@PhDThesis{HernándezPardo:2021:ReWaMi,
author = "Hern{\'a}ndez Pardo, Lianet",
title = "On the representation of warm-phase microphysical processes in
numerical simulations of cumulus clouds",
school = "Instituto Nacional de Pesquisas Espaciais (INPE)",
year = "2021",
address = "S{\~a}o Jos{\'e} dos Campos",
month = "2020-11-19",
keywords = "cloud microphysical processes, microphysics parameterizations,
convective clouds, aerosol-cloud-precipitation interactions,
droplet size distributions, processos de microf{\'{\i}}sica de
nuvens, parametriza{\c{c}}{\~o}es de microf{\'{\i}}sica de
nuvens, nuvens convectivas, intera{\c{c}}{\~o}es
aeross{\'o}is-nuvens-precipita{\c{c}}{\~a}o,
distribui{\c{c}}{\~a}o de tamanho de gotas.",
abstract = "This research analyzes the role of microphysical processes in the
evolution of convective clouds, with emphasis on their
representation in atmospheric models, taking advantage of in-situ
measurements obtained as part of the ACRIDICON-CHUVA field
campaign near the Amazon basin. Firstly, cloud sensitivities to
changes in aerosol properties, such as total number concentration
(Na), size distribution and composition, are explored. This
analysis focuses on the droplet number concentration and effective
diameter at the top of a shallow cumulus simulated with a
bin-microphysics single-column model, for initial conditions
typical of the Amazonian region. The impact of considering bin
versus bulk aerosol size distributions is evaluated, in order to
investigate the influence of entrainment and activation scavenging
on the derived sensitivities. It is shown that the evolution of
cloud-top droplet size distributions (DSDs) is generally very
sensitive to changes in aerosol parameters, but the sign and
absolute value of this sensitivity depends on the position within
the multidimensional aerosol parameter space considered, as well
as on the treatment of entrainment and aerosol scavenging.
Following, idealized two- and three-dimensional Weather Research
and Forecasting model simulations of an isolated warm cumulus
cloud were performed to assess the role of DSD broadening
mechanisms in bin microphysics parameterizations. As expected,
collision-coalescence is a key process broadening the modeled
DSDs. In-cloud droplet activation also contributes substantially
to DSD broadening, whereas evaporation has only a minor effect and
sedimentation has little effect. Cloud dilution (mixing of
cloud-free and cloudy air) also broadens the DSDs considerably,
whether or not it is accompanied by evaporation. Artificial
broadening from vertical numerical diffusion appears to be
relatively unimportant overall for these cumulus simulations.
Finally, the spatial distribution of DSD relative dispersion ()
within observed and modeled convective clouds is analyzed, taking
into account changes in Na. The spatial distribution of DSD shape
in aircraft measurements of growing cumuli near the Amazon basin
shows distinctive patterns depending on the aerosol loading. In
cleaner clouds (Na < 900 cm\−3), varies between 0.1 and
0.6, overall being inversely related to the ratio of the cloud
water content (qc) and the adiabatic water content (qa). In
polluted clouds (Na > 2000 cm\−3), generally has values in
the range 0.25 \− 0.45, with no evident dependence on
either height above cloud base or qc/qa. Bin-microphysics
numerical simulations confirm that these contrasting behaviors are
associated with the predominance of collision-coalescence in
cleaner clouds, and in-cloud droplet activation in polluted ones.
These findings can be useful for parameterizations of the shape
parameter (\μ) of droplet gamma size distributions in
bulk-microphysics cloud-resolving models. It is shown that
emulating the observed \μ\−qc/qa relationship
improves the estimation of the collision-coalescence rate in bulk
microphysics simulations. RESUMO: Esta pesquisa analisa o papel
dos processos de microf{\'{\i}}sica na evolu{\c{c}}{\~a}o das
nuvens convectivas, com {\^e}nfase na sua
representa{\c{c}}{\~a}o em modelos atmosf{\'e}ricos, usando
observa{\c{c}}{\~o}es do experimento ACRIDICON-CHUVA na
Amaz{\^o}nia. Em primeiro lugar, {\'e} explorada a sensibilidade
das nuvens a mudan{\c{c}}as nas propriedades dos aeross{\'o}is,
como concentra{\c{c}}{\~a}o (Na), distribui{\c{c}}{\~a}o de
tamanho e composi{\c{c}}{\~a}o. Esta an{\'a}lise est{\'a}
focada na concentra{\c{c}}{\~a}o total e o di{\^a}metro efetivo
das got{\'{\i}}culas no topo de cumulus rasos simulados com um
modelo 1D usando um esquema de microf{\'{\i}}sica bin, para
condi{\c{c}}{\~o}es iniciais tipicas da Amaz{\^o}nia. {\'E}
avaliado o impacto de considerar enfoques bin ou bulk para as
distribui{\c{c}}{\~o}es de tamanho dos aeross{\'o}is, com o
objetivo de investigar a influ{\^e}ncia do entranhamento e a
ativa{\c{c}}{\~a}o nas sensibilidades obtidas. Mostra-se que a
evolu{\c{c}}{\~a}o da distribui{\c{c}}{\~a}o de tamanho de
gotas (DSD) {\'e} geralmente muito sens{\'{\i}}vel a
mudan{\c{c}}as nos par{\^a}metros dos aeross{\'o}is, mas o
sinal e o valor absoluto da sensibilidade depende da
posi{\c{c}}{\~a}o considerada no espa{\c{c}}o multidimensional
dos par{\^a}metros dos aeross{\'o}is, assim como do tratamento
do entranhamento e do remo{\c{c}}{\~a}o de aeross{\'o}is por
ativa{\c{c}}{\~a}o. Seguidamente, s{\~a}o utilizadas
simula{\c{c}}{\~o}es idealizadas de cumulus rasos em 2D e 3D com
o modelo Weather Research and Forecasting para estudar o papel dos
processos que intervem no alargamento das DSDs em
parametriza{\c{c}}{\~o}es de microf{\'{\i}}sica bin. Como
esperado, a colis{\~a}o-coalesc{\^e}ncia {\'e} um processo
chave no alargamento das DSDs no modelo. A ativa{\c{c}}{\~a}o de
gotas no interior das nuvens tamb{\'e}m contribui
significativamente para o alargamento das DSDs, enquanto a
evapora{\c{c}}{\~a}o tem um efeito menor e a influ{\^e}ncia da
sedimenta{\c{c}}{\~a}o {\'e} negligenci{\'a}vel. A
dilui{\c{c}}{\~a}o da nuvem (mistura com o ar na
vizinhan{\c{c}}a) tamb{\'e}m alarga as DSDs consideravelmente,
independentemente da ocorr{\^e}ncia de evapora{\c{c}}{\~a}o. O
alargamento artificial das DSDs devido {\`a} difus{\~a}o
num{\'e}rica na dire{\c{c}}{\~a}o vertical tem relativamente
pouca import{\^a}ncia em geral nessas simula{\c{c}}{\~o}es.
Finalmente, analisa-se a distribui{\c{c}}{\~a}o espacial da
dispers{\~a}o relativa das DSDs () em nuvens convectivas
observadas e simuladas, levando em conta mudan{\c{c}}as em Na. A
distribui{\c{c}}{\~a}o espacial da forma das DSDs em
medi{\c{c}}{\~o}es diretas de cumulus em crescimento na
Amaz{\^o}nia cont{\'e}m padr{\~o}es distintivos dependendo do
conte{\'u}do de aeross{\'o}is. Em nuvens limpas (Na < 900
cm\−3), varia entre 0.1 e 0.6, dependendo principalmente
raz{\~a}o entre o conte{\'u}do de {\'a}gua de nuvem (qc) e
conte{\'u}do de {\'a}gua l{\'{\i}}quida adiab{\'a}tico (qa).
Em nuvens polu{\'{\i}}das (Na > 2000 cm\−3), generalmente
toma valores no intervalo 0.25 \− 0.45, e {\'e}
praticamente independente da altura acima da base na nuvem e de
qc/qa. Simula{\c{c}}{\~o}es com o esquema de
microf{\'{\i}}sica bin confirmam que estes comportamentos
contrastantes est{\~a}o associados {\`a} predomin{\^a}ncia de
colis{\~a}o-coalesc{\^e}ncia em nuvens limpas, e
ativa{\c{c}}{\~a}o de gotas no interior de nuvens
polu{\'{\i}}das. Estes resultados podem ser {\'u}teis para
parametrizar o par{\^a}metro de forma (\μ) de
distribui{\c{c}}{\~o}es gama de tamanho de gotas em esquemas de
microf{\'{\i}}sica do tipo bulk em modelos de nuvem resolvida.
{\'E} mostrado que a inclus{\~a}o da rela{\c{c}}{\~a}o
\μ\−qc/qa de acordo com as observa{\c{c}}{\~o}es
melhora a estimativa da colis{\~a}o-coalesc{\^e}ncia em
simula{\c{c}}{\~o}es de microf{\'{\i}}sica bulk.",
committee = "Sapucci, Luiz Fernando (presidente) and Machado, Luiz Augusto
Toledo (orientador) and Vendrasco, {\'E}der Paulo and Morrison,
Hugh and P{\"o}hlker, Mira",
englishtitle = "An{\'a}lise da representa{\c{c}}{\~a}o dos processos de
microf{\'{\i}}sica de nuvens correspondentes {\`a} fase quente
em simula{\c{c}}{\~o}es num{\'e}ricas de nuvens cumulus",
language = "en",
pages = "135",
ibi = "8JMKD3MGP3W34R/43FSLES",
url = "http://urlib.net/ibi/8JMKD3MGP3W34R/43FSLES",
targetfile = "publicacao.pdf",
urlaccessdate = "27 abr. 2024"
}