Performance report of the RHUM-RUM ocean bottom seismometer network around La R&
Title:Performance report of the RHUM-RUM ocean bottom seismometer network around La Réunion, western Indian Ocean
Authors:; ; ; ; ; ; ; ; ;
Affiliation:AA(Dept. of Earth Sciences, Ludwig-Maximilians-Universit?t München, Theresienstrasse 41, 80333 Munich, G Leibniz-Institute for Baltic Sea Research, Seestrasse 15, 18119 Rostock, Germany ), AB(Dept. of Earth Sciences, University of Oxford, South Parks Road, Oxford, OX1 3AN, UK; Dept. of Earth Sciences, Ludwig-Maximilians-Universit?t München, Theresienstrasse 41, 80333 Munich, Germany ), AC(Dept. of Earth Sciences, Ludwig-Maximilians-Universit?t München, Theresienstrasse 41, 80333 Munich, Germany ), AD(Institut de Physique du Globe de Paris, Sorbonne Paris Cité, UMR7154 - CNRS, Paris, France ), AE(Laboratoire GéoSciences Réunion, Université de La Réunion, Institut de Physique du Globe de Paris, Sorbonne Paris Cité, UMR7154 - CNRS, Université Paris Diderot, Saint Denis CEDEX 9, France ), AF(Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Alten Hafen 26, 27568 Bremerhaven, Germany ), AG(Dept. of Earth Sciences, University of Oxford, South Parks Road, Oxford, OX1 3AN, UK; Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Alten Hafen 26, 27568 Bremerhaven, Germany ), AH(Laboratoire GéoSciences Réunion, Université de La Réunion, Institut de Physique du Globe de Paris, Sorbonne Paris Cité, UMR7154 - CNRS, Université Paris Diderot, Saint Denis CEDEX 9, F Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Alten Hafen 26, 27568 Bremerhaven, Germany ), AI(Institut de Physique du Globe de Paris, Sorbonne Paris Cité, UMR7154 - CNRS, Paris, France ), AJ(Institut de Physique du Globe de Paris, Sorbonne Paris Cité, UMR7154 - CNRS, Paris, France )
Publication:Advances in Geosciences, Volume 41, 2016, pp.43-63 ()
Publication Date:02/2016
Bibliographic Code:
RHUM-RUM is a German-French seismological experiment based on the sea
floor surrounding the island of La Réunion, western Indian Ocean
(Barruol and Sigloch, 2013). Its primary objective is to clarify the
presence or absence of a mantle plume beneath the Reunion volcanic
hotspot. RHUM-RUM's central component is a 13-month deployment (October
2012 to November 2013) of 57 broadband ocean bottom seismometers (OBS)
and hydrophones over an area of 2000 × 2000 km2
surrounding the hotspot. The array contained 48 wideband OBS from the
German DEPAS pool and 9 broadband OBS from the French INSU pool. It is
the largest deployment of DEPAS and INSU OBS so far, and the first joint
experiment.
This article reviews network performance and
data quality: of the 57 stations, 46 and 53 yielded good seismometer and
hydrophone recordings, respectively. The 19 751 total deployment days
yielded 18 735 days of hydrophone recordings and 15 941 days of
seismometer recordings, which are 94 and 80 % of the theoretically
possible yields.
The INSU seismic sensors stand away from
their OBS frames, whereas the DEPAS sensors are integrated into their
frames. At long periods (& 10 s), the DEPAS seismometers are affected
by significantly stronger noise than the INSU seismometers. On the
horizontal components, this can be explained by tilting of the frame and
buoy assemblage, e.g. through the action of ocean-bottom currents, but
in addition the DEPAS intruments are affected by significant self-noise
at long periods, including on the vertical channels. By comparison, the
INSU instruments are much quieter at periods & 30 s and hence better
suited for long-period signals studies.
The trade-off of
the instrument design is that the integrated DEPAS setup is easier to
deploy and recover, especially when large numbers of stations are
involved. Additionally, the wideband sensor has only half the power
consumption of the broadband INSU seismometers. For the first time, this
article publishes response information of the DEPAS instruments, which
is necessary for any project where true ground displacement is of
interest. The data will become publicly available at the end of 2017.
&&& (see )
Find Similar Abstracts:
Abstract Text
Query Results
Return &&&items starting with number
Query Form
arXiv e-printsA Coupled Regional Climate Model Simulation of Typhoon Son-Tinh with Increased Horizontal Resolutions--《Periodical of Ocean University of China》2015年03期
A Coupled Regional Climate Model Simulation of Typhoon Son-Tinh with Increased Horizontal Resolutions
WAN Xiu-QMA Wei-WWU De-XLI Ming-KOcean University of China,College of Physical and Environmental OOcean University of China,The Key Laboratory of Physical Oceanography,Ministry of E
Numerical model is one of the main means to study typhoon process,and how to improve the precision of typhoon simulation has become more important.Although considerable progress has been made in our understanding of the dynamic and thermodynamic mechanisms of typhoon process,there is still long way to go to simulate and predict typhoon in reality.Based on a regional coupled model developed recently in Ocean University of China,we set up a series of numerical experiments to simulate typhoon Son-Tinh occurred in South China Sea in 2012 with increased horizontal model resolutions.The NCEP FNL Operational Global Analysis data and HYCOM Global Analysis data are applied as initial and boundary conditions to all the experiments.The results indicate that air-sea coupling and higher resolution can improve the model capability to simulate typhoon process.The oceanic responses to typhoon are also discussed.
【Fund】：
【CateGory Index】：
supports all the CNKI
only supports the PDF format.
【Citations】
Chinese Journal Full-text Database
ZHANG Xiaoyan WANG B JI Zhongzhen Qingnong XIAO , and ZHANG Xin
LASG, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029 Mesoscale and Microscale Meteorology, National Center for Atmospheric Research, Boulder, Colorado, USA;[J];大气科学进展(英文版);2003-04
Chen Lianshou and Meng Zhiyong (Chinese Academy of Meteorological Setences, Beijing 100081);[J];Chinese Journal of Atmospheric S2001-03
ZENG Zhi-hua1-2,CHEN Lian-shou3(1.Nanjing University of Information Science and Technology,Nanjing 210044,C 2.Shanghai Typhoon Institute,Laboratory of Typhoon Forecast Technique/China Meteorological Administration, Shanghai 200030,C 3.Chinese Academy of Meteorological Sciences,China Meteorological Administration,Beijing 100081,China);[J];Plateau M2011-06
Liu Ningwei Wang Feng'an(Institute of Atmospheric Environment,CMA,Shenyang 110016);[J];Meteorological Science and T2006-04
Chen Yuling 1 Zhou Jun 2 Ma Fenhua 2 (1 Department of Physical Sciences,NIM ,Nanjing 210044) (2 Department of Atmospheric Sciences,NIM,Nanjing 210044);[J];Scientia Meteorologica S2005-03
Fei Jianfang1 Huang Xiaogang1 Cheng Xiaoping1 Zheng Jing2(1 Institute of Meteorology,PLA University of Science and Technology,Nanjing 211101,China)(2 73061Troop of PLA,Jiangsu Xuzhou 221008,China);[J];Scientia Meteorologica S2010-05
Zhao Ping (Chinese Academy of Meteorological Sciences, Beijing 100081);[J];Acta Meteorologica S2003-06
HA HyeKyeong1,2,3,WANG Zhen-hui1,2,KIM Jeoung-Yun1,4,3,NIU Sheng-Jie1,SUH AeSook3 (1.Jiangsu Key Laboratory of Meteorological Disaster,NUIST,Nanjing .School of Remote Sensing,NUIST,Nanjing .Korea Meteorological Administration,Seoul 156-720,K4.Key Laboratory of Atmospheric Physics and Atmospheric Environment,NUIST,Nanjing 210044,China);[J];Journal of Tropical M2009-04
LIN Fang-ni1,2,WAN Qi-lin1,GUAN Zhao-yong1 (1.Nanjing university of information Science & Technology,Nanjing 210044,C 2.Institute of Tropical and Marine Meteorology,CMA,Guangzhou 510080,China);[J];Journal of Tropical M2011-02
LIAO Jing-biao1,2,WANG Xue-mei1,XIA Bei-cheng1,WANG Ti-jian2,WANG Zhi-ming1(1.School of Environmental Science and Engineering,Sun Yat-Sen University,Guangzhou 510275,China)(2.School of Atmospheric Sciences,Nanjing University,Nanjing 210093,China);[J];Journal of Tropical M2012-04
【Co-citations】
Chinese Journal Full-text Database
PAN Jun et al (Atmospheric Science Department,Physics School of Peking University,Beijing 100871);[J];Journal of Anhui Agricultural S2010-22
WANG Yuan-chao et al(Yulin Meteorological Bureau in Guangxi,Yulin,Guangxi 537000);[J];Journal of Anhui Agricultural S2010-34
LU Bing 1,WANG Bin 1,and ZHAO Ying 2 1 State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics,Institute of Atmospheric Physics,Chinese Academy of Sciences,Beijing 100029,China 2 Institute of Sciences,University of Science and Technology of the Chinese People's Liberation Army,Nanjing 211101,C[J];大气和海洋科学快报(英文版);2011-04
WANG Shu-Dong 1,2,LIU Juan-Juan 2,and WANG Bin 2 1 Meteorological Observation Centre,China Meteorological Administration,Beijing 100081,China 2 Laboratory of Numerical Modeling for Atmospheric Sciences & Geophysical fluid dynamics (LASG),Institute of Atmospheric Physics (IAP),Chinese Academy of Sciences,Beijing 100029,C[J];大气和海洋科学快报(英文版);2011-05
ZHANG Qinghong
1)Department of Atmospheric Science,Peking University, Beijing 100871;
2)Weather Centre,Northeast Air Traffic Administration Bureau,Shenyang,110169);[J];Acta Scicentiarum Naturalum Universitis P2004-01
DAI Zhanpeng,ZHANG Qinghong? Department of Atmospheric and Oceanic Sciences,School of Physics,Peking University,Beijing 100871;?;[J];Acta Scientiarum Naturalium Universitatis P2011-02
CHEN Zhao-hui1,CHENG Shui-yuan1,SU Fu-qing2,GAO Qing-xian2(1.College of Environmental and Energy Engineering,Beijing University of Technology,Beijing .Chinese Research Academy of Environmental Sciences,Beijing 100012,China);[J];Journal of Beijing University of T2010-02
HOU Jian-zhong1,NING Zhi-qian1,XIE Shuang-ting1,DENG Ai-jun2 (1.Shaanxi Bureau of Meteorology,Xi'an .Pennsylvanian State University,PA 16802,US);[J];Journal of Chengdu University of Information T2007-05
HOU Jian-zhong1,XU Xin-tian1,ZHANG Xiao-ling2,LI Ming1,DUAN Chang-hui1(1.Shaanxi Bureau of Meteorology,Xi'an .National Meteorological Center,Beijing 100081,China);[J];Journal of Chengdu University of Information T2008-05
ZENG Xin-xin, HUANG Xin-qing, TENG Dai-gao(Meteorological Observatory of Zhejiang Province, Hangzhou 310017, China);[J];Journal of Marine S2010-01
China Proceedings of conference Full-text Database
HU Shu~(1,2) LI Ying~1 WEI Na~2 1 Oceanic Hydrologic Center of Navy,Beijing
State Key Laboratory of Severe Weather,Chinese Academy of Meteorological Sciences,Beijing 100082;[A];[C];2012
Huizhong He1 Minghu Cheng2 Fengxian Zhou3 1 State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing
Chinese Academy of Meteorological Sciences, Beijing
Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100080;[A];[C];2005
Wei Ying-zhi1,2, Tang Da-zhang2 Xu Jian-min3, Wu Chen-feng1
(1. Xiamen Meteorological Bureau, Xiamen 361012, Fujian C 2. Nanjing University of Information Science and Technology, Nanjing 210044, C 3. National Satellite Meteorological Center, Beijing 100081, China );[A];[C];2005
Ye Chengzhi1 Chenri2 Xu Aihua3 Pan Zhixiang4 (1. Institute of Heavy Rain,CMA,Wuhan) (2.LASG; 3. Jiangxi Meteorological Office, 4. Hunan Meteorological Office,Changsha );[A];[C];2005
ZHANG Hong, Liang Shengjun, Hou Jianzhong (Shaanxi Meteorological Observatory, Xian, Shaanxi 710015);[A];[C];2005
Yu Zhenshou1 Ji Chunxiao2 Wang Zhongdong1 Lou Liyin1 Wu Mengchun1 (1.Wenzhou Meteorological Bureau, Wenzhou,325003) (2.Zhejiang Research Institute of Meteorology,Hangzhou,310002);[A];[C];2005
Zhou Fengcai1 Zhang Xijun2
1 Shanghai Meteorological Center, Air Traffic Management Bereau of East China, Shanghai
Institute of Meteorology, PLA University of Science and Engineering, Nanjing 211101;[A];[C];2006
Huang Yanyan, Yan Jinghua, Meng Weiguang, Wan Qilin
(Institute of Tropical and Marine Meteorology, CMA, Guangzhou 510080 );[A];[C];2008
ZHANG Jian-Hai YU Zhong-Kai (Shaoxing Meteorological bureau,shaoxing 312000, China);[A];[C];2008
LI Gang1, WANG Tie1 TAN Yan-ke1 FENG Wen-tian2 (1. Institute of Meteorology, PLA Univ.of Sci&Tech., Nanjing 211101,C(2. No.66393 Troops of PLA,071000,China);[A];[C];2009
【Secondary Citations】
Chinese Journal Full-text Database
Fei Jianfang and Lu HanchengAir Force Institute of Meteorology,Nanjing 211101;[J];大气科学进展(英文版);1996-04
ZHANG Xiaoyan WANG B JI Zhongzhen Qingnong XIAO , and ZHANG Xin
LASG, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029 Mesoscale and Microscale Meteorology, National Center for Atmospheric Research, Boulder, Colorado, USA;[J];大气科学进展(英文版);2003-04
ZENG Zhihua 1,2 , Yuqing WANG 3 , DUAN Yihong 4 , CHEN Lianshou 5 , and GAO Zhiqiu 2 1 Laboratory of Typhoon Forecast Technique, Shanghai Typhoon Institute/ China Meteorological Administration, Shanghai
The State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing
International Pacific Research Center and Department of Meteorology, School of Ocean and Earth Science and Technology, University of Hawaii at Manoa, Honolulu, HI 96822, USA 4 National Meteorological Center, China Meteorological Administration, Beijing
Chinese Academy of Meteorological Sciences, China Meteorological Administration, Beijing 100081;[J];大气科学进展(英文版);2010-02
Chen Lianshou (Chinese Academy of Met6orological Science,Beijing 100081)and Luo Zhexian(Naming institute of Meteorology,Naming 210044);[J];大气科学进展(英文版);1995-02
Wang Donghai and Zhou Xiaoping(Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100080);[J];Chinese Journal of Atmospheric S1994-01
Zhu Hongyan, Chen Lianshou and Xu Xiangde (Chinese Academy of Meteorological Sciences, Beijing 100081);[J];SCIENTIA ATMOSPHERICA SINICA;2000-05
Chen Lianshou and Meng Zhiyong (Chinese Academy of Meteorological Setences, Beijing 100081);[J];Chinese Journal of Atmospheric S2001-03
Luo Zhexian (Naming Institute of Meteorology, Nanjing 210044);[J];SCIENTIA ATMOSPHERICA SINICA;1994-05
Zhai G Gao K Yu Zhangxiao and Tu Caihong (Department of Geography, Hangzhou University, Hangzhou 310028);[J];SCIENTIA ATMOSPHERICA SINICA;1995-04
Luo ZChen Lianshou (Nanjing Institute of Meteorology, Nanjing 210044) (Academy of Meteorological Science, Beijing 100081 );[J];SCIENTIA ATMOSPHERICA SINICA;1995-06
Similar Journals
(C)2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reservedA truncated Gaussian filter for data assimilation with inequality constraints: Application to the hydrostatic stability condition in ocean models - ScienceDirect
Export JavaScript is disabled on your browser. Please enable JavaScript to use all the features on this page., 2009, Pages 1-17Author links open overlay panelShow moreAbstractIn many data assimilation problems, the state variables are subjected to inequality constraints. These constraints often contain valuable information that must be taken into account in the estimation process. However, with linear estimation methods (like the Kalman filter), there is no way to incorporate optimally that kind of additional information. In this study, it is shown that an optimal filter dealing with inequality constraints can be formulated under the assumption that the probability distributions are truncated Gaussian distributions. The statistical tools needed to implement this truncated Gaussian filter are described. It is also shown how the filter can be adapted to work in a reduced dimension space, and how it can be simplified following several additional hypotheses. As an application, the truncated Gaussian assumption is shown to be adequate to deal with the condition of hydrostatic stability in ocean analyses. First, a detailed evaluation of the method is made using a one-dimensional z-coordinate model of the mixed layer: particular attention is paid to the parameterization of the probability distribution, the accuracy of the estimation and the sensitivity to the observation system. In a second step, the method is applied to a three-dimensional hybrid coordinate ocean model (HYCOM) of the Bay of Biscay (at a 1/15° resolution), to show that it is efficient enough to be applied to real size problems. These examples also demonstrate that the algorithm can deal with the hydrostatic stability condition in isopycnic coordinates as well as in z-coordinates.KeywordsData assimilationInequality constraintsReduced-order Kalman filterHybrid coordinate ocean modelsChoose an option to locate/access this article:Check if you have access through your login credentials or your institution.ororRecommended articlesCiting articles (0)The bifurcation of the western boundary... (PDF Download Available)
See all >3 CitationsSee all >34 ReferencesSee all >12 Figures
4.86Ipebj - Instituto Paulista de Estudos Bioéticos e Jurídicos7.38Federal University of Rio de Janeiro+ 225.31None3.04Universidade Federal da BahiaShow more authorsAbstractThe results of two high-resolution ocean global circulation models-OGCMs (Hybrid Coordinate Ocean Model-HYCOM and Ocean Circulation and Climate Advanced Modeling Project-OCCAM) are analyzed with a focus on the Western Boundary Current (WBC) system of the South Atlantic Ocean. The volume transports are calculated for different isopycnal ranges, which represent the most important water masses present in this region. The latitude of bifurcation of the zonal flows reaching the coast, which leads to the formation of southward or northward WBC flow at different depths (or isopycnal levels) is evaluated. For the Tropical Water, bifurcation of the South Equatorial Current occurs at 13o-15oS, giving rise to the Brazil Current, for the South Atlantic Central Water this process occurs at 22oS. For the Antarctic Intermediate Water, bifurcation occurs near 28o-30oS, giving rise to a baroclinic unstable WBC at lower latitudes with a very strong vertical shear at mid-depths. Both models give similar results that are also consistent with previous observational studies. Observations of the South Atlantic WBC system have previously been sparse, consequently these two independent simulations which are based on realistic high-resolution OGCMs, add confidence to the values presented in the literature regarding flow bifurcations at the Brazilian coast.Discover the world's research14+ million members100+ million publications700k+ research projectsFigures
ii“main” —
— 12:06 — page 241 — #1iiiiiiRevista Brasileira de Geof??sica (): 241-257(C) 2014 Sociedade Brasileira de Geof??sicaISSN Xwww.scielo.br/rbgTHE BIFURCATION OF THE WESTERN BOUNDARY CURRENT SYSTEMOF THE SOUTH ATLANTIC OCEANJanini Pereira1,3 , Mariela Gabioux2, Martinho Marta-Almeida3, Mauro Cirano1,3,Afonso M. Paiva2and Alessandro L. Aguiar4ABSTRACT. The results of two high-resolution ocean global circulation models – OGCMs (Hybrid Coordinate Ocean Model – HYCOM and Ocean Circulation andClimate Advanced Modeling Project – OCCAM) are analyzed with a focus on the Western Boundary Current (WBC) system of the South Atlantic Ocean. The volumetransports are calculated for different isopycnal ranges, which represent the most important water masses present in this region. The latitude of bifurcation of the zonalflows reaching the coast, which leads to the formation of southward or northward WBC flow at different depths (or isopycnal levels) is evaluated. For the Tropical Water,bifurcation of the South Equatorial Current occurs at 13o-15oS, giving rise to the Brazil Current, for the South Atlantic Central Water this process occurs at 22oS.For the Antarctic Intermediate Water, bifurcation occurs near 28o-30oS, giving rise to a baroclinic unstable WBC at lower latitudes with a very strong vertical shearat mid-depths. Both models give similar results that are also consistent with previous observational studies. Observations of the South Atlantic WBC system havepreviously been sparse, consequently these two independent simulations which are based on realistic high-resolution OGCMs, add confidence to the values presentedin the literature regarding flow bifurcations at the Brazilian coast.Keywords: Southwestern Atlantic circulation, water mass, OCCAM, HYCOM.RESUMO. Resultados de dois modelos globais de alta resoluc,~ao (HYCOM e OCCAM) s~ao analisados focando o sistema de Corrente de Contorno Oeste do OceanoAtl^antico Sul. Os transportes de volume s~ao calculados para diferentes n??veis isopicnais que representam as principais massas de ?agua da regi~ao. ?E apresentada aavaliac,~ao da latitude de bifurcac,~ao do fluxo zonal que atinge a costa, permitindo a formac,~ao dos fluxos da Corrente de Contorno Oeste para o sul e para o norte emdiferentes n??veis de profundidades (ou isopicnal). Para a ?Agua Tropical, a bifurcac,~ao da Corrente Sul Equatorial ocorre entre 13o-15oS, originando a Corrente do Brasil,eparaa?Agua Central do Atl^antico Sul ocorre em 22oS. A bifu rcac,~ao da ?Agua Interme di?aria Ant?artica ocorre pr?oximo de 28o-30oS, dand o um aumento na in stabilidadebarocl??nica da Corrente de Contorno Oeste em baixas latitudes e com um forte cisalhamento vertical em profundidades intermedi?arias. Ambos os modelos apresentamresultados similares e consistentes com estudos observacionais pr?evios. Considerando que as o bservac,~oes do sistema de Corrente de Contorno Oeste do Atl^antico Suls~ao escassas, essas duas simulac,~oes independentes com modelos globais de alta resoluc,~ao adicionam confianc,a aos valores apresentados na literatura, relacionadosaos fluxos das bifurcac,~oes na costa do Brasil.Palavras-chave: circulac,~ao do Atl^antico Sudoeste, massas de ?agua, OCCAM, HYCOM.1Universidade Federal da Bahia – UFBA, Departamento de F??sica da Terra e do Meio Ambiente, Rua Bar~ao de Jeremoabo, s/n,
Salvador, BA, Brazil.Phone: +55(71) ; Fax: +55(71)
– E-mails: janinipereira@ufba. mcirano@ufba.br2Programa de Engenharia Oce^anica – COPPE/UFRJ, Av. Hor?acio Macedo, 2030, Bloco C, sala 203, Centro de Tecnologia – Cidade Universit?aria, Ilha do Fund~ao,P.O. Box 68.508, Rio de Janeiro, RJ, Brazil – E-mails: mariela@oceanica.ufrj. afonso@oceanica.ufrj.br3Rede de Modelagem e Observac,~ao Oceanogr?afica-REMO/UFBA e Grupo de Oceanografia Tropical – GOAT, Brazil – E-mail: m.martalmeida@gmail.com4Universidade Federal da Bahia – UFBA, Programa de P?os-Graduac,~ao em Geof??sica, Rua Bar~ao de Jeremoabo, s/n,
Salvador, BA, Brazil – E-mail:alexlaguiar@yahoo.com.br
ii“main” —
— 12:06 — page 242 — #2iiiiii242 THE BIFURCATION OF THE WESTERN BOUNDARY CURRENT SYSTEM OF THE SOUTH ATLANTIC OCEANINTRODUCTIONThe conspicuous circulation in the slope waters of the West-ern South Atlantic Ocean characterized by a southward surfaceflow that corresponds to the Brazil Current (BC) and closes thesubtropical gyre (Peterson & Stramma, 1991), which overlies anorthward flow at intermediate levels in the form of an Intermedi-ate Western Boundary Current (IWBC). This flow primarily trans-ports the Intermediate Antarctic Water (AAIW) (Boebel et al., 1997;Schmid & Garzoli, 2009; Legeais et al., 2013). In reality, the WBCsystem at the South Atlantic exhibits a complex baroclinic struc-ture with reversal flows at different latitudes for different levelsof the water column and transports distinct water masses eitherequator- or pole-ward (Muller et al., 1998; Stramma & England,1999; Silveira et al., 2008). A schematic view of these reversalflows was presented by Stramma & England (1999) (their Fig. 4).The BC originates south of 10oS, where the southern limb ofthe westward flowing South Equatorial Current (SEC) bifurcates atthe Brazilian coast (Stramma et al., 1990; Rodrigues et al., 2007).While most of its transport (~12 Sv) flows northward and even-tually feeds the North Brazil Current (NBC), a smaller contribution(~4 Sv) gives rise to a shallow BC (Stramma et al., 1990), whichcarries the Atlantic Tropical Water (TW) southward. This warmhigh-salinity surface water is located in the upper 200 m of depth(θ&20oCandS&36). However, the exact position of this sur-face bifurcation has not been clearly established but was locatedapproximately at 16oS by Stramma & England (1999). Followingthe work by these authors, SEC bifurcation occurs below this sur-face layer at higher latitudes, at levels that are roughly correspon-dent to the South Atlantic Central Waters (SACW) (6o&θ&20oCand 34.6&S&36, according to (Sverdrup et al., 1942) thermoha-line limits) at a depth range from 200 to 600 m near the Victoria-Trindade submarine mountain chain, which is located at 20oS.This finding suggests that part of the SACW flows toward theequator along the coast at lower latitudes, while the other por-tion is carried southward by the BC and recirculates within thesubtropical gyre, as corroborated by other studies (Reid, 1989).At intermediate levels, between 800 and 1300 m of depth, theAAIW (3o&θ&6oC; 34.2&S&34.6, (Sverdrup et al., 1942)) alsorecirculates within the gyre, reaching the Brazilian coast and bifur-cating somewhere between 25oS and 28oS (Muller et al., 1998;Boebel et al., 1997, 1999). Direct measurements, geostrophic cal-culations and subsurface floats, corroborate the idea that, uponreaching the coast, part of the AAIW flows southward with the BCat latitudes higher than 28oS, while another part flows northwardat latitudes lower than 25oS, forming an intermediate boundarycurrent (IWBC) (Muller et al., 1998; Boebel et al., 1999; Schmid& Garzoli, 2009). Numerical and observational studies (Silveiraet al., ; Mano et al., 2009) have shown that the baro-clinic instability and the associated mesoscale activity of the BCis strongly correlated with this flow at intermediate levels. Be-low these levels, near 2000 m of depth, a relatively intense andwell organized flow carries the North Atlantic Deep Water (NADW)(3&θ&4oC and 34.6&S&35, (Miranda, 1985)) southward asa deep WBC, this flow continues along the entire length of theBrazilian coast (Reid, 1989; Stramma & England, 1999; Hogg &Thurnherr, 2005).Based on a literature review (Silveira et al., 2000; Strammaet al., 1990; Cirano et al., 2006), estimates of BC transport rangefrom nearly 6 Sv at 15oS to approximately 16 Sv at 28oSwheresignificant variability and many uncertainties are present in theseestimates. Values as low as 1-2 Sv have been reported for inter-mediate latitudes, indicating that the rate of growth is not con-stant with latitude. Stramma & England (1999) calculated thegeostrophic transport of CB at 20oS to be approximately 1.6 Sv,while Souza (2000) presented transport values of ~2 Sv mea-sured at 25oS. However, these calculations were performed byconsidering only the flows at the TW and SACW levels in con-trast, one would expect more intense transport growth for higherlatitudes if the southward flows at the AAIW and NADW also areconsidered.While flow direction reversal at different levels creates a vari-able vertical shear that impacts the current instability, it also indi-cates the presence of a comple x flow pattern in the upper an di nter-mediate water masses. Along the continental slope and within theWBC system, a single water mass can move either southward ornorthward depending on the latitude under consideration. There-fore, understanding the flow bifurcations in different regions ofthe water column, as well as the associated water mass transportbehaviors is relevant for understanding the contribution of theSouth Atlantic Ocean to the upper limb of the Meridional Over-turning Cell (MOC) (Garzoli et al., 2013), which has further im-plications for the Earth’s present climate and its associated low-frequency variability (Ganachaud, 2003; Talley, 2003).The objective of the present work is tocontribute to the knowl-edge of the WBC system in the South Atlantic. In particular, thelatitude of bifurcation and the flow transports within distinct watermasses are determined and analyzed.Because previous observations of the South Atlantic WBCsystem are sparse, this study is carried out within the framework ofnumerical ocean modeling, and the analysis is performed basedon the results obtained from high-resolution global simulationsusing two distinct ocean global circulation models – OGCMs:the Ocean Circulation and Climate Advanced Modeling Project(OCCAM), and the Hybrid Coordinate Ocean Model (HYCOM).Revista Brasileira de Geof??sica, Vol. 32(2), 2014
ii“main” —
— 12:06 — page 243 — #3iiiiiiPEREIRA J, GABIOUX M, MARTA-ALMEIDAM, CIRANO M, PAIVA AM & AGUIAR AL 243We expect that our results, which are based on two independentmodeling runs (using different factors of vertical discretization,boundary forcing, and others), add confidence to the values pre-sented in the literature regarding flow bifurcations at the Braziliancoast.The study region covers part of the western South and Equa-torial Atlantic Ocean, ranging from 35oS to the Equator in latitudeand from 52oWto25oW in longitude. This domain is illustratedin Figure 1, which presents a map of the western South AtlanticOcean from Smith & Sandwell (1997) in terms of the global seafloor topography. This database is used in the OCCAM simula-tions.Figure 1 – The bottom topographyin the western South Atlantic Ocean, as usedin the OCCAM simulations based on Smith & Sandwell (1997). The radial linesindicate the location of the zonal sections (latitudes of 5o,13o,22oand 30oS)analyzed later in the present study.To avoid ambiguities regarding the water mass transports thatare inherent to calculations made for different ranges of waterdepth, such as those present in (Peterson & Stramma, 1991), thetransport and bifurcations points are computed for distinct densityranges which better represent the various water masses present inthe South Atlantic Ocean. Therefore, it is important to note, thatthe two models have different vertical discretizations: while theOCCAM is executed in a geopotential or z-level model, the HY-COM is primarily an isopycnal model within the density range ofinterest, or more precisely, the HYCOM has hybrid coordinates,which include isopycnal, geopotential and terrain-following coor-dinates. It is interesting to observe how these two distinct models“view” the ocean in different ways, i.e. along fixed vertical levelsor along moving density layers, yet both also perceive the samephysical phenomenon.The paper is organized as follows. In Section 2, the modelcharacteristics and configurations used in the different simula-tions are presented. A discussion of the flow computation per-mormed for different density ranges is presented in Section 3.In Section 4, the results regarding the depth range of each wa-ter mass, the locations of the bifurcations, the meridional velocitywithin the WBC and the transports in each water mass are ana-lyzed and discussed. Section 5 presents the conclusions of thisstudy.METHODOLOGYNumerical Model SimulationsIn this study, the results from two high-resolution (1/12o)eddy-resolving global numerical simulations are analyzed, whereone simulation is executed in z-coordinates (the OCCAM isavailable at http://www.noc.soton.ac.uk/JRD/OCCAM/) and theother, with hybrid-coordinates (the HYCOM is available athttp://www.hycom.org/dataserver/glb-simulation).The OCCAM is a fixed-grid z-coordinate ocean circulationmodel based on the Bryan-Cox-Semtner general ocean circula-tion model (Bryan, 1969; Semtner, 1974; Cox, 1984). The OC-CAM is configured with 66 levels in the vertical direction. Thesimulation was initialized with potential temperature and salinityinterpolated fields based on the WOCE SAC climatology (Gouret-ski & Janke, 1996). The surface forcing input data for the periodfrom 1985 to 2003 were supplied by NCAR and are describedin Large et al. (1997). The zonal and meridional wind compo-nents, air temperature and specific humidity were obtained fromthe NCEP reanalysis (Kalnay et al., 1996). The model topogra-phy was derived from a composite global bathymetric datasetconstructed from a uniform-gridded version of the Smith &Sandwell (1997). A detailed description of the OCCAM and itsconfiguration can be found in Coward & Cuevas (2005). Themonthly mean temperature and the salinity and velocity fieldsderived from the model results for the period from 1985 to 2004are used in the present analysis.The HYCOM is the hybrid-coordinate ocean circulation model(fixed-depth zor pressure p-coordinates, isopycnic ρ-densitytracking coordinates, and terrain-following σcoordinates). Thevertical coordinate adjustment was designed so that the isopy-cnal vertical coordinates present in the ocean interior allow asmoth transition to z-coordinates in the near-surface, well-mixedregions to sigma (terrain-following) coordinates in shallow wa-ter regions and than back to z-coordinates in very shallow wa-ters, which prevents the layers from becoming too thin (Blecket al., 2002). The HYCOM was configured with 32σ2layersBrazilian Jou rnal of Geophysics, Vol. 32(2),2014
ii“main” —
— 12:06 — page 244 — #4iiiiii244 THE BIFURCATION OF THE WESTERN BOUNDARY CURRENT SYSTEM OF THE SOUTH ATLANTIC OCEAN(Table 1). The simulation was initialized with the January clima-tology produced by the Generalized Digital Environment Modelversion 3.0 (GDEM3) (Carnes, 2009), which was developed bythe Naval Oceanography Office (NAVOCEANO). The surface forc-ing was obtained from the Navy Operational Global AtmosphericPrediction System (NOGAPS) (HYCOM, 2007) and includes thewind and heat flux. The topography used in the model was derivedfrom a quality controlled NRL DBDB2 bathymetry dataset (HY-COM, 2007). The model was integrated from 01/2003 to 04/2007.The monthly mean temperature and the salinity and velocity fieldsused in the present analysis were derived from model snapshotstaken every three days from 2003 to 2005.Table 1 – Vertical discretization in the layers of σ2used in the HYCOM for theWestern South Atlantic. Column 3 presents the water masses with a better repre-sentation by layer (TW – Tropical W SACW – South Atlantic Central WatAAIW – Antarctic Intermediate W NADW – North Atlantic Deep Water).Layer σ2Water Mass128.10228.90329.70 layers used for a better430.50 discretization of the mixed layer530.95631.50732.05832.60933.15 TW10 33.7011 34.2512 34.75 SACW13 35.1514 35.5015 35.8016 36.0417 36.20 AAIW18 36.3819 36.5220 36.6221 36.7022 36.7723 36.8324 36.8925 36.97 NADW26 37.0227 37.0628 37.1029 37.1730 37.3031 37.42 ——32 37.48To characterize the southwestern South Atlantic current sys-tem and to investigate SEC bifurcation, volume transports werecomputed within the density ranges corresponding to the mainwater masses in this region, for both the OCCAM and HYCOM.Several authors, such as Mamayev (1975), Reid (1989), Zemba(1991), Tomczak (1981), Memery et al. (2000), Ganachaud (2003)and You (2006), discuss different methods for characterizing thecore and the vertical limits of the water masses. In the presentwork, these limits are based on the thermohaline indexes, inaddition to the corresponding potential density surfaces pre-sented in the literature. For OCCAM, σθ=25.70 (Mamayev,1975; Stramma & England, 1999) separates the TW from SACW,σθ=26.80 (Mamayev, 1975; Schott et al., 2005; Rodrigueset al., 2007) separates the SACW from AAIW, and σθ=27.53(Stramma & England, 1999; Memery et al., 2000; Rodrigueset al., 2007) separates the AAIW from NADW. For the HYCOM,each water mass is represented by a group of σ2isopycnal lay-ers, as listed in Table 1. The volume transports were calculated byintegrating the velocities within the defined density ranges for theOCCAM and within the model layers for the HYCOM.RESULTS AND DISCUSSIONIn the present study, it is important to note the correct definitionof the density limits that define each water mass. These limits aredetermined from the thermohaline indexes of each water mass, aspresented in the methodology. Consequently, it is essential to es-timate how the vertical water mass structure is represented in theinformation under analysis. Thus, we estimated the mean fields oftemperature and salinity for each simulation, where these valueswere compared with the climatology of the WOA05 (World OceanAtlas 2005) (Locarnini et al., 2006; Antonov et al., 2006). It isworth noting that the present analyses are qualitative and that thesimulations were validated in previous studies using the OCCAM(Lee & Coward, 2003; Marsh et al., 2005; Lee et al., 2007) andthe HYCOM (Gabioux, 2008; Krelling, 2010).The thermohaline structure of the southwest South Atlantic iscorrectly represented in both cases (the OCCAM and HYCOM),and the main features of the TS diagrams compare well with theWOA05 climatology (Fig. 2). In both simulations, the TW is char-acterized by high values of potential temperature (θ&20oC)and salinity (S&36), and the SACW by a typical linear θ-S rela-tionship (Sverdrup et al., 1942) between the thermohaline limits(6&θ&20oC and 34.6&S&36.4). Both simulations representthe WOA05 data dispersion for central and intermediate waters(3&θ&6oC and 34.2&S&36.4).Revista Brasileira de Geof??sica, Vol. 32(2), 2014
ii“main” —
— 12:06 — page 245 — #5iiiiiiPEREIRA J, GABIOUX M, MARTA-ALMEIDAM, CIRANO M, PAIVA AM & AGUIAR AL 245Figure 2 – Scatter diagrams of the annual means of the potential temperature(θ) and salinity for the OCCAM 1/12o(blue dot), HYCOM 1/12o(red dot) andWOA05 1o(green dot) at the longitude of 30oW and within the latitude regionfrom 1.5oSto35oS.The vertical distribution of the water masses is also repre-sented in the two simulations. Figure 3 illustrates a meridionalsection of the temperature and salinity along 30oW (between40oS and Equator) based on the OCCAM, HYCOM and WOA05.The locations in terms of depth of the water mass cores are prop-erly simulated. Near the surface, at locations close to tropical lat-itudes, the TW reaches temperature values of 28oC and a salin-ity value of 37.5. At mid-depth, the SACW is represented by thetemperature range of 6&θ&20oC, with salinity values between34.6 and 36, as found in the WOA05 dataset (Fig. 3a,d). Thesame features are presented in both high-resolution simulations(Fig. 3b,e,c,f). Beneath the SACW, the AAIW for WOA05 extendsto ~1600 m of depth at 40oS and grows shallower toward thenorth, reaching a depth of ~1000 m at the Equator (Fig. 3d).Below 1500 m of depth, the high salinity water of the NADW isfound with a temperature below 4oC and salinities higher than34.6. With respect to the intermediate and deep waters, this pat-tern is also represented by the OCCAM and HYCOM simulations.Figure 3 also shows the isopycnals σθand σ2, which definethe water mass limits for the OCCAM (as well as WOA05) andHYCOM, respectively. It is interesting to note that despite severaldifferences, the water mass limits computed with σθfor the OC-CAM and σ2for the HYCOM are generally in agreement over thedomain. For all water masses, the difference between the base de-fined by the depth of a surface isopycnal σθor the depth of a layerσ2is less than ~15%.Another approach used to assess the representativeness ofthe thermohaline fields simulated by the OCCAM and HYCOMis calculating the spatial distribution of the annual mean isopyc-nal layer depths, based on the limits presented in the respectivemethodology (Figs. 4, 5 and 6). The TW base (Fig. 4a-c) lies be-tween 60 and 200 m of depth. The TW base reaches deeper val-ues near 18oS for the OCCAM, namely, at approximately 170 m(Fig. 4b), where this value also is close to that of the WOA05dataset (Fig. 4a). A similar pattern is observed for the HYCOM,where the maximum depths (~165 m) are located north of 28oS(Fig. 4c). In the southern part of the BC region, between 25oSand 35oS, the TW base reaches depths of approximately 120 min the HYCOM. The same feature is observed in the OCCAM until~30oS (Fig. 4b).In the case of the SACW/AAIW interface, both models agreewith the climatology (Fig. 5). The SACW base shows a re-gion of maximum depth between 25oS and 30oSandare-gion of minimum depth between 5oS and 10oS. The maximumdepths simulated in the OCCAM (480 m, Fig. 5b) and HYCOM(~500 m, Fig. 5c) are consistent with the observed values inWOA05 (Fig. 5a). In the northern part of the domain (beyond15oS), the depths reach values below 400 m. For the HYCOMthe minimum depth is ~280 m, and for the OCCAM is ~240 m.In intermediate waters, the main characteristics of the spatialpattern of the AAIW base, as observed in the climatology (Fig. 6a),are represented in the OCCAM (Fig. 6b). For both cases, thedepth of the isopycnal σθ=27.53 varies from
m,increasing toward higher latitudes. North of 24oS, the depth ofthe isopycnal presents at shallower values, namely, 1200 m,in contrast south of 30oS, it presents very deep levels of ap-proximately 1500 m. However, in the HYCOM, the AAIW base(Fig. 6c) presents a pattern of maximum depth at both the northand south boundaries (1300 m and 1400 m, respectively) anda minimum depth in the center of the domain at approximately20oS(~1200 m). It is important to note that the differences be-tween the WOA05 and HYCOM AAIW base depths are less than20% of the total thickness of the AAIW ~1000 m of depth, that is,differences as great as 200 m of depth appear only in a restrictedregion in the northern part of the domain.Once verified, the numerical results with respect to thethermohaline field define the limits between the densities of thewater masses and are used to characterize the southwesternSouth Atlantic current system, as well as to investigate the SECbifurcation. Accordingly, the annual, seasonal and monthly meanvolume transport streamfunctions are computed for the defineddensity ranges (Figs. 7 to 10). In this part of the analysis, in ad-dition to the numerical results, other data-bases are also used,such as the SODA reanalysis from 1985 to 2004. The SODA oceanreanalysis was carried out based on the Parallel Ocean ProgramBrazilian Jou rnal of Geophysics, Vol. 32(2),2014
ii“main” —
— 12:06 — page 246 — #6iiiiii246 THE BIFURCATION OF THE WESTERN BOUNDARY CURRENT SYSTEM OF THE SOUTH ATLANTIC OCEANFigure 3 – The meridional section at 30oW of the annual mean temperature (upper panels) and salinity (lower panels) for theWOA05 dataset are given in (a) and (d),for the OCCAM 1/12oin (b) and (e), and for the HYCOM 1/12oin (c) and (f). The black solid lines indicate the isopycnal levels of σθ25.7, 26.8 and 27.53,and thedashed lines indicate the isopycnal layer depths of σ233.7, 35.5 and 36.62. The vertical axis for z≤1600 m is expanded for visualization.Figure 4 – The mean depth of the TW/SACW interface is represented by the isopycnal σθ25.7 for the WOA05 (a) and OCCAM 1/12o(b) and by σ233.7 for theHYCOM 1/12o(c). The bold line contour represents a depth of 100 m.POP-1.4 model, which was 40 levels in the vertical direction anda0.4×0.25 degree displaced pole grid (Carton et al., 2000a,b).Flow bifurcations are defined by the position of the zerostreamfunction near the coast. For the TW, SACW and AAIW, thepresence of an approximately zonal flow indicates that the southequatorial current corresponds to the northern branch of the sub-tropical gyre (Stramma & England, 1999).At the TW level, the SEC bifurcation is located at ~15oSinthe HYCOM and SODA results. In contrast, the bifurcation ap-pears near 13oS in the OCCAM, based on the annual means.Revista Brasileira de Geof??sica, Vol. 32(2), 2014
ii“main” —
— 12:06 — page 247 — #7iiiiiiPEREIRA J, GABIOUX M, MARTA-ALMEIDAM, CIRANO M, PAIVA AM & AGUIAR AL 247Figure 5 – The same as Figure 4 but for the SACW/AAIW interface (σθ=26.8 and σ2=35.5). The bold line contour represents a depth of 400 m.Figure 6 – The same as Figure 4 but for the AAIW/NADW interface (σθ=27.53 and σ2=36.62). The bold line contour represents a depth of 1200 m.Figure 7 – The mean volume transport of the TW, between the surface and the isopycnal σθ≤25.7 for SODA (a), OCCAM 1/12o(b) and between the surface andthe isopycnal σ2≤33.7 for HYCOM 1/12o(c). The unit of transport is Sv. The contour interval is 1 Sv, with labels at every 2 Sv.Brazilian Jou rnal of Geophysics, Vol. 32(2),2014
ii“main” —
— 12:06 — page 248 — #8iiiiii248 THE BIFURCATION OF THE WESTERN BOUNDARY CURRENT SYSTEM OF THE SOUTH ATLANTIC OCEANThe seasonal variability of the SEC bifurcation for the TW levelis summarized in Table 2 as monthly means for the OCCAM andHYCOM. In the OCCAM simulation, the SEC bifurcation reachesits northernmost position at 12oS in December and its southern-most position at 16oS in August. However, for the HYCOM, thenorthernmost position occurs in February and reaches 12.5oS,while the southernmost position occurs at 17oS during the aus-tral winter, which is in July. These results are consistent with thefindings of Rodrigues et al. (2007) who used a reduced-gravityprimitive equation for the OGCM and CTD data and found thatthe SEC bifurcation latitude reaches its northernmost position inNovember (~13oS in the uppermost 200 m) and its southernmostposition in July (~17oS in the uppermost 200 m). The authors’explanation for the seasonal variability of the bifurcation latitudein the upper thermocline is related primarily to variations in windforcing based on the combined effect of local Ekman pumping andremotely forced Rossby waves.South of the bifurcation latitude, a part of the SEC creates asouthward flowing limb, which gives rise to the BC. At 25oS, theTW transport carried by the BC is 2 Sv for the OCCAM and SODA(Fig. 7b, a) and is 4 Sv for the HYCOM (Fig. 7c). North of the bi-furcation latitude, the northern branch of the SEC feeds the NBC.Silveira et al. (1994) estimated the geostrophic transport of theNBC to be 6.5 Sv (the section located at 5oS between 34o30’W–32o00’W from the surface to 100 m of depth). At the same lat-itude, the TW transport is 9 Sv for the OCCAM simulation, 6 Svfor the HYCOM and 10 Sv for SODA.Below the TW, at the SACW level, the SEC bifurcation isshifted southward and is located at approximately 22oS for bothhigh-resolution simulations (Fig. 8b and c) and at 23oS for SODA(Fig. 8a). These values are closer to the SEC bifurcation latitudecalculated by Stramma & England (1999) (~20oS, from approx-imately 100 m to 500 m) and Rodrigues et al. (2007) (~21oSat400 m). The southernmost position of the SEC bifurcation occursin December (21oS for the OCCAM) and in January (21oSfortheHYCOM).At the SEC bifurcation, a part of the SACW flows northwardwith the North Brazil Undercurrent (NBUC), while another partflows southward with the BC. At latitudes north of 15oS, the NBUCmerges with the surface flow, forming the NBC system, which ap-pears as an intensified northwestward flow as it moves beyond theEquator. At 5oS, the SACW transport carried by the NBUC is es-timated as 10 Sv for the OCCAM simulation and SODA (Fig. 8band a) and as 9 Sv for the HYCOM (Fig. 8c).In intermediate waters, the SEC bifurcation is located at ap-proximately 30oS for the OCCAM simulation (Fig. 9b) and at28oS for SODA and the HYCOM (Fig. 9a,c). For the mean flowfield of the AAIW, Stramma & England (1999), Boebel et al. (1999)and Schmid & Garzoli (2009) represent the northern limb of thesubtropical gyre reaching the Brazilian continent at approximately28oS. Rodrigues et al. (2007) found that the annual mean SEC bi-furcation latitude, which occurs at 900 m, is approximately 26oS.In the OCCAM simulation, the SEC bifurcation presents a sea-sonal shift from 35oS in June and 28oS in December. In theHYCOM, the seasonal variability exhibits its southernmost po-sition in July at 32.5oS, while the northernmost position appearsin November at 25oS (Table 2). The seasonal variability of theSEC bifurcation simulated by Rodrigues et al. (2007), at 900 m,presents the southernmost position of the bifurcation at 27oSinJuly, while in October, the northern position occurs at 26oS. Atthis level (the northernmost position of the bifurcation), the AAIWis carried northward by the IWBC. The northward flow was es-timated from direct observations for April 1983 along 22-23oSby Silveira et al. (2004) to be 3.6 Sv. The values for the OC-CAM and HYCOM are 4 Sv near the same region and 6 Sv forSODA. Schmid & Garzoli (2009) estimated a mean transport ofthe IWBC between 28oSand6oS of 2.8 Sv with maximum valuesof ~10 Sv near 20oS. The simulation results from the OCCAM,HYCOM and SODA analyses present transport values near thisregion that are approximately ~8Sv.The latitudinal variation of the SEC bifurcation with depthis consistent with the values found in the literature. Wienderset al. (2000) estimated the SEC bifurcation from hydrographicdata obtained in January-March 1994 and found that the bifurca-tion ranges downward from approximately 14oS at the surface to28oS at a depth of 600 m . Based on annual mean CT D observationdata, Rodrigues et al. (2007) found that the bifurcation occurs at21oS at a depth of 400 m, representing a southward displacementof 7 degrees from its surface value (14oS).A relatively intense and well organized flow carries the NADWsouthward along the Brazilian continental margin as a Deep West-ern Boundary Current (DWBC), and this pattern is well repre-sented in all simulations, as well as in SODA (Fig. 10). This flowpresents an eastward turning between 20oS and 22.5oSthatisrelated to the presence of the Vit?oria-Trindade Ridge, which is abathymetric obstacle that resides perpendicular to the continentalslope at 20oS-21oS (Fig. 1). The eastward turning was also ob-served by Memery et al. (2000) at the A17 WOCE line. The NADWtransport is estimated at 11oSas21SvinSODAand18SvintheOCCAM and HYCOM simulations and at 19oSas18SvinSODA,15 Sv in the OCCAM and 12 Sv in the HYCOM. These values un-derestimate the NADW transport described in Ganachaud (2003).Revista Brasileira de Geof??sica, Vol. 32(2), 2014
ii“main” —
— 12:06 — page 249 — #9iiiiiiPEREIRA J, GABIOUX M, MARTA-ALMEIDAM, CIRANO M, PAIVA AM & AGUIAR AL 249Table 2 – The seasonal variability of the SEC bifurcation for the OCCAM 1/12oand the HYCOM 1/12o.OCCAM 1/12oHYCOM 1/12oMonth TW SACW AAIW TW SACW AAIWJanuary 13oS21.5oS30oS13oS21oS27.5oSFebruary 13oS21.5oS30oS12.5oS22oS27.5oSMarch 14oS22oS31oS14.5oS22.5oS30oSApril 14oS23oS32oS15oS22.5oS27oSMay 15oS24oS32oS15oS22oS27.5oSJune 15.5oS25oS35oS15.5oS23oS30oSJuly 16oS24oS34oS17oS23oS32.5oSAugust 16oS23oS32oS15.5oS22oS26oSSeptember 14.5oS22oS32oS15oS21.5oS26oSOctober 13oS22oS31oS15oS22oS25.5oSNovember 12.5oS22oS29oS14.5oS22oS25oSDecember 12oS21oS28oS14.5oS22oS26oSFigure 8 – The same as Figure 7 but for the SACW, between the isopycnal intervals of 25.7 &σθ≤26.8 for (a), (b); and 33.7 &σ2≤35.5 for (c).Figure 9 – The same as Figure 7 but forthe AAIW, between the isopycnal intervals of 26.8 &σθ≤27.53 for (a), (b); and 35.5 &σ2≤36.62 for (c).Brazilian Jou rnal of Geophysics, Vol. 32(2),2014
ii“main” — 2014/12/26 — 12:06 — page 250 — #10iiiiii250 THE BIFURCATION OF THE WESTERN BOUNDARY CURRENT SYSTEM OF THE SOUTH ATLANTIC OCEANFigure 10 – The same as Figure 7 but for the NADW, between the isopycnal intervals of σθ&27.53 and the bottom for (a), (b) and between the isopycnals σ2&36.62and the bottom (c). The contour interval is 3 Sv.Figure 11 – The mean meridional velocity (m s-1) at the latitude of 5oS for SODA (a), OCCAM 1/12o(b), and HYCOM 1/12o(c). The white contour representsthe zero velocity, while values higher than 0.6 are shown as dotted contours. The black dashed lines indicate the isopycnal levels of σθ25.7, 26.8 and 27.53 in (a), (b)and σ233.7, 35.5 and 36.62 in (c). The vertical axis for z≤1000 m is expanded for better visualization. The location of the section is indicated in Figure 1. Positivevalues for velocity and transport indicate a northward flow.According to the author, the NADW transport values calculatedduring the WOCE period () for sections at 11oS(A8)and 19oS (A9) are 23 Sv at both lines.To complement the analysis of the southwestern Atlantic re-gion, we examine the annual mean vertical velocity structure ofthe WBC in conjunction with the volume transport for each watermass. This computation is performed for zonal sections at key lo-cations along the western boundary. The location of each sectionis shown in Figure 1, and the transports for annual and seasonalare summarized in Tables 3, 4 and 5 for the OCCAM, HYCOM andSODA analyses, respectively.At 5oS, in all simulations, although less apparent for SODA,the WBC system presents a bimodal structure with a northwardflow in the first 1300 m of depth, identified as the NBUC, and asouthward flow underneath, known as the DWBC (Fig. 11). TheNBUC shows a significant core of annual mean meridional veloci-tiesaslargeas0.6ms-1fr om ~50 m to 400 m of dep th, reachingmaximum values of 1.12 ±0.07 m s-1,0.82±0.15 m s-1and0.93 ±0.10 m s-1for the OCCAM (Fig. 11b), SODA (Fig. 11a)and HYCOM (Fig. 11c) analyses, respectively. These values areslightly higher than the velocity of 0.8 m s-1calculated by Schottet al. (2005). Below the NBUC, between 1200 and 3300 m ofdepth, the DWBC presents maximum meridional velocities of 0.30±0.05 m s-1for the OCCAM and 0.20 ±0.04 m s-1for theHYCOM. These values also are in agreement with the velocity of0.2 m s-1calculated by Schott et al. (2005).At this section and in the uppermost 1000 m of depth, theNBUC transport is associated with the TW, SACW and AAIW(Fig. 11). The NBUC mean annual transport calculated as a sum-mation of the TW, SACW and AAIW transports is 30.2 Sv for theOCCAM (Table 3), 22.4 Sv for the HYCOM (Table 4) and 33.8 Svfor SODA (Table 5). At the same latitude, Schott et al. (2005)Revista Brasileira de Geof??sica, Vol. 32(2), 2014
ii“main” — 2014/12/26 — 12:06 — page 251 — #11iiiiiiPEREIRA J, GABIOUX M, MARTA-ALMEIDAM, CIRANO M, PAIVA AM & AGUIAR AL 251estimated a mean transport of 26.5 ±3.7 Sv above the σ1=32.15 isopycnal (~1000 m depth). Below the AAIW layer, theNADW transport values calculated for the OCCAM, HYCOM andSODA are 17.8 ±4.1 Sv, 16.1 ±3.5 Sv and 18.0 ±3.5 Sv, re-spectively. Schott et al. (2005) found a mean southward DWBCtransport of 25.5 ±8.3 Sv for nine measured sections during, ranging between σ1=32.15 and σ4=45.90 westof 33.5oW. In relation to the seasonal transport variation of theNBUC, the maximum values are observed in winter for all of theanalyzed results (Tables 3, 4 and 5). In this case, the DWBC trans-port maximum arises during summer.At 13oS (Fig. 12), the main circulation feature is the IWBC,which is located in the upper 1300 m of depth. Legeais et al.(2013) present a IWBC mean velocity at 15oSof~0.08 m s-1that increases to 0.2 m s-1at 10oS with a maximum valueof 0.7 m s-1. In both high-resolution simulations, the IWBCcore is confined between 200-800 m of depth (Fig. 12b, c). Themaximum velocities are 0.42 ±0.10 m s-1for the OCCAM,0.44 ±0.09 m s-1for the HYCOM and 0.29 ±0.08 m s-1for SODA. In this section, we found the origin of the BC near thesurface (~the first 50 m of depth) for in the OCCAM simula-tion, although this was not observed for either the HYCOM orSODA analysis. This difference is a result of the bifurcations inthe HYCOM and SODA analyses south of 13oS (at approxi-mately 15oS – Fig. 7a, c).Below the IWBC appears the DWBC, which transports theNADW. This boundary current presents maximum velocities of0.19 ±0.05 m s-1for the OCCAM, 0.10±0.04 m s-1for theHYCOM and 0.07 ±0.02 m s-1for SODA (Fig. 12).In terms of transport, the IWBC appears as a boundary cur-rent constituted mainly of the SACW and the AAIW, transportingnorthward flows of ~20 Sv, ~23 Sv and ~24 Sv in the OCCAM,HYCOM and SODA analyses, respectively. In this same section,the southern flow of the NADW is estimated to be ~26 Sv forthe OCCAM, ~31 Sv for the HYCOM and ~24 Sv for SODA.Here, the maximum values of the SACW, AAIW and NADW trans-ports are observed in winter based on the OCCAM, HYCOM andSODA analyses (Tables 3, 4 and 5). At 11oS, Schott et al. (2005)observed a northward NBUC maximum in July and minimum inOctober-November. For NADW transport at the same latitude, theauthors observed maximum values in November and minimumvalues in July.At 22oS, the TW and a fraction of the SACW are transportedsouthward by the BC, which resides in the upper levels (Fig. 13).West of 39.3oW, the BC appears well developed in all simula-tions and SODA (Fig. 13a-c), extending through the upper 300 mof depth. The maximum annual meridional velocities found forthe BC are 0.55 m ±0.14 s-1,0.54±0.15 m s-1and 0.43±0.11 m s-1for the OCCAM, HYCOM and SODA analyses,respectively. Near this latitude, in a transect at 22.75oS, Oliveiraet al. (2009) found the BC velocity to be 0.39 ±0.23 m s-1usingsurface drifter data.Below 300 m of depth, part of the SACW is carried north-ward. This occurs due to shifting of the SEC bifurcation withdepth, which is located approximately at this level at 22oS(Fig. 8). The northward SACW and AAIW flows depict a IWBCwith a defined core that is confined between 400 and 1000 m,the associated velocities are 0.26 ±0.09 m s-1for the OCCAM,0.54 ±0.15 m s-1for the HYCOM and 0.43 ±0.11 m s-1forSODA. At this latitude, for all simulations, the DWBC corepresents maximum velocities as great as 0.1 m s-1and is locatedat 38.5oW between the depths of 2000 m and 2500 m. The DWBCmaximum velocity for the OCCAM is 0.17 ±0.04, for the HYCOMis 0.12 ±0.04 m s-1and for SODA is 0.05 ±0.02 m s-1.Based on observations that the mean transport value of theBC is 8.6 ±4.1 Sv at 24oS and 19.4 ±4.3 Sv at 35oS, Garzoliet al. (2013) found that the BC increases toward the south. Froma velocity cross-section developed during the Transport of theBrazil Current Experiment (TRANSCOBRA, ), carriedout at 22o–23oS, Silveira et al. (2004) estimated a BC transportof 5.6 ±1.4 Sv and a IWBC transport of 3.6 ±0.8 Sv. To comparethese values, we computed the BC and the IWBC transport coresin the section at 22oS, located between land and the longitude of39.3oW,and we obtained BC transports of 2.99 Sv and 3.07 Sv forthe OCCAM and HYCOM, respectively. For the IWBC, we foundtransports of 5.70 Sv and 9.20 Sv for the OCCAM and HYCOM(not shown). Below the IWBC, the NADW transports were com-puted as ~12 Sv, 11 Sv and 20 Sv for the OCCAM, HYCOM andSODA analyses, respectively. In this section, the seasonal cyclesfor the BC, IWBC and DWBC are less clear than in the previoussections.Finally, in the southern part of our domain, at 30oS, the BCgrows deeper (reaching approximately 400-500 m of depth in allsimulations and SODA, as shown in Fig. 14) because the bifur-cation of the intermediate flow occurs near this latitude (Fig. 9).The BC maximum velocities for the OCCAM, HYCOM and SODAanalyses are 0.34 ±0.07 m s-1,0.33±0.15 m s-1and 0.23 ±0.08 m s-1, respectively. Unlike the BC, the IWBC in this sectionis slightly less intense for SODA than for the other simulations.In the section at 30oS, the seasonal cycles of the BC, IWBC andDWBC transport also are less evident.Brazilian Jou rnal of Geophysics, Vol. 32(2),2014
ii“main” — 2014/12/26 — 12:06 — page 252 — #12iiiiii252 THE BIFURCATION OF THE WESTERN BOUNDARY CURRENT SYSTEM OF THE SOUTH ATLANTIC OCEANTable 3 – The transport (Sv) for each water mass of the OCCAM 1/12oat the analyzed sections for the annual mean, summer and winter.OCCAM 1/12oSections Mean TW SACW AAIW NADWAnnual 8.9 ±1.7 –0.1 ±0.3 11.0 ±1.0 –0.5 ±0.5 10.3 ±2.7 –3.8 ±2.6 3.4 ±1.6 –17.8 ±4.15oSSummer 8.2 ±1.6 –0.0 ±0.1 10.4 ±0.9 –0.4 ±0.4 9.5 ±2.0 –5.1 ±2.8 2.3 ±1.1 –20.6 ±2.9Winter 8.8 ±1.2 –0.3 ±0.4 11.5 ±0.9 –0.5 ±0.5 11.0 ±2.8 –2.1 ±1.5 3.8 ±1.6 –14.2 ±3.6Annual 3.0 ±1.4 –1.3 ±1.0 7.6 ±1.6 –0.3 ±0.4 12.3 ±2.6 –4.1 ±2.7 9.8 ±7.0 –26.5 ±7.613oSSummer 2.2 ±1.0 –2.2 ±1.2 6.4 ±1.1 –0.4 ±0.6 11.0 ±2.6 –4.1 ±3.0 7.7 ±6.1 –23.2 ±5.9Winter 4.1 ±1.1 –0.5 ±0.5 8.9 ±1.3 –0.2 ±0.3 13.7 ±2.0 –3.8 ±2.6 11.3 ±8.6 –28.4 ±9.3Annual 1.1 ±0.8 –3.0 ±0.8 3.3 ±1.3 –2.2 ±1.1 10.9 ±3.0 –3.5 ±2.9 2.9 ±2.0 –12.3 ±1.522oSSummer 1.2 ±0.8 –3.2 ±0.7 3.3 ±1.3 –2.4 ±1.2 10.9 ±3.3 –3.5 ±3.4 2.9 ±1.8 –11.1 ±2.6Winter 0.8 ±0.6 –2.5 ±0.8 2.9 ±1.1 –1.8 ±0.9 10.2 ±2.9 –3.5 ±2.5 2.2 ±2.0 –12.4 ±2.1Annual 0.8 ±0.8 –2.9 ±1.2 1.8 ±1.3 –5.5 ±1.5 3.2 ±2.1 –4.8 ±2.2 0.6 ±0.5 –10.9 ±2.730oSSummer 0.8 ±0.9 –3.5 ±1.1 1.8 ±1.4 –6.1 ±1.4 2.7 ±2.2 –5.0 ±2.4 0.6 ±0.6 –10.5 ±2.6Winter 0.9 ±0.9 –2.5 ±1.3 1.7 ±1.1 –4.9 ±1.4 3.7 ±2.1 –4.9 ±1.9 0.5 ±0.4 –11.5 ±2.4Table 4 – The transport (Sv) for each water mass of the HYCOM 1/12oat the analyzed sections for the annual mean, summer and winter.HYCOM 1/12oSections Mean TW SACW AAIW NADWAnnual 5.9 ±0.9 –0.8 ±0.5 9.4 ±0.4 –0.7 ±0.2 7.1 ±1.6 –3.9 ±2.1 2.8 ±0.8 –16.1 ±3.55oSSummer 4.7 ±1.1 –0.8 ±0.7 9.0 ±0.6 –0.7 ±0.1 6.4 ±1.4 –6.2 ±1.4 2.1 ±0.5 –20.3 ±1.6Winter 5.7 ±0.2 –1.3 ±0.4 9.6 ±0.6 –0.9 ±0.1 7.9 ±1.2 –2.1 ±1.0 3.7 ±0.8 –13.1 ±1.8Annual 4.0 ±1.2 –2.1 ±0.8 8.7 ±0.8 –0.7 ±0.4 14.1 ±1.2 –6.1 ±2.7 11.2 ±3.4 –31.2 ±4.813oSSummer 2.3 ±0.2 –2.1 ±0.8 7.4 ±0.4 –0.3 ±0.1 12.5 ±0.7 –4.6 ±2.0 8.7 ±2.6 –29.8 ±3.4Winter 5.1 ±0.6 –2.5 ±1.2 9.8 ±0.1 –0.7 ±0.3 14.7 ±0.6 –5.0 ±1.9 11.3 ±5.6 –35.5 ±1.1Annual 1.1 ±0.4 –3.4 ±0.5 2.6 ±0.7 –2.4 ±0.5 9.1 ±1.4 –3.1 ±1.0 4.2 ±1.1 –11.0 ±1.222oSSummer 0.9 ±0.1 –3.7 ±0.5 1.9 ±0.1 –2.4 ±0.5 7.7 ±1.6 –3.6 ±0.5 3.2 ±0.9 –10.5 ±0.9Winter 1.4 ±0.3 –3.0 ±0.3 3.7 ±0.6 –2.2 ±0.4 10.6 ±0.5 –2.9 ±0.8 4.0 ±0.7 –11.5 ±0.9Annual 2.4 ±0.8 –6.8 ±1.1 4.2 ±1.5 –10.3 ±1.2 4.8 ±1.4 –7.5 ±1.1 2.6 ±0.7 –12.0 ±2.130oSSummer 1.7 ±0.7 –8.2 ±0.8 2.8 ±0.9 –11.1 ±0.8 3.1 ±0.4 –8.8 ±0.7 2.3 ±0.6 –11.1 ±1.3Winter 3.2 ±0.7 –6.1 ±0.2 4.6 ±1.4 –9.8 ±0.6 6.4 ±0.9 –7.4 ±0.6 3.5 ±0.3 –14.5 ±1.1Table 5 – The transport (Sv) for each water mass of SODA at the analyzed sections for the annual mean, summer and winter.SODASections Mean TW SACW AAIW NADWAnnual 9.9 ±1.5 0.0 ±0.5 10.9 ±0.9 –0.6 ±0.3 13.0 ±2.5 –7.1 ±2.8 0.0 ±0.7 –18.0 ±3.55oSSummer 8.8 ±1.1 0.0 ±0.3 10.1 ±0.6 –0.6 ±0.2 14.2 ±3.1 –7.7 ±2.6 0.1 ±0.5 –18.9 ±2.8Winter 11.0 ±1.3 0.0 ±0.4 11.1 ±0.8 –0.9 ±0.3 14.7 ±2.4 –6.8 ±2.4 0.1 ±0.7 –16.5 ±3.1Annual 2.0 ±1.4 –1.3 ±0.8 9.2 ±1.6 –0.6 ±0.4 15.0 ±2.5 –3.5 ±2.2 4.5 ±4.0 –24.1 ±6.213oSSummer 0.7 ±1.3 –1.3 ±0.5 8.1 ±1.3 –0.6 ±0.2 15.2 ±2.4 –2.9 ±1.8 4.1 ±3.5 –20.2 ±5.8Winter 4.2 ±0.9 –2.8 ±0.7 10.8 ±1.4 –1.1 ±0.4 16.9 ±2.2 –4.7 ±2.0 5.0 ±3.7 –26.4 ±5.7Annual 1.0 ±0.5 –3.1 ±0.8 2.5 ±1.1 –1.1 ±0.6 8.5 ±2.3 –0.4 ±1.3 0.8 ±1.0 –20.1 ±3.322oSSummer 0.9 ±0.5 –3.6 ±0.5 2.1 ±0.9 –1.6 ±0.4 8.4 ±2.1 –0.1 ±1.2 0.6 ±0.8 –17.5 ±3.0Winter 1.4 ±0.4 –2.9 ±0.6 2.8 ±0.8 –0.7 ±0.3 8.7 ±2.0 –0.5 ±0.9 0.9 ±0.9 –21.2 ±3.1Annual 0.5 ±0.6 –3.5 ±1.1 1.8 ±1.0 –7.8 ±1.6 2.1 ±1.2 –6.5 ±2.6 1.1 ±0.6 –17.0 ±3.630oSSummer 0.5±0.4 –4.8 ±1.0 2.6 ±1.0 –8.3 ±1.5 4.0 ±0.9 –7.2 ±2.1 1.9 ±0.4 –15.5 ±2.4Winter 0.5 ±0.5 –0.9 ±1.1 1.8 ±0.9 –4.2 ±0.2 1.5 ±1.0 –4.8 ±1.2 0.7 ±0.5 –15.6 ±3.5Revista Brasileira de Geof??sica, Vol. 32(2), 2014
ii“main” — 2014/12/26 — 12:06 — page 253 — #13iiiiiiPEREIRA J, GABIOUX M, MARTA-ALMEIDAM, CIRANO M, PAIVA AM & AGUIAR AL 253Figure 12 – The same as Figure 11, but at the latitude of 13oS.Figure 13 – The same as Figure 11, but at the latitude of 22oS.Figure 14 – The same as Figure 11, but at the latitude of 30oS.Brazilian Jou rnal of Geophysics, Vol. 32(2),2014
ii“main” — 2014/12/26 — 12:06 — page 254 — #14iiiiii254 THE BIFURCATION OF THE WESTERN BOUNDARY CURRENT SYSTEM OF THE SOUTH ATLANTIC OCEANIn this section, the southward NADW transport simulated bythe OCCAM is 10.9 ±2.7 Sv and gives values of 12.0 ±2.1 Svfor the HYCOM and 17.0 ±3.6 Sv for SODA (Tables 3, 4 and5). In all simulations and for SODA, the NADW transport valuesunderestimate the value of 23 ±3 Sv calculated by Ganachaud(2003), which aggregates the return of the Antarctic Bottom Waterand the AAIW as they progress southward.CONCLUSIONSThe results of two eddy resolving (1/12o) OGCMs were used toinvestigate the flow bifurcations of the Western Boundary Currentsystem (WBC) and the water masses of the Western South AtlanticOcean. Particular attention was given to the latitude of bifurcationof the currents feeding into the WBC system, with a focus on theflow associated with the different water masses. Considering thatobservations of the South Atlantic WBC system are sparse, in ad-dition to the fact that previous knowledge regarding the WBC bi-furcation is based on limited data observations and coarse numer-ical simulations, these two independent simulations with realistichigh-resolution OGCMs offer an interesting source of additionalinformation.The numerical models used were the OCCAM in z-coor-dinates and the HYCOM in hybrid-coordinates. Modeling com-parisons have long been performed, for example, the DYNAMO(Willebrand et al., 2001) and the DAM?EE-NAB (Chassignet et al.,2000) projects, with a predominant focus on model performance.The DYNAMO project, for example, investigated the role of differ-ent vertical model discretizations using the GFDL-MOM (level),MICOM (isopycnal) and SPEM (sigma). The project contributedto our understanding of the circulation in the