@PhDThesis{Aguiar:2012:ExHeIn,
author = "Aguiar, Ana Paula Dutra de",
title = "Modelagem de mudan{\c{c}}a do uso da terra na Amaz{\^o}nia:
explorando a heterogeneidade intrarregional",
school = "Instituto Nacional de Pesquisas Espaciais",
year = "2012",
address = "S{\~a}o Jos{\'e} dos Campos",
month = "2006-06-08",
keywords = "Amaz{\^o}nia, desflorestamento, modelagem de mudan{\c{c}}a de
uso da terra, CLUE, estat{\'{\i}}stica espacial, sensoriamento
remoto, Amazon, region, deforestation, land use change modeling,
CLUE, spatial statistcs, remote sensing.",
abstract = "Este trabalho descreve os resultados da aplica{\c{c}}{\~a}o do
arcabou{\c{c}}o de modelagem din{\^a}mica CLUE para explorar
como fatores alternativos, pol{\'{\i}}ticas p{\'u}blicas e
condi{\c{c}}{\~o}es de mercado influenciam o processo de
ocupa{\c{c}}{\~a}o da Amaz{\^o}nia. Trabalhos anteriores na
Amaz{\^o}nia enfatizaram aspectos como dist{\^a}ncia a estradas,
quase sempre desconsiderando a enorme heterogeneidade
biof{\'{\i}}sica e socioecon{\^o}mica da regi{\~a}o. A
an{\'a}lise estat{\'{\i}}stica realizada utiliza um banco de
dados espacial (c{\'e}lulas de 100 x 100 Km\$^{2}\$ e 25 x 25
km\$^{2}\$) com 40 vari{\'a}veis ambientais, demogr{\'a}ficas,
de estrutura agr{\'a}ria, tecnol{\'o}gicas, e indicadores de
conectividade a mercados como vari{\'a}veis independentes, e
vari{\'a}veis de uso e cobertura da terra (desflorestamento,
pastagem, agricultura tempor{\'a}ria e permanente) como
vari{\'a}veis dependentes. Os fatores determinantes dos
padr{\~o}es de uso da terra foram identificados usando modelos de
regress{\~a}o (\textit{spatial lag} e regress{\~a}o linear
m{\'u}ltipla) para toda a regi{\~a}o e tr{\^e}s
sub-regi{\~o}es. Os resultados dos modelos de regress{\~a}o
demonstram que a import{\^a}ncia relativa dos fatores
determinantes apresenta grande varia{\c{c}}{\~a}o na
regi{\~a}o. Modelos de regress{\~a}o enfatizam fatores
associados a pol{\'{\i}}ticas p{\'u}blicas, em especial
vari{\'a}veis de acessibilidade e {\'a}reas protegidas. Os
modelos foram utilizados para realizar diferentes
explora{\c{c}}{\~o}es de cen{\'a}rios de mudan{\c{c}}a de uso
at{\'e} 2020. Explora{\c{c}}{\~o}es incluem a an{\'a}lise da
influ{\^e}ncia de diferentes fatores na din{\^a}mica das novas
fronteiras na Amaz{\^o}nia, de poss{\'{\i}}veis impactos
pol{\'{\i}}ticas p{\'u}blicas, e aumento e
diminui{\c{c}}{\~a}o da demanda. As principais conclus{\~o}es
s{\~a}o: (a) conex{\~a}o a mercados nacionais {\'e} o fator
mais importante para capturar os padr{\~o}es espaciais das novas
fronteiras; (b) a intera{\c{c}}{\~a}o entre os fatores de
conex{\~a}o e demais fatores biof{\'{\i}}sicos e
socioecon{\^o}micos que influencia a din{\^a}mica intrarregional
heterog{\^e}nea; (c) estas diferen{\c{c}}as levam a impactos
diferenciados das pol{\'{\i}}ticas p{\'u}blicas na regi{\~a}o.
Este trabalho reflete a import{\^a}ncia da explora{\c{c}}{\~a}o
de cen{\'a}rios como uma ferramenta para auxiliar o entendimento
do processo de ocupa{\c{c}}{\~a}o da Amaz{\^o}nia e a
formula{\c{c}}{\~a}o de pol{\'{\i}}ticas para a regi{\~a}o.
ABSTRACT: This work describes the results of applying the CLUE
modeling framework to explore how alternative determining factors,
policies and market constraints influence the process of land
occupation in Amazonia. Previous work regarding deforestation in
Amazonia has emphasized aspects such as distance to roads, and has
disregarded the regionīs enormous biophysical and socio-economical
heterogeneity. The statistical analysis uses a spatially-explicit
database (cells of 100 x 100 km\$^{2}\$ and 25 x 25
km\$^{2}\$) with 40 environmental, demographical, agrarian
structure, technological, and market connectivity indicators as
independent variables, and land-use (pasture, temporary and
permanent crops, non-used agricultural land) patterns as dependent
variables. The determinant factors of land patterns were
identified using regression models (spatial lag and multiple
linear regressions) at multiple spatial resolutions for the whole
region and for three sub-regions. Regression models results showed
that the relative importance and significance of land use
determining factors greatly vary across the Amazon. Models
emphasize policy-relevant factors, especially those related to
connection to national markets, connection to ports, and protected
areas. The models were used to build different exploration
scenarios of land use change until 2020. Explorations include an
analysis of the influence of different factors in the new Amazon
frontiers dynamics, the possible impacts public policies, and
increasing or decreasing demand. The main conclusions drawn from
the scenario explorations are: (a) connection to national markets
is the most important factor for capturing the spatial patterns of
the new Amazonian frontiers; (b) it is the interaction between
connectivity and other factors that influence the heterogeneous
intra-regional dynamics; (c) these differences led to
heterogeneous impact of policies across the region. This work
reflects the importance of scenario exploration as a tool to
understand the process of occupation in the Brazilian Amazonia,
and subside public policies for the region.",
committee = "Valeriano, Dalton de Morisson (presidente) and C{\^a}mara,
Gilberto (orientador) and Freitas, Corina da Costa and Escada,
Maria Isabel Sobral and Moran, Emilio Frederico and Veldkamp,
Antonie",
copyholder = "SID/SCD",
englishtitle = "Modeling land use change in the Brazilian Amazon: exploring
intra-regional heterogeneity",
language = "pt",
pages = "206",
ibi = "6qtX3pFwXQZGivnJSY/M7t7e",
url = "http://urlib.net/ibi/6qtX3pFwXQZGivnJSY/M7t7e",
targetfile = "publicacao.pdf",
urlaccessdate = "05 maio 2024"
}