inct AmbTropic
Ambientes Marinhos Tropicais: Heterogeneidade Espaço-Temporal e Respostas à Mudanças Climáticas
Apresentação
As mudanças climáticas deverão afetar as características físicas, biológicas e biogeoquímicas das zonas costeiras e oceanos, modificando sua estrutura ecológica, suas funções e os diferentes serviços prestados ao Homem. Estas mudanças tem o potencial de causar sérios impactos sócio-econômicos nas escalas local (center parcs), regional (plataforma e mares rasos) e global (oceano).
As respostas dos ambientes marinhos às mudanças climáticas irão também depender da variabilidade natural destes sistemas e de outras mudanças introduzidas pelo homem como resultado dos diferentes usos dos recursos marinhos.
O bem-estar das comunidades humanas depende intrinsecamente da disponibilidade dos serviços que os ecossistemas costeiros e marinhos provêem. Isto é particularmente importante para a região norte e nordeste, que apresenta em alguns dos seus municípios costeiros, algumas das densidades populacionais mais elevadas do Brasil.
O INCT AmbTropic tem como objetivo central unificador a avaliação de como a heterogeneidade espaço-temporal dos ambientes marinhos tropicais poderá determinar os padrões de resposta destes ambientes e sua resiliência às mudanças climáticas que afetarão o norte-nordeste do Brasil neste século. Este aspecto é de grande importância estratégica para a região.
Outros objetivos deste INCT incluem:
(i) Gerar uma sólida base conceitual sobre os processos, a dinâmica e o funcionamento da zona costeira, plataforma e oceano tropicais do Brasil e sua variabilidade espaço-temporal.
(ii) Dotar a região norte/nordeste do Brasil de uma infraestrutura de pesquisa moderna e adequada para o avanço e consolidação das atividades em Ciências do Mar nesta região.
(iii) Expandir e consolidar a incipiente estrutura de ensino, treinamento e formação de pessoal, para fazer frente ao conjunto de desafios atuais e futuros a serem enfrentados por esta região principalmente como conseqüência das mudanças climáticas.
(iv) Criar uma rede de excelência para dar suporte na solução de problemas prementes que afetam os ambientes marinhos da região norte e nordeste.
(v) Criar mecanismos que possibilitem a transferência ativa de conhecimentos para os principais atores sociais de maneira descentralizada além de garantir acesso irrestrito a todos os dados e informações pretéritas e a serem gerados, pelo INCT.
O INCT AmbTropic é uma iniciativa conjuntas das Universidades Federais da Bahia e de Pernambuco e engloba aproximadamente 200 cientistas distribuídos em mais de 20 instituições de ensino e pesquisa que compartilham esta visão. Os recursos para sua implantação são oriundos do CNPq, FAPESB e CAPES.
O INCT AmbTropic é coordenado pelos professores José Maria Landim Dominguez (UFBA) e Sigrid Neumann Leitão (UFPE)
A sede do INCT AmbTropic está hospedada no Instituto de Geociências da UFBA.
Grandes Linhas
Como já foi apontado pelo IPCC as mudanças globais no clima deverão afetar as características físicas, biológicas e biogeoquímicas das zonas costeiras e oceanos, modificando sua estrutura ecológica, suas funções e os diferentes serviços prestados ao Homem. Estas mudanças tem o potencial de causar sérios impactos sócio-econômicos nas escalas local (zona costeira), regional (plataforma e mares rasos) e global (oceano). Neste contexto, mudanças nas escala local e regional podem ser até mesmo mais relevantes para os Estados costeiros quando comparadas àquelas em escala global.
A região norte e nordeste do Brasil, devido às suas características, apresenta uma oportunidade única para se avaliar de que maneira a heterogeneidade espacial e temporal dos ambientes marinhos tropicais influencia os padrões de resposta destes ambientes e sua resiliência às mudanças climáticas que afetarão a região neste século. Nesta região, compreendida entre os estados do Espírito Santo e Amapá, encontram-se as principais construções recifais do oceano Atlântico Sul Ocidental, os principais deltas brasileiros, uma das áreas mais extensas de manguezais do mundo, uma plataforma continental que varia da mais estreita a mais larga do Brasil, as principais ilhas e montes submarinos, variações extremas nos fluxos de sedimentos e nutrientes, além da sua importância inegável na transferência de calor e massa interhemisférica. Heterogeneidade é assim a característica singular e ao mesmo tempo integradora do INCT AmbTropic. Três escalas espaciais de abordagem estão contempladas:
Zona Costeira (Local) – é uma área de grande heterogeneidade física e biológica e a interface de interação entre as forçantes naturais e antropogênicas.
Plataforma Continental (Regional) – é uma área também de grande heterogeneidade, pouco compreendida e cada vez mais intensamente utilizada pelo Homem. É também uma área que oferece um contexto para interpretação de mudanças e tendências observadas na escala local.
Oceano (Global) – é um componente integral do sistema Terra influenciado por transporte de massa e por suas interações com a atmosfera.
Cada uma destas escalas espaciais engloba uma série de Grupos de Trabalho, que abordam aspectos específicos dentro de cada uma destas escalas
Zona Costeira (Local)
Plataforma Continental (Regional)
Oceano (Global)
Grupos de Trabalho
Cada uma das escalas espaciais de investigação do INCT AmbTropic inclui vários Grupos de Trabalho (GTs). Estes GTs investigarão aspectos específicos dentro de cada uma destas escalas espaciais. A escolha dos temas dos Grupos de Trabalho e suas áreas específicas de investigação foram balizadas pelos seguintes principios:
(i) contemplar a heterogeneidade espacial e temporal nas suas três escalas principais de investigação, privilegiando gradientes fisico-quimicos e dinâmicos (ondas, marés, CO2, circulação etc) e ambientes marinhos intrinsecamente associados à região tropical do Brasil, tais como recifes de corais, manguezais, recursos vivos, ilhas oceânicas, processos oceânicos etc.,
(ii) construir sobre a experiência dos grupos de pesquisa locais,
(iii) concentrar os esforços dos GTs em areas geográficas comuns, não só para minimizar custos, como também estimular a sinergia entre os GTs, e
(iv) avaliação de recursos naturais (vivos e não vivos) específicos da região.
Coordenadores
- José Maria Landim Dominguez
- Sigrid Neumann Leitão
- Eduardo Siegle
- José Maria Barbosa Filho
- Zelinda Margarida de Andrade Nery Leã
- oMário Luiz Gomes Soares
- Marcelo Cancela Lisboa Cohen
- Carlos Augusto França Schettini
- Gilvan Yogui
- Vanessa Hatje
- Helenice Vital
- Alessandro Luvizon
- Fabio Hazin
- Ruy Kenji P. de KikuchiMoacyr Araújo
- Nathalie Lefèvre
- Alex Bastos
- Pedro Pereira
- Dóris Veleda
- George Emmannuel C. de Miranda
- Ralf Schwamborn
- Rodrigo A. Torres
- Mônica Lúcia Adams
- Thierry Frédou
- Marcus André Silva
- Illa Fair
Organograma
Publicações
Contribuição 1 - Veleda, D., M. Araujo, R. Zantopp, and R. Montagne (2012), Intraseasonal variability of the North Brazil Undercurrent forced by remote winds, J. Geophys. Res., 117, C11024, doi:10.1029/2012JC008392.
Intraseasonal variability of the North Brazil Undercurrent forced by remote winds
D. Veleda,1,2 M. Araujo,1,2 R. Zantopp,3 and R. Montagne4
Received 27 July 2012; revised 9 October 2012; accepted 10 October 2012; published 21 November 2012.
[1] Intraseasonal signals with periods of 2 to 3 weeks in near-surface alongshore current
measurements are detected from four moorings (K1–K4) deployed from 2000 to 2004
at the 11 S section close to the Brazilian coast as part of the German CLIVAR Tropical
Atlantic Variability Project. This section crosses the path of the North Brazil Undercurrent,
the most powerful western boundary current in the South Atlantic Ocean. We investigate the
origin of this intraseasonal variability of the North Brazil Undercurrent by relating
the oceanic oscillation of the alongshore currents to its atmospheric counterpart, the
meridional wind stress. On average, the results indicate a well-defined lagged (10 days)
correlation (0.6) structure between meridional wind stress and alongshore currents.
The oceanic region with the highest cross-correlations is identified as a relatively narrow
band along the Brazilian coast, from 22 –36 S and 40 –50 W, bounded in the north by an
eastward change in coastline orientation. The cross-wavelet transform establishes the
common power between the time series of meridional wind stress and alongshore currents,
predominantly during austral winter and spring. These signals propagate equatorward with
an alongshore speed of 285 63 km day1
, consistent with Coastal Trapped Wave theory.
Citation: Veleda, D., M. Araujo, R. Zantopp, and R. Montagne (2012), Intraseasonal variability of the North Brazil Undercurrent
forced by remote winds, J. Geophys. Res., 117, C11024, doi:10.1029/2012JC008392.
- Introduction
[2] Over the past few decades, the western boundary currents of the tropical South Atlantic and their associated scales
of variability have been investigated by numerous researchers [Molinari, 1983; Stramma, 1991; Schott and Böning,
1991; Mayer and Weisberg, 1993; Schott et al., 1993;
Rhein et al., 1995; Stramma et al., 1995; Dengler et al., 2004;
Schott et al., 2005]. While the Deep Western Boundary
Current (DWBC) transports cold North Atlantic Deep Water
far into the southern hemisphere, the North Brazil Undercurrent (NBUC) and North Brazil Current (NBC) serve as a
northward warm-water conduit as part of the thermohaline
overturning cell [Gordon, 1986; Schmitz, 1995]. The South
Atlantic Central Water (SACW), located at depths between
100 and 500 m, is transported westward within the southern
band of the South Equatorial Current (sSEC) until it reaches
the Brazilian shelf. After the sSEC bifurcates, its southward
limb becomes the Brazil Current (BC) and merges into the
South Atlantic subtropical gyre system. The northward limb
feeds into the NBUC [Stramma and Schott, 1999; Stramma
et al., 2005] and - as a western boundary current - carries
warm waters of South Atlantic origin across the equator and
into the northern hemisphere. It also supplies the eastward
flow of the South Equatorial Countercurrent (SECC) which
partially recirculates into the central band of the SEC (cSEC).
In addition, in the 50–300 m depth range, the NBUC seems to
play an important role in the Atlantic Equatorial Gyre [Schott
et al., 2005] and in the coupled ocean-atmosphere system.
This strong western boundary current has a nearshore core
position approximately 50 km from the Brazilian coast,
reaching down to about 900 m in depth, with a maximum
speed of about 65 cm s1 at 180 to 250 m on average. At 11 S,
the mean flow structure of the NBUC is already well developed, indicating that the bifurcation of the sSEC is located well
south of this section with a maximum northward NBUC flow
in July and a minimum in the October–November period
[Schott et al., 2005]. The sSEC bifurcation has a southernmost
position in July and a northernmost position in November
[Rodrigues et al., 2007; Silva et al., 2009].
[3] While the dominant fluctuation in the NBUC core,
approx. 50 km from the coast, has a period of about two
months [Schott et al., 2005], biweekly signals are the dominant modes along the western (onshore) flank of the NBUC,
about 10 km from the coast [von Schuckmann, 2006].
[4] The cause of these intraseasonal fluctuations in the
upper-ocean circulation may be found in several dynamic
processes, such as local wind-forcing, remote wind-forcing via
waveguide dynamics, mean flow instability, and resonance
1
Laboratório de Oceanografia Física Estuarina e Costeira, Departamento
de Oceanografia, Universidade Federal de Pernambuco, Recife, Brazil. 2
CEERMA-Centro de Estudos e Ensaios em Risco e Modelagem
Ambiental, Universidade Federal de Pernambuco. 3
GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany.
4
Departamento de Física, Universidade Federal Rural de Pernambuco,
Recife, Brazil.
Corresponding author: D. Veleda, Laboratório de Oceanografia Física
Estuarina e Costeira, Departamento de Oceanografia, Universidade Federal
de Pernambuco, Av. Arquitetura s/n, 50740-550, Cidade Universitária,
Recife, PE, Brazil. (doris.veleda@ufpe.br)
©2012. American Geophysical Union. All Rights Reserved.
0148-0227/12/2012JC008392
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 117, C11024, doi:10.1029/2012JC008392, 2012
C11024 1 of 10
due to the coastline geometry [von Schuckmann, 2006]. Furthermore, if these fluctuations are in reasonable proximity of
the coast, the motion of oceanic waters over the continental
shelf and slope may also be influenced by the earth’s rotation,
density stratification, the offshore current regime, sloping
bottom topography and the presence of the coastline [Allen,
1980], causing coastal-trapped waves (hereafter CTWs). The
existence of CTWs depends entirely on the presence of a
shallow shelf between the coast and the deep ocean. CTWs
propagate along the continental slopes and shelves of the
world’s coastlines, with the coast to the right in the Northern
Hemisphere and to the left in the Southern Hemisphere. They
have periods of days to 2–3 weeks and wavelengths on the
order of 2000 km, determined by the atmospheric weather
patterns. The impact of CTWs on shelf currents and sea level
varies with seasonal changes in stratification over the coast.
For arbitrary topography and stratification, there is a free-wave
mode (similar to the barotropic Kelvin wave) plus an infinite
set of higher mode, more slowly propagating waves [Brink,
1991]. These perturbations induce variations in sea level and
alongshore currents over the continental shelf for typical periods ranging from the inertial period to about 20 days. Winds
and tides in coastal regions at sub-inertial frequencies force
these waves with periods ranging from days to weeks. Since
the continental shelf is a transition zone between the coastal
waters and the open ocean, the CTWs transport materials (e.g.,
nutrients and pollutants) from regions near the coastline to and
from the open ocean, and along the continental margins. Thus,
CTWs modulate the upwelling system and primary productivity in the coastal regions.
[5] This paper investigates the relationship between
observed intraseasonal fluctuations in alongshore current
measurements near the Brazilian shelf and the large-scale
wind stress field over the South Atlantic through the use of
cross-correlations and continuous wavelet transforms.
- Data
2.1. Moored Array Observations at 11 S
[6] The ocean data set used here was obtained between
March 2000 and August 2004 as part of the German CLIVAR Tropical Atlantic Variability Project [Schott et al.,
2005]. The data were collected in an array of five moorings
(K1–K5) stretching 225 km across the NBUC (10 S–11.5 S)
(Figure 1). Data sources, positions and sampling periods are
presented in Table 1. Since the offshore mooring K5 was
deployed for only two years, and observations were confined
to depths below 1500 m. this mooring was not used in the
present analysis. Additional details of this data set are given
by Schott et al. [2005].
Figure 1. Schematic representation of the Atlantic Subtropical Cell (STC) circulation, including subduction (blue) and upwelling (green) zones. Current branches involved in the STC flows are NEC, SEC,
sSEC, NECC and EUC; NEUC, SEUC = North and South Equatorial Undercurrent; NBC, NBUC = North
Brazil Current and Undercurrent; GD, AD = Guinea and Angola domes. Interior equatorward thermocline
pathways dotted. Adapted from Schott et al. [2004]. The mooring array at 11 S is shown by the red line.
Table 1. Location and Sampling Periods for Moorings and Wind
Stress Data
Data Source Latitude/Longitude Sampling Period
Moorings, German CLIVAR Project
Station K1 10 16.0′S
35 51.7′W
02/2002–01/2003
06/2003–08/2004
Station K2 10 22.8′S
35 40.8′W
02/2002–05/2004
Station K3 10 36.7′S
35 23.4′W
05/2003–08/2004
Station K4 10 56.5′S
34 59.5′W
01/2002–08/2004
Wind stress data, NOAA/NCDC 40 S–0 N
60 W–20 E
01/2002–12/2004
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
2 of 10
2.2. Wind Stress Data
[7] The wind stress data set used in this study was obtained
from NOAA/NCDC (National Oceanic and Atmospheric
Administration/National Climatic Data Center) and is available
at www.ncdc.noaa.gov/oa/rsad/seawinds.html. The domain
used here covers the area 40 S to 0 N and 60 W to 20 W
(Figure 2). In this database, surface wind stresses (N m2
) are
estimated from blended sea surface wind speeds at 10 m above
sea level, generated from six satellites, on a global 0.25 regular
grid and for several time resolutions [Zhang et al., 2006]. Wind
speeds are converted to (u, v) components using the NCEP
Reanalysis 2 (NRA-2). In this work, the wind stress data has a
temporal resolution of 12 h.
[8] The mean zonal and meridional components of the
wind stress are shown in Figures 2a–2d for austral summer
(DJF) and winter (JJA). These charts capture the main spatial
structure of the observed means, in agreement with the published literature [Trenberth et al., 1990; Harrison, 1989;
Castelão and Barth, 2006]. During austral summer, the zonal
wind is negative (easterlies) over nearly the entire domain,
while positive (northward) meridional wind stress is
restricted to equatorial latitudes in the western part of the
South Atlantic Ocean and negative (southward) for almost
the entire western South Atlantic (Figures 2a and 2b). The
zonal components of the wind stress are strongest in June,
July and August. Northward meridional wind stress is
observed in the South Atlantic during JJA, expanding equatorward and reaching 20 S in the western basin (Figures 2c
and 2d). Negative (southward) meridional wind stress is
found between 20 and 30 S, and positive (northward) to the
south of it.
- Results
3.1. Current Structure at 11 S
[9] The current meter time series were smoothed using a
40-h lowpass filter to eliminate tidal and inertial effects.
Current vectors were divided into alongshore and cross-shore
components by rotating clockwise by 36 , i.e., parallel to the
coastline. Figure 3 shows the mean section of alongshore
velocities at 11 S for the entire deployment length (2000–
2004); instrument locations for moorings K1–K4 are marked
by black dots. The strongest contribution to the NBUC
transports occurs near 250 m depth at moorings K1 and K2
[Schott et al., 2002; Stramma et al., 2003; Schott et al.,
2005], indicated by the strong subsurface core in the upper
left corner of Figure 3. Below the NBUC, we find the colder
water masses of the DWBC, with the core centered at 1500–
3500 m depth along mooring K3 (dark blue colors). Mooring
K4 is located outside the equatorward boundary current, with
a weak mean southward flow between 500 and 3500 m depth.
[10] Fluctuations about these means are shown as anomalies of the 100 m current vectors (i.e., demeaned), as at this
level the measurements are available for all moorings K1–K4
(Figure 4). The dominant intraseasonal signals have periodicities of 2–3 weeks and 2–3 months (see also Figure 5).
The highest amplitudes at periods of 2–3 weeks are found in
the coastal boundary region at moorings K1 and K2, gradually decreasing away from the NBUC core. Figure 5 shows
Figure 2. (a, c) Mean zonal and (b, d) mean meridional components of wind stress for austral summer
(DJF) and winter (JJA). The zero contours are represented by dashed lines.
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
3 of 10
Figure 4. Vector time series of 40-h low-pass filtered alongshore current anomalies at moorings K1–K4
at 100 m depth, from March 2000 to August 2004.
Figure 3. Mean of alongshore velocity at 11 S, derived from moorings K1–K4. Instrument locations are
marked by black dots. Note the equatorward NBUC above 1000 m and the southward flowing DWBC
between 1500 and 3500 m (adapted from von Schuckmann [2006]).
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
4 of 10
the spectral distribution of currents between 50 and 300 m
depth for moorings K1 to K4. High-frequency fluctuations
(10–30 days) are more prominent near the coast, at mooring
K1. Farther offshore, at moorings K2 to K4, these fluctuations have the highest energy in the upper layers and scale
down with increasing distance from the NBUC current core.
Note that the variance levels at mooring K2 are nearly identical between 150 and 300 m.
3.2. Cross-Correlation Between Wind Stress
and Currents
[11] We used cross-correlation analysis between the
meridional wind stress and alongshore currents to spatially
localize the remote forcing across the entire ocean basin.
Previous work indicates that the meridional component of
wind stress is the prime generator of such coastal-trapped
waves [Brink, 1991; Csanady, 1997], and the periods
involved correspond to those of the weather systems. The
meridional wind stress components are cross-correlated with
alongshore currents at 100 m depth at 11 S for moorings
K1–K4 (Figure 6) as this depth level is available for all
moorings (see Figure 5).
[12] For mooring K1, two separate intervals are shown,
caused by a gap in the available current measurements. Since
we are interested in the intraseasonal components of the signal
in the 10–30 day band, we applied a 10–30 day band-pass
filter to both the atmospheric and oceanic data. The correlation
of the meridional wind stress for each grid point with the K1
alongshore current at 100 m depth is mapped for the entire
ocean basin (Figure 6a), and the lag between the two signals is
shown in the maps of the right panels. Identical procedures
were used for the other moorings, displayed in Figures 6b–6e.
Clearly, the lag between the wind stress of a given grid point
and the 11 S currents decreases with decreasing distance to
the moorings, showing a systematic propagation along the
shore. For the first interval of available data at mooring K1,
the maximum cross-correlation between meridional wind
and alongshore current is found between 22 and 36 S, which
is approximately 1500–2000 km from the 11 S moorings
(Figure 6a). The corresponding lag is 6.5 days. For the
2003–2004 period of mooring K1 (Figure 6b), there are
Figure 5. Variance preserving spectra of kinetic energy [cm2 s
2
] (solid line) for moorings K1–K4, at 50
to 300 m depth, from March 2000 to August 2004, with the respective 95% confidence limits.
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
5 of 10
weaker cross-correlations, with alongshore currents lagging
the wind stress by 10 days. Subsequently, we find a lag of
9 days at mooring K2 (Figure 6c), a lag of 7 days at
mooring K3 (Figure 6d), and a lag of 9.5 days for mooring
K4. Here is also where we find the highest cross-correlation
values between meridional wind stress and the alongshore
current at about 36 S (Figure 6e).
[13] Table 2 summarizes the cross-correlation results of
Figure 6 between the meridional wind stress and alongshore
current at 11 S. The coordinates of cross-correlation maxima, the distances to the mooring sites, and the lag period
clearly indicate a strong connection between the alongshore
current signal at 11 S and its remote atmospheric forcing
located at about 22 S to 36 S (box area in Figure 6).
[14] To further investigate this connection, we spatially
averaged the meridional wind stress components across the
area defined by the highest cross-correlations shown in
Figure 6. The resulting time series were cross-correlated
with the corresponding time series of meridional currents at
11 S (moorings K1–K4) at different depths, 50 to 1000 m
(Figure 7). The highest correlations are found in the nearsurface layers of the NBUC core, gradually weakening
downward and outward toward the open ocean. The right
panel shows the corresponding lags, where currents lag the
wind-forcing by 8 to 10 days (Figure 7, right).
3.3. Time-Frequency Common Power Between Wind
Stress and Currents
[15] Most traditional mathematical methods to examine
periodicities in frequency space, such as the Fourier analysis, implicitly assume that the underlying processes are stationary in time. However, wavelet transforms can be used
to analyze a non-stationary time series at many different
frequencies and scales by expanding the time series into
time-frequency space in order to find localized intermittent
periodicities. The wavelet transform is based on a practical
guide to wavelet analysis of Torrence and Compo [1998].
One of the problems with the resulting “global wavelet
spectra” is a distortion of the energy distribution [Liu et al.,
2007; Veleda et al., 2012], with high-frequency peaks
shown lower than their low-frequency counterparts. In this
work, we construct rectified Cross Wavelet Transform
(XWT) by normalizing the wavelet transform by the square
root of the scale, based on Veleda et al. [2012].
[16] An XWT analysis was applied to meridional wind
stress and alongshore currents at 100 m depth (moorings
K1–K4). The XWT identifies the regions in time-frequency
space with a large common power between two time series.
The individual cross-wavelet spectra, from 2002 to 2004, are
averaged to form a seasonal ensemble cross-wavelet spectrum, for austral winter (June to November) and austral
summer (December to May), respectively. In this case, we
used the wind stress data restricted to the region where the
highest cross-correlations between the wind stress and
alongshore currents were found (see Figure 6).
Figure 6. (left) The cross-correlation between meridional
wind stress and alongshore currents at 100 m depth for
(a) K1 mooring, 3/2002–1/2003 and (b) 6/2003–8/2004,
© K2 mooring, 3/2002–5/2004, (d) K3 mooring, 6/2003–
8/2004 and (e) K4 mooring, 3/2002–5/2004. (right) The
corresponding lags (currents lagging winds) for the crosscorrelation. Box in cross-correlation maps marks the area
of maximum cross-correlation.
Table 2. Coordinates of Maximum Cross-Correlation Between Meridional Wind Stress and Alongshore Currents, Corresponding
Distances and Lag Periodsa
Mooring Maximum Cross-Correlation Position Distance (km) Lag Period (days)
K1_I 0.55 (0.076) 25 S–44 W 1550 6.5
K1_II 0.44 (0.067) 34 S–47 W 2550 10
K2 0.52(0.049) 28 S–45 W 1890 8.5
K3 0.64 (0.065) 33 S–48 W 2440 7
K4 0.39 (0.046) 36 S–50 W 2780 9.5
a
The 95% confidence limits are shown in parentheses.
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
6 of 10
[17] The seasonal ensemble averages of the XWT between
the meridional wind stress averaged over the high crosscorrelation area [22 –36 S] and the corresponding alongshore currents within the matching time interval are shown
in Figure 8, with the blocked out “white” margins indicating
the “Cone of Influence” (COI), where edge artifacts become
important [Grinsted et al., 2004]. The color scale for
Figure 8 shows maxima in dark red and minima in dark blue.
[18] Figure 8 clearly indicates that the XWT between currents and wind stress is strongest for the 16-day period during
the month of September, linking this periodicity band of
the NBUC, at 11 S for mooring K1 (winter) at 100 m depth,
to the wind stress forcing at 22 –36 S close to the Brazilian
coast. The ensemble average XWT for the winter season
shows a stronger or more prominent response of the 16-day
period of the alongshore currents to the wind stress.
[19] An identical analysis was done for moorings K2, K3
and K4 (Figure 8). The result shows that during austral
winter, the 16-day periodicity of the 100 m currents at
moorings K1–K4, along the western and eastern flank of the
NBUC at 11 S, is linked to the wind-forcing at 22 –36 S
along the coast of Brazil. At mooring K1, the highest energy
spans the period from July through October, except for
August. For mooring K2, the highest energy is centered in
July and weaker in September and November. For moorings
K3 and K4, the highest energy extents from July through
October.
- Discussion
[20] Previous studies in other regions have confirmed the
presence of CTWs with typical periods of 5–20 days. Brink
[1983] associated these waves with remote wind-forcing,
while Battisti and Hickey [1984] determined that the pressure and alongshore velocity fields in the Northwest Pacific
may be a response of the wind-forced CTW. The variance in
sea surface pressure off Oregon and Washington was found
to be generated by wind forcing between San Francisco and
Cape Mendocino, California. The signal propagates 900 to
1300 km northward to Washington-Oregon, arriving there
3 to 4 days later. Brink [1982] also pointed out the agreement
between CTW theory and observations off Peru in 1977,
indicating free wave phase speeds of about 200 km day1
,
as well as sea level and alongshore velocity fluctuations in
the 5–10 day period band. Spillane et al. [1987] found
oscillations with intraseasonal periods of 36–73 days close to
the coast of Peru, with poleward phase propagation of 150–
200 km day1
. Enfield [1987] established that the intraseasonal sea level variations pointed out by Spillane et al. [1987]
are forced in the western equatorial Pacific by atmospheric
oscillations, consistent with previous studies with propagation speeds of 216–259 km day1
.
[21] Smith [1978] found a persistent poleward propagation
of fluctuations in currents and sea level along the Peruvian
coast between 10 and 15 S, with a wave speed of about
200 km day1
. Enfield and Allen [1980] also estimated the
wave phase speed as the ratio of alongshore station separation to the corresponding lag. Along the Peruvian coast,
Camayo and Campos [2006] used wavelet analysis to study
the intraseasonal current oscillations, suggesting remotely
forced baroclinic Kelvin waves with periods between 10 and
20 days, with velocities of approx. 200 km day1
.
[22] In this work, we establish the propagation of Coastal
Trapped Waves along the Brazilian coast as a suitable
mechanism for explaining the strong correlation between the
meridional wind stress at latitudes of 22 S–36 S and the
Figure 7. Cross-correlation and lag (in days) between meridional wind stress averaged in the area along
the Brazilian coast [22 S–36 S]–[40 W–50 W] and alongshore currents at the 11 S section.
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
7 of 10
alongshore currents at 11 S. Kelvin waves are low-frequency
gravity waves which occur where the deflection caused by
the Coriolis force is either constrained by coastlines or goes
to zero at the equator. These waves have typical amplitudes
of several tens of meters in the thermocline region, and
lengths are thousands of kilometers.
[23] Spectral analysis of the current velocities identified
high-frequency components present near the surface and
gradually decreasing with depth. A fluctuation at 10–30 day
periodicity is observed in the four moorings at the jointly
available 100 m depth level. At 11 S, the thermocline is located
at about 65 m depth [Schott et al., 2005; von Schuckmann,
2006], so that the core of the NBUC is located just below
these strong density and temperature gradients, making this
flow regime potentially favorable for internal Kelvin wave
propagation.
[24] Internal coastal-trapped Kelvin waves may be generated by an abrupt change in the winds and depend on the
existence of a coast against which they can lean. Our work
showed maximum cross-correlations for the nearshore wind
data located between 22 S and 36 S, a region along the
Brazilian coast usually referred to in the literature as the
Southeast Brazil Bight (SBB) [Castro Filho, 1985; Stech and
Lorenzzetti, 1992; Campos et al., 1995; Cirano and Campos,
1996]. The northern boundary of this area corresponds to a
relatively narrow shelf close to Cabo Frio (22 S), where an
abrupt change in coastline orientation exists. Furthermore,
the SBB is also under the influence of synoptic and
mesoscale frontal low pressure systems which interact
with the western reaches of the South Atlantic High
(or “St. Helena High”). Strong meridional winds from the
northeast (NE) occur before the passage of a low pressure
system, then rapidly changing to southwesterly winds (SW).
These frontal systems have frequencies of 3 to 4 per
month [Rodrigues et al., 2004], with average speeds of
500 km day1
, thereby crossing the SBB region in about
2 days [Stech and Lorenzzetti, 1992]. As a result, these atmospheric anomalies induce drastic changes in wind direction
and cause significant disturbances in the ocean, such as mean
sea level changes, generation of surface waves and currents
[Castro Filho, 1985; Campos et al., 1995]. We argue that the
combination of local wind stress variability and the abrupt
change in coastline direction reveals the SBB as a preferred
locus for CTW generation along the Brazilian shore. The
distance between the currents at 11 S and the area where
high cross-correlations were found is about 1600–2700 km.
In addition, the ratio between these distances and their
corresponding lag periods yields an equatorward propagation
speed of 285 63 km day1 along the Brazilian shore.
Coastal trapped waves certainly have major effects on sea
level and currents locally, but their far-reaching consequences
may not yet be fully explored: As the waves approach the
equator, they continue eastward as equatorial Kelvin waves,
only to be dispersed again as coastal Kelvin waves along the
eastern boundary (off Africa) and also reflected westward as
equatorial Rossby waves. The conclusions regarding our
Figure 8. The seasonal ensemble average of the Cross-Wavelet Transform between the meridional wind
stress averaged in the area along the Brazilian coast [22 S–36 S]–[40 W–50 W] and the respective alongshore current.
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
8 of 10
observations may have significant effects on the entire equatorial waveguide system.
- Summary
[25] The western boundary regime of the tropical South
Atlantic is well known for its complex dynamics and multiple
scales of variability, with the North Brazil Undercurrent
(NBUC) playing an important role in the Atlantic Subtropical
Cell (STC) and in the coupled ocean-atmosphere system.
[26] Previous studies of these moorings (K1–K4) at 11 S
section have shown that the NBUC core, about 50 km offshore, has a dominant fluctuation with a periodicity of two
months [Schott et al., 2005; von Schuckmann, 2006]. In this
work we investigate the forcing mechanism behind additional
intraseasonal periodicities of the near-surface alongshore current measurements at 2 to 3 weeks. Cross-correlation analysis
between currents and meridional wind stress points to the wind
field located between 22 and 36 S close to Brazilian coast as
the driving mechanism.
[27] The distances between the wind-forcing area and the
mooring locations, as well as the corresponding lags, indicate
an equatorward propagation speed of 285 63 km day1
along the Brazilian coast. Such distances and travel times
suggest Coastal Trapped Waves (CTWs) as the prime candidates for accomplishing this task. Therefore, two of the
main ingredients for the existence of CTW are readily
available, namely wind stress variability and abrupt change
in coastline direction.
[28] We investigated the relationship between the NBUC,
wind stress and CTWs by using several statistical and mathematical tools. High-frequency variability is the dominant
mode along the western flank of the NBUC. Spectral analysis
of the current velocities for K1 (the mooring closest to the
coast) shows a spectral peak at 14 and 30 days at 100 m depth.
This peak decreases as we move farther offshore. Crosscorrelation analysis of currents and wind stress along the entire
Brazilian coast reveals the precise origin of these fluctuations.
In addition, the Cross-Correlation Wavelet Transform accurately identifies the main structures in the frequency-time
space where the atmospheric and oceanic time series both had
their largest common power.
[29] The correlations between meridional wind stress, from
22 to 36 S near the Brazilian coast, and alongshore currents
at 11 S are strongest during austral winter and spring. As
shown in Enfield and Allen [1980], these signals propagate
equatorward with the coast to the left (in the southern hemisphere) and have alongshore speeds consistent with wave
propagation processes. Furthermore, the area of origin for the
forcing of intraseasonal current signals is dominated by the
first baroclinic mode over the continental slope. It has also
been shown that this area has stronger density stratification
during austral winter than austral summer [Campos et al.,
2000], making it more favorable for the propagation of
internal Kelvin waves.
[30] Acknowledgments. The authors would like to thank the scientific team of the German Climate Variability and Predictability (CLIVAR)
program and, in particular, our colleagues from GEOMAR for providing
full access to the K1–K4 mooring data. We also thank the editor and the
two anonymous reviewers for their valuable suggestions and comments.
The first author thanks Peter Brandt and Lothar Stramma for their assistance
during her one-year stay at GEOMAR in Kiel. She also wishes to thank
CAPES/DAAD (Coordination for the Improvement of Higher Education
Staff/Deutscher Akademischer Austauschdienst) for scholarship support.
R.M. acknowledges financial support from the Pernambuco State Agency
FACEPE (APQ-0871-1.05/08).This work was carried out under the National
Institute on Science and Technology in Tropical Marine Environments -
INCT-AmbTropic (CNPq process 565054/2010-4), and under the project
BIO-NE (CNPq process 558143/2009-1.
References
Allen, J. S. (1980), Models of wind-driven currents on the continental
shelf, Annu. Rev. Fluid Mech., 12, 389–433, doi:10.1146/annurev.
fl.12.010180.002133.
Battisti, D. S., and B. M. Hickey (1984), Application of remote windforced coastal trapped wave theory to the Oregon and Washington coasts,
J. Phys. Oceanogr., 14, 887–903, doi:10.1175/1520-0485(1984)014<0887:
AORWFC>2.0.CO;2.
Brink, K. H. (1982), A comparison of long coastal trapped wave theory with
observations off Peru, J. Phys. Oceanogr., 12, 897–913, doi:10.1175/
1520-0485(1982)012<0897:ACOLCT>2.0.CO;2.
Brink, K. H. (1983), The near-surface dynamics of coastal upwelling, Prog.
Oceanogr., 12, 223–257, doi:10.1016/0079-6611(83)90009-5.
Brink, K. H. (1991), Coastal-trapped waves and wind-driven currents over
the continental shelf, Annu. Rev. Fluid Mech., 23, 389–412, doi:10.1146/
annurev.fl.23.010191.002133.
Camayo, R., and E. J. D. Campos (2006), Application of wavelet transform
in the study of coastal trapped waves off the west coast of South America,
Geophys. Res. Lett., 33, L22601, doi:10.1029/2006GL026395.
Campos, E. J. D., J. E. Gonçalves, and Y. Ikeda (1995), Water mass characteristics and geostrophic circulation in the South Brazil Bight: Summer
of 1991, J. Geophys. Res., 100(C9), 18,537–18,550, doi:10.1029/
95JC01724.
Campos, E. J. D., D. Velhote, and I. C. A. da Silveira (2000), Shelf break
upwelling driven by Brazil current cyclonic meanders, Geophys. Res.
Lett., 27, 751–754, doi:10.1029/1999GL010502.
Castelão, R. M., and J. A. Barth (2006), Upwelling around Cabo Frio, Brazil:
The importance of wind stress curl, Geophys. Res. Lett., 33, L03602,
doi:10.1029/2005GL025182.
Castro Filho, B. M. (1985), Subtidal response to wind forcing in the South
Brazil Bight during winter, PhD thesis, 211 pp., Univ. of Miami, Miami,
Fla.
Cirano, M., and E. J. D. Campos (1996), Numerical diagnostic of the circulation in the Santos Bight with COROAS hydrographic data, Braz.
J. Oceanogr., 44(2), 105–121.
Csanady, G. T. (1997), On the theories that underlie our understanding of
continental shelf circulation, J. Oceanogr., 53, 207–229.
Dengler, M., F. A. Schott, C. Eden, P. Brandt, J. Fischer, and R. Zantopp
(2004), Break-up of the Atlantic deep western boundary current into
eddies at 8 S, Nature, 432, 1018–1020, doi:10.1038/nature03134.
Enfield, D. B. (1987), The intraseasonal oscillation in eastern Pacific sea
levels: How is it forced?, J. Phys. Oceanogr., 17, 1860–1876,
doi:10.1175/1520-0485(1987)017<1860:TIOIEP>2.0.CO;2.
Enfield, D. B., and J. S. Allen (1980), On the structure and dynamics of
monthly mean sea level anomalies along the Pacific coast of North and
South America, J. Phys. Oceanogr., 10, 557–578, doi:10.1175/1520-
0485(1980)010<0557:OTSADO>2.0.CO;2.
Gordon, A. L. (1986), Interocean exchange of thermocline water, J. Geophys.
Res., 91, 5037–5046, doi:10.1029/JC091iC04p05037.
Grinsted, A., J. C. Moore, and S. Jevrejeva (2004), Application of the crosswavelet transform and wavelet coherence to geophysical time series,
Nonlinear Processes Geophys., 11, 561–566, doi:10.5194/npg-11-561-
Harrison, D. E. (1989), On climatological monthly mean wind stress and
wind stress curl fields over the world ocean, J. Clim., 2, 57–70,
doi:10.1175/1520-0442(1989)002<0057:OCMMWS>2.0.CO;2.
Liu, Y., X. S. Liang, and R. H. Weisberg (2007), Rectification of the bias in
the wavelet power spectrum, J. Atmos. Oceanic Technol., 24(12), 2093–
2102, doi:10.1175/2007JTECHO511.1.
Mayer, D. A., and R. H. Weisberg (1993), A description of COADS
surface meteorological fields and the implied Sverdrup transports for the
Atlantic Ocean from 30 S to 60 N, J. Phys. Oceanogr., 23, 2201–2221,
doi:10.1175/1520-0485(1993)023<2201:ADOCSM>2.0.CO;2.
Molinari, R. L. (1983), Observations of near-surface currents and temperature in the central and western tropical Atlantic Ocean, J. Geophys. Res.,
88, 4433–4438.
Rhein, M., L. Stramma, and U. Send (1995), The Atlantic Deep Western
Boundary Current: Water masses and transports near the equator,
J. Geophys. Res., 100, 2441–2451, doi:10.1029/94JC02355.
Rodrigues, M. L. G., D. Franco, and S. Sugahara (2004), Climatologia de
frentes frias no litoral de Santa Catarina, Braz. J. Geophys., 22(2), 135–151.
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
9 of 10
Rodrigues, R. R., L. M. Rothstein, and M. Wimbush (2007), Seasonal variability of the South Equatorial Current bifurcation in the Atlantic Ocean:
A numerical study, J. Phys. Oceanogr., 37, 16–30, doi:10.1175/
JPO2983.1.
Schmitz, W. J., Jr. (1995), On the interbasin-scale thermohaline circulation,
Rev. Geophys., 33, 151–173, doi:10.1029/95RG00879.
Schott, F., and C. W. Böning (1991), The WOCE model in the western
equatorial Atlantic: Upper-layer circulation, J. Geophys. Res., 96,
6993–7004, doi:10.1029/90JC02683.
Schott, F., J. Fischer, J. Reppin, and U. Send (1993), On mean and seasonal
currents and transports at the western boundary of the equatorial Atlantic,
J. Geophys. Res., 98, 14,353–14,368, doi:10.1029/93JC01287.
Schott, F., P. Brandt, M. Hamann, J. Fisher, and L. Stramma (2002),
On the boundary flow off Brazil at 5–10 S and its connection to the interior tropical Atlantic, Geophys. Res. Lett., 29(17), 1840, doi:10.1029/
2002GL014786.
Schott, F. A., J. P. McCreary, and G. C. Johnson (2004), Shallow overturning circulation of the tropical-subtropical oceans, in Earth’s Climate: The
Ocean Atmosphere Interaction, Geophys. Monogr. Ser., vol. 147, edited
by C. Wang, S.-P. Xie, and J. A. Carton, pp. 261–304, AGU, Washington,
D. C., doi:10.1029/147GM15
Schott, F., M. Dengler, R. Zantopp, L. Stramma, J. Fischer, and P. Brandt
(2005), The shallow and deep western boundary circulation of the South
Atlantic at 5 –11 S, J. Am. Meteorol. Soc., 35, 2031–2053.
Silva, M. A., M. Araújo, J. Servain, P. Penven, and C. Lentini (2009), Highresolution regional ocean dynamics simulation in the southwestern tropical
Atlantic, Ocean Modell., 30, 1–29, doi:10.1016/j.ocemod.2009.07.002.
Smith, R. L. (1978), Poleward propagating perturbations in currents and
sea levels along the Peru coast, J. Geophys. Res., 83, 6083–6092,
doi:10.1029/JC083iC12p06083.
Spillane, M. C., D. B. Enfield, and J. S. Allen (1987), Intraseasonal oscillations in sea level along the west coast of the Americas, J. Am. Meteorol.
Soc., 17, 313–325.
Stech, J. L., and J. A. Lorenzzetti (1992), The response of the South Brazil
Bight to the passage of wintertime cold fronts, J. Geophys. Res., 97(C6),
9507–9520, doi:10.1029/92JC00486.
Stramma, L. (1991), Geostrophic transport of the South Equatorial
Current in the Atlantic, J. Mar. Res., 49, 281–294, doi:10.1357/
002224091784995864.
Stramma, L., and F. Schott (1999), The mean flow field of the tropical
Atlantic Ocean, Deep Sea Res., Part II, 46, 279–303, doi:10.1016/
S0967-0645(98)00109-X.
Stramma, L., J. Fischer, and J. Reppin (1995), The North Brazil Undercurrent, Deep Sea Res., Part I, 42(5), 773–795, doi:10.1016/0967-0637(95)
00014-W.
Stramma, L., J. Fischer, P. Brandt, and F. Schott (2003), Circulation, variability and near-equatorial meridional flow in the central tropical Atlantic,
in Interhemispheric Water Exchanges in the Atlantic Ocean, edited by
G. J. Goni and P. Malanotte-Rizzoli, pp. 1–22, Elsevier, Amsterdam.
Stramma, L., M. Rhein, P. Brandt, M. Dengler, C. Böning, and M. Walter
(2005), Upper ocean circulation in the western tropical Atlantic in boreal
fall 2000, Deep Sea Res., Part I, 52, 221–240, doi:10.1016/j.dsr.2004.
07.021.
Torrence, C., and G. P. Compo (1998), A practical guide to wavelet analysis, Bull. Am. Meteorol. Soc., 79(1), 61–78, doi:10.1175/1520-0477
(1998)079<0061:APGTWA>2.0.CO;2.
Trenberth, K. E., W. G. Large, and J. G. Olson (1990), The mean annual
cycle in global ocean wind stress, J. Phys. Oceanogr., 20, 1742–1760,
doi:10.1175/1520-0485(1990)020<1742:TMACIG>2.0.CO;2.
Veleda, D., R., Montagne, and M. Araújo (2012), Cross-wavelet bias
corrected by normalizing scales, J. Atmos. Oceanic Technol., 29,
1401–1408, doi:10.1175/JTECH-D-11-00140.1.
von Schuckmann, K. (2006), Intraseasonal variability in the southwestern
and central tropical Atlantic Ocean, PhD thesis, 140 pp., Christian
Albrechts Univ. Kiel, Kiel, Germany.
Zhang, H.-M., J. J. Bates, and R. W. Reynolds (2006), Assessment of composite global sampling: Sea surface wind speed, Geophys. Res. Lett., 33,
L17714, doi:10.1029/2006GL027086.
C11024 VELEDA ET AL.: ISV OF THE NBUC FORCED BY REMOTE WINDS C11024
10 of 10