@InProceedings{CamargoPezzHerd:2010:ObNuSt,
author = "Camargo, Ricardo de and Pezzi, Luciano Ponzi and Herdies, Dirceu
Luis",
affiliation = "Atmospheric Sciences, Institute of Astronomy, Geophysics and
Atmospheric Sciences of University of S{\~a}o Paulo, S{\~a}o
Paulo, Brazil and {Instituto Nacional de Pesquisas Espaciais
(INPE)} and {Instituto Nacional de Pesquisas Espaciais (INPE)}",
title = "Observational and numerical studies of SST variability at South
Atlantic using the regional CPTEC oceanic data assimilation system
(R-CODAS)",
year = "2010",
organization = "The Meeting of the Americas.",
publisher = "AGU",
keywords = "numerical modeling, ocean data assimilation, reanalysis, ocean
predictability, prediction.",
abstract = "The Sea Surface Temperature (SST) variability can play a decisive
role on the atmospheric systems evolution either on climate or
weather time scales. There are lots of efforts being applied to
adequately represent SST features and variability, even
considering uncoupled models of both atmosphere and ocean. This
work aims to investigate SST features and variability at South
Atlantic, based on Optimal Interpolation SST version 2 (OISSTv2)
and numerical experiments performed with Regional Ocean Modeling
System (ROMS) with and without data assimilation. This was made
through the use of Regional CPTEC Oceanic Data Assimilation System
(R-CODAS) based on Local Kalman Ensemble Filter Transform (LETKF)
technique. On the observational aspects, space EOFs showed modes
of SST variability at South Atlantic that can be connected to
Equatorial Pacific SST anomalies linked to El-Niņo Southern
Oscilation (ENSO) and also to South Pacific SST variations related
to extratropical signature of Pacific-South America atmospheric
teleconnection patterns (PSA). For the modeling stage, ROMS
simulations were performed in a 1/2 degree resolution grid
covering Tropical and South Atlantic Ocean, using both bulk-fluxes
formulation and NCEP/Reanalysis fluxes, being the latter with heat
and fresh water flux corrections. These numerical results could
show the model sensitivity to atmospheric forcing, mainly over
higher latitudes of Western South Atlantic. Spin-up stages of 20
years with each of those forcing files reached dynamically
balanced fields that were used as initial conditions for long
integrations in the period 1980-2009. Once again, differences
between bulk-formulation and NCEP/Reanalysis fluxes are more
significant over Argentinean shelf than in other regions of the
domain. Despite this fact, both simulations could reasonable
represent SST features and variability, and could be considered
for the data assimilation stage. R-CODAS is still under
development/tuning and preliminary results assimilating only
OISSTv2 data showed that the number of members as well as the
inflation coefficient are key-parameters for the simulations,
reflecting directly on the capacity of the method on assimilating
real data. New analyzed fields after the assimilation cycle were
closer to the reality when compared with no-data-assimilation
results. The next step is the consideration of Sea Surface Height
from satellite data (SSH) as well as ocean profiles to be included
in the assimilation procedure, besides spatial and time analyses
of the numerical fields.",
conference-location = "Foz do Igua{\c{c}}u",
conference-year = "8-12 Aug. 2010",
language = "en",
targetfile = "Camargo_Observational.pdf",
urlaccessdate = "20 set. 2024"
}