@PhDThesis{Oliveira:2016:FlRaÁg,
author = "Oliveira, Gabriel de",
title = "Fluxos de radia{\c{c}}{\~a}o, {\'a}gua e carbono em
ecossistemas terrestres na Amaz{\^o}nia oriental com base em
dados MODIS",
school = "Instituto Nacional de Pesquisas Espaciais (INPE)",
year = "2016",
address = "S{\~a}o Jos{\'e} dos Campos",
month = "2016-07-22",
keywords = "sensoriamento remoto, balan{\c{c}}o de radia{\c{c}}{\~a}o,
efici{\^e}ncia de uso da {\'a}gua, MODIS, Amaz{\^o}nia, remote
sensing, net radiation, water-use efficiency, Amazonia.",
abstract = "O objetivo principal desta pesquisa foi obter estimativas dos
componentes do balan{\c{c}}o de radia{\c{c}}{\~a}o (Rn),
evapotranspira{\c{c}}{\~a}o (ET), produtividade prim{\'a}ria
bruta (GPP), e efici{\^e}ncia de uso da {\'a}gua (WUE) em escala
regional, abrangendo sete bacias na por{\c{c}}{\~a}o leste da
Amaz{\^o}nia, entre 2001 e 2013, a partir da
utiliza{\c{c}}{\~a}o de dados do sensor MODIS. Os objetivos
espec{\'{\i}}ficos inclu{\'{\i}}ram: i) desenvolver uma
abordagem para estimativa do Rn e seus componentes sob diversas
condi{\c{c}}{\~o}es do c{\'e}u na regi{\~a}o amaz{\^o}nica
utilizando apenas dados de sensoriamento remoto (MODIS) e
rean{\'a}lise (GLDAS), atrav{\'e}s da adapta{\c{c}}{\~a}o do
modelo SEBAL; ii) calcular a WUE a partir de dados de ET e GPP
provenientes dos algoritmos MOD16 e MOD17, respectivamente; iii)
analisar a din{\^a}mica espacial e temporal dos fluxos de
radia{\c{c}}{\~a}o, {\'a}gua e carbono na {\'a}rea de estudo;
iv) avaliar a magnitude e variabilidade temporal dos fluxos de
radia{\c{c}}{\~a}o, {\'a}gua e carbono em diferentes tipos de
uso e cobertura da terra verificados na regi{\~a}o; e v) avaliar
os impactos de mudan{\c{c}}as de uso e cobertura da terra nos
fluxos de energia e carbono. A compara{\c{c}}{\~a}o das
estimativas obtidas pelo m{\'e}todo proposto para
determina{\c{c}}{\~a}o dos fluxos radiativos com
observa{\c{c}}{\~o}es de torres do LBA mostrou erros m{\'e}dios
de \$\sim\$14 e 13\% para o Rn instant{\^a}neo e di{\'a}rio,
respectivamente. Os produtos MOD16 (ET) e MOD17 (GPP) apresentaram
erros, em m{\'e}dia, de \$\sim\$14 e 30\%, enquanto as
estimativas de WUE tiveram um erro m{\'e}dio de \$\sim\$32\%.
A variabilidade espacial do Rn, ET, GPP, e WUE esteve associada
principalmente ao gradiente de uso e cobertura da terra ao longo
da regi{\~a}o. Os maiores valores dessas vari{\'a}veis foram
observados na por{\c{c}}{\~a}o centro-norte (sul do estado do
Par{\'a}), ao passo que os menores foram verificados na
por{\c{c}}{\~a}o centro-sul (norte do estado do Mato Grosso).
Isso esteve relacionado ao fato de que o setor centro-norte, mesmo
apresentando {\'a}reas com forte ocorr{\^e}ncia de desmatamento,
possui maiores {\'a}reas de floresta (prim{\'a}ria ou
secund{\'a}ria) em rela{\c{c}}{\~a}o ao setor centro-sul.
Temporalmente, o Rn, ET, GPP, e WUE na {\'a}rea de estudo
mostraram um padr{\~a}o sazonal bem pronunciado, variando com as
esta{\c{c}}{\~o}es {\'u}mida (janeiro a junho) e seca (julho a
dezembro) na regi{\~a}o. Os ecossistemas de floresta tropical
prim{\'a}ria (FP), vegeta{\c{c}}{\~a}o secund{\'a}ria (VS) e
cerrado (CE) apresentaram menores valores de temperatura do ar e
da superf{\'{\i}}cie, albedo, radia{\c{c}}{\~a}o infravermelha
termal incidente (L\$\downarrow\$) e emitida
(L\$\uparrow\$), e maiores valores de Rn, GPP, ET, e WUE em
rela{\c{c}}{\~a}o {\`a} soja (SO) e pastagem (PA). O
comportamento temporal dos fluxos de radia{\c{c}}{\~a}o,
{\'a}gua e carbono nos diferentes ecossistemas terrestres esteve
diretamente associado com a sazonalidade clim{\'a}tica da
regi{\~a}o, ou seja, {\`a}s caracter{\'{\i}}sticas
intr{\'{\i}}nsecas aos per{\'{\i}}odos chuvoso e seco. Tais
caracter{\'{\i}}sticas influenciaram no comportamento dos
par{\^a}metros biof{\'{\i}}sicos da vegeta{\c{c}}{\~a}o e,
consequentemente, nos fluxos de superf{\'{\i}}cie. Constatou-se
que a convers{\~a}o de FP para SO possui um maior impacto no
clima regional da por{\c{c}}{\~a}o leste da Amaz{\^o}nia na
compara{\c{c}}{\~a}o com a altera{\c{c}}{\~a}o de FP para PA.
Ainda, foi verificado que, de um modo geral, em ambas as
situa{\c{c}}{\~o}es (FP para PA e FP para SO) os maiores
impactos nas magnitudes dos valores ocorrem durante a
esta{\c{c}}{\~a}o seca. Em suma, a utiliza{\c{c}}{\~a}o de
dados do sensor MODIS em conjunto com rean{\'a}lises foi adequada
no sentido de quantificar e mapear de maneira consistente a
distribui{\c{c}}{\~a}o espacial dos fluxos de
radia{\c{c}}{\~a}o, {\'a}gua e carbono na por{\c{c}}{\~a}o
oriental da Amaz{\^o}nia durante um per{\'{\i}}odo de 13 anos.
Esse conjunto de dados possibilitou a aquisi{\c{c}}{\~a}o de
informa{\c{c}}{\~o}es tanto sob o contexto regional quanto sob
ecossistemas terrestres espec{\'{\i}}ficos existentes nesse
setor da regi{\~a}o amaz{\^o}nica. Tais informa{\c{c}}{\~o}es
s{\~a}o extremamente importantes no sentido de melhor compreender
o acoplamento entre os ciclos hidrol{\'o}gicos e de carbono na
Amaz{\^o}nia, bem como proporcionar um conhecimento mais
aprofundado da variabilidade clim{\'a}tica nesse ambiente em base
regional. ABSTRACT: This study aimed to estimate the components of
net radiation (Rn), evapotranspiration (ET), gross primary
productivity (GPP), and water-use efficiency on a regional scale,
encompassing seven basins in the eastern part of Amazonia, between
2001 and 2013, through the utilization of MODIS sensor data. The
specific aims included: i) to develop an approach to estimate Rn
and its components under all-sky conditions for the Amazon region
through the SEBAL model utilizing only remote sensing (MODIS) and
reanalysis data (GLDAS); ii) to calculate WUE by the use of ET and
GPP data obtained through MOD16 and MOD17 products, respectively;
iii) to analyze the spatio-temporal dynamics of the fluxes of
radiation, water and carbon over the study area; iv) to evaluate
the magnitude and temporal variability of the fluxes of radiation,
water and carbon in different land use and land cover types
observed in the region; and v) to evaluate the impacts caused by
land use and land cover changes on the energy and carbon fluxes.
Comparison between estimates obtained by the proposed method to
retrieve the radiative fluxes and observations from the LBA towers
showed average errors of \$\sim\$14 and 13\% for instantaneous
and daily Rn, respectively. The MOD16 and MOD17 products presented
errors, on average, of \$\sim\$14 and 30\%, respectively,
whilst WUE estimates showed an average error of \$\sim\$32\%.
The spatial variability of Rn, ET, GPP, and WUE were primarily
related to the gradient of land use and land cover along the
region. The highest values were observed in the central-north
portion (south of Par{\'a} state), whilst the lowest values were
observed in the central-south portion (north of Mato Grosso
state). It was related to the fact that the central-north part,
even with areas of high occurrence of deforestation, has larger
areas of forest (primary or secondary) than the central-south
part. Temporally, Rn, ET, GPP, and WUE have shown a strongly
seasonal pattern, varying according with wet (January to June) and
dry (July to December) seasons in the region. The ecosystems of
primary tropical forest (FP), secondary succession (VS) and
cerrado (CE) had lower air and surface temperatures, albedo,
thermal infrared incoming (L\$\downarrow\$) and outgoing
(L\$\uparrow\$) radiation, and higher Rn, GPP, ET, and WUE
compared to soybean (SO) and pasture (PA). The temporal behavior
of the fluxes of radiation, water and carbon under the different
terrestrial ecosystems was directly associated with the climatic
seasonality of the region, that is, to the intrinsic
characteristics of the rainy and dry seasons. These
characteristics affected the behavior of biophysical parameters of
vegetation and hence the surface fluxes. It was observed that the
conversion from FP to SO has a greater impact on regional climate
of the eastern portion of Amazon compared to the change from FP to
PA. In addition, it was found that, in general, in both situations
(FP to PA and FP to SO) the major impact on magnitude of the
values occurs during the dry season. In summary, the use of MODIS
data combined with reanalysis products was adequate to quantify
and consistently map the spatial distribution of the fluxes of
radiation, water and carbon in the eastern portion of Amazonia
over a period of 13 years. This dataset allowed the acquisition of
information both in the regional context and under different types
of terrestrial ecosystems existent in this part of the Amazon.
Such information are extremely important in order to better
understand the coupling between the hydrologic and carbon cycles
in Amazonia, as well as to provide a deeper knowledge of climate
variability in this environment on a regional basis.",
committee = "Moraes, Elisabete Caria (presidente/orientadora) and Brunsell,
Nathaniel Alan (orientador) and Arag{\~a}o, Luiz Eduardo Oliveira
e Cruz de and Shimabukuro, Yosio Edemir and Oliveira, Gilvan
Sampaio de and Victoria, Daniel de Castro and Zeri, Luiz Marcelo
Mattos",
copyholder = "SID/SCD",
englishtitle = "Fluxes of radiation, water and carbon over terrestrial ecosystems
in eastern Amazon based on MODIS data",
language = "pt",
pages = "178",
ibi = "8JMKD3MGP3W34P/3M4EGMP",
url = "http://urlib.net/ibi/8JMKD3MGP3W34P/3M4EGMP",
targetfile = "publicacao.pdf",
urlaccessdate = "27 abr. 2024"
}