@PhDThesis{Quispe:2022:EsNu,
author = "Quispe, David Pareja",
title = "Import{\^a}ncia de fatores f{\'{\i}}sicos na ocorr{\^e}ncia de
dois eventos h{\'{\i}}dricos extremos na regi{\~a}o
Amaz{\^o}nica: Um estudo num{\'e}rico",
school = "Instituto Nacional de Pesquisas Espaciais (INPE)",
year = "2022",
address = "S{\~a}o Jos{\'e} dos Campos",
month = "2021-11-29",
keywords = "eventos extremos, seca, chuvoso, separa{\c{c}}{\~a}o de fatores,
RegCM4, extreme events, drought, flood, factor separation.",
abstract = "O modelo RegCM4 foi usado para investigar a influ{\^e}ncia de
tr{\^e}s fatores f{\'{\i}}sicos na ocorr{\^e}ncia de dois
eventos extremos (seco de 2004-05 e chuvoso de 2008-09) na bacia
Amaz{\^o}nica (BA). Para examinar o papel da temperatura da
superf{\'{\i}}cie do mar (TSM), gases do efeito estufa (GEE),
umidade do solo (SM) e suas sinergias sobre a
precipita{\c{c}}{\~a}o, o m{\'e}todo de separa{\c{c}}{\~a}o
de fatores foi usado. Para isso, foi feita uma
simula{\c{c}}{\~a}o continua de 34 anos com o RegCM4 sobre o
dom{\'{\i}}nio CORDEX da Am{\'e}rica do Sul (AS). O modelo foi
for{\c{c}}ado com a rean{\'a}lise ERA-Interim (EIN15). A
princ{\'{\i}}pio, avaliamos o desempenho do RegCM4, sobretudo as
vari{\'a}veis associadas a circula{\c{c}}{\~a}o
atmosf{\'e}rica, ao balan{\c{c}}o de radia{\c{c}}{\~a}o {\`a}
superf{\'{\i}}cie e ao ciclo hidrol{\'o}gico. Apesar de algumas
defici{\^e}ncias, o modelo reproduziu bem o padr{\~a}o geral das
vari{\'a}veis analisadas em compara{\c{c}}{\~a}o com as
observa{\c{c}}{\~o}es (EIN15, CLARA2 e CRU/PERSIANN). Na segunda
parte, o per{\'{\i}}odo de dezembro-fevereiro foi selecionado
para avaliar os dois eventos extremos. A avalia{\c{c}}{\~a}o foi
feita a partir de uma rodada de longo prazo, chamado de
experimento controle. Os resultados mostraram que o RegCM4 pode
reproduzir a distribui{\c{c}}{\~a}o espacial e temporal das
anomalias negativas da precipita{\c{c}}{\~a}o no
per{\'{\i}}odo de seca entre 2004-05 e as anomalias positivas no
per{\'{\i}}odo chuvoso de 2008-09 sobre a BA, o que indica que o
RegCM4 {\'e} {\'u}til para simular eventos hidro-clim{\'a}ticos
extremos. Finalmente, os experimentos de sensibilidade
considerando a import{\^a}ncia relativa dos tr{\^e}s fatores
diretos e seus efeitos combinados foram realizados. Para cada
experimento, a m{\'e}dia do conjunto de cinco membros foi usado
na an{\'a}lise. Os resultados para o evento extremo seco, no
ver{\~a}o austral de 2004-05, mostraram que os fatores diretos da
TSM e GEE e o sinergismo entre os tr{\^e}s fatores desempenham um
papel muito importante na ocorr{\^e}ncia da seca na BA. O fator
TSM {\'e} o mais importante contribuindo com aumentos de
precipita{\c{c}}{\~a}o superiores a 50%, enquanto o fator GEE
contribui com aumentos pr{\'o}ximos a 30%. Isto mostra que o
fator TSM foi o principal fator que favoreceu a ocorr{\^e}ncia do
epis{\'o}dio seco 2004-05 na BA. A sinergia dos tr{\^e}s fatores
tamb{\'e}m contribu{\'{\i}}ram com aumentos de
precipita{\c{c}}{\~a}o pr{\'o}ximo a 50% nas regi{\~o}es do
centro e sul da BA. Al{\'e}m disso, sugere-se que as
condi{\c{c}}{\~o}es mais {\'u}midas do solo podem reduzir a
intensidade dos eventos secos. Os resultados para o evento extremo
chuvoso, no ver{\~a}o austral de 2008-09, mostraram que o efeito
direto do fator GEE foi o mais importante, apresentando grandes
extens{\~o}es com redu{\c{c}}{\~o}es de
precipita{\c{c}}{\~a}o de at{\'e} -30% na BA. O fator TSM
tamb{\'e}m contribui com redu{\c{c}}{\~o}es de
precipita{\c{c}}{\~a}o superiores a 20%, por{\'e}m, apresentam
menor extens{\~a}o aos observados no experimento do fator GEE. O
efeito combinado dos tr{\^e}s fatores apresentaram
redu{\c{c}}{\~o}es de precipita{\c{c}}{\~a}o de at{\'e} -30%,
com extens{\~a}o semelhante aos observados no experimento do
fator GEE. Isto indica que no caso chuvoso, a sinergia dos
tr{\^e}s fatores (TSM, GEE e SM) {\'e} importante junto ao fator
GEE. Em todos os experimentos de sensibilidade, a T2m apresentou
efeitos de sinal positivo nas regi{\~o}es onde
redu{\c{c}}{\~o}es de precipita{\c{c}}{\~a}o prevaleceram, e o
sinal contr{\'a}rio foi observado em regi{\~o}es com aumentos de
precipita{\c{c}}{\~a}o. Uma vez que as intera{\c{c}}{\~o}es
n{\~a}o lineares n{\~a}o podem ser bem avaliadas, mais estudos
s{\~a}o necess{\'a}rios para entender melhor os mecanismos
dominantes envolvidos. ABSTRACT: The RegCM4 model was used to
investigate the influence of three physical factors on the
occurrence of two extreme events (2004-05 drought and 2008-09
flood) in the Amazon basin (AB). To examine the role of sea
surface temperature (SST), greenhouse gases (GHG), soil moisture
(SM) and their synergies on precipitation the factor separation
method was used. For this purpose, a 34-years simulation was
performed with RegCM4 on the CORDEX domain of South America (SA).
The model was forced with ERA-Interim reanalysis (EIN15). At
first, we assess the RegCM4 performance, especially the variables
associated with atmospheric circulation, the surface radiation
budget and the hydrological cycle. Although some deficiencies, the
model reproduced well the general pattern of the variables
analyzed in comparison with the observations (EIN15, CLARA2 and
CRU/PERSIANN). In the second part, the period December-February
was selected to evaluate the two extreme events. The evaluation
was made from a long-term run, called a control experiment. The
results showed that RegCM4 can reproduce the spatial distribution
and temporal evolution of negative precipitation anomalies in the
2004-05 drought event and positive anomalies in the 2008-09 flood
event over AB, which indicates that RegCM4 is useful for
simulating hydro-climatic extreme events. Finally, sensitivity
experiments considering the relative importance of the three
direct factors and their combined effects were performed. For each
experiment, the five-member ensemble-averaged is used in the
analysis. The results, for the extreme drought event in the
austral summer of 2004-05, showed that the direct factors of SST
and GHG and the synergism between the three factors play a very
important role in the occurrence of drought in AB. The SST factor
is the most important contributing with increases in precipitation
above 50%, while the GHG factor contributes with increases around
30%. This shows that the SST factor was the main factor that
favored the occurrence of the 2004-05 extreme drought episode in
AB. The synergy of the three factors also contributed to
precipitation increases close to 50% in the central and southern
regions of AB. Moreover, it is suggested that wetter soil
conditions can reduce the intensity of drought events. The results
for the extreme flood event in the austral summer 2008-09, showed
that the direct effect of the GHG factor was the most important,
presenting large extensions with precipitation reductions in AB
with values of up to -30%. The SST factor also contributes with
precipitation reductions greater than 20%, however, they are less
extensive than those observed in the GHG factor experiment. The
combined effect of the three factors showed precipitation
reductions of up to -30%, with a similar extent to those observed
in the GHG factor experiment. This indicates that in the extreme
flood case, the synergy of the three factors (SST, GHG and SM) is
important together with the GHG factor. In all sensitivity
experiments, T2m showed positive sign effects in regions where
precipitation reductions prevailed, and the opposite sign was
observed in regions with precipitation increases. Since nonlinear
interactions are not yet well knowledge, further studies are
needed to better understand the dominant mechanisms involved.",
committee = "Gon{\c{c}}alves, Luis Gustavo Gon{\c{c}}alves de (presidente)
and Fernandez, Julio Pablo Reyes (orientador) and Gan, Manoel
Alonso and Rocha, Rosmeri Porfirio da and Reboita, Michelle
Sim{\~o}es",
englishtitle = "Importance of physical factors in the occurrence of two extreme
hydrological events in the Amazon region: A numerical study",
language = "pt",
pages = "159",
ibi = "8JMKD3MGP3W34T/45R3SM8",
url = "http://urlib.net/ibi/8JMKD3MGP3W34T/45R3SM8",
targetfile = "publicacao.pdf",
urlaccessdate = "23 maio 2024"
}