@PhDThesis{Fonseca:2020:EvGrPr,
author = "Fonseca, Let{\'{\i}}cia d'Agosto Miguel",
title = "Evapotranspiration and gross primary productivity seasonality in a
floodplain forest at the Bananal island region",
school = "Instituto Nacional de Pesquisas Espaciais (INPE)",
year = "2020",
address = "S{\~a}o Jos{\'e} dos Campos",
month = "2020-03-27",
keywords = "floodplain, Amazon, Bananal, gross primary productivity,
evapotranspiration, plan{\'{\i}}cie de inunda{\c{c}}{\~a}o,
Amaz{\^o}nia, Bananal, produtividade prim{\'a}ria bruta,
evapotranspira{\c{c}}{\~a}o.",
abstract = "The Amazon forest plays an important role in regulating the local,
regional and global climate, due to the high potential for
absorbing carbon in its biomass and transferring large amounts of
water from the land surface to the atmosphere through
evapotranspiration. Despite the several studies that have explored
the mechanisms of seasonal vegetation control, a small number of
them have focused on flooded forests. In the Amazon basin, it
corresponds to c.a. 14% of the basin. This study was performed in
a floodplain forest located at the transition area between the
Amazon and Cerrado (Savana) biomes, near the Bananal (BAN) Island
region seeking to understand the mechanisms of vegetation control
during the dry and flooded periods. The seasonality of gross
primary productivity (GPP) and evapotranspiration (ET) from eddy
covariance measurements were assessed, along with environmental
drivers and phenological patterns, obtained from the field (leaf
litter mass) and satellite measurements (enhanced vegetation index
(EVI) from the Moderate Resolution Imaging
Spectroradiometer/multi-angle implementation correction
(MODIS/MAIAC)). ET measurements presented many gaps, and a
statistical model (the Generalized Additive Model - GAM) was used
to reconstruct the records from 2004 to 2017, using the ERA5
reanalysis climate data. Moreover, the remote sensing product
(MOD16A2) was acquired to analyze the reliability of this product
in describing seasonal ET. The long-term change on the
hydrological pattern at the BAN region was analyzed through the
Gravity Recovery and Climate Experiment (GRACE) satellite product.
The results revealed that Gross primary production is limited by
soil moisture during the flooded period due to the excess water,
while GPP is positively associated with soil moisture during
non-flooded months. Besides, GPP is maximized when the accumulated
water deficit (CWD) increases, indicating that it depends on the
amount of water input in the environment. EVI was positively
associated with leaf litter mass and GPP, suggesting the synchrony
between leaf production and the photosynthetic capacity of the
canopy, decreasing at the peak of the flooded period and at the
end of the dry season. The EVI was also able to describe the
interannual variations of the canopy in relation to environmental
factors, such as during the extreme drought of the El Niņo year
(2015/2016). The main ET drivers were identified during the model
calibration process, which are: vapour pressure deficit, radiation
and soil moisture. The seasonal ecosystem productivity and
evapotranspiration are not synchronized in this Southern Amazon
forest during the flooded period, because the free water
evaporation mainly drives ET. However, during non-flooded months
ET is governed by forest transpiration, as indicated by the
association with the carbon, phenological and meteorological
seasonal patterns. The flood pulse regulates the soil volume water
content, and consequently, the water availability for plants
during non-flooded months. These findings highlighted the
vulnerability of this forest facing extreme dry years, given the
decreased flood pulse trend reported here, which consequently
diminished the total water storage in this region during 2016,
assessed through GRACE product. The Amazon forest plays an
important role in regulating the local, regional and global
climate, due to the high potential for absorbing carbon in its
biomass and transferring large amounts of water from the land
surface to the atmosphere through evapotranspiration. Despite the
several studies that have explored the mechanisms of seasonal
vegetation control, a small number of them have focused on flooded
forests. In the Amazon basin, it corresponds to c.a. 14% of the
basin. This study was performed in a floodplain forest located at
the transition area between the Amazon and Cerrado (Savana)
biomes, near the Bananal (BAN) Island region seeking to understand
the mechanisms of vegetation control during the dry and flooded
periods. The seasonality of gross primary productivity (GPP) and
evapotranspiration (ET) from eddy covariance measurements were
assessed, along with environmental drivers and phenological
patterns, obtained from the field (leaf litter mass) and satellite
measurements (enhanced vegetation index (EVI) from the Moderate
Resolution Imaging Spectroradiometer/multi-angle implementation
correction (MODIS/MAIAC)). ET measurements presented many gaps,
and a statistical model (the Generalized Additive Model - GAM) was
used to reconstruct the records from 2004 to 2017, using the ERA5
reanalysis climate data. Moreover, the remote sensing product
(MOD16A2) was acquired to analyze the reliability of this product
in describing seasonal ET. The long-term change on the
hydrological pattern at the BAN region was analyzed through the
Gravity Recovery and Climate Experiment (GRACE) satellite product.
The results revealed that Gross primary production is limited by
soil moisture during the flooded period due to the excess water,
while GPP is positively associated with soil moisture during
non-flooded months. Besides, GPP is maximized when the accumulated
water deficit (CWD) increases, indicating that it depends on the
amount of water input in the environment. EVI was positively
associated with leaf litter mass and GPP, suggesting the synchrony
between leaf production and the photosynthetic capacity of the
canopy, decreasing at the peak of the flooded period and at the
end of the dry season. The EVI was also able to describe the
interannual variations of the canopy in relation to environmental
factors, such as during the extreme drought of the El Niņo year
(2015/2016). The main ET drivers were identified during the model
calibration process, which are: vapour pressure deficit, radiation
and soil moisture. The seasonal ecosystem productivity and
evapotranspiration are not synchronized in this Southern Amazon
forest during the flooded period, because the free water
evaporation mainly drives ET. However, during non-flooded months
ET is governed by forest transpiration, as indicated by the
association with the carbon, phenological and meteorological
seasonal patterns. The flood pulse regulates the soil volume water
content, and consequently, the water availability for plants
during non-flooded months. These findings highlighted the
vulnerability of this forest facing extreme dry years, given the
decreased flood pulse trend reported here, which consequently
diminished the total water storage in this region during 2016,
assessed through GRACE product. RESUMO: A Floresta Amaz{\^o}nica
desempenha um importante papel na regula{\c{c}}{\~a}o do clima
em escala local, regional e global, devido ao elevado potencial de
absor{\c{c}}{\~a}o de carbono em sua biomassa e da
transfer{\^e}ncia de {\'a}gua para a atmosfera atrav{\'e}s da
evapotranspira{\c{c}}{\~a}o. Apesar dos in{\'u}meros estudos
que exploraram os mecanismos de controle sazonal da
vegeta{\c{c}}{\~a}o, poucos se concentraram nas florestas
inundadas, as quais correspondem a aproximadamente 14% da bacia
Amaz{\^o}nica. Em busca de compreender os mecanismos de controle
da vegeta{\c{c}}{\~a}o durante os per{\'{\i}}odos secos e
inundados, o presente trabalho foi realizado em uma floresta
sazonalmente alagada localizada na {\'a}rea de
transi{\c{c}}{\~a}o entre os biomas Amaz{\^o}nia e Cerrado,
pr{\'o}ximo {\`a} regi{\~a}o da Ilha do Bananal (BAN). Foi
realizada a an{\'a}lise da sazonalidade da produtividade
prim{\'a}ria bruta (PPB) e da evapotranspira{\c{c}}{\~a}o (ET)
obtidas atrav{\'e}s da torre micrometeorol{\'o}gica, juntamente
com vari{\'a}veis clim{\'a}ticas e padr{\~o}es
fenol{\'o}gicos, estes, obtidos em campo (massa foliar) e
atrav{\'e}s do {\'{\i}}ndice de vegeta{\c{c}}{\~a}o melhorado
(EVI). O EVI foi calculado a partir de imagens com
resolu{\c{c}}{\~a}o moderada e corre{\c{c}}{\~a}o angular e
atmosf{\'e}rica (MODIS / MAIAC)). As medidas de ET apresentaram
muitas lacunas e um modelo estat{\'{\i}}stico (do ingl{\^e}s,
Generalized Aditive Model - GAM) foi usado para reconstruir a
s{\'e}rie temporal desse dado de 2004 a 2017, dados
clim{\'a}ticos de rean{\'a}lise (ERA5) foram utilizados como
vari{\'a}veis preditoras nesse modelo. Al{\'e}m disso, o produto
de sensoriamento remoto MOD16A2 foi adquirido para analisar a
potencialidade do mesmo em representar a sazonalidade da ET. As
mudan{\c{c}}as no padr{\~a}o hidrol{\'o}gico da regi{\~a}o
foram analisadas por meio do produto do sat{\'e}lite GRACE (do
ingl{\^e}s, Gravity Recovery and Climate Experiment). Os
resultados revelaram que a produtividade do ecossistema {\'e}
limitada pela umidade do solo de duas maneiras, durante o
per{\'{\i}}odo alagado, o excesso de {\'a}gua limita a PPB,
enquanto a PPB est{\'a} associada positivamente {\`a} umidade do
solo durante os meses n{\~a}o alagados. Al{\'e}m disso, a PPB
{\'e} maximizada quando o d{\'e}ficit acumulado de {\'a}gua
(CWD) aumenta, indicando a depend{\^e}ncia da quantidade de
{\'a}gua que entra no sistema. O EVI associou-se positivamente
{\`a} massa foliar e {\`a} PPB, sugerindo a sincronia entre a
produ{\c{c}}{\~a}o foliar e a capacidade fotossint{\'e}tica do
dossel, diminuindo no pico do per{\'{\i}}odo de
inunda{\c{c}}{\~a}o e no final da esta{\c{c}}{\~a}o seca. O
EVI tamb{\'e}m foi capaz de descrever as varia{\c{c}}{\~o}es
interanuais do dossel em rela{\c{c}}{\~a}o aos fatores
ambientais, como durante a seca extrema do ano de El Niņo
(2015/2016). Os principais fatores clim{\'a}ticos que influenciam
na sazonalidade da ET s{\~a}o: O d{\'e}ficit de press{\~a}o de
vapor, a radia{\c{c}}{\~a}o e a umidade do solo, identificados
durante o processo de calibra{\c{c}}{\~a}o do modelo
estat{\'{\i}}stico. A produtividade sazonal do ecossistema e a
evapotranspira{\c{c}}{\~a}o s{\~a}o dissociadas nesta floresta,
pois as altas taxas de ET est{\~a}o relacionadas principalmente
com a evapora{\c{c}}{\~a}o de {\'a}gua livre durante o
alagamento, enquanto durante os meses n{\~a}o inundados, a
transpira{\c{c}}{\~a}o da floresta {\'e} o principal regulador
da ET, como indicado pela sazonalidade do carbono, da fenologia e
dos dados clim{\'a}ticos. O pulso de inunda{\c{c}}{\~a}o regula
o volume de {\'a}gua no solo e, consequentemente, a
disponibilidade de {\'a}gua para as plantas durante os meses
n{\~a}o inundados. Esses resultados evidenciam a vulnerabilidade
dessa floresta em anos extremos de seca, dada {\`a}
tend{\^e}ncia de diminui{\c{c}}{\~a}o da amplitude do
alagamento identificada nesse estudo, que consequentemente
diminuiu o armazenamento total de {\'a}gua na regi{\~a}o durante
o ano de 2016, observado atrav{\'e}s do produto do sat{\'e}lite
GRACE.",
committee = "Randow, Celso von (presidente) and Borma, Laura de Simone
(orientadora) and Renn{\'o}, Camilo Daleles and Rocha, Humberto
Ribeiro da and Silva, Thiago Sanna Freire",
englishtitle = "Sazonalidade da evapotranspira{\c{c}}{\~a}o e da produtividade
prim{\'a}ria bruta em uma floresta alag{\'a}vel na regi{\~a}o
da ilha do Bananal",
language = "en",
pages = "89",
ibi = "8JMKD3MGP3W34R/428U3L5",
url = "http://urlib.net/ibi/8JMKD3MGP3W34R/428U3L5",
targetfile = "publicacao.pdf",
urlaccessdate = "28 abr. 2024"
}