@PhDThesis{Zepka:2011:PrDeAt,
author = "Zepka, Gisele dos Santos",
title = "Previs{\~a}o de descargas atmosf{\'e}ricas usando o modelo de
mesoescala WRF",
school = "Instituto Nacional de Pesquisas Espaciais",
year = "2011",
address = "S{\~a}o Jos{\'e} dos Campos",
month = "2011-07-12",
keywords = "raios, modelo de mesoescala WRF, previs{\~a}o,
parametriza{\c{c}}{\~o}es f{\'{\i}}sicas, lightning, WRF
mesoscale model, forecasting, physical options.",
abstract = "Identificar, com razo{\'a}vel anteced{\^e}ncia e confiabilidade,
condi{\c{c}}{\~o}es favor{\'a}veis {\`a} ocorr{\^e}ncia de
descargas atmosf{\'e}ricas {\'e} essencial para a emiss{\~a}o
de alertas, bem como antecipar a ado{\c{c}}{\~a}o de
estrat{\'e}gias, que eliminem ou minimizem o impacto negativo
deste fen{\^o}meno meteorol{\'o}gico. O objetivo deste trabalho
{\'e} estabelecer uma metodologia para prever a ocorr{\^e}ncia
de raios a partir da combina{\c{c}}{\~a}o de vari{\'a}veis
meteorol{\'o}gicas obtidas de simula{\c{c}}{\~o}es
num{\'e}ricas de alta resolu{\c{c}}{\~a}o espacial com o modelo
de mesoescala Weather Research and Forecasting (WRF). Este estudo
tamb{\'e}m vislumbra melhor identificar e compreender
princ{\'{\i}}pios f{\'{\i}}sicos inerentes {\`a}s descargas
atmosf{\'e}ricas que ocorrem durante o ver{\~a}o na regi{\~a}o
Sudeste do Brasil. Primeiramente foram selecionados doze casos de
tempestades e, para eles, realizada uma avalia{\c{c}}{\~a}o da
destreza do WRF, comparando par{\^a}metros de superf{\'{\i}}cie
com dados de esta{\c{c}}{\~o}es observacionais da CETESB e do
CPTEC. A seguir, seis vari{\'a}veis do modelo (CAPE,
{\'{\i}}ndice de levantamento, {\'{\i}}ndice K, temperatura
potencial equivalente, velocidade vertical m{\'e}dia e integrado
da raz{\~a}o de mistura de gelo) participaram de um m{\'e}todo
proposto para buscar as parametriza{\c{c}}{\~o}es de
convec{\c{c}}{\~a}o e microf{\'{\i}}sica que melhor
representam as condi{\c{c}}{\~o}es atmosf{\'e}ricas quando da
ocorr{\^e}ncia de raios. A an{\'a}lise dos resultados permitiu a
sele{\c{c}}{\~a}o de Grell-Devenyi, como esquema cumulus, e
Thompson, de microf{\'{\i}}sica, nas simula{\c{c}}{\~o}es
finais com o WRF. Dois m{\'e}todos capazes de inferir
qualitativamente a ocorr{\^e}ncia de raios foram sugeridos, o
M{\'e}todo Linear e o M{\'e}todo Normalizado. Ambos t{\^e}m por
base cinco das seis vari{\'a}veis mencionadas acima, {\`a}s
quais foram atribu{\'{\i}}dos {\'{\i}}ndices
classificat{\'o}rios que correspondem a probabilidades de
ocorr{\^e}ncia de descargas. O M{\'e}todo Normalizado foi
considerado o mais adequado para a previs{\~a}o de raios, por
conseguir representar satisfatoriamente as condi{\c{c}}{\~o}es
da atmosfera quando os raios ocorrem e, com isso, apontar as
regi{\~o}es mais prop{\'{\i}}cias. Os resultados finais deste
trabalho se mostram promissores e indicam que a metodologia
proposta traz avan{\c{c}}os significativos neste campo
multidisciplinar do conhecimento. ABSTRACT: Identify, with
reasonable advance and reliability, favorable conditions for the
lightning occurrence is essential for alerts as well as adoption
of strategies which eliminate or minimize the negative impact of
this phenomenon. The objective of this work is to propose a
methodology to forecast the lightning occurrence using a
combination of meteorological variables obtained from high
resolution numerical simulations with Weather Research and
Forecasting (WRF) mesoscale model. The study also envisages
identifying and understanding the relevant physical principles
related to lightning discharges which occur during the summer
season in Southeastern Brazil. Twelve thunderstorms cases were
selected and the evaluation of the WRF results was performed using
surface meteorological stations data of CETESB and CPTEC. Six
model variables (CAPE, Lifted Index, K-Index, equivalent potential
temperature, average vertical velocity and integrated ice mixing
ratio) have been taken as part in the proposed method to find the
convective and microphysical parameterizations which best
represent the atmospheric conditions on the lightning occurrence.
From the analysis, Grell-Devenyi and Thompson respectively were
the cumulus and microphysical schemes chosen to the simulations
with the WRF model. Two methods to infer qualitatively the
lightning occurrence have been suggested, the Linear Method and
Normalized Method. Both are based on five of the six variables
mentioned above, to which have been attributed classificatory
indices corresponding to the probabilities of discharges
occurrence. The Normalized Method was considered the most
appropriate for the lightning forecasting, because it is able to
represent satisfactorily the atmospheric conditions when lightning
occurs and, therefore, to identify the most favorable regions. The
results of the study are promising and indicate that the
forecasting method provides significant advances in this
multidisciplinary field of knowledge.",
committee = "Lago, Alisson Dal (presidente) and Souza, Jonas Rodrigues de and
Pinto Junior, Osmar (orientador) and Pinto, Iara Regina Cardoso de
Almeida and Yamazaki, Yoshihiro and Fernandes, Widinei Alves",
copyholder = "SID/SCD",
englishtitle = "Lightning forecasting using WRF mesoscale model",
language = "pt",
pages = "176",
ibi = "8JMKD3MGP7W/3A4EGUB",
url = "http://urlib.net/ibi/8JMKD3MGP7W/3A4EGUB",
targetfile = "publicacao.pdf",
urlaccessdate = "02 maio 2024"
}