Microsoft Word - 17-Agra_47993-PG


1638 
Bioscience Journal  Original Article 

Biosci. J., Uberlândia, v. 36, n. 5, p. 1638-1644, Sept./Oct. 2020 
http://dx.doi.org/10.14393/BJ-v36n5a2020-47993 

RELATIONSHIP BETWEEN VEGETATION INDICES AND AGRONOMIC 
PERFORMANCE OF MAIZE VARIETIES UNDER DIFFERENT NITROGEN 

RATES  
 

RELAÇÃO ENTRE ÍNDICES DE VEGETAÇÃO E DESEMPENHO AGRONÔMICO 
DE VARIEDADES DE MILHO SOB DIFERENTES DOSES DE NITROGÊNIO 

 
Marcela da Silva FLORES¹; Willian Menitti PASCHOALETE¹; Fabio Henrique Rojo BAIO¹; 

Cid Naudi Silva CAMPOS¹; Ariane de Andréa PANTALEÃO¹;  
Larissa Pereira Ribeiro TEODORO¹; Carlos Antônio da SILVA JÚNIOR²;  

Paulo Eduardo TEODORO¹* 
1. Universidade Federal de Mato Grosso do Sul, Chapadão do Sul, MS, Brasil; 2. Universidade do Estado do Mato Grosso, Sinop, MT, 

Brasil. *eduteodoro@hotmail.com 
 
ABSTRACT: Precision agriculture is a set of techniques that assist the monitoring of the agronomic 

performance of the maize crop by using vegetation indices. This study aimed to verify the relationship between 
vegetation indices, plant height, leaf N content, and grain yield of three maize varieties, grown under high and 
low N as topdressing. The experiment was carried out at the Fundação de Apoio à Pesquisa Agropecuária de 
Chapadão (Fundação Chapadão), located in the municipality of Chapadão do Sul, during the 2017/2018 season. 
The experiment consisted of a randomized block design with four replications, arranged in a 3x2 split-plot 
scheme. The first factor (plots) corresponded to three open-pollinated maize varieties (BRS 4103, BRS 
Gorotuba, and SCS 154), and the second factor (subplots) consisted of two N rates applied as topdressing (80 
and 160 kg- 1). All the evaluated variables showed varieties x N interaction. Vegetation indices in maize 
varieties were influenced by the N rate applied as topdressing. Normalized Difference Vegetation Index (NDVI) 
and Soil-adjusted Vegetation Index (SAVI) showed a higher correlation with plant height. At the same time, 
Normalized Difference Red Edge (NDRE) had a stronger association with leaf N content.  

 
KEYWORDS: NDRE. NDVI. SAVI. Remote sensing. Zea mays. 
 

INTRODUCTION 
 
Maize (Zea mays L.) is the world's largest 

cereal crop, accounting for 981 million tons of grain 
in 2019. Brazilian maize production in the 
2018/2019 season was estimated in 100 million 
tons. of which 73.2 million tons were produced in 
the off-season, with a mean yield of 5,682 kg ha-1 
(CONAB, 2020).  The production of off-season 
maize has increased each year. The area cultivated 
with maize crops in the country is 17.5 million 
hectares. The state of Mato Grosso do Sul is the 
third-largest national producer of off-season maize, 
accounting for about 9.6 million tons in the 
2019/2020, with a planted area of approximately 1.8 
million hectares (CONAB, 2020). 

Currently, the production of off-season 
maize is higher than the actual season. Initially, 
maize crops were grown with little investment and 
had low yield, causing producers dissatisfaction. 
However, due to the low cost of the off-season 
maize production, producers have insisted on this 
activity. The germplasm source available for use 
consists of single, double, and triple hybrids and 

open-pollinated varieties. The use of these materials 
depends on the technological level of production 
and the climatic risks of the growing region. In the 
region of Chapadões, in the Midwest Brazilian 
Cerrado, maize is mostly grown in the off-season, 
after soybean or cotton harvest. For being highly 
rustic, maize varieties are an interesting alternative 
due to the climatic instability of this region. 
Moreover, their seeds are cheaper when compared 
with hybrids. Currently, precision agriculture has 
been used to monitor the crop throughout the 
cultivation. Precision agriculture is a set of 
techniques that allows for the specific management 
of crops and the optimization of production 
expenses by knowing the variability of different 
factors, such as soil fertility, pest and disease 
incidence, and the physiological condition of the 
plant (XU; SU, 2017; GAO et al., 2018; 
SCHWALBERT et al., 2020). One of the main 
techniques of precision agriculture is remote sensing 
(RS), a non-destructive and accurate method for 
monitoring and characterizing areas or objects.  

Vegetation indices (VIs), obtained by the 
relation between different wavelengths captured by 

Received: 08/04/19 
Accepted: 30/12/19 



1639 
Relationship between…  FLORES, M. S. et al. 

Biosci. J., Uberlândia, v. 36, n. 5, p. 1638-1644, Sept./Oct. 2020 
http://dx.doi.org/10.14393/BJ-v36n5a2020-47993 

a sensor, is one of the RS techniques most widely 
used in the evaluation of vegetation cover, vigor and 
growth dynamics, nutritional status, among other 
applications (XU; SU, 2017; BAIO et al., 2018; 
SILVA JUNIOR et al., 2018). Some VIs that show a 
high correlation with plant height and leaf nitrogen 
content (LN) can be obtained from the data 
collected by active optical sensors.  For instance, the 
Normalized Difference Vegetation Index (NDVI) is 
highly correlated to plant height when compared 
with the Red-Edge Vegetation Index (VI Red Edge). 
On the other hand, the Normalized Difference Red 
Edge (NDRE) is correlated with the N content 
(RAPER; VARCO, 2015). Baio et al. (2018) used 
different vegetation indices to estimate the 
application rate and leaf deposition in maize, and 
their result allows better planning of several 
agricultural operations. Thus, this study aimed to 
verify the relationship between vegetation indices, 
plant height, leaf N content, and grain yield of three 
maize varieties, grown under high and low N. 
 
MATERIAL AND METHODS 

 
The experiment was carried out at Fundação 

de Apoio à Pesquisa Agropecuária de Chapadão 
(Fundação Chapadão), located in the municipality of 
Chapadão do Sul (18°41'33''S, 52°40'45''W, 810 m 
high), Mato Grosso do Sul. The climate of the 
region is characterized as Savana Tropical (Aw). 
The soil is classified as Clayey Dystrophic Red 
Latosol. The experiment consisted of a randomized 
block design with four replications, arranged in a 
3x2 split-plot scheme. The first factor (plots) 
corresponded to three open-pollinated maize 
varieties (BRS 4103, BRS Gorotuba, and SCS 154), 
and the second factor (subplots) consisted of two N 
rates applied as topdressing (80 and 160 kg- 1).  

The experiment was installed in November 
2017, using a 0.45 m spacing between rows and 2.5 
plants m-1, totaling a stand of 55,555 plants ha-1. 
Each experimental unit (subplot) was composed of 
five 5.5-m rows. A rate of 300 kg ha-1 of NPK (04-
20-20) was used at the experiment installation. 
When plants were at the V4 stage, 80 kg ha-1 of N 
was applied as topdressing to the total area, using 
urea as N source. For the other treatments with a 
high N rate (160 kg ha-1), another topdressing was 
applied, using 80 kg ha-1 of N at the V6 stage. 

When the plants were at full bloom, the leaf 
N content, plant height, and vegetation index were 
measured. For analysis of the leaf N content, the 
middle third of five leaves were collected according 
to the recommendation and procedures described in 
Malavolta et al. (1989). The vegetation indices 

Normalized Difference Vegetation Index (NDVI), 
Normalized Difference Red Edge (NDRE), and 
Soil-adjusted Vegetation Index (SAVI) were 
measured using the fixed-wing Unmanned Aerial 
Vehicle (UAV) Senseflye Bee RKT, with take-off, 
flight plan, and landing autonomous control. eBee 
was equipped with the Sensefly Sequoia 
multispectral sensors. This sensor collects 
reflectance at green (550 nm), red (660 nm), near-
infrared (735 nm), and infrared (790 nm) 
wavelength, with a brightness sensor allowing 
calibration of the values collected. 

The aerial survey was carried out using the 
RTK (Real Time Kinematics) technology, which 
estimates the position of the camera at the time of 
image collection, with 0.025m accuracy. The area 
was overflown at 100m of local altitude, providing a 
spatial resolution of 0.120 m. The images were 
mosaiced and orthorectified by the Pix4Dmapper 
software. Grain yield was obtained by harvesting the 
central rows of each plot and extrapolating the value 
to kg ha-1, after correcting the grain moisture to 
13%.  

Data were subject to analysis of variance 
(test F) to verify the presence of varieties x N rates 
interaction, and means were compared by the 
Tukey’s test at the 5% of probability. Subsequently, 
the multivariate analysis of the canonical variables 
was performed to verify the interrelation between 
the treatments used and the variables evaluated. 
These analyses were carried out in the Rbio 
software (BHERING, 2017). The VI maps and 
prescription maps were processed by the 
Geographic Information Systems (GIS) ESRI 
ArcGis 10.5. 

 
RESULTS AND DISCUSSION 

 
Table 1 shows the F values calculated for 

the variables leaf nitrogen content (LN), plant height 
(PH), vegetation indices NDVI, NDRE, and SAVI, 
and grain yield (Y). The effect of varieties (V), 
Topdressing Nitrogen (N), and VxN interaction was 
significant (p <0.05) for all variables evaluated. The 
estimates of the coefficient of variation (CV) were 
lower than 10% for all variables evaluated and 
denote high experimental precision according to the 
criteria established by Pimentel-Gomes (2009). The 
high experimental precision obtained by vegetation 
indices, whose CV estimates were below 2%, is 
fundamental. The present results are inferior to 
those reported by Baio et al. (2018). The authors 
evaluated the vegetation indices NDVI and NDRE 
in maize and obtained estimates between 8 and 
12%. 



1640 
Relationship between…  FLORES, M. S. et al. 

Biosci. J., Uberlândia, v. 36, n. 5, p. 1638-1644, Sept./Oct. 2020 
http://dx.doi.org/10.14393/BJ-v36n5a2020-47993 

Table 1. F values calculated for leaf nitrogen content (LN), plant height (PH), NDVI, NDRE, and SAVI 
vegetation indices, and grain yield (GY) evaluated in three varieties of open-pollinated maize grown 
under high (160 kg ha-1) and low (80 kg ha-1) nitrogen as topdressing. 

Source of Variation LN PH NDVI NDRE SAVI GY 
Varieties (V) 7.05* 15.05* 67.15* 87.30* 233.49* 7.09* 
Topdressing Nitrogen (N) 14.5* 6.07* 5.82* 138.79* 6.25* 44.54* 
VxN 6.44* 7.01* 8.79* 23.95* 13.59* 10.46 
Coefficient of variation of the plot (%) 5.46 7.50 0.39 1.00 0.93 9.26 
Coefficient of variation of the subplot (%) 3.42 9.90 0.40 1.05 1.35 8.62 
*: significant at the 5% of probability by the F test. 
 

Table 2 shows the unfolding of the 
significant interaction between open-pollinated 
maize varieties and N rates applied as topdressing 
for the variables evaluated. The increase in the N 
rate as topdressing increased the concentration of 

leaf N and grain yield for the varieties assessed, 
except for SCS154 regarding leaf N. Likewise, 
Majerowicz et al. (2002), Machado et al. (2003), 
and Borges et al. (2006) also verified response of 
maize varieties to high N rate as topdressing. 

 
Table 2. Unfolding of the significant interaction between nitrogen rates as topdressing and open-pollinated 

maize varieties for leaf N content (LN), plant height (PH), NDVI, NDRE, and SAVI vegetation 
indices, and grain yield (GY). 

Topdressing N BRS4103 BRSGorotuba SCS154 

Leaf N (%) 

High 27.63 aB 30.83 aA 28.49 aB 
Low 25.65 bB 28.31 bA 28.12 aA 

Plant Height (m) 

High 2.12 aA 1.90 aB 2.15 aA 
Low 1.75 bC 1.85 aB 1.90 bA 

NDVI 

High 0.81 aB 0.80 aC 0.82 aA 
Low 0.80 bB 0.80 aB 0.81 bA 

NDRE 

High 0.33 aA 0.31 aB 0.33 aA 
Low 0.30 bB 0.28 bB 0.32 aA 

SAVI 

High 0.63 aB 0.59 aC 0.66 aA 
Low 0.61 bB 0.60 aB 0.65 aA 

Grain Yield (kg ha-1) 
High 11095.53 aB 13207.12 aA 11136.43 aB 
Low 8605.07 bB 9829.30 bA 8314.57 bB 
Lowercase letters in the same column and uppercase letters in the same row do not differ by the Tukey’s test at the 5% probability level. 
 

The PH value of varieties BRS 4103 and 
SCS 154 was higher at a higher nitrogen rate (Table 
2). Similarly, varieties had a higher NDVI index at a 
higher N rate. However, BRS Gorotuba showed no 
difference in leaf N content and NDVI in relation to 
N rates applied as topdressing. These results suggest 
a positive linear association between plant height 
and NDVI and agree with the results of Baio et al. 
(2018), who verified a high linear correlation 
between NDVI and plant height in maize plants. 

Regarding NDRE, the N rates led to differences for 
varieties BRS 4103 and BRS Gorotuba, revealing a 
strong association with the results obtained for the 
leaf N content. Differences between varieties were 
observed for all vegetation indices evaluated in this 
study. These results reinforce the possibility of 
using precision agriculture in maize breeding 
programs since the evaluated vegetation indices 
could separate the spectral behavior of each 
evaluated variety. 



1641 
Relationship between…  FLORES, M. S. et al. 

Biosci. J., Uberlândia, v. 36, n. 5, p. 1638-1644, Sept./Oct. 2020 
http://dx.doi.org/10.14393/BJ-v36n5a2020-47993 

Precision agriculture has become a 
necessary practice to increase yield and product 
quality. Besides providing more profitable 
conditions, PA allows the monitoring of the crop 
area, generating data through field analysis and 
equipment, such as impact penetrometer 
(SUDDUTH et al., 1994), penetroLOG, and UAVs, 
to improve the efficiency and durability of 
production systems. 

The canonical variables analysis was used to 
investigate the simultaneous association between the 
treatments and the variables evaluated. This 
multivariate statistical analysis allows reducing the 
dimensionality of data obtained in the experiment, 
seeking for combinations of these varied traits in 

which the correlation of these variables is ignored 
(KHATTREE; NAIK, 2000). Therefore, this 
procedure expresses the variation between the 
treatments and the variables, considering the 
residual dispersion of each variable. This technique 
is similar to the principal components analysis but 
should be preferred when the study uses an 
experimental design (replications). Each canonical 
variable is a linear combination of the variables 
used. To represent the scores on a two-dimensional 
graph, the percentage of variance accumulated in the 
first two canonical variables must be greater than 
80.0% (MINGOTI, 2005). In this work, the variance 
accumulated in the first two canonical variables was 
88.0%, allowing a precise interpretation. 

 

 
Figure 1. Canonical variables analysis demonstrating the association between the treatments and the variables 

evaluated. V1: BRS 4103; V2: BRS Gorotuba; V3: SCS 154; LN: leaf nitrogen; PH: plant height; 
GY: grain yield. 

 
The canonical variables analysis shows that 

variety BRS Gorotuba was the most yielding and 
had higher leaf N content, regardless of the N rate. 
Results revealed a high linear association between 
PH and the vegetation indices used in this study 
since the angle between the vectors of these 
variables was lower than 90º. Conversely, LN is 
highly correlated with Y due to the proximity 
between their vectors. NDRE was the VI that 
correlated the most with these variables due to the 

angle between their vectors, which was lower than 
90º.  

Nitrogen participates directly in the 
production of chemical compounds, such as proteins 
and chlorophylls, which are essential for plant 
metabolism (ANDRADE et al., 2003). Thus, this 
nutrient is fundamental at the initial stages of plant 
growth, when absorption precision is intense 
(BASSO; CERETTA 2000). Maize is one of the 
most demanding crops when it comes to fertilizers, 
especially nitrogen-based. This cereal responds 



1642 
Relationship between…  FLORES, M. S. et al. 

Biosci. J., Uberlândia, v. 36, n. 5, p. 1638-1644, Sept./Oct. 2020 
http://dx.doi.org/10.14393/BJ-v36n5a2020-47993 

expressively to this nutrient, influencing the final 
production. According to Melgar et al. (1991), N is 
the nutrient that has the most significant effect on 
final maize production.  

Similarly, Raper and Varco (2015) observed 
that NDVI is more associated with plant height, 
while NDRE is associated with leaf N content. 
According to these authors, NDRE correlated more 

strongly to leaf N status and total plant N content 
when compared with indices that rely on reflectance 
in green or red regions, such as NDVI. Among the 
indices evaluated here, NDRE shows the highest 
relation with grain yield. The spatial variability of 
this index is shown in Figure 2, where the N 
application resulted in higher values of NDRE, 
regardless of the evaluated variety. 

 
Figure 2. Spatial variability of NDRE evaluated in three maize varieties grown under high and low N rates 

applied as topdressing. 
 

High NDVI and NDRE values are related to 
healthy and dense vegetation, which has low 
reflectance values in the red wavelength (red and 
red-edge) and high reflectance values in the NIR 
wavelength (XUE; SU, 2017). The red edge is 
considered one of the most sensitive regions to 
changes in chlorophyll and N status (FITZGERALD 
et al., 2010; LI et al., 2014). Therefore, the results 
obtained here reveal the greater efficiency of NDRE 
in the assessment of the leaf N content in maize, 
allowing the discrimination of maize varieties 
grown under high and low N level conditions. Such 

findings can contribute to support agricultural 
decisions related to managing risks within crop 
production to increase the maize yield. 

 
CONCLUSION 

 
The vegetation indices analyzed for the 

different maize varieties were influenced by the N 
rate applied as topdressing. NDVI and SAVI indices 
showed a higher correlation with plant height, while 
NDRE had a stronger association with leaf N 
content.

 
 
RESUMO: A agricultura de precisão é um conjunto de técnicas que auxiliam no monitoramento do 

desempenho agronômico da cultura do milho utilizando índices de vegetação. Este trabalho teve como objetivo 
verificar a relação entre índices de vegetação, altura de planta, teor de N foliar e rendimento de grãos de três 
variedades de milho, cultivadas sob alto e baixo N, em cobertura. O experimento foi realizado na Fundação de 
Apoio à Pesquisa Agropecuária de Chapadão, localizada no município de Chapadão do Sul, na safra 
2017/2018. O experimento consistiu de um delineamento de blocos casualizados com quatro repetições, 
dispostos em esquema de parcelas subdivididas 3x2. O primeiro fator (parcelas) correspondeu a três variedades 
de milho de polinização aberta (BRS 4103, BRS Gorotuba e SCS 154), e o segundo fator (subparcelas) 
consistiu de duas doses de N aplicadas como cobertura (80 e 160 kg-1). Todas as variáveis avaliadas 
apresentaram interação variedades x N. Os índices de vegetação nas variedades de milho foram influenciados 
pela dose de N aplicada como cobertura. Os índices NDVI e SAVI mostraram uma maior correlação com a 
altura da planta, enquanto o NDRE apresentou uma associação mais forte com o conteúdo de N foliar. 

 
PALAVRAS-CHAVE: NDRE. NDVI. SAVI. Sensoriamento remoto. Zea mays. 
 
 



1643 
Relationship between…  FLORES, M. S. et al. 

Biosci. J., Uberlândia, v. 36, n. 5, p. 1638-1644, Sept./Oct. 2020 
http://dx.doi.org/10.14393/BJ-v36n5a2020-47993 

REFERENCES 
 
ANDRADE, A. C.; ANDRADE, A. C., FONSECA, D. D., QUEIROZ, D. S.; SALGADO, L. T.; CECON, P. 
R. Adubação nitrogenada e potássica em capi m-elefante (Pennisetum purpure um Sc hu m.cv. Napier). 
Ciência e Agrotecnologia, v. 27, p. 1643-1651, 2003. 
 
BAIO, F. H. R.; SILVA, E. E.; SOUZA, M. A. V.; ZANIN, A. R. A.; TEODORO, P. E. Vegetation indices to 
estimate spray application rates of crop protection products in corn. Agronomy Journal, v. 110, p. 1-15, 2018. 
https://doi.org/10.2134/agronj2017.12.0718 
 
BASSO, C. J.; CERETTA, C. A. Manejo do nitrogênio no milho em sucessão a plantas de cobertura de solo, 
sob plantio direto. Revista Brasileira de Ciência do Solo, v. 24, n. 4, p. 905-915, 2000. 
https://doi.org/10.1590/S0100-06832000000400022 
 
BHERING, L. L. Rbio: A tool for biometric and statistical analysis using the R platform. Crop Breeding and 
Applied Biotechnology, v. 17, p. 187-190, 2017. https://doi.org/10.1590/1984-70332017v17n2s29 
 
BORGES, E.A.; FERNADES, M.S.; LOSS, A.; SILVA, E.V.; SOUZA, S.R. Acúmulo e remobilização de 
nitrogênio em variedades de milho. Caatinga, v. 19, n. 3, p. 278-286, 2006. 
 
CONAB. Acompanhamento da safra brasileira de grãos: safra 2019/20, sexto levantamento. Companhia 
Nacional de Abastecimento (CONAB), Brasília. Disponível em: <https://www.conab.gov.br/info-
agro/safras/graos/boletim-da-safra-de-graos>. Acesso em 05 de abril de 2020. 
 
FITZGERALD, G.; RODRIGUEZ, D.; O’LEARY, G. Measuring and predicting canopy nitrogen nutrition in 
wheat using a spectral index – the canopy chlorophyll content index (CCCI). Field Crops Research, v. 116, p. 
318–324, 2010. https://doi.org/10.1016/j.fcr.2010.01.010 
 
GAO, F.; ANDERSON, M.; DAUGHTRY, C.; JOHNSON, D. Assessing the Variability of Corn and Soybean 
Yields in Central Iowa Using High Spatiotemporal Resolution Multi-Satellite Imagery. Remote Sensing, v. 10, 
p. 1489, 2018. https://doi.org/10.3390/rs10091489 
 
LI, F., MIAO, Y., FENG, G., YUAN, F., YUE, S., GAO, X.; LIU, Y.; LIU, B.; USTIN, S.L.; CHEN, X. 
Improving estimation of summer maize nitrogen status with red edge-based spectral vegetation indices. Field 
Crops Research, v. 157, p. 111-123, 2014. https://doi.org/10.1016/j.fcr.2013.12.018 
 
KHATTREE, R.; NAIK, D.N. Multivariate data reduction and discrimination with SAS software. Cary, 
NC, USA: SAS Institute Inc., 2000. 558p. 
 
MACHADO, A. T.; MACHADO, C. T. T.; MIRANDA, G. V.; COELHO, C. H. M.; GUIMARÃES, L. J. M. 
Resposta de variedades de milho a níveis e fontes de nitrogênio. Boletim de Pesquisa e Desenvolvimento. 
Planaltina: Embrapa Cerrados, 2003. 27p. 
 
MAJEROWICZ, N.; PEREIRA, J. M. S.; MEDICI, L. O.; BISON, O.; PEREIRA, M. B.; SANTOS JUNIOR, 
U. M. Estudo da eficiência de uso do nitrogênio em variedades locais e melhoradas de milho. Revista 
Brasileira Botânica, v. 25, p. 129-136, 2002. https://doi.org/10.1590/S0100-84042002000200002 
 
MALAVOLTA, E.; VITTI, G.C.; OLIVEIRA, A. S. Avaliação do estado nutricional das plantas: princípios 
e aplicações. Piracicaba: Associação Brasileira para Pesquisa da Potassa e do Fosfato, 1989. 201p. 
 
MELGAR, R. J.; SIMITH, T. J.; CRAVO, M. S.; SÁNCHEZ, P. A. Rates and dates of nitrogen fertilizer 
application for maize on a latossol in the central Amazonia region. Revista Brasileira de Ciência do Solo, v. 
15, n. 3, p. 289-296, 1991. 
 



1644 
Relationship between…  FLORES, M. S. et al. 

Biosci. J., Uberlândia, v. 36, n. 5, p. 1638-1644, Sept./Oct. 2020 
http://dx.doi.org/10.14393/BJ-v36n5a2020-47993 

MOLIN, J. P.; JUNIOR, P. A. V.; NETO, D. D.; FAULIN, G. D. C.; MASCARIN, L. Variação espacial na 
produtividade de milho safrinha devido aos macronutrientes e à população de plantas. Revista Brasileira de 
Milho e Sorgo, v. 6, n. 3, 2010. https://doi.org/10.18512/1980-6477/rbms.v6n3p309-324 
 
PIMENTEL GOMES, F. Curso de estatística experimental. 15.ed. Piracicaba: FEALQ, 2009. 451p. 
RAPER, T. B.; VARCO, J. J. Canopy-scale wavelength and vegetative index sensitivities to cotton growth 
parameters and nitrogen status. Precision Agriculture, v. 56, p. 345-353, 2015. 
 
SCHWALBERT, R. A.; AMADO, T.; CORASSA, G.; POTT, L. P.; PRASAD, P. V.; CIAMPITTI, I. A. 
Satellite-based soybean yield forecast: Integrating machine learning and weather data for improving crop yield 
prediction in southern Brazil. Agriculture and Forest Meteorology, v. 284, 107886, 2020. 
https://doi.org/10.1016/j.agrformet.2019.107886 
 
SILVA JUNIOR, C. A.; NANNI, M. R.; SHAKIR, M.; TEODORO, P. E.; DE OLIVEIRA-JÚNIOR, J. F.; 
CEZAR, E.; SHIRATSUCHI, L. S. Soybean varieties discrimination using non-imaging hyperspectral sensor. 
Infrared Physics and Technology, v. 89, p. 338-350, 2018. https://doi.org/10.1016/j.infrared.2018.01.027 
 
SUDDUTH, K. A.: BORGELT, S. C: BIRRELL, S. J.; KITCHEN, N. R. Withinfield location anel sensing 
technology. Columbia: University of Missouri. 1994. 12p. 
 
XUE, J.; SU, B. Significant remote sensing vegetation indices: A review of developments and applications. 
Journal of Sensors, v. 2017, p. 1-17, 2017. https://doi.org/10.1155/2017/1353691