The spatial dynamics of agriculture in Brazil in 2008 and 2018
Marcio Aloisio da Silva Júnior
1
*; Patrícia de Siqueira Ramos
2
; Lincoln Frias
3
DOI: https://doi.org/10.35699/2447-6218.2021.35385
Abstract
Given the great influence of agriculture on Brazilian society and economy, it is important to understand the spatial dynamics
of different agricultural cultures as a way to identify strategic producing regions, a key information for the development of
public policies. Therefore, the purpose of this work is to carry out, through the exploratory analysis of spatial data, a
comparative analysis of the spatial dependence of the production of different crops in Brazil, in the years 2008 and 2018. This
is done using data from the Municipal Agricultural Production (PAM) by the Brazilian Institute of Geography and Statistics
(IBGE) and focusing on the 12 crops with the highest total production values in 2018 (cotton, rice, banana, coffee, sugarcane,
beans, tobacco, orange, cassava, corn, soybeans and tomato). The comparative analysis of cultures evidenced the existence of
four groups in relation to the spatial dynamics of the high-high groupings in the two years studied. The first group is formed
by cultures (cotton, rice, coffee, sugarcane, orange and tobacco) that tend to be concentrated in specific areas of the territory
and did not show relevant changes in the clusters between the years under study, while the second group includes crops that
expanded their areas bet- ween the two years (beans, corn and soybeans). The third group is formed by crops with smaller
clusters and more spaced apart among themselves (banana and tomato). Finally, cassava was not included in any of the groups
due to its dispersed production throughout the territory and also because it suffered a reduction in some regions.
Keywords:
Agricultural Zoning. Exploratory spatial data analysis. Moran’s I.
A dinâmica espacial da agricultura no Brasil em 2008 e 2018
Resumo
Dada a grande influência da agricultura na sociedade e na economia brasileira, é importante compreender a dinâmica espacial de
diferentes culturas agrícolas para identificar regiões produtoras estratégicas, servindo como subsídio para políticas públicas. Por
isso, o objetivo deste trabalho é realizar, por meio da análise exploratória de dados espaciais, uma análise comparativa da
dependência espacial da produção de diferentes culturas agrícolas no Brasil, nos anos de 2008 e 2018. São utilizados dados
da Produção Agrícola Municipal (PAM) do IBGE, da qual foram selecionadas as 12 culturas com os maiores valores totais da
produção em 2018 (algodão, arroz, banana, café, cana-de-açúcar, feijão, fumo, laranja, mandioca, milho, soja e tomate). A
análise comparativa das culturas evidenciou a formação de quatro grupos, em relação à dinâmica espacial dos agrupamentos
alto-alto nos dois anos estudados. O primeiro grupo é formado pelas culturas que tendem a se concentrar em áreas específicas
do território e não apresentaram relevan- tes alterações nos agrupamentos (algodão, arroz, café, cana-de-açúcar, laranja e
fumo), enquanto o segundo grupo engloba as culturas que expandiram suas áreas entre os dois anos (feijão, milho e soja). O
terceiro grupo é formado por culturas que apresentaram agrupamentos com menores extensões e mais espaçados entre si
(banana e tomate). Por fim, a mandioca não foi incluída em nenhum dos grupos por ter sua produção dispersa em todo o
território e por apresentar redução nos agrupamentos em algumas regiões.
Recebido para publicação em 26 de Julho de 2021. Aceito para publicação 22 de novembro de 2021.
e-ISSN: 2447-6218 /
ISSN: 2447-6218. Atribuição CC BY.
CADERNO DE CIÊNCIAS AGRÁRIAS
Agrarian Sciences Journal
2
Silva Júnior, M. A. et al.
Palavras-chave:
Análise exploratória de dados espaciais. I de Moran. Zoneamento agrícola.
Introduction
The influence and importance of agriculture in
society comes from crop production for human and ani- mal
food and also from its participation in the economic
development of countries and regions. The spatial distri-
bution of agricultural crops in certain regions depends on
climatic conditions, soil types, topography, among other
factors, including those related to public policies (for
example, agricultural zoning and agricultural insurance
subsidies).
conditions, soil types and topography. Furthermore, bet- ween
different producing regions, the effect of interde-
pendence
manifests itself in different ways, including the
spatial
diffusion of phenomena that influence neighbors and the
processes of spatial competition, whether in the
expansion of
the agricultural frontier or in the formation
of agricultural
belts.
Spatial analysis comprises a set of techniques
capable of measuring properties and relationships accor-
ding
to the location of a given phenomenon. In general, it aims to
incorporate space into the analysis (CÂMARA et al., 2004).
In Brazil, agriculture is one of the main agents
of
economic development. However, this sector has gone
through
a series of political changes since 1990, inclu- ding greater
external opening, market deregulation and new public
policies for the sector. After that, the sector developed a
more competitive structure, both internally and externally
(PEROBELLI et al., 2007).
The analysis of the dependency structure between
the
values observed in the different areas of the study is done by
the so-called spatial autocorrelation function. It is called
autocorrelation because it takes into account
the correlation
with the variable itself at a different point
in space
(ANDRADE et al., 2007).
The occurrence of unsatisfactory results in agricul-
tural activity can cause damage to societies. Though, the
development of agriculture contributed to reducing the risks
of such results, since in addition to being assumed, such risks
started to be quantified based on probabilistic calculations
(SANTOS; MARTINS, 2016).
According to Almeida (2012), positive spatial
autocorrelation indicates that, in general, regions that
present high values for the variable of interest tend to have
around them regions that also present high values for this
variable (high-high type regions) or that regions with low
values tend to be surrounded by regions that also have low
values (low-low). On the other hand, ne- gative spatial
autocorrelation indicates the dissimilarity between the values
of the variable of interest and its location (high-low or low-
high).
An important milestone for the sector in the cou- ntry
began in 1973 with the institution of the Agricultural
Activity
Guarantee Program (PROAGRO), which began its
operations in 1975. The central aim of this initiative was to
guarantee the activity of rural producers in cases where the
costs invested in their projects were affected by adverse
natural phenomena, given that the unpredic- tability of
climate variability is one of the main risks to agricultural
activity (CUNHA; ASSAD, 2001).
The purpose of this paper is to carry out, through
the
exploratory analysis of spatial data, a comparative analysis
of the spatial dependence of the production of different
crops in Brazil, in the years 2008 and 2018. The remainder
of this paper is organized as follows. The next section
presents the dataset and the methods in exploratory spatial
data analysis. Section 3 contains a description and a
discussion of the main findings from the exploratory and
spatial analysis. Finally, some con- clusions are presented in
Section 4.
With the beginning of the Program, it became
evident that PROAGRO would have to undergo serious
structural and operational changes to maintain itself as an
instrument of agricultural policy, given that it had a deficit
history throughout its first 22 years of existence (ROSSETI,
2001).
As of the 1996 harvest, the Program underwent
important changes in its structure, the main one being
the
implementation of the Agricultural Zoning. With this,
the
Program started to be guided by new rules such as
not
covering the multiplicity of risks and encouraging the
use of
technologies, including climate risk zoning, the
indicated
cultivars and direct planting (CUNHA; ASSAD,
2001).
Material e methods
The present work uses data from the Municipal
Agricultural Production of the Brazilian Institute of Geo-
graphy and Statistics (IBGE), which presents information
on
31 crops from temporary crops and 33 crops from
permanent crops (IBGE, 2019). The 12 crops with the
highest total production values in Brazil, in reais, in 2018
were
selected. The Python programming language was used to
process the data and apply the methods.
According to Almeida et al. (2008), the develop-
ment of crops in space is heterogeneous as it depends on
factors such as different production techniques, climatic
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385
3
The spatial dynamics of agriculture in Brazil in 2008 and 2018
The selected crops are presented in Table 1,
ordered according to the total production value in 2018.
In addition to this value, the total quantities produced, in
tons, in 2008 and 2018 are also present in the table.
Table 1 – Total quantities produced and total values of production of agricultural crops ordered according to the total value of
production in the year 2018
total quantity produced
1
total production value
2
crop
2008
2018
2018
Soybeans
Sugarcane Corn
(in grain)
Coffee (in grain total)
Cotton (seed) Cassava
Orange
Rice (in paddy)
Banana (in bunch)
Tobacco (in leaf)
Beans (in grain)
Tomato
59,833,105
645,300,182
58,933,347
2,796,927
3,983,181
26,703,039
18,538,084
12,061,465
6,998,150
851,058
3,416,194
3,867,655
117,887,672
746,828,157
82,288,298
3,556,638
4,956,044
17,644,733
16,713,534
11,749,192
6,752,171
762,266
2,915,030
4,110,242
127,549,867
52,238,542
37,644,731
22,623,368
12,790,580
9,718,965
9,450,570
8,650,626
6,975,536
6,510,625
5,693,442
5,088,543
1 Total quantity produced in tons
2 Total production value in reais in 2018.
To compare the quantities produced by the mu-
nicipalities in 2008 and 2018, thematic maps construc- ted
according to four ranges of values obtained by the Fisher-
Jenks method were used, a method in which the variance
within the class is minimized while maximizing the variance
between classes. The municipalities that did not present
production in the year in question were disregarded. In order
to establish a direct comparison
between the maps, such
intervals were obtained for each
crop according to the year
2018, applying this division also to the 2008 data.
absolute values presented in Table 1 must also be taken into
account.
Subsequently, summary measures were obtained on
the amount produced (in tons) of Brazilian municipa- lities.
Again, for each agricultural crop, the municipalities
that did
not have production in the years under study were
disregarded. Thus, Table 2 presents the number of producing
municipalities and the sample means, mini- mum values,
maximum values and the quartiles of the
distribution of
production by municipality for each of the
selected crops.
The spatial weighting matrix is used as a crite- rion
for its definition of the neighborhood relationship between
the municipalities. The contiguity convention
adopted was
the queen convention, and the normalization
of the weighting
matrix in relation to the lines was also performed.
It is observed in Table 2 that corn, beans and cas-
sava had the largest number of producing municipalities (over
4,700 municipalities), while the lowest numbers were
observed for cotton and tobacco (722 and 925). Data for
2018 are presented in Table 3.
Comparing the statistics between 2008 and 2018, it
can be highlighted that, between the years, the five crops
with
the highest number of producing municipalities in 2008 also
had the highest values in the year 2018. Ho- wever, with the
exception of corn, there were changes in
the ordering of the
numbers of producing municipalities. Regarding the smaller
numbers of producers, in addition
to the fact that cotton and
tobacco crops also presented the lowest values in 2018, these
values were even lower than at the beginning of the
comparison.
Results and discussion
Figure 1 shows the participation of regions in the
country in the production of each crop as a percentage, in order
to illustrate the proportion of the quantity produced
by these
regions in 2008 and 2018. It can be seen that certain crops
presented most of their products in specific
regions of the
country, such as cotton in the Midwest, rice and tobacco in the
South and coffee, sugarcane and orange
in the Southeast. It
should be noted that since these are percentage values of the
totals produced, changes in the
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385
4
Silva Júnior, M. A. et al.
Figure 1 – Percentage of the amount produced, in tons, of crops in each region of the country in 2008 and 2018
Table 2 – Summary measures of the quantity produced, in tons, of selected agricultural crops in 2008
x
–
Crop
n
Min.
q1(25%)
q2(50%)
q3(75%)
Max.
Banana
Bean
Cassava
Coffee
Corn
Cotton
Orange
Rice
Soybeans
Sugarcane
Tobacco
Tomato
3,494
4,741
4,726
1,838
5,322
722
3,028
3,417
1,83
3,727
925
1,903
2,002.90
730.06
5,650.24
1,521.72
11,073.53
5,516.87
6,122.22
3,529.84
32,695.69
173,141.99
920.06
2,032.40
2.00
1.00
6.00
1.00
1.00
1.00
1.00
1.00
1.00
2.00
1.00
3.00
80.00
44.00
400.00
22.00
349.25
12.00
52.00
28.00
1,024.50
834.00
19.00
60.00
220.00
136.00
1,362.50
124.00
1,536.00
54.00
165.00
148.00
6,240.00
3,600.00
104.00
200.00
834.50
450.00
4,477.50
913.75
7,732.50
360.00
700.00
945.00
26,415.00
44,730.00
612.00
800.00
158,400.00
123,840.00
592,000.00
40,315.00
997,440.00
534,342.00
556,160.00
650,642.00
1,794,000.00
10,260,000.00
23,650.00
239,400.00
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385
5
The spatial dynamics of agriculture in Brazil in 2008 and 2018
Table 3 – Summary measures of the quantity produced, in tons, of selected agricultural crops in 2018
x
–
Crop
n
Min.
q1(25%)
q2(50%)
q3(75%)
Max.
Banana
Bean
Cassava
Coffe Corn
Cotton
Orange
Rice
Soybeans
Sugarcane
Tobacco
Tomato
3,399
4,33
4,73
1,448
5,069
223
2,42
1,925
2,318
3,345
651
1,788
1,986.52
673.22
3,730.39
2,456.24
16,233.64
22,224.41
6,906.42
6,103.48
50,857.49
223,267.01
1,170.92
2,298.79
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
2.00
2.00
1.00
4.00
51.00
24.00
170.00
16.00
185.00
14.00
40.00
10.00
1,800.00
420.00
20.00
58.00
160.00
84.00
607.50
135.00
1,108.00
390.00
155.00
78.00
10,800.00
1,800.00
110.00
172.50
748.00
298.00
2,280.00
1,616.25
6,840.00
10,482.50
902.50
720.00
43,557.50
75,000.00
690.50
800.00
210,975.00
63,870.00
373,809.00
82,830.00
2,851,200.00
756,891.00
510,000.00
755,486.00
2,232,000.00
8,118,000.00
26,790.00
356,000.00
In general, it is noted that, between 2008 and
2018,
cotton, rice, coffee, sugarcane, tobacco and orange
crops had a
reduction in the number of producing muni-
cipalities, while
their average quantities produced increa-
sed. On the other
hand, the cassava crop had a greater
number of producing
municipalities and a reduction in the average amount produced
in 2018. Corn and tomato crops
had higher averages in the last
year and a reduction in
the number of producing
municipalities. In 2018, banana
and bean crops presented
smaller numbers of producing municipalities and smaller
average quantities produced,
average. It should be reminded that the closer to 1 the value
of the culture statistics, the greater its spatial con- centration.
Among the changes in the value of the statistics
between the two years, there is an increase in the value
for
the crops coffee, sugar cane, beans, cassava and corn,
and the
reduction in the spatial concentration for the crops cotton,
rice, banana, tobacco, orange, soybeans
and tomato. Here, it
is worth noting that the tomato crop
had the lowest global
spatial autocorrelation statistic in the two years analyzed,
having even decreased in 2018.
while soybeans suffered the opposite movement. In both
years,
there
is
a
large
discrepancy
between
the
mean
(x
–) and
median (q2) values for all cultures, indicating that the mean
value is highly influenced by high values. This
is an early
indication that some municipalities have much higher
production than others, concentrating most of the
production.
With these results in hand, a more detailed
analysis of the location of spatial regimes in each of the
cultures was carried out using thematic maps and LISA
maps (local indicators of spatial association). Thus, it was
possible to identify the location of spatial clusters and
spatial outliers.
To identify the existence or absence of spatial
autocorrelation in cultures, the global Moran statistic was
used
as a first step. The significance of such values was verified
by means of the random permutation test, with their
respective pseudo-probability values present in the p-value
columns of Table 4. The cultures were sorted in descending
order, according to the statistical values in 2008.
Through the data already presented and the
comparison of maps prepared for each of the cultures, it is
possible to consider the existence of four groups of cultures
according to their spatial dynamics between 2008 and 2018.
The first group is formed by cultures that tend to
be concentrated in a few areas, namely: cotton, rice, coffee,
sugar cane, orange and tobacco. Such cultures
were initially
identified by the predominant participation
of certain regions
in the total quantities produced in the country. In addition,
they showed similar behavior in relation to changes in
summary measures between 2008 and 2018, namely the
reduction in the number of producing municipalities and
higher average quantities
produced. The thematic maps of the
cultures of this group
are shown in Figure 2.
At a significance level of 5%, all cultures showed
spatial autocorrelation in both years, which indicates that
the amount produced is spatially autocorrelated in Brazilian
municipalities. In addition to that, given the positive signs
of the statistics, there is evidence that mu- nicipalities with
quantities produced above the average tend to be neighbors
of municipalities with quantities produced above the
average and municipalities with quantities produced below
the average, in general, are also neighbors of municipalities
with quantities below
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385
6
Silva Júnior, M. A. et al.
Table 4: Global Moran’s I for crops in the years 2008 and 2018 and pseudo-p-values
2008
2018
Crops
Moran’s I
p-value
Moran’s I
p-value
Tobacco
Coffee
Sugarcane
Soybeans
Rice
Orange
Corn
Banana
Cassava
Bean
Cotton
Tomato
0.614
0.599
0.599
0.546
0.536
0.501
0.489
0.341
0.337
0.305
0.266
0.150
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.511
0.601
0.621
0.540
0.498
0.418
0.545
0.281
0.413
0.429
0.252
0.088
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.002
Figure 2 – Thematic maps of the cultures of the first group for the quantity produced, in tons, in the years 2008 and 2018
The thematic maps show that the municipalities
with
the largest quantities of crops produced are very con- centrated
in certain locations, in addition to the complete lack of
production of these crops in certain regions of the
country. In
the case of cotton, municipalities with large quantities
produced can be observed in the Center-West and Northeast
of the country, while municipalities with the largest
quantities of rice and tobacco produced were concentrated
mainly in the South.
of quantity produced for orange and sugarcane crops,
especially in the state of São Paulo.
The LISA maps of the crops in the first group are
shown in Figure 3.
The LISA maps of the crops in the first group show
high-high clusters in certain locations in the country, but
mainly around the municipalities with the largest quanti- ties
produced highlighted in the thematic maps. Between
the two
years, it is noted that the high-high groupings did not have
major changes, with the exception of rice crop clusters
located in the Midwest and North, which
reduced their
extension, and sugar cane, which expanded
the extent of the
cluster in the Southeast.
Coffee had the municipalities with the largest
quantities produced in the Southeast region in both years.
In the
North region, such values were only observed in 2018, due
to the increase in the amount produced in
municipalities in
the state of Rondônia. In the Southeast, the largest
municipalities were also concentrated in terms
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385
7
The spatial dynamics of agriculture in Brazil in 2008 and 2018
The second group of crops is formed by beans, corn
and soybeans, which expanded their areas between 2008 and
2018 due to increased production in neighbo- ring
municipalities of areas that already were major
producers and, consequently, the change or significance
of
their spatial regimes. The thematic maps of the cultures
of the
second group are shown in Figure 4.
Figure 3 – LISA maps of crops in the first group for the amount produced, in tons, in 2008 and 2018
Figure 4 – Thematic maps of the cultures of the second group for the quantity produced, in tons, in the years 2008 and 2018
It can be noted on such maps that the municipali- ties
that produce such crops occupy a greater extension of the
territory, not limited to certain regions of the country.
It is also
observed that for the three crops in the group
there was an increase in the amount produced in the vicinity
of the municipalities that stood out in 2008. The LISA maps
of the crops in the second group are shown in Figure 5.
Figure 5 – LISA maps of crops in the first group for the amount produced, in tons, in 2008 and 2018
For the cultures of the second group, the ex-
pansion of high-high clusters is evident, especially in the
Midwest region, and it is important to highlight the changes
in the spatial regime of the municipalities from
low-high to
high-high. Such changes indicate an increase
in the level of
crop production for municipalities that
in 2008 had their produced quantities considered to be
different from neighboring municipalities. In addition to the
change in the spatial regime, it is worth noting that such
expansion was also due to the significance of the spatial
regimes of municipalities that in 2008 were considered non-
significant.
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385
8
Silva Júnior, M. A. et al.
According to Trentin et al. (2018) there were
important technological advances in the last 40 years that
brought a significant evolution in agriculture, with emphasis
on soybean, wich is the main crop in Brazil,
standing out for its high production capacity and profi-
tability.
The third is formed by banana and tomato, and
the
thematic maps of such cultures are presented in Figure
6.
Figure 6 – Thematic maps of the cultures of the third group for the amount produced, in tons, in the years 2008 and 2018
These thematic maps show that the areas con-
taining the municipalities with the largest quantities
produced have smaller extensions and are more distant
from each other. In 2018, there was an increase in the
amount produced in some municipalities, while in others
there
was a reduction in the amount produced.
Figure 7 – LISA maps of crops in the third group for the amount produced, in tons, in 2008 and 2018
LISA maps also show the smaller extension and
distance between the high-high clusters. In this sense, in
relation to the banana crop, it is important to stress that
the
areas highlighted in the North region present greater extension
due to the number of municipalities, but mainly
due to their
larger area in the region. In addition, the North region was
responsible for only about 12% and
13% of the total produced in the country in 2008 and 2018,
respectively, as shown in Figure 1.
Finally, there is the cassava crop, which, as it does
not have characteristics in common with the cul- tures of
the other groups, will have its results presented individually.
Its thematic maps are shown in Figure 8.
Figure 8 – Cassava thematic maps for the amount produced, in tons, in 2008 and 2018
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385
9
The spatial dynamics of agriculture in Brazil in 2008 and 2018
The maps show that the municipalities producing
cassava are found in almost the entire territory of the
country, with a concentration of those with the largest
quantities produced in the North region and between the
South
and Midwest of the country. Between 2008 and
2018, there was a reduction in the amount produced in the
Northeast region due to the change in the first range of values
in municipalities that in 2008 had higher pro- duction.
Figure 9 shows the LISA maps for this crop.
Figure 9 – LISA maps of the cassava crop for the amount produced, in tons, in 2008 and 2018
Next, some considerations will be presented in
order to identify possible explanations for the main changes
observed in the cultures under study.
According to Silva and Wander (2013), con-
cerning the first harvest, the South is traditional in the
cultivation of the crop, occupying in the period 2006-2011
the
first place in relation to area and production. Then come the
Southeast, Midwest, Northeast and North. In relation to this
kind of harvest, the authors stress that
the Midwest had the
highest national productivity among the regions, with evidence
that producers have improved varieties of common beans, in
addition to high technology.
For rice, it was observed that between 2008 and 2018
there was an increase in the participation of the South
region and a reduction in the Midwest and Nor- theast
regions of the country, without major changes in the total
quantity produced, which went from 12 tons to 11,7 tons in
2018, as shown in Table 1. To understand these changes, it
is necessary to know the different me- thods of cultivation of
the crop in the country.
Concerning the second harvest, in the same pe-
riod, 2006-2011, there was a reduction in the planted
area
and, as a result, a reduction in production. However, there is an
evolution in yield levels due to the adoption of technologies
that enable greater productivity, especially in
the states of
Paraná, São Paulo, Minas Gerais and Goiás. According to the
authors, the highlight of second crop production is also in
the South region.
Rice cultivation in Brazil takes place through two
types of production systems: flood-irrigated and highland
systems. The first of them is concentrated in the south of the
country and in tropical floodplains, especially in the states of
Rio Grande do Sul, Santa Catarina, Maranhão, Tocantins and
Mato Grosso do Sul. In 2014, these states together accounted
for around 85% of rice production in the country. The
highland system, on the other hand, represents about 50% of
the national area occupied by rice cultivation and is spread
throughout the national territory (WANDER; SILVA,
2014).
Finally, the third harvest comes from the possibi- lity
of using irrigation in times of low rainfall, in addition
to the
favorable altitudes of some regions, especially in the cerrado
region. This harvest occurs mainly in the states of Goiás,
Mato Grosso, Tocantins, northwestern
Minas Gerais,
Espírito Santo, São Paulo and part of Bahia.
In the period from
2006 to 2011, significant advances in the level of
productivity were observed nationwide,
mainly due to
increases in the Southeast and Center-West
regions.
The participation in the total supply of the highland
system has decreased in recent years, as the
areas occupied
by the crop, mostly involving the states of Goiás, Distrito
Federal, Mato Grosso and Tocantins, have given way to crops
such as soybeans, corn and sugarcane
(WANDER; SILVA,
2014).
Thus, the presence and importance of the Sou-
theast and Center-West regions for beans production in its three
kinds of harvest should be mentioned, but mainly in
the third
one, which together with the high productivity of the region
could be one of the reasons that justify the increase in its
participation in the production of beans in the country
between 2008 and 2018.
As for the common bean crop, the increased share of
production in the Midwest region may be related to the
fact that
it is considered an atypical crop, as it allows for
three harvests
to be obtained throughout the agricultural
year. In the present
work, the edible beans crop covers the total production of
the three crops.
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385
10
Silva Júnior, M. A. et al.
In the case of corn, which had an increase in the
participation of the Center-West region and a reduc- tion in
the South and Southeast of the country, one of the possible
explanations is presented by Souza et al. (2018). According
to the authors, the shift over the last four decades of corn
production from the South to the North of the country,
especially to the Midwest, has been
brought about by the
increase in demand for the grain and
by the availability of
cheaper land. In addition, greater growth in terms of
production has been observed in the Midwest. However, the
South region stands out in terms of productivity.
analyze not only the effects of interdependence between
regions in the same culture, but also the processes of spatial
competition between different cultures.
The first group (cotton, rice, coffee, sugar cane,
orange and tobacco) has crops that tend to be concentra-
ted in
specific areas of the country and did not present
major
changes between 2008 and 2018. The second group
(beans,
corn and soybean) has the crops that increased
their area
between the two years, whereas the third group (banana and
tomato) contains those with smaller clusters
and with clusters
more spaced apart among themselves. In addition, the
cassava crop was not included in any of
the groups since it
has producers throughout the national territory and it presented
a reduction in high-high clusters
in the North and Northeast
regions and an increase in the South and Midwest of the
country.
Conclusion
The analysis of agricultural crops throughout the
national territory made it possible to understand certain
patterns of spatial association between different crop
profiles in the years under study. Thus, were identified those
regions that did not show major changes in the production
of their municipalities, and also those that were influenced
by, or influenced, their neighbors.
Finally, it should be noted that this paper is limi- ted
to the comparison of changes related to crops with greater
economic value. Future studies could focus on different
categories of cultures and analyze other mea-
ningful
variables, such as productivity, in order to enable
the analysis
of the sector’s peculiarities from different points of view.
The comparative analysis between different cul-
tures
allowed the identification of cultures with similar
characteristics, either by the concentration of munici-
palities in certain regions by the expansion of clusters
observed from 2008 to 2018 or even by the extent and
proximity between the clusters. Furthermore, the com-
parative analysis of different crops made it possible to
Financing
This study was financed in part by the Coorde-
nação de Aperfeiçoamento de Pessoal de Nível Superior
– Brasil (CAPES) – Finance Code 001.
References
Almeida, E. Econometria Espacial Aplicada. Campinas: Alínea, 2012.
Rossetti, L. Zoneamento agrícola em aplicações de crédito e securidade rural
no Brasil: aspectos atuariais e de política agrícola. Revista Brasileira
de
Agrometeorologia, v. 9, n. 3, p. 386–399, 2001.
Almeida, E.; Perobelli, F.; Ferreira, P. Existe convergência espacial da
produtividade agrícola no Brasil? Revista de Economia e Sociologia Rural,
v. 46, n. 1, p. 31–52, 2008.
Santos, W.; Martins, J. O zoneamento agrícola de risco climático e sua
contribuição à agricultura brasileira. Revista de Política Agrícola, v. 25, n.
3, p. 73–94, 2016.
Andrade, A.; Monteiro, A.; Barcellos, C.; Lisboa, E.; Acosta, L. Almeida,
M.;
Brito, M.; Carvalho, M.; Santos, M.; Cruz, O.; Santos, R.; Flores, R.;
Silva, S.; Santos, S.; Correia, V.; Souza, W. Introdução à estatística espacial
para a saúde pública. Brasília: Ministério da Saúde, 2007.
Silva, O.; Wander, A. O feijão-comum no Brasil: passado, presente e
futuro. Embrapa Arroz e Feijão, 2013.
Câmara, G.; Monteiro, A.; Fucks, S.; Carvalho, M. Análise espacial e
geoprocessamento. In: Druks, S.; Carvalho, M.; Câmara, G.; Monteiro,
A. (Ed.). Análise espacial de dados geográficos. Brasília: Embrapa, 2004.
Souza, A.; Reis, J.; Raymundo, J.; Pinto, R. Estudo da produção de milho
no Brasil. South American Development Society Journal, v. 4, n. 11, p.
182, 2018.
Trentin, R.; Modolo, A.; Vargas, T.; Campos, J.; Adami, P.; Baesso, M.
Soybean productivity in rhodic hapludox compacted by the action of
furrow openers. Acta Scientiarum. Agronomy, v. 40, p. 1-9, 2018.
Cunha, G.; Assad, E. Uma visão geral do número especial da RBA sobre
zoneamento Agrícola no Brasil. Revista Brasileira de Agrometeorologia,
v. 9,
n. 3, p. 377–385, 2001.
Instituto Brasileiro de Geografia e Estatística, IBGE. 2019. Produção
agrícola municipal. Sistema IBGE de Recuperação Automática - SI- DRA,
Tabela 5457, 2019. Disponível em: https://sidra.ibge.gov.br/ tabela/5457.
Wander, A.; Silva, O. Rentabilidade da produção de arroz no Brasil.
Embrapa Arroz e Feijão, In: Campos, S.; Torres, D.; Ponchio, A.; Bar- ros,
G., Sustentabilidade e sustentação de alimentos: o desafio da rentabilidade
na produção. Brasília, DF: Centro de Gestão e Estudos Estratégicos, 2014.
Perobelli, F.; Almeida, E.; Alvim, M.; Ferreira, P. Produtividade do setor
agrícola brasileiro (1991-2003): uma análise espacial. Nova Economia,
v. 17,
n.1, p. 65–91, 2007.
Cad. Ciênc. Agrá., v. 13, p. 01–10, DOI: https://doi.org/10.35699/2447-6218.2021.35385