@PhDThesis{Ferreira:2021:UsAsDa,
author = "Ferreira, Rute Costa",
title = "Uso da assimila{\c{c}}{\~a}o de dados de radar e descargas
el{\'e}tricas na previs{\~a}o de curt{\'{\i}}ssimo prazo no
Sul do Brasil",
school = "Instituto Nacional de Pesquisas Espaciais (INPE)",
year = "2021",
address = "S{\~a}o Jos{\'e} dos Campos",
month = "2021-02-25",
keywords = "WRFDA, GSI, BrasilDAT, GLM.",
abstract = "As tempestades que atuam no Sul do Brasil causam diversos danos
econ{\^o}micos e sociais. Muitas vezes, estas tempestades
est{\~a}o associadas a tempo severo, com rajadas de ventos,
granizo e descargas el{\'e}tricas atmosf{\'e}ricas, e at{\'e}
mesmo tornados. O uso da modelagem num{\'e}rica e dados
observados {\'e} fundamental para aumentar o conhecimento do
ambiente no qual estas se formam e, a partir da{\'{\i}},
melhorar sua previsibilidade. Neste contexto, este trabalho
estudou tempestades com ocorr{\^e}ncia de descargas
el{\'e}tricas do ano de 2017 e 2018 com foco na
assimila{\c{c}}{\~a}o de dados de radar e de descargas
el{\'e}tricas. As simula{\c{c}}{\~o}es foram feitas a partir
dos dados de refletividade e velocidade radial dos radares da
For{\c{c}}a A{\'e}rea Brasileira, com foco nos dados de Santiago
(RS). Os dados de descargas el{\'e}tricas utilizados foram da
rede de detec{\c{c}}{\~a}o BrasilDAT, e dados do instrumento GLM
(Geostationary Lightning Mapper) do sat{\'e}lite
geoestacion{\'a}rio GOES-16. Para comparar o campo de
precipita{\c{c}}{\~a}o acumulada foram utilizados campos de
estimativa de precipita{\c{c}}{\~a}o do produto MERGE do
CPTEC/INPE. O modelo atmosf{\'e}rico utilizado foi o Weather
Research and Forecasting model (WRF) e sua componente de
assimila{\c{c}}{\~a}o dados (WRFDA), para
assimila{\c{c}}{\~a}o de dados de radar e o sistema Gridpoint
Statistical Interpolation (GSI) para assimila{\c{c}}{\~a}o dos
dados de rel{\^a}mpagos. Os resultados foram analisados em
fun{\c{c}}{\~a}o do impacto da assimila{\c{c}}{\~a}o nos
campos atmosf{\'e}ricos e evolu{\c{c}}{\~a}o na previs{\~a}o
de um sistema frontal em 2017, e um sistema convectivo de meso
escala relacionado {\`a} instabilidade termodin{\^a}mica em
2018, ocorridos no Rio Grande do Sul. Em ambas
condi{\c{c}}{\~o}es atmosf{\'e}ricas, os experimentos com
assimila{\c{c}}{\~a}o dos dados da BrasilDAT ou GLM indicaram um
aumento dos hidrometeoros distribu{\'{\i}}dos em toda a
troposfera e, consequentemente, da precipita{\c{c}}{\~a}o. A
r{\'a}pida taxa de convers{\~a}o microf{\'{\i}}sica foi notada
principalmente na primeira hora de previs{\~a}o. Para o caso de
precipita{\c{c}}{\~a}o decorrente de instabilidade
termodin{\^a}mica, os experimentos com os dados do GLM e radar
assimilados mostraram melhor posicionamento dos n{\'u}cleos
convectivos e desempenho ao prever os sistemas estudados.
Entretanto, a assimila{\c{c}}{\~a}o de dados apenas de descargas
el{\'e}tricas para simula{\c{c}}{\~a}o da frente fria indicou
limita{\c{c}}{\~o}es no GSI ao gerar grandes incrementos e
superestimativa de precipita{\c{c}}{\~a}o, por{\'e}m a
assimila{\c{c}}{\~a}o de dados de radar foi capaz de minimizar
tal limita{\c{c}}{\~a}o ao serem assimilados em algumas das
simula{\c{c}}{\~o}es. Com isso, a assimila{\c{c}}{\~a}o de
dados de radar p{\^o}de complementar poss{\'{\i}}veis
limita{\c{c}}{\~o}es encontradas no novo operador de descargas
el{\'e}tricas no GSI, se tornando complementares de acordo com as
informa{\c{c}}{\~o}es que cada observa{\c{c}}{\~a}o pode
fornecer ao modelo. ABSTRACT: The storms observed in the South of
Brazil cause several economic and social damages. These storms are
often associated with severe weather conditions, with wind gusts,
hail, lightning, and even tornadoes. The use of numerical modeling
and observed data is fundamental to increase the knowledge of the
environment in which storms are formed and improve their
predictability. In this context, this work studied storms with the
occurrence of lightning in the years 2017 and 2018 with a focus on
radar and lightning data assimilation. The simulations were made
based on data of reflectivity and radial velocity from the radars
of the Brazilian Air Force, focusing on data from Santiago (RS).
The data of lightning used were from the BrasilDAT detection
network, and data from the GLM (Geostationary Lightning Mapper)
instrument of the geostationary satellite GOES-16. To compare the
accumulated precipitation field, estimation fields of
precipitation of the MERGE product from CPTEC / INPE were used.
The Weather Research and Forecasting (WRF) model - together with
its data assimilation component (WRFDA) - was used to assimilate
radar data. The Gridpoint Statistical Interpolation (GSI) system
was used to assimilate lightning data. The results were analyzed
according to the impact of assimilation in the atmospheric fields
and evolution in the forecast of studied cases. The cases were: a
frontal system in 2017, and a mesoscale convective system related
to thermodynamic instability in 2018, both occurred in Rio Grande
do Sul. In both atmospheric conditions, the experiments with
assimilation of data from BrasilDAT or GLM indicated an increase
in hydrometeors distributed throughout the troposphere and,
consequently, in precipitation. The rapid rate of microphysical
conversion was noticed mainly in the first hour of forecast. For
the case of precipitation due to thermodynamic instability, the
experiments with the data from the GLM and assimilated radar
showed better positioning of the convective nuclei and performance
when predicting the studied systems. However, the assimilation of
data only from lightning for the simulation of the cold front
indicated limitations in the GSI when generating large increments
and overestimation of precipitation, however, the assimilation of
radar data was able to minimize such limitation when they were
assimilated in some of the simulations. With this, the
assimilation of radar data could complement possible limitations
found in the new lightning operator in GSI, becoming complementary
according to the information that each observation can provide to
the model.",
committee = "Vila, Daniel Alejandro (presidente) and Herdies, Dirceu Luis
(orientador) and Vendrasco, {\'E}der Paulo (orientador) and
Arav{\'e}quia, Jos{\'e} Antonio and Quadro, M{\'a}rio Francisco
Leal de and Beneti, Cesar Augustus Assis",
englishtitle = "Use of radar and lightning data assimilation in very short-term
forecast in Southern Brazil",
language = "pt",
pages = "171",
ibi = "8JMKD3MGP3W34R/44CQ5PL",
url = "http://urlib.net/ibi/8JMKD3MGP3W34R/44CQ5PL",
targetfile = "publicacao.pdf",
urlaccessdate = "05 maio 2024"
}