

fotossintéticamente ativa (IPAR) e luminância incidentes em viveiros florestais suspensos sob diferentes telas de

sombreamento. Além disso foram avaliadas as frações radiométricas e obtidas equações estatísticas de estimativas

baseadas na incidência externa. As medidas instantâneas de IG, IPAR e luminância ocorreram no exterior e interior dos

viveiros (alinhados no sentido Leste-Oeste), nos solstícios (22/12/2018 e 22/06/2019), equinócios (21/03/2019 e

21/09/2019) e nas culminações zenitais locais (18/02/2019 e 20/10/2019), entre às 7h00min e 17h00min. Para

avaliação das estimativas, os dados foram agrupados em duas bases, compostas por 70 e 30% para geração e validação

das regressões, para cada variável, respectivamente. Para avaliação do desempenho estatístico das regressões foram

empregados os indicativos estatísticos: coeficiente de determinação (R2), erro absoluto médio (MBE), raiz quadrada

do erro quadrático médio (RMSE) e índice de Willmott (d). IG, IPAR e luminância apresentaram dinâmicas semelhantes

na evolução diurna e sazonal sob as telas de sombreamento em relação as condições externas, sendo a transmissividade

influenciados por condições ambientais (sazonalidade hídrica e declinação solar) e intrínsecas a tela (porosidade e cor).

A transmissão e absorção de IG, IPAR e luminância foram afetadas pela cor e porosidade líquida, enquanto que a reflexão

apenas pela cor. Os valores do índice de Willmott foram superiores a 0,9975 e 0,9973 para as telas pretas e foto-

seletivas, respectivamente, sendo considerado como ótimos, indicando que as equações ajustadas permitem boas

estimativas de IG, IPAR e luminância para aplicação em diferentes regiões. A escolha da tela de sombreamento para o

cultivo de plantas fica dependente das necessidades de composição espectral e transmissividade da IPAR de cada espécie.

Palavras-chave: propriedades radiométricas de telas foto-seletivas; radiação fotossinteticamente ativa; irradiâncias;

luminância; indicadores estatísticos.

Borella et al.

Nativa, Sinop, v. 9, n. 5, p. 612-627, 2021.

613

1. INTRODUCTION

Solar radiation is the main source of energy that regulates

physical, biochemical, and physiological processes of earthly

components; it determines the microclimate, mainly

modulating temperatures, air humidity, and soil moisture, is

responsible for energy exchanges in the water-soil-plant-

atmosphere system, and is essential for ecophysiological

responses of plants, which reflects in crop yields and product

quality.

The availability of energy that reaches the earth surface is

dependent on astronomical (solar declination), atmospheric

(cloudiness, air humidity, and atmospheric turbidity), and

geographical factors that determine spatial and temporal

variations in solar radiation incidence (TERAMOTO et al.,

2019). In addition, the atmosphere composition (gases,

aerosols, water vapor, dust, and particulate matter) affects the

transmissivity of solar radiation, and the clouds are the main

reductors because they absorb specific wavelengths (infrared)

and reflect and diffuse (anisotropically) most solar radiation

(SOUZA, et al., 2016; PALÁCIOS et al., 2018).

The transitional region between the Cerrado and Amazon

biomes in northern state of Mato Grosso, Brazil, has high

mean monthly global radiation, from 16.56 ± 2.82 MJ m-2

day-1 (February, rainiest month in the region) to 21.17 ± 0.83

MJ m-2 day-1 (October), with higher atmospheric

transmissivity and solar radiation in the dry season, between

May and October (SOUZA et al., 2016). It is estimated that

a fraction of this global radiation (40% to 45%) consists of

photosynthetically active radiation (between 400 and 700

nm), which corresponds to a good part of the visible range

of the electromagnetic spectrum (BERGAMASCHI;

BERGONCI, 2017) and is responsible for activating

photosynthetic process in plants (WANG et al., 2015; WU et

al., 2019).

Excess or lack of solar radiation can be harmful to

different groups of plant species; moreover, they affect flows

of latent heat (evapotranspiration) and sensitive heat (air

temperature) (AHMED et al., 2016). Direct incidence of

global solar radiation on plants can cause significant changes

to their biochemical, physiological, and morphogenic

processes (ZHANG; ZHANG, 2017; WU et al., 2019),

oxidative stress, compromised photosynthetic activity and

structural and metabolic changes in chloroplasts are some of

the damage caused by the combination of high light and heat

stress (BALFAGÓN et al., 2019).Therefore, the use of

protected environments for agricultural crops and forest

species in regions of adverse climate conditions has been

increasingly studied, focused on improving yields and quality

of species that present difficulties for production in specific

seasons of the year or regions (HOLCMAN; SENTELHAS,

2012; AHMED et al., 2019; TANG et al., 2020; BORELLA

et al., 2020a).

The use of white plastic screens in greenhouses with

artificial or natural ventilation (AHMED et al., 2019), in

nebulization or evaporative cooling systems (AHMED et al.,

2016), and in energy system with solar photovoltaic modules

(TANG et al., 2020) predominates among the protected crop

systems. However, other plastic materials with different

physical characteristics (chemical composition, porosity,

color, and density) have been used alone or combined with

plastic screens (KOTILAINEN et al., 2018; CHOAB et al.,

2019).

Some ecophysiological studies have investigated

microclimate dynamics and effects of using photoselective

screens (aluminized or thermo reflectors, red, blue, green,

and black) on the growth and development of plants and

found promising results (HOLCMAN; SENTELHAS, 2012;

MONTEIRO et al., 2016; MAHMOOD et al., 2018;

SABINO et al., 2020; BORELLA et al., 2020a,b).

Nevertheless, a better understanding of microclimate

dynamics in these protected environments is important for

different regions and seasons of the year, since the choosing

of the adequate type of screen and percentage of shading

(porosity) is dependent on the species, cultivar, and local

climate conditions (ABDEL-GHANY, 2015; AHMED et al.,

2016; STATUTO et al., 2020).

Information on micrometeorological dynamics within

protected environments is essential for the planning and

development of hydro-agricultural activities, crop

management, and selection of agricultural and forest species

better adapted to local environmental conditions, focusing

on reducing costs, saving water, and increasing production.

The use of shading screens in hot regions decreases the

harmful effects caused by high irradiance on plants

(AHMED et al., 2016; ZHANG; ZHANG, 2017;

BORELLA et al., 2020b), providing a more uniform

distribution of temperature and relative air humidity under

shaded environments (AHMED et al., 2019; BORELLA et

al., 2021), reducing the water vapor pressure deficit (CHOAB

et al., 2019) and, consequently, the water demand of plants

(MONTEIRO et al., 2016; BORELLA et al., 2020a). In

addition, it forms a physical protection barrier against insects-

pest (MAHMOOD et al., 2018). Thus, shading is a simple

and low-cost method regarding implementation and

maintenance (ABDEL-GHANY et al., 2015).

However, controlling climate variables in these

environments is a complex and dynamic process that

depends on external conditions (HOLCMAN;

SENTELHAS, 2012) and a monitoring routine. Moreover,

the implementation costs of monitoring routine systems with

sensors and data acquisition systems (dataloggers) can be

high. Thus, micrometeorological information under

protected environments with no sensors can be obtained by

using simplified statistical models based on weather variables

under full-sun environmental conditions and that allow the

estimation of a variable of interest, such as solar radiation,

with a high degree of accuracy (SOUZA et al., 2017; ROSSI

et al., 2018; TERAMOTO et al., 2019).

Therefore, the objectives of this work were to describe

the diurnal and seasonal evolution, determine the radiometric

ratios, and fit statistic models for estimating global irradiance

(IG), photosynthetically active irradiance (IPAR), and

luminance (Lux) through shading screens with different

physical and spectral characteristics. The evaluations were

carried out in different crop seasons, in a tropical climate

region of Brazil, to obtain tools to subsidize agricultural and

forest production projects.

2. MATERIAL AND METHODS

2.1. Study region and implementation

The study was conducted in a transitional region between

the Cerrado and Amazon biomes, in Sinop, Mato Grosso

(MT), Brazil (11°51'50"S, 55°29'08"W and 384 meters of

altitude). The region presents an Aw climate (tropical hot and

wet), according to the Köppen classification (SOUZA et al.,

Solar radiation incidence under different shading screens in tropical climate: diurnal evolution and estimates

Nativa, Sinop, v. 9, n. 5, p. 612-627, 2021.

614

2013), with mean annual air temperature and relative air

humidity of 25.9 °C and 74.0%, respectively (Figure 1A and

1B). The mean daily global radiation and insolation of the

region are 17.5 to 21.2 MJ m-2 day-1 and 8.2 to 9.7 hours day-

1 in the dry season, and 16.8 to 18.6 MJ m-2 day-1 and 4.9 to

6.3 hours day-1 in the rainy season (Figure 1C).

The climate seasonality of the region is defined by two

hydrological seasons: rainy (October to April) and dry (May

to September) (Figure 1D) (SOUZA et al., 2013). The mean

annual rainfall depth is 1,945.0 mm, which is more than

1,700.0 mm in the spring-summer season, whereas the

reference evapotranspiration range is from 105.0 to 170.0

mm month-1 (3.5 to 5.5 mm day-1) between the rainy and dry

periods in the region (Figure 1D).

The solar radiation components were measured in

aboveground forest nurseries arranged in an East-West

direction, with dimensions of 3.0 × 1.0 × 1.0 m (length,

width, and height), and at 1.0 m above the ground. The full-

sun conditions were used as reference; the top, front, and

lateral sides of the experimental units were covered with

black polyolefin screens with 35%, 50%, 65%, and 80%

shading, thermo-reflector screen (Aluminet®, 50% shading),

and red, blue (Chromatinet®, 50% shading) and green

(Frontinet®, 50% shading) polyolefin screens.

Figure 1. Monthly means and standard deviations for air temperature (A), relative air humidity (B), global radiation and insolation (C),

rainfall and reference evapotranspiration (D) between September 01, 2010 and December 31, 2019 in Sinop, MT, Brazil.

Figura 1. Médias mensais e desvio-padrão da temperatura do ar (A), umidade relativa do ar (B), radiação global e insolação (C), precipitação

e evapotranspiração de referência (D), entre 01/09/2010 e 31/12/2019, em Sinop, MT, Brasil.

2.2. Measurements of IG, IPAR, and Luminance

Instantaneous measurements of incident and reflected

global irradiance (IG – W m-2), photosynthetically active

irradiance (IPAR – μmol m-2 s-1), and luminance (lux) were

measured in the protected environments (nurseries) covered

with shading screens and in the environments at full-sun

conditions. The following sensors were used: i) pyranometer

MP-200 (spectral reading range of 360 to 1,120 nm;

directional response (cosine effect): 5% up to 75° of zenith

angle; temperature response; -0.04 ± 0.04% per °C; response

time: minimum of 1.0 m s-1; non-linearity: below 1% for

measures above 1,750 W m-2); ii) pyranometer MQ-200

(spectral reading range: 410 to 655 nm (considering a

maximum of 50% wavelengths in this range); directional

response (cosine effect): 5% up to 75° of zenith angle;

temperature response; 0.06 ± 0.06% per °C; response time:

minimum of 1.0 m s-1; non-linearity: below 1% for measures

above 3,000 μmol m-2 s-1 of Apogee; and iii) lux meter (LD-

200 - Instrutherm). These sensors were fixed in a leveled

metal platform at 1.50 m height inside and at 0.50 m above

each unit (aboveground nursery).

Solar radiation was measured in the summer solstice

(December 22, 2018) and winter solstice (June 22, 2019),

when the solar declination (δ) is equal to -23.45 and 23.45°;

in the autumnal equinox (March 21, 2019) and spring equinox

(September 21, 2019), when δ = 0°; and in the local zenithal

culmination (𝜙 -11.85°) on February 18, 2019 and October

20, 2019, between 7h00min and 17h00min (local solar time),

in external (above) and internal (inside) conditions of each

aboveground nursery, with maximum intervals of 2 minutes

from each other to minimize the hour angle effects. Three

readings (replications) were carried out for each time of the

day, date, and protected environment.

Jan

Feb

MarApr

May

Jun

Jul

AugSep

Oct

NovDec

100

JanFeb

Mar

Apr

MayJun Jul

AugSepOct

Nov

Dec

Jan

Feb

MarApr

May

Jun

Jul

AugSep

Oct

NovDec

100

150

200

250

300

350

400

450

500

Rainfall Reference evapotranspiration

Months

Rainfall (mm month

-1

)

100

120

140

160

180

200

220

Reference evapotranspiration (mm month

-1

)

Jan

Feb

MarApr

May

Jun

Jul

AugSep

Oct

NovDec

Maximum Mean Minimum

Air temperature (°C)

Air relative humidity (%)

Global radiation

Global radiation (MJ m

-2

day

-1

)

Months

Insolation (hours day

-1

)

Insolation

Borella et al.

Nativa, Sinop, v. 9, n. 5, p. 612-627, 2021.

615

The data were analyzed for consistency, and different

values were excluded due to reading errors generated by the

data acquisition system or by atmospheric instability (cloudy

sky), as the case of some times in September and October.

The irradiance at the top of the atmosphere (I0) was obtained

and, then, the coefficient of atmospheric transmissivity (KT)

was determined by the ratio between global irradiance (IG)

and irradiance at the top of the atmosphere (I0).

The transmissivity of IG, IPAR, and luminance under the

polyolefin screens were obtained by the ratio between

readings of the variable inside and outside the protected

environments. The percentage of transmission, reflection,

and absorption of the irradiances were calculated based on

the incident and reflected data found for each shading screen.

2.3. Statistical models to estimate IG, IPAR, and

luminance

The hourly instantaneous values of IG, IPAR, and

luminance in shading conditions were grouped into two

databases: one with 70% of the total data for calibration of

statistical coefficients of models (with dates of December 22;

March 21; September 21, and October 20), and other with

30% of the data for validation of statistical performance of

estimation models (with the dates of February 18 and June

22). Simple linear regressions (y = a + b x) were used between

internal IG, IPAR, and luminance (dependent variable) and

external IG, IPAR, and luminance (independent variable) for

each shading condition.

The statistical performance of the generated models was

evaluated using the following indicators: mean square error

(MSE) (Eq. 1), root mean square error (RMSE) (Eq. 2) and

fit of the Willmott index (dw) (Eq. 3) (SOUZA et al., 2017).

MSE = ∑ |  |



  (01)

𝑅MSE = 󰇣∑( )



 󰇤. (02)

𝑑w = 1 − ∑( )





∑(|󰆒 

| |󰆒 

|)



 (03)

where: 𝑃 is the estimated values; 𝑂 is the measured values; 𝑛 is the

number of observations; |𝑃′ | is the absolute value of the

difference 𝑃′− 𝑂

; and |𝑂′| is the absolute value of the difference

𝑂′− 𝑂

.

The fractions of IPAR and luminance (independent

variable) in relation to IG (variable dependent) were

determined using linear regressions with grouping of data

from black screens with different porosities and from

photoselective screens with the same porosity.

3. RESULTS

The atmospheric conditions in the zenithal culmination

dates for the latitude -11.85° are presented in the Table 1.

There was 0.2 mm rainfall at the evening of October 20,

2019, when the atmospheric transmissivity was lower than

0.35, i.e., a cloudy sky (ESCOBEDO et al., 2009). This

atmospheric condition hindered the instantaneous

measurements of IG, IPAR, and luminance between 15h00min

and 17h00min in that date (Figures 2 to 4K-L).

Table 1. Daily rainfall, air temperature, relative air humidity, global radiation, and insolation in different dates of solar declination, in Sinop,

MT, Brazil.

Tabela 1. Valores diários de precipitação, temperatura do ar, umidade relativa do ar, radiação global e insolação nas diferentes datas de

declinação solar, em Sinop, MT, Brasil.

Date Rainfall Air temperature (°C) Relative air humidity (%) Global radiation Insolation

(mm) Mean

Maximum Minimum Mean

Maximum Minimum (MJ m-2 d-1) (hours)

12/22/2018

0 27.55 36.50 21.50 71.19 95.00 34.00 18.67 7.10

02/18/2019 0 25.81 32.46 21.56 83.50 99.20 49.94 20.85 9.30

03/21/2019

0 26.20 33.29 22.23 87.10 97.72 60.81 14.65 5.00

06/22/2019

0 29.94 34.66 20.43 46.80 94.94 32.94 17.65 10.40

09/21/2019 0 28.35 38.97 22.45 63.87 94.94 32.94 15.72 5.80

10/20/2019 0.2 27.49 35.74 22.35 71.98 94.90 39.70 17.74 5.30

3.1. Diurnal and seasonal evolution of solar radiation

through shading screens

The global and photosynthetically active irradiances and

the luminance presented similar dynamics throughout the

day, regardless of the microenvironment (full-sun conditions

and under shading screens) and the different solar

declinations for the same latitude (-11.85°). The highest

peaks of incidence of solar radiation occurred by 12h00min

and the lowest at sunrise and sunset, 07:00 and 17h00min,

respectively (Figures 2, 3, and 4), due to variations in the

zenith angle.

The highest incident energy levels in full-sun conditions

at approximately the solar mid-day (IG > 1000 W m-2, IPAR >

2000 μmol m-2 s-1, and luminance > 60.000 lux) were found

during the rainy season, in the summer solstice, autumnal

equinox, and at the local solar culmination, whereas the

lowest energy levels were found in the dry and dry-rainy

seasons (winter solstice and spring equinox, respectively).

The increase in black polyolefin screen shading level

gradually decreased the IG, IPAR, and luminance inside the

microenvironments in all dates (declinations) considered in

this study. Contrastingly, in the qualitative analysis, the black

polyolefin screen, thermo-reflector screen, and red, blue, and

green polyolefin screens, all with 50% shading, presented

similar IG transmissivity values; a higher transmissivity of IPAR

and luminance was found for the black polyolefin screen

when compared to the other colored screens with the same

shading percentage (Figures 3 and 4).

3.2. Solar radiation transmission, reflection, and

absorption through shading screens

The values of coefficient of atmospheric transmissivity

(KT) of global radiation ranged from 0.43 in October to 0.69

in February (rainy season), and from 0.67 in June to 0.54 in

September (dry season), confirming the results of Souza et al.

(2016), who found intervals of monthly KT ranging from 0.43