Content uploaded by Yuji Kashino
Author content
All content in this area was uploaded by Yuji Kashino on Jun 23, 2015
Content may be subject to copyright.
Content uploaded by Yuji Kashino
Author content
All content in this area was uploaded by Yuji Kashino on Jun 23, 2015
Content may be subject to copyright.
Content uploaded by Yuji Kashino
Author content
All content in this area was uploaded by Yuji Kashino on Jun 23, 2015
Content may be subject to copyright.
REVIEW
doi:10.1038/nature14504
Pacific western boundary currents and
their roles in climate
Dunxin Hu
1
, Lixin Wu
2
, Wenju Cai
2,3
, Alex Sen Gupta
4
, Alexandre Ganachaud
5
,BoQiu
6
, Arnold L. Gordon
7
, Xiaopei Lin
2
,
Zhaohui Chen
2
, Shijian Hu
1
, Guojian Wang
3
, Qingye Wang
1
, Janet Sprintall
8
, Tangdong Qu
9
, Yuji Kashino
10
,
Fan Wang
1
& William S. Kessler
11
Pacific Ocean western boundary currents and the interlinked equatorial Pacific circulation system were among the first
currents of these types to be explored by pioneering oceanographers. The widely accepted but poorly quantified
importance of these currents—in processes such as the El Nin
˜
o/Southern Oscillation, the Pacific Decadal Oscillation
and the Indonesian Throughflow—has triggered renewed interest. Ongoing efforts are seeking to understand the heat
and mass balances of the equatorial Pacific, and possible changes associated with greenhouse-gas-induced climate
change. Only a concerted international effort will close the observational, theoretical and technical gaps currently
limiting a robust answer to these elusive questions.
W
estern boundary currents (WBCs) are swift, narrow oceanic
currents found in all major oceanic gyres. Within the Pacific
Ocean, the subtropical gyre WBCs are the Kuroshio Current
in the Northern Hemisphere, and the East Australian Current (EAC) in
the Southern Hemisphere (Fig. 1a). The Pacific low-latitude, tropical
belt WBCs include the Mindanao Current in the Northern Hemisphere
and the New Guinea Coastal Undercurrent (NGCUC) south of the
Equator, both of which are directly connected to the equatorial Pacific
circulation system. Much of modern wind-driven ocean circulation
theory was derived from a quest to understand these Pacific Ocean
currents. Knowledge of the effect of the Earth’s rotation on WBCs,
and of the Ekman transport (see Box 1), led to ground-breaking
advances: that wind stress (Fig. 1b) is a driving agent of ocean currents,
but it is the horizontal gradient rather than the absolute strength that is
important; that latitudinal gradients in the effect of Earth’s rotation on
the horizontal motion cause a flow intensification towards the west of
the ocean basins; and that the Pacific wind-forced ocean currents
include the equatorial current system, the low-latitude and subtropical
WBCs.
Impacts of the Pacific WBCs on the global ocean circulation and
climate variability are manifold. First, in winter, the subtropical
WBCs are associated with the largest supply of heat and moisture into
the atmosphere in the Pacific basin, and are coupled to North Pacific
storm tracks
1
. As cold and dry air comes into contact with the warm
water carried poleward by the subtropical WBCs, heat and moisture are
extracted from the surface. Second, the Indonesian Throughflow (ITF),
the only low-latitude inter-ocean current, flows from the Pacific to the
Indian Ocean. The throughflow plays a role in the return branch of the
global thermohaline circulation
2
. The ITF sources its water mostly from
the North Pacific Ocean via the Mindanao Current with the remainder
sourced from the South Pacific Ocean via the New Guinea Coastal
Current (NGCC), implicating these Pacific WBCs as important
players in the global climate system. Third, the Pacific is home to the
El Nin
˜
o/Southern Oscillation (ENSO), the most prominent source of
global climate variability on interannual timescales, which severely dis-
rupts global weather patterns, affecting ecosystems, agriculture, tropical
cyclones, drought, bushfires, floods and other extreme weather events
worldwide
3
. The Pacific low-latitude WBCs are agents for transporting
mass into the equatorial Pacific, therefore critically influencing the
Western Pacific Warm Pool (a region of sea surface temperatures
(SSTs) warmer than 28.5 uC) and the life cycle of ENSO, as well as the
East Asian monsoon and the Indian/Southeast Asian monsoon. Finally,
the EAC, through its outflow to the Indian Ocean, participates in the
Southern Hemisphere supergyre circulation that links the three subtrop-
ical gyres
4–6
, providing a subtropical gateway for the Pacific’s influence
on global climate.
Research and observations over the past 15 years have enabled a string
of advances in our understanding of the structure of individual compo-
nents of the Pacific WBCs, interactions amongst them, and their
climatic impacts. In particular, several multi-national programmes
are underway. Included in these are the Northwestern Pacific Ocean
Circulation and Climate Experiment
7
, the Southwest Pacific Ocean
Circulation and Climate Experiment
8
, and the ITF monitoring pro-
gramme
9
, providing an intensive observational focus in Pacific
low-latitude regions. These process studies, together with targeted mod-
elling and recent advances in broad-scale sampling of ocean temper-
ature and salinity, upper oceanic currents, winds, precipitation and sea
surface heights, as well as satellite retrievals, have led to a significant
advance in our understanding of the physical structure and dynamics
of the Pacific WBCs, and their possible changes under greenhouse
warming.
Here we review the current state of understanding of the structure and
variability of the Pacific WBCs, their climatic impacts, and how they
may be affected by greenhouse warming. We show that the entire Pacific
WBC system moves northwards or southwards concurrently, on sea-
sonal and interannual timescales. The climatic impacts of the Pacific
1
Key Laboratory of Ocean Circulation and Waves, Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China.
2
Physical Oceanography Laboratory, Qingdao Collaborative Innovation
Center of Marine Science and Technology, Ocean University of China, Qingdao 266003, China.
3
CSIRO Oceans and Atmosphere Flagship, Aspendale, Victoria 3195, Australia.
4
Australian Research Council
(ARC) Centre of Excellence for Climate System Science, Mathews Building, The University of New South Wales, Sydney 2052, Australia.
5
Institut de Recherche pour le Developpement (IRD), UMR5566-
LEGOS, UPS (OMP-PCA), 31400 Toulouse, France.
6
Department of Oceanography, University of Hawaii at Manoa, 1000 Pope Road, Honolulu, Hawaii 96822, USA.
7
Lamont-Doherty Earth Observatory, Earth
Institute at Columbia University, Palisades, New York 10964, USA.
8
Scripps Institution of Oceanography, 9500 Gilman Drive, La Jolla, California 92037, USA.
9
IPRC, Department of Oceanography, SOEST,
University of Hawaii, Honolulu, Hawaii 96822, USA.
10
Center for Earth Information Science and Technology, Japan Agency for Marine-Earth Science and Technology (JAMSTEC) 3173-25 Showa-machi
Kanazawa-ku, Yokohama 236-0001, Japan.
11
NOAA/Pacific Marine Environmental Laboratory, Seattle, Washington 98115, USA.
G
2015 Macmillan Publishers Limited. All rights reserved
18 JUNE 2015 | VOL 522 | NATURE | 299
WBCs are far-reaching, and are exerted in many ways, including
through interactions with the South China Sea circulation, the
Tropical Pacific Warm Pool, and the ITF; through longevity of SST
anomalies along the WBCs and their feedback to the atmosphere; and
through contribution to the global thermohaline circulation. Under
greenhouse warming, the response of Pacific WBCs is uncertain,
although changes are not expected to resemble anomalies associated
with climate variability.
−0.1 0.1 0.2
−1 0 1 2
40° N
20° N
0°
20° S
40° S
140° E 140° W 100° W
EAC
NECC
EUC
TO
ACC
ACC
NEC
KC
MC
ITF
SEC
SECC
SEC
GS
OC
40° N
20° N
0°
20° S
40° S
Wind stress curl (10
–7
N m
–3
)
Wind stress (Nm
-2
)
–0.8
C
C
N
100° E
10° S
0°
10° N
20° N
30° N
140° E 160° E
40° S
30° S
20° S
10° S
0°
ba
dc
2,000
Bottom de
p
th
(
m
)
Borneo
Taiwan St
Luzon St
Makassar St
South
China
Sea
KC
KC
MC
NGCUC
SEC
NECLUC
NEUC
MUC
NECC
N
GCC
SEC
Solomo
n St
St Georges St
Vitiaz St
NGCUC
NICU
GPC
NQC
NVJ
SEC
NCJ
TF
EAUC
EAC
TO
SECC
Solomon
Sea
Coral
Sea
Gulf of
Papua
New
Caledonia
Tasman
Sea
Tasmania
Australia
?
New
Zealand
Karimata St.
Sibutu P.
Molucca Sea
Halmahera Sea
Papua-
N.Guinea
SJC
180° 0
140° E120° E
0502005001,0001,500
Wind stress curl (10
–7
N m
–3
)
0.80.3–0.3
NGCU
NGCU
Figure 1
|
Pacific Ocean circulation and boundary currents. a, Schematic of
major currents and observed surface wind vectors (orange dashed lines) and
wind stress curl (shading: positive curl in red; negative in blue). The Intertropical
Convergence Zone and the South Pacific Convergence Zone are indicated by
thick grey-dashed lines. Red and blue shading in the Southern Hemisphere
corresponds to anticyclonic and cyclonic curls, respectively, and vice versa in the
Northern Hemisphere. b, Pacific zonally averaged zonal wind stress (grey curve)
and wind stress curl (red curve) for the present day (solid lines) and the latter
half of the twenty-first century (dashed lines). Projected changes are calculated
as the multi-model mean difference between 2050–2100 future winds under
RCP8.5 and 1900–2000 historical winds. OC, Oyashio Current; KC, Kuroshio
Current; NEC, North Equatorial Current; MC, Mindanao Current; NECC,
North Equatorial Counter Current; NEC, North Equatorial Current; EUC,
Equatorial Undercurrent; NGCUC, New Guinea Coastal Undercurrent; ITF,
Indonesian Throughflow; SECC, South Equatorial Counter Current; SEC, South
Equatorial Current; EAC, East Australian Current; TO, Tasman Outflow; ACC,
Antarctic Circumpolar Current; GS, Gulf Stream. c, d, Topographic and
boundary current systems in the Northwest Pacific (c) and Southwest Pacific
(d), as indicated by the dashed white boxes in a. Depths shallower than 2,000 m
are colour-shaded. Grey shading indicates depths shallower than 50 m. The
main currents, integrated from 1,000 m to the surface, are indicated by light blue
arrows; the red curved arrow indicates the ‘freshwater plug’; red straight arrows
indicate the main surface-trapped counter currents; purple arrows indicate
undercurrents; question mark indicates uncertainty in flow path. Panel c is
adapted from ref. 8 (John Wiley and Sons). LUC, Luzon Undercurrent; MUC,
Mindanao Undercurrent; NGCC, New Guinea Coastal Current; NEUC, North
Equatorial Undercurrent; NICU, New Ireland Coastal Current; NVJ, North
Vanuatu Jet; NCJ, North Caledonia Jet; SCJ, South Caledonia Jet; NQC, North
Queensland Current; GPC, Gulf of Papua Current; TF, Tasman Front; EAUC,
East Auckland Current; TO, Tasman Outflow.
RESEARCH REVIEW
G
2015 Macmillan Publishers Limited. All rights reserved
300 | NATURE | VOL 522 | 18 JUNE 2015
Structure of the Pacific WBC system
Driven by the large-scale pattern of wind stress curl (see Box 1), the Pacific
WBC system is characterized by the unique presence of two intense
equatorward WBCs—the Mindanao Current and the NGCUC. These
serve as subtropical-to-tropical oceanic pathways that modulate the mass
and heat balances of the Western Pacific Warm Pool, and the ventilation
of the equatorial Pacific thermocline. As such, low-latitude WBCs are a
part of the life cycle of ENSO and Pacific decadal variability
10,11
.
The Mindanao Current and the NGCUC originate at the bifurcations
of zonal currents arriving at the western boundaries: the westward-flowing
North Equatorial Current (NEC) in the North Pacific, and the South
Equatorial Current (SEC) in the South Pacific. Upon reaching the
Philippine coast, the NEC bifurcates to feed the poleward Kuroshio
Current
12
and the equatorward Mindanao Current
13
(Fig. 1c). In a sim-
ilar manner, the westward-flowing SEC bifurcates at the Australian
coast into the northward-flowing Gulf of Papua Current
14
and south-
ward-flowing EAC
15
. The Gulf of Papua Current feeds the NGCUC,
which exits through narrow straits at the northern boundary of the
Solomon Sea
16
, connecting the South Pacific to the Equator and the
ITF (Fig. 1c, d).
The convergence of the Mindanao Current and NGCUC near the
Equator (Fig. 2) feeds the Western Pacific Warm Pool, the surface to
lower thermocline of the ITF
17
, and the eastward-flowing Equatorial
Undercurrent (EUC)
18
. In this way, the two low-latitude WBCs together
with the strong mixing in the Indonesian Seas largely determine the
water mass characteristics of the Pacific equatorial current systems
and of the ITF
18
. Also involved in the mass balance of the Western
Pacific are counter currents, which are associated with the
Intertropical Convergence Zone and the South Pacific Convergence
Zone, the two prominent rainbands in the Pacific. These convergence
zones themselves, together with local winds they have altered and the
associated wind stress curl, produce two eastward-flowing, surface-
intensified counter currents, known as the North Equatorial Counter
Current
19
and the South Equatorial Counter Current
20
(Fig. 1a).
The Pacific low-latitude current system has a rich vertical structure,
and over the years there has been conflicting evidence as to the nature of
some currents
21–29
. In both hemispheres, the bifurcation latitude
increases with depth. For example, the long-term average NEC bifurca-
tion latitude is approximately 14u N at the surface, but 20u N at a depth
of 1,000 m. The Luzon Undercurrent
22,23
, observed beneath the
Kuroshio, with a core at depths around 650 m near 18u N, intensifies
towards the south before turning to the east to feed the North Equatorial
Undercurrent. By contrast, despite the suggestion of a northward-
flowing Mindanao Undercurrent with a centre at around 600 m (refs 21,
24–26; Fig. 2a), debate persisted as to whether it is transient in nature
27
,
a quasi-permanent flow but affected by thermocline variability
28
,
or a subthermocline eddy
29
. The presence of the undercurrent at
depths below 400 m has been confirmed by recent measurements
(Fig. 2a; ref. 25).
In the South Pacific, the range of the SEC bifurcation latitudes as a
function of depth is large, varying between 13u and 25u S. The surface
SEC flow turns southward feeding the EAC, while the subsurface flow, the
Great Barrier Reef Undercurrent (part of the Gulf of Papua Current
system), veers northward feeding the NGCUC
20–32
. The Great Barrier
Reef Undercurrent, the South Pacific counterpart of the Luzon
Undercurrent, has been linked to the basin-scale circulation of the sub-
tropical gyre that contracts poleward with depth
33
. The NGCUC was
recently measured
31
as a strong, permanent undercurrent against the
coast of Papua New Guinea (Fig. 2b), with its core at about 400 m carrying
the bulk of the South Pacific low-latitude WBC transport that splits in the
different Solomon Sea straits then flows towards the Equator or the ITF
8
.
Because the seasonal wind stress curl variability that controls the
bifurcation latitude is due to seasonal movement of trade winds, the
NEC and SEC bifurcations move synchronously and in the same meri-
dional direction on seasonal timescales. At the surface, the NEC bifurca-
tion latitude moves from 15u N in boreal summer, to 17u N in boreal
winter
34
, in conjunction with the seasonal reversal of the South East
Asia monsoon. This seasonal movement is accompanied by a stronger
NEC leading to both a stronger Mindanao Current and Kuroshio
Current when the bifurcation occurs at lower latitudes in boreal sum-
mer, increasing the tropical water injection into the North Pacific
subtropical gyre
13
. As such, the NEC, the Mindanao Current and the
Kuroshio Current transport east of Luzon are all at their seasonal max-
imum during boreal summer (June–July) when bifurcation occurs
at the southernmost latitude
34,35
but minimum during boreal winter
(November–December) when bifurcation occurs at the northernmost
BOX 1
Processes associated with the
Pacific surface WBCs
Ekman transport and oceanic currents. In the tropical Pacific, the
main currents, westward flowing over the upper ocean, are forced by
the easterly trade winds. In the first few tens of metres of water depth,
immediately away from the Equator, deflection by the rotation of the
Earth occurs to the right in the Northern Hemisphere, and left in the
Southern Hemisphere. This ‘Coriolis’ force causes ‘Ekman transport’,
whereby tropical surface waters move poleward in both hemispheres.
Likewise, poleward of 25u Nand25u S, the prevailing westerly winds
induce Ekman transports towards the Equator, generating a
convergence between 15u–30u Nand15u–30u S, with high sea levels
around 10u–15u Nand10u–15u S, respectively. The sea-level slope
drives two broad geostrophic westward flows: the North Equatorial
Current (NEC) and the South Equatorial Current (SEC).
Bifurcation and WBCs. Upon approaching the western boundaries,
equatorial currents split into two branches: equatorward low-latitude
WBCs; and poleward subtropical WBCs. In the Pacific, the NEC
bifurcates into the equatorward Mindanao Current and the poleward
Kuroshio Current, and the SEC bifurcates into the equatorward Gulf of
Papua Current and poleward East Australia Current (EAC).
Rossby wave adjustment. The above ocean wind–current
relationship assumes a steady wind forcing the ocean for a long time.
Winds vary on the large scale, and the ocean adjusts by generating
large-scale anomalies that move in the form of oceanic Rossby waves.
Such waves propagate along the thermocline, a boundary between the
warm upper layer and the cold deeper part of the ocean, and always
propagate westward because of the latitudinal gradient of the
tangential speed of the Earth’s rotation.
Wind stress curl. It is not the absolute magnitude of the wind stress
but rather its curl (the horizontal gradient of the stress) that forces the
north–south transport of the interior ocean. At some locations, the curl
is reduced to zero and there is no north–south transport (although
there might be flows at depth which cancel when summed up). Lines
along which the curl is zero provide natural boundaries that separate
circulation into gyres. Such lines in subtropical latitudes are often used
as an approximation to locate the bifurcation latitude.
Kuroshio Extension. At some latitude, the Kuroshio Current
separates from the coast and flows eastward into the ocean interior. At
this eastward excursion, the Kuroshio Extension has some of the
largest air–sea fluxes found across the North Pacific basin. It is a region
with a great quantity of latent,sensible, andnet surfaceheat loss, and is
co-located with the Pacific storm track.
Tasman Front and Tasman Outflow. In the South Pacific, part of the
EAC separates from the coast and veers eastward as the Tasman
Front; the residual continues southward passing through the Tasman
Sea. A portion of the Tasman Front attaches to the northern coast of
New Zealand, forming the East Auckland Current. The portion flowing
through the Tasman Sea reaches Tasmania and turns westward into
the eastern Indian Ocean as Tasman Outflow.
REVIEW RESEARCH
G
2015 Macmillan Publishers Limited. All rights reserved
18 JUNE 2015 | VOL 522 | NATURE | 301
latitude. The SEC bifurcation shifts northward towards the Equator in
austral summer (November–December), and is southernmost in austral
winter (June–July)
15
. When the SEC is closer to the Equator, its transport
strengthens, feeding more water into both the EAC and the NGCUC
15
.
The EAC attains a maximum transport (up to 36 Sv; 1 Sv 5 10
6
m
3
s
21
)
and extends furthest poleward in austral summer with a large Tasman
Outflow into the Indian Ocean. In austral winter, the EAC transport
weakens (to 27 Sv), and much of this veers eastward as the Tasman Front
(Fig. 1d; ref. 36).
Consistent with dynamical argument, observations have shown that
the South Pacific subtropical gyre and the Tasman Outflow south of
Australia are part of a system of nested anticyclonic gyres encompassing
the entire Southern Hemisphere subtropics, connecting the Atlantic,
Indian and Pacific basins to the global thermohaline circulation
4–6,37,38
.
The Atlantic and Indian Oceans are connected south of Africa by a
westward Agulhas leakage and the eastward South Atlantic Current
38
,
while the Pacific and Indian Oceans are connected south of Tasmania by
the westward Tasman Outflow
4,5,37
, a narrow boundary flow of the west-
ern Pacific gyre that turns westward around the south of Tasmania and
‘leaks’ to the Indian Ocean basin
4
. Tasman leakage occurs primarily as a
subsurface or intermediate water circulation
6
, in contrast to the ITF
route that largely occurs in the upper 300 m. The linkage of the
Tasman leakage to the supergyre provides a mechanism whereby
Antarctic Intermediate Water, which transits to the southwest Pacific,
is distributed between the ocean basins before it spreads northward into
the Pacific, Indian and Atlantic oceans
5,6
.
Variability of the Pacific WBCs
Pacific WBCs are subject to strong variability on intraseasonal
26,27,39
,
interannual
30,40–44
, decadal and longer timescales
42,43
. However, as a
result of possible eddy aliasing
30,44
, and the discrete short-term nature
of in situ surveys that may be difficult to place in an appropriate climate
context, there is large uncertainty in the volume transport estimate of
these WBCs. For example, the Mindanao Current transport ranges from
13 to 39 Sv depending on the study
12,27,41
.
The NEC bifurcation latitude and its partitioning into the Mindanao
Current and the Kuroshio Current are affected by many factors, includ-
ing local monsoonal winds and buoyancy fluxes, and remote forcing
from the Pacific interior and North Pacific along the western bound-
ary
35,40
. The main forcing, though, is ENSO. The response of the WBCs
to ENSO, in turn, influences the life cycle of ENSO. As an El Nin
˜
o
develops, the entire Pacific circulation system ‘breathes’ together, bifurc-
ating at a more northerly latitude (Fig. 3a, b), as manifested through a
positive correlation of ENSO with the north and south bifurcation lati-
tudes. Accompanying the more northerly bifurcation are a stronger
NEC and North Equatorial Counter Current, a stronger Mindanao
Current, a more intense EUC at its westernmost part, but a weaker
Kuroshio Current and ITF
28,30,40
(Fig. 3a). To the south, the NGCUC
intensifies during, or several months after, the peak of an El Nin
˜
o
30
, and
the surface southward NGCC disappears and turns northwards
(Fig. 3b), as occurred from July to February during the 1997–98
extreme El Nin
˜
o
45
. Although the EAC transport displays variability
on interannnual timescales, only a weak ENSO signal is evident in
observations
36
.
The responses of the Mindanao Current and the NGCUC to ENSO as
described above are integral to the ENSO discharge/recharge cycle
46
.
The increased equatorward WBC transport during an El Nin
˜
o creates
a strong confluence of relatively cooler water in the tropical western
Pacific, and constitutes a compensatory flow for the associated discharge
of the equatorial Pacific warm water, conducted in the interior Pacific
through the Sverdrup process, whereby mass is pumped out of the
equatorial west Pacific in response to wind forcing.
Recent observations revealed a rich spatial structure and a large res-
ponse of the SEC transport to ENSO
30–32
. Variability of the NGCUC
inflow to the Solomon Sea can reach 100% of the mean transport, with
the changes mainly occurring in the upper 250 m. During an El Nin
˜
o
event, the SEC transport between New Caledonia and the Solomon
Islands increases
21,31,32
, particularly in the North Vanuatu Jet, with an
enhanced transport entering the Solomon Sea (Fig. 3b). This leads to an
increased equatorward mass transport through the Solomon Strait and
200
400
600
0 90 180 270
Distance from coast (km)
360 440
600
400
200
100
50
0
–50
–100
40
20
0
–20
1 July 20121 Jan 20121 July 20111 Jan. 2011
Depth (m)Depth (m)
Velocity (cm s
–1
) Velocity (cm s
–1
)
Date
b
a
Figure 2
|
Characteristics of the two equatorward low-latitude WBCs in the
tropical Pacific. Such WBCs are a unique feature of the Pacific Ocean.
a, Moored current meter data at 8u N, 127u E showing evolution of meridional
velocities, the Mindanao Current (blue, negative values) and its undercurrent
(red, positive values) from boreal winter 2010 to boreal winter 2012, adapted
from ref. 26 (John Wiley and Sons). Shown are daily ADCP data smoothed by a
three-day running mean filter. Moorings were anchored in 6,100 m deep water,
with two 75K RDI ADCP current meters at ,400 m scanning up and down,
respectively. The current meters move vertically in the water column, as strong
currents interact with the mooring system. b, Velocity section across the South
Solomon Sea, from a compilation of glider observations adapted from ref. 31
(AMS). Units are cm s
21
. Red shading indicates equatorward flow, and blue
poleward flow. The NGCUC is characterized by a strong core, centred at about
400 m, against the coast of Papua New Guinea, to the left of the section. The
x-axis represents an equivalent eastward distance from the coast of Papua New
Guinea.
RESEARCH REVIEW
G
2015 Macmillan Publishers Limited. All rights reserved
302|NATURE|VOL522|18JUNE2015
through the New Ireland Coastal Undercurrent, in addition to the
increased transport via the Vitiaz Strait
45,47
(see Fig. 3b). The transport
of heat and salt through the Solomon Sea can change by up to a factor of
two between extreme El Nin
˜
o and La Nin
˜
a conditions
47
.
Connections between the Pacific WBCs and the ITF are complex,
particularly when ENSO is considered. Synthesis of observational data
and model simulations indicates that the depth and velocity of the
ITF core vary with ENSO, with the ITF slowing and shoaling
during El Nin
˜
o events
48
. Modelling and observational studies indicate
that, on average, the proportion of the ITF water derived from the
North and South Pacific depends sensitively on ENSO phases.
Climatologically, approximately 80% of the ITF originates from the
North Pacific and 20% from the South Pacific via the NGCUC
48–50
.As
a La Nin
˜
a develops, the ITF increases due to stronger easterly trade
winds, which induces a pressure gradient between the West Pacific
and the eastern Indian Ocean. Climate models suggest a greater
amount of ITF water originates from the South Pacific during La
Nin
˜
a compared to during El Nin
˜
o
49
. During the development of La
Nin
˜
a, a seasonal and ENSO-dependent inflow from the Luzon Strait
to the Indonesian Seas forced by large-scale winds in the Pacific,
called the South China Sea Throughflow
51
, weakens. Further, the
Mindanao Current transport decreases, though a greater portion of
this current goes into the ITF at the expense of the North Equatorial
Counter Current
50
.
Interactions between the Pacific WBCs and the Pacific Decadal
Oscillation (PDO) occur on decadal timescales in a manner somewhat
different to the interannual timescales for ENSO, particularly in the
North Pacific
42,43
. The post-1990 cold PDO phase is linked to a south-
ward shift of the NEC bifurcation latitude, and a strengthened NEC and
North Equatorial Counter Current, as a result of intensified easterlies
over the Western Pacific
42,43
. Both the Kuroshio Current and the
Mindanao Current have also strengthened, inconsistent with ENSO-
induced changes, which would imply a decreased Mindanao Current.
In the Southern Hemisphere, observations and ocean reanalysis suggest
that a cold phase of the PDO is associated with a deeper southward EAC
extension
36
, while the Tasman Front weakens, which suggests that there
is a gating between these two currents
52
. In the two decades before 1993,
the PDO was in a warm phase, whereby the trade winds across the
tropical Pacific were weaker, consistent with a weaker atmospheric
Walker circulation
53
. In response, the NEC and its counter currents were
also weaker and shifted northward
42,43
.
Multidecadal fluctuations can also modulate the interannual relation-
ship between ENSO and the NEC bifurcation latitude, such that the
relationship between ENSO and the northwestern Pacific WBCs is
non-stationary
13,43
. The NEC mean bifurcation latitude is situated fur-
ther to the south during a cold PDO phase, and from this more south-
ward position the NEC bifurcation latitude can ‘jump’ a considerable
distance to the north during an El Nin
˜
o. As such, interannual NEC
bifurcation variability is more strongly correlated to the Nin
˜
o-3.4 index
during the cold phase of the PDO. Thus, the bifurcation latitude of the
NEC which is determined by wind forcing in the 12u–14u N band also
contains variability not solely represented by the Nin
˜
o-3.4 index
42
.
Interactions with climate
The Pacific WBCs interact with physical climate on synoptic, interann-
ual and long-term climate scales, and such interactions are better under-
stood for the North Pacific than for the South Pacific. In boreal winter,
strong advection of warm water by the Kuroshio Current into regions of
40° S
30° S
20° S
10° S
0°
Low salinity
thermocline
water
injection
Luzon
+
Sibutu
+
MC
+
NEC
+
NECC
+
Karimata
+
Makassar
transport
inhibited
KC
–
MK
–
–
NICU
+
NGCUC
+
NVJ
+
SEC
+
Equatorward
transport enhanced
through NICU
and NGCUC
Bifurcation latitude change
Current anomaly
Mean north bifurcation
ba
2,000
Bottom de
p
th
(
m
)
0
502005001,0001,500
100° E 140° E 160° E
140° E
120° E
10° S
0°
10° N
20° N
30° N
Figure 3
|
Impact of El Nin
˜
o on the Pacific WBC system. a, b,DuringanEl
Nin
˜
o, the bifurcation moves north in both the Northern (a) and the Southern
(b) Hemispheres, as indicated by the black arrows. The ocean depth is indicated
as on Fig. 1c, d. Increasing (green boxes) and decreasing (orange boxes) flow
anomalies during El Nin
˜
o are indicated. Some currents are enhanced, while
others are weakened, as indicated by the anomalies (large dark blue/red
arrows). From the developing phase to approximately three-to-four months
before the peak of an El Nin
˜
o, the Kuroshio Current (KC) decreases while the
Mindanao Current increases. The ITF weakens as the Pacific-Indian pressure
gradient reduces. The South China Sea Throughflow intensifies, setting up a
freshwater plug. Towards the peak of an El Nin
˜
o, equatorward transport
enhances through NICU and NGCUC. LUC, Luzon Undercurrent; MUC,
Mindanao Undercurrent; NGCC, New Guinea Coastal Current; NEUC, North
Equatorial Undercurrent; NICU, New Ireland Coastal Current; NVJ, North
Vanuatu Jet; NCJ, North Caledonia Jet; SCJ, South Caledonia Jet; NQC,
North Queensland Current; GPC, Gulf of Papua Current; TF, Tasman Front;
EAUC, East Auckland Current; TO, Tasman Outflow.
REVIEW RESEARCH
G
2015 Macmillan Publishers Limited. All rights reserved
18 JUNE 2015 | VOL 522 | NATURE | 303
colder air temperatures, such as the Kuroshio/Oyashio extension region,
results in large losses of latent and sensible heat to the atmosphere
54,55
.
This heat loss triggers oceanic and atmospheric deep convection, and
fuels storms for their recurrent development, while contributing to the
formation of mode waters
55
. In the mixed water region north of the
Kuroshio Extension, turbulent heat fluxes of more than 600 W m
22
are observed during wintertime northerly wind events
56
, where the con-
trast between the WBC warm core and adjacent waters gives rise to
particularly large SST gradients: more than 10 uC over 200 km across
the Oyashio Front. The presence of such oceanic fronts can generate
atmospheric instability, and increased vertical momentum exchange, as
well as winds induced by changes in air density
57,58
. This tends to exert a
positive feedback, whereby larger turbulent heat fluxes from the ocean to
the atmosphere over the warm core of the current generate even stronger
winds
57,58
. Atmospheric eddy heat fluxes associated with synoptic sys-
tems act to reduce the sharp meridional air temperature gradients
1
, but
the large thermal inertia of the ocean mixed-layer and oceanic advection
rapidly restore the SST front and the air temperature gradient
59
.An
impact of the warm water maintained in this manner is that typhoons
tend to strengthen as they pass over the warm core of the Kuroshio
Current
60
. During summer time, heat and moisture supply from the
warm Kuroshio, though less than in winter, is effective in retaining
the convectively unstable stratification of low-level northwestward
monsoonal airflow from the tropics, enhancing the convective heavy
rainfall along the Kuroshio
61
.
Beyond synoptic scales, the North Pacific WBCs exert a considerable
impact on Asia summer monsoons
62,63
. One such impact is on the South
China Sea summer monsoon, which marks the commencement of the
Asia summer monsoon. In post-La Nin
˜
a boreal spring and early sum-
mer, the western tropical Pacific heat content is anomalously high, as a
part of the heat recharge associated with stronger trade winds. In asso-
ciation with a stronger Walker circulation, convection over the eastern
Indian Ocean and the Western Pacific remains strong, resulting in
stronger-than-normal westerlies over the tropical Indian Ocean deliver-
ing the crucial moisture source favourable for the South China Sea
summer monsoon
63
. Consequently, the South China Sea summer mon-
soon commences earlier than normal, leading to droughts along the
middle-lower Yangtze River basin and floods in southern and northern
China
61
. Thus, the heat content of the Western Pacific Warm Pool can
be a predictor of the South China Sea summer monsoon onset and
strength, with a high heat content associated with an early onset and
strong intensity
63
. Conversely, in the late boreal spring-early summer
after an El Nin
˜
o, when the warm pool is still cooler than normal, the
South China Sea monsoon takes place later than normal, causing floods
along the middle-lower Yangtze River basin and drought in southern
and northern China
63
. A similar link exists between the Australian
summer monsoon and ENSO via SST anomalies in the central and
Western Pacific
64
; however, the timing is different. The Australian sum-
mer monsoon tends to be weakened, concurrently, by an El Nin
˜
o, and
strengthened by a La Nin
˜
a.
Interactions between the North Pacific WBCs and the South China
Sea Throughflow
65
on seasonal and interannual timescales constitute
another source of climatic influence. During the boreal winter, when
the northwest monsoon prevails over the Indonesian Seas, or during El
Nin
˜
o periods, the wind drives buoyant, low-salinity surface water
derived from the South China Sea via the Karimata Strait (Fig. 1c) into
the southern Makassar Strait
65
and from the Sibutu Passage into the
northern Makassar Strait
50
. This creates an anomalous pressure gradient
in the surface layer of the Makassar Strait (red curved arrow, Fig. 1c) that
inhibits the warm surface water from the Mindanao Current from enter-
ing into the ITF. This surface-layer ‘freshwater plug’, originally due to
heavy rainfall and large runoff from Southeast Asia, inhibits the Pacific
warm surface water from flowing southward into the Indian Ocean,
leading to a reduced, cooler ITF in the Makassar Strait during
December–January and during El Nin
˜
o, and inducing a cooler upper
Indian Ocean, which may in turn weaken the Asian monsoon
65
. During
the boreal summer, when winds in the Indonesian Seas reverse to the
southeast monsoon, or during La Nin
˜
a periods, the obstructing pressure
gradient is reduced, reducing the impact of the freshwater plug, support-
ing an earlier and stronger South China Sea monsoon in the late spring,
or early summer after a La Nin
˜
a event
50
.
Variations in the North Pacific WBCs also exert a climatic impact
through the longevity of SST anomalies they generate
54,55,66–70
, influ-
encing the large-scale atmospheric circulation and storm-track activ-
ities, particularly near the multiple fronts of the WBC extension
1,71,72
.
These extension fronts are fertile grounds for strong SST variability.
Because of the strong horizontal gradients, a slight frontal movement
can lead to a large change in SSTs and is an effective process for gen-
erating SST variability
66,69
. Frontal movement can be generated by
intrinsic variability of the WBCs as well as by remote ocean and atmo-
sphere forcing via extratropical ocean Rossby waves
67–70
. The Kuroshio
Extension fronts, for example, are strongly influenced by large-scale
changes in the subtropical gyre driven by wind stress changes across
the North Pacific
42,73
, and by ENSO-induced basin-scale wind stress curl
anomalies through atmospheric teleconnections
67,68
, emanating from
the tropical Pacific, remotely forcing SST variability in the Kuroshio
Extension region via oceanic Rossby waves
73,74
.
The longevity of SST anomalies along the Kuroshio Extension can be
further enhanced by processes that either maintain or reinforce the
anomalies
75–80
. For example, in boreal summer, the North Pacific
Ocean cools the warm moist air advected from the south, inducing
the formation of low-level stratus and fog within the boundary layer
on the northern sides of the Kuroshio Extension fronts
75,76
. These clouds
(in particular low clouds) reduce solar insolation because of their high
albedo, in turn maintaining low SSTs
76,77
. This positive ‘cloud–SST feed-
back’ acts to reinforce cold SST anomalies, and can enhance and prolong
summertime cold SST anomalies near the North Pacific subarctic front
zone
78
. In boreal winter, through anomalous geostrophic heat advection,
the axial migration of the Kuroshio Current and the associated frontal
movement generate subsurface temperature anomalies that can persist
over multiple years. Shoaling of the North Pacific mixed layer in boreal
spring and deepening in autumn results in mid-latitude SST anomalies
recurring from one winter to the next without persisting through the
intervening summer
79,80
. This ‘re-emergence mechanism’ operates when
SST anomalies spread throughout the deep winter mixed layer, but
remain beneath the mixed layer in spring, isolated from the ‘damping’
surface fluxes. When the mixed-layer deepens again in the following
autumn, the temperature anomalies are brought out into the surface
layer and influence SST
80
. Along the Kuroshio Extension, advection
by the mean currents transports these anomalies by as much as 4,000
km over the course of a year
79,80
. These processes are favourable for
decadal SST variability such that it is particularly strong along such
fronts
55,69,81
.
Climatic impacts of the EAC are less well understood, but there is
suggestion of an influence on Australian and New Zealand climates.
Weaker EAC transport is linked to anomalous cool conditions in New
Zealand
82
, and a stronger EAC is shown to promote rainfall along the
east Australia coast
83
. However, little is known about the impact of the
EAC on, for example, storm tracks or atmospheric circulations.
The Pacific WBCs exert substantial influence on timescales longer
than decadal, and on climate remote from the Pacific regions, through
their linkage to the Southern Hemisphere supergyre circulation via the
Tasman Outflow
4–6,38
and the ITF
9,48,84–87
. The Tasman Outflow influ-
ences downstream ocean climate through intermediate-depth water
4
,
but the influence through the ITF is conducted through the upper
300 m (ref. 9). Increased transport within the warmer upper layer of
the ITF modifies the heat budget of the Indian Ocean
84,85
as well as the
residence time of tropical Indian Ocean waters
86
, thermocline depth and
SST patterns
87
. Further downstream, the ITF feeds the Agulhas leakage,
and ultimately provides an important heat source for supporting the
global thermohaline circulation
2
.
RESEARCH REVIEW
G
2015 Macmillan Publishers Limited. All rights reserved
304 | NATURE | VOL 522 | 18 JUNE 2015
Observed changes and future projection
Over the past century, there has been direct and indirect evidence sup-
porting changes having occurred in the subtropical gyres. For example,
there appears to be a synchronous surface ocean warming along the path
of all subtropical WBCs that is two-to-three times greater than the global
mean
88
. The enhanced warming is associated with a poleward shift and/
or intensification of global subtropical WBCs in conjunction with a
systematic change in winds over both hemispheres. Specifically, the
zero-curl line shifted poleward in the North Pacific, the North
Atlantic and the South Atlantic Oceans; in the South Pacific and
South Indian Oceans, high-latitude westerly winds strengthened while
mid-latitude southeasterly trade winds weakened
5,37
, consistent with a
positive trend of the Southern Annular Mode
5,38
, leading to accelerated
warming and an intensified supergyre.
Over the past 60 years, there has been a synchronous long-term
southward shift of the NEC and the SEC bifurcations
89
in the tropical
Pacific (Fig. 4a, b), similar to their synchronous movement on inter-
annual timescales. Over the 60-year period, the overall long-term south-
ward shift of the NEC is modulated by a slight northward migration
from 1970 to 1992 (ref. 90) and a southward shift in the post-2000
period, associated with the warm and cold PDO phases, respectively.
The long-term southward shift of the latitude of the NEC bifurcation
suggests a strengthening Kuroshio Current, consistent with the observed
accelerated warming (Fig. 4a). By contrast, the southward shift of the
SEC bifurcation has no bearing on the transport partition, but is accom-
panied by a substantial southward extension of the EAC
5,37
. The relative
importance of greenhouse warming and decadal variability for the
North Pacific changes is difficult to quantify. However, about two-thirds
of the poleward intensification of the Southern Hemisphere supergyre
circulation, including the southward EAC extension and the strength-
ening of the Tasman Outflow
91,92
, has been attributed to wind changes
induced by Antarctic ozone depletion
93
, and one-third to increasing
atmospheric CO
2
(refs 91–93).
Projected changes in the Pacific WBCs depend on emission scenarios
of greenhouse gases. Antarctic ozone is projected to recover by around
the middle of this century, providing a mechanism to offset the impacts
of a projected continuous increase in CO
2
. As such, climate models
under the business-as-usual emission scenario, which currently matches
the observed emission rates, simulate no further trend in the Southern
Hemisphere westerly winds before 2045, but an acceleration in the post-
2045 period
94
. In the following two paragraphs, we discuss projected
changes averaged over the 2050–2100 period referenced to the mean
conditions over the twentieth century, under this emission scenario.
Climate models consistently indicate that the Tasman Outflow trans-
port will continue to increase (from ,2 Sv in the twentieth century to
,6 Sv over 2050–2100) and that the latitude of maximum EAC transport
will shift further south (,1.5u latitude over 2050–2100; ref. 94); however,
the projected EAC maximum transport shows little change (Fig. 5). In the
North Pacific, the Kuroshio Current is projected to weaken at its origin by
approximately 10%, although the maximum Kuroshio transport shows
little change. The EUC tends to strengthen in the eastern and central
Pacific (Fig. 5) and become shallower, particularly in the west
95,96
. This
tendency for strengthening occurs despite a weakening of the equatorial
trade winds
53
. The projected EUC intensification is driven by a projected
strengthening of the NGCUC that results from a wind stress curl change
that is negative, associated with a weakening of the equatorial trade winds
and a strengthening of the southeasterly trade winds
96
. A strengthening of
the NGCUC supporting a strong EUC is consistent with a present-day
observation that much of the NGCUC transport goes to support the
EUC
97
, although the range in model projections is rather large.
Both the Mindanao Current and the ITF are projected to weaken
(Fig. 5). The reduced ITF is not fully accounted for by the projected
wind-induced changes through Sverdrup dynamics that exclude an
influence from upwelling deep water. This exclusion is an invalid
assumption because such upwelling is important on timescales relevant
for climate change. It is conceivable that a projected slowdown of the
global thermohaline circulation
98
would drive a weakening in upwelling
of deep water in the South Pacific, contributing to the projected ITF
decrease, and following the notion that the ITF is in part supported by
the upwelled deep water associated with the compensating flow for the
global thermohaline circulation
2
.
Uncertain outcome
There is little doubt that the mean climate over the Pacific will continue to
change in the coming century owing to past and future emissions of
greenhouse gases. Climate models suggest that the tropical trade winds
are likely to weaken
53
, and that the southern mid-latitude westerlies are
likely to undergo a poleward shift and intensification
91–94
, and contribute
to the intensification of the WBCs, although changes in North Pacific
winds are less clear
99
. These changes, together with a projected weakening
in the global thermohaline circulation
98
, may simultaneously affect the
Pacific WBCs. Further, such changes may not resemble anomalies assoc-
iated with variability on interannual or inter-decadal timescales
96
.
Although climate models agree on many aspects of the projected future
changes, such as the weakening of the NEC and the NECC and the
strengthening of the NGCUC and the NQC, the simulation of WBCs is
generally poor because these coarse-resolution simulations do not resolve
the terrain of marginal seas, complex flow structure of WBCs, bathymetry
and eddies. Certain aspects of future change in high-resolution models
remain consistent with the lower-resolution models
100
, but uncertainty in
projected changes remains high. Another challenge is the lack of informa-
tion about the flow structure and amplitude of the low-latitude Pacific
WBCs in the real world, necessary for benchmarking climate models. For
example, despite an extensive observational campaign, the ITF mass bud-
get is not closed; the estimated outflow exceeds the inflow by ,2Sv
(ref. 9). It is unknown whether a contribution from the NGCUC through
the Molucca and Halmahera Seas might account for the discrepancy. This
mass imbalance is linked to many long-standing issues, including whether
Antarctic Intermediate Water, through the NGCUC and the Mindanao
12.5
13
13.5
14
14.5
15
15.5
16
16.5
17
17.5
Latitude (°N)
a
1950 1960 1970 1980 1990 2000 2010
17.5
17
16.5
16
15.5
15
14.5
14
13.5
13
12.5
Latitude (°S)
Year
b
Figure 4
|
Trend in the bifurcation latitude of the North and South
Equatorial Currents. a, b, Time series of annual mean NEC bifurcation
latitude integrated over the upper 381 m (a; red curve, adapted from ref. 90
(John Wiley and Sons)), and of SEC bifurcation latitudes integrated over the
upper 200 m (b; blue line, adapted from ref. 89 (John Wiley and Sons)), based
on SODA version 2.2.4. The bifurcation in each layer of SODA data is defined
as where the meridional velocity averaged within a 2u longitude band off the
Philippine/Australian coast is zero. Correlation between the NEC bifurcation
latitude and Nin
˜
o-3.4 is r 5 0.55, and between the SEC bifurcation latitude and
Nin
˜
o-3.4 is r 5 0.40, both statistically significant above the 95% confidence
level. Both trend lines are above the 95% confidence level.
REVIEW RESEARCH
G
2015 Macmillan Publishers Limited. All rights reserved
18 JUNE 2015 | VOL 522 | NATURE | 305
Undercurrent, reaches east of the Luzon northern tip
19
, or directs east-
ward between 10u–12u NundertheNEC
24,28
. Without such ground-truth
information, our confidence in model performance will remain low.
With a sustained community effort, our understanding of the Pacific
WBCs and their impacts will continue to improve. The ultimate goal is to
produce a reliable projection of the Pacific WBCs and their climatic
impacts that is consistent with our physical theoretical understanding,
as well as with what observations show. To this end, coordinated observa-
tions, studies of processes, and modelling should be bolstered. Our models
need to resolve the complex terrain of marginal seas and the detailed
structure of the WBCs, together with bathymetry, mesoscale eddies, and
other currently unresolved processes, especially turbulent mixing. An
international WBC intermodel comparison project may offer a good ave-
nue for a substantial advance. These modelling efforts need to be sup-
ported by a coordinated observation system integrating high-density
mooring arrays across all Pacific WBCs, satellite observations, Argo floats
and platforms that enable near-coast measurements, to provide a three-
dimensional picture of the coast-to-deep-water WBC structure. In the
meantime, we need to combine all sources of information to make
reliable predictions of variability and robust assessments of the response
of the Pacific WBCs to greenhouse warming and the associated climatic
impacts.
Received 13 August 2014; accepted 8 April 2015.
1. Nakamura, H., Sampe, T., Tanimoto, Y. & Shimpo, A. Observed associations
among stormtracks, jet streams and midlatitude oceanic fronts. AGU Geophys.
Monogr. Ser. 147, 329–346 (2004).
2. Gordon, A. L. Interocean exchange of thermocline water. J. Geophys. Res. 91,
5037–5046 (1986).
3. Cai, W. et al. More extreme swings of the South Pacific convergence zone due to
greenhouse warming. Nature 488, 365–369 (2012).
4. Ridgway, K. R. & Dunn, J. R. Observational evidence for a Southern Hemisphere
oceanic supergyre. Geophys. Res. Lett. 34, L13612 (2007).
5. Cai, W. Antarctic ozone depletion causes an intensification of the Southern
Ocean super-gyre circulation. Geophys. Res. Lett. 33, L03712 (2006).
6. Speich, S., Blanke, B. & Cai, W. Atlantic meridional overturning circulation and the
Southern Hemisphere supergyre. Geophys. Res. Lett. 34, L23614 (2007).
7. Hu, D. et al. Northwestern Pacific Ocean Circulation and Climate Experiment
(NPOCE) Science/Implementation Plan (China Ocean Press, 2011).
Details of NPOCE, outlining its goal and scope with a comprehensive literature
review on ocean circulation and climate in the northwestern Pacific Ocean.
8. Ganachaud, A. S. et al. The Southwest Pacific Ocean Circulation and Climate
Experiment (SPICE). J. Geophys. Res. 119, 2642–2657 (2014).
9. Gordon, A. L. et al. The Indonesian Throughflow during 2004–2006 as observed
by the INSTANT program. Dyn. Atmos. Oceans 50, 115–128 (2010).
10. McCreary, J. P. & Lu, P. On the interaction between the subtropical and the
equatorial ocean circulations: the subtropical cell. J. Phys. Oceanogr. 24,
466–497 (1994).
11. Gu, D. F. & Philander, S. G. H. Interdecadal climate fluctuations that depend on
exchanges between the tropics and extratropics. Science 275, 805–807 (1997).
12. Stommel, H. & Yoshida, K. (eds) Kuroshio – Its Physical Aspects (Univ. Tokyo Press,
1972).
13. Gordon, A. L., Flament, P., Villanoy, C. & Centurioni, L. The nascent Kuroshio of
Lamon Bay. J. Geophys. Res. 119, 4251–4263 (2014).
14. Burrage, D. Naming a western boundary current from Australia to the Solomon
Sea. CLIVAR Newsl. Exchanges 58, 28, http://www.clivar.org/node/238 (2012).
15. Kessler, W. S. & Gourdeau, L. The annual cycle of circulation of the south-west
subtropical Pacific,analysed in an ocean GCM.J. Phys. Oceanogr. 37, 1610–1627
(2007).
16. Lindstrom, E. et al. The western equatorial Pacific Ocean circulation study. Nature
330, 533–537 (1987).
17. Gordon, A. & Fine, R. Pathways of water between the Pacific and Indian oceans in
the Indonesian seas. Nature 379, 146–149 (1996).
18. Fine, R. A., Lukas, R., Bingham, F. M., Warner, M. J. & Gammon, R. H. The western
equatorial Pacific is a water mass crossroads. J. Geophys. Res. 99, 25063–25080
(1994).
19. Johnson, G., Sloyan, B., Kessler, W. & McTaggart, K. Direct measurements of
upper ocean currents and water properties across the tropical Pacific during the
1990s. Prog. Oceanogr. 52, 31–61 (2002).
20. Gouriou, Y. & Toole, J. Mean circulation of the upper layers of the western
equatorial Pacific Ocean. J. Geophys. Res. 98, 22495–22520 (1993).
21. Hu, D. & Cui, M. The western boundary current of the Pacific and its role in the
climate. Chin. J. Oceanology Limnol. 9, 1–14 (1991).
22. Qu, T., Kagimoto, T. & Yamagata, T. A subsurface countercurrent along the east
coast of Luzon. Deep Sea Res. Part I 44, 413–423 (1997).
23. Hu, D. et al. Direct measurements of the Luzon Undercurrent. J. Phys. Oceanogr.
43, 1417–1425 (2013).
24. Lukas, R. et al. Observations of the Mindanao Current during the Western
Equatorial Pacific Ocean Circulation study (WEPOCS). J. Geophys. Res. 96,
7089–7104 (1991).
25. Qu, T. & Lindstrom, E. J. Northward Intrusion of Antarctic Intermediate Water in
the Western Pacific. J. Phys. Oceanogr. 34, 2104–2118 (2004).
26. Zhang, L., Hu, D., Hu, S., Wang, F. & Yuan, D. Mindanao Current/Undercurrent
measured by a subsurface mooring. J. Geophys. Res. Oceans 119, 3617–3628
(2014).
Confirms the existence of the Mindanao Undercurrent and reveals its strong
60–80-day variability using observations spanning two years.
a
b
−80
−60
−40
−20
0
20
40
60
−10
−5
0
5
10
15
21 13 16 14 18 16 20 13 21 20 19 20 19 17 16
Average transport (Sv)Transport change (Sv)
EAC extension (35°–40° S)
EAC (maximum)
EAC northern (20°–25° S)
SEC (170° E)
NQC (12.5°–17.5° S)
SECC (170° E)
NGCUC (2°–10° S)
EUC (170° E)
ITF
MC (5°–10° N)
NECC (200° E)
NEC (170° E)
KC southern (15°–25° N)
KC (maximum)
KC northern (32°–37° N)
ACCESS1-0
NorESM1-M
MRI-ESM1
MRI-CGCM3
MPI-ESM-MR
MIROC-ESM
MIROC-ESM-CHEM
IPSL-CM5A-MR
IPSL-CM5A-LR
HadGEM2-ES
HadGEM2-CC
GISS-E2-R
GFDL-ESM2M
GFDL-ESM2G
GFDL-CM3
CanESM2
CSIRO-Mk3-6-0
CNRM-CM5
CESM1-CAM5
CCSM4
ACCESS1-3
Figure 5
|
Modelled transports of annual-mean Pacific WBCs and their
projected changes. a, Historical mean transport averaged over the twentieth
century (positive values indicate northward, and negative, southward
transports), and b, projected change from the historical mean, averaged over
the 2050–2100 period under RCP8.5, for selected Pacific currents (listed at
bottom). Positive changes of positive mean transport indicate strengthening,
and vice versa. Shown are individual models (markers; see key at top), multi-
model mean (horizontal black lines) and multi-model medians (horizontal red
lines). Values at the top of each column in b indicate the number of models (out
of a total of 21) that agree on the sign of the change. Based on a binomial
distribution, agreement of 13, 14, 15 and 16 models (out of 21) corresponds to
significance levels of 81%, 90.5%, 96% and 99%, respectively. Mean EUC
transports for IPSL-CM5A-MR/LR and CNRM-CM5 are greater than 90 Sv,
and are not shown in a.
RESEARCH REVIEW
G
2015 Macmillan Publishers Limited. All rights reserved
306 | NATURE | VOL 522 | 18 JUNE 2015
27. Wijffels, S., Firing, E. & Toole, J. The mean structure and variability of the
Mindanao Current at 8uN. J. Geophys. Res. 100, 18421–18435 (1995).
28. Kashino, Y., Ishida, A. & Kuroda, Y. Variability of the Mindanao Current: mooring
observation results. Geophys. Res. Lett. 32, L18611 (2005).
29. Firing, E., Kashino, Y. & Hacker, P. Energetic subthermocline currents observed
east of Mindanao. Deep Sea Res. Part II 52, 605–613 (2005).
30. Kessler, W. S. & Cravatte, S. ENSO and short-term variability of the south
equatorial current entering the Coral Sea. J. Phys. Oceanogr. 43, 956–969 (2013).
31. Davis, R. E., Kessler, W. S. & Sherman, J. T. Gliders measure western boundary
current transport from the South Pacific to the equator. J. Phys. Oceanogr. 42,
2001–2013 (2012).
32. Gasparin, F., Ganachaud, A., Maes, C., Marin, F. & Eldin, G. Oceanic transports
through the Solomon Sea: the bend of the New Guinea Coastal Undercurrent.
Geophys. Res. Lett. 39, L15608 (2012).
A hydrographic survey across the southern Solomon Sea allows a complete
estimate of the NGCU transports down to 2,000 m, as well as counter currents,
inflows and outflows between the Solomon and Coral Seas.
33. Qu, T. & Lindstrom, E. A climatological interpretation of the circulation in the
western South Pacific. J. Phys. Oceanogr. 32, 2492–2508 (2002).
34. Qu, T. & Lukas, R. The bifurcation of the North Equatorial Current in the Pacific. J.
Phys. Oceanogr. 33, 5–18 (2003).
First report of the findings of vertical distribution and seasonal variation of the
NEC bifurcation,withstrong relevance to seasonal variationof the South China
Sea throughflow in the Luzon Strait found by later studies.
35. Qiu, B. & Lukas, R. Seasonal and interannual variability of the North Equatorial
Current, the Mindanao Current and the Kuroshio along the Pacific western
boundary. J. Geophys. Res. 101, 12315–12330 (1996).
36. Ridgway, K. R. Long term trend and decadal variability of the southward
penetration of the East Australia Current. Geophys. Res. Lett. 34, L13613 (2007).
37. Roemmich, D. et al. Decadal spinup of the South Pacific Subtropical Gyre. J. Phys.
Oceanogr. 37, 162–173 (2007).
38. Beal, L. M. On the role of the Agulhas system in ocean circulation and climate.
Nature 472, 429–436 (2011).
39. Qiu, B., Mao, M. & Kashino, Y. Intraseasonal variability in the Indo-Pacific
throughflow and the regions surrounding the Indonesian seas. J. Phys. Oceanogr.
29, 1599–1618 (1999).
40. Kim, Y. et al. Seasonal and interannual variations of the North Equatorial Current
bifurcation in a high-resolution OGCM. J. Geophys. Res. 109, C03040 (2004).
41. Kashino, Y. et al. Observations of the North Equatorial Current, Mindanao
Current, and the Kuroshio Current system during the 2006/7 El Nin
˜
o and 2007/
08 La Nin
˜
a. J. Oceanogr. 65, 325–333 (2009).
42. Qiu, B. & Chen, S. Interannual-to-decadal variability in the bifurcation of the north
equatorial current off the Philippines. J. Phys. Oceanogr. 40, 2525–2538 (2010).
Shows a decadal modulation in the characteristics of the NEC bifurcation,
which is determined by wind forcing in the 12
6
–14
6
N band that contains
variability not fully representable by the Nin
˜
o-3.4 index.
43. Wu, C.-R. Interannual modulation of the Pacific Decadal Oscillation (PDO) on the
low-latitude western North Pacific. Prog. Oceanogr. 110, 49–58 (2013).
44. Qiu, B., Kessler, W. S. & Chen, S. Source of the 70-day mesoscale eddy variability
in the Coral Sea and North Fiji Basin. J. Phys. Oceanogr. 39, 404–420 (2009).
45. Ueki, I., Kashino, Y. & Kuroda, Y. Observation of current variations off the New
Guinea coast including the 1997–1998 El Nin
˜
o period and their relationship
with Sverdrup transport. J. Geophys. Res. 108 (C7), 3243 (2003).
46. Jin, F. F. An equatorial ocean recharge paradigm for ENSO. Part I: conceptual
model. J. Atmos. Sci. 54, 811–829 (1997).
47. Melet, A., Gourdeau, L., Verron, J. & Djath, B. Solomon Sea circulation and water
mass modifications: response at ENSO timescales. Ocean Dyn. 63, 1–19 (2013).
48. Sprintall, J. et al. The Indonesian seas and their role in the coupled ocean–climate
system. Nature Geosci. 7, 487–492 (2014).
Reports recent progress in our understanding of oceanography in the
Indonesian Seas and their climatic impact through variations with ENSO,
focusing on ocean heat content, sea level, winds and precipitation in the
tropical Indian Ocean region.
49. van Sebille, E. et al. Pacific-to-Indian Ocean connectivity: Tasman leakage,
Indonesian Throughflow, and the role of ENSO. J. Geophys. Res. Oceans 119,
1365–1382 (2014).
50. Gordon, A. L. et al. South China Sea throughflow impact on the Indonesian
throughflow. Geophys. Res. Lett. 39, L11602 (2012).
51. Qu, T., Yan, D. & Hideharu, S. SouthChina Sea throughflow: a heat and freshwater
conveyor. Geophys. Res. Lett. 33, L23617 (2006).
52. Hill, K. L., Rintoul, S. R., Ridgway, K. R. & Oke, P. R. Decadal changes in the South
Pacific western boundary current system revealed in observations and ocean
state estimates. J. Geophys. Res. 116, C01009 (2011).
53. Vecchi, G. A. et al. Weakening of tropical Pacific atmospheric circulation due to
anthropogenic forcing. Nature 441, 73–76 (2006).
54. Kelly, K. A. et al. Western boundary currents and frontal air–sea interaction: Gulf
Stream and Kuroshio Extension. J. Clim. 23, 5644–5667 (2010).
55. Kwon, Y. O. et al. Role of the Gulf Stream and Kuroshio–Oyashio systems in large-
scale atmosphere–ocean interaction: a review. J. Clim. 23, 3249–3281 (2010).
56. Konda, M. H., Ichikawa, H., Tomita, H. & Cronin, M. F. Surface heat flux variations
across the Kuroshio Extension as observed by surface flux buoys. J. Clim. 23,
5206–5221 (2010).
57. Chelton, D. B., Schlax, M., Freilich, M. & Milliff, R. Satellite measurements reveal
persistent small-scale features in ocean winds. Science 303, 978–983 (2004).
58. Tokinaga, H. et al. Ocean frontal effects on the vertical development of clouds
over the western North Pacific: in situ and satellite observations. J. Clim. 22,
4241–4260 (2009).
59. Hotta, D. & Nakamura, H. On the significance of the sensible heat supply from the
ocean in the maintenance of the mean baroclinicity along storm tracks. J. Clim.
24, 3377–3401 (2011).
60. Wu, C.-R. et al. Air-sea interaction between tropical cyclone Nari and Kuroshio.
Geophys. Res. Lett. 35, L12605 (2008).
61. Sasaki, n., Minobe, S., Asai, T. & Inatsu, M. Influence of the Kuroshio in the East
China Sea on the early summer (baiu) rain. J. Clim. 25, 6627–6645 (2012).
62. Huang, R. & Li, W. Influence of the heat source anomaly over the tropical western
Pacific on the subtropical high over East Asia and its physical mechanism.
Chinese J. Atmos. Sci. 14, 95–107 (1988).
63. Feng, J. & Hu, D. How much does heat content of the western tropical Pacific
Ocean modulate the South China Sea summer monsoon onset in the last four
decades? J. Geophys. Res. Oceans 119, 4029 (2014).
64. Holland, G. J. Interannual variability of the Australian summer monsoon at
Darwin: 1952–82. Mon. Weath. Rev. 114, 594–604 (1986).
65. Gordon, A. L., Susanto, R. D. & Vranes, K. Cool Indonesian throughflow as a
consequence of restricted surface layer flow. Nature 425, 824–828 (2003).
Shows that a stronger low-salinity South China Sea throughflow during boreal
winter sets up a gradient, which limits the Mindanao Current inflow into the
Indonesian Seas and thus leads to cold surface water in the Indonesian Seas.
66. Seager, R., Kushnir, Y., Naik, N. H., Cane, M. A. & Miller, J. Wind-driven shifts in the
latitude of the Kuroshio–Oyashio Extension and generation of SST anomalies on
decadal timescales. J. Clim. 14, 4249–4265 (2001).
67. Lau, N.-C. & Nath, M. J. Impact of ENSO on SST variability in the North Pacific and
NorthAtlantic: seasonaldependence and role of extratropical sea–aircoupling. J.
Clim. 14, 2846–2866 (2001).
68. Alexander, M. A. The atmospheric bridge: the influence of ENSO teleconnections
on air–sea interaction over the global oceans. J. Clim. 15, 2205–2231 (2002).
69. Nakamura, H. & Kazmin, A. S. Decadal changes in the North Pacific oceanic
frontal zones as revealed in ship and satellite observations. J. Geophys. Res. 108,
3078 (2003).
70. Taguchi, B. et al. Decadal variability of the Kuroshio Extension: observations and
an eddy-resolving model hindcast. J. Clim. 20, 2357–2377 (2007).
71. Frankignoul, C. & Senne
´
chael, N. Observed influence of North Pacific SST
anomalies on the atmospheric circulation. J. Clim. 20, 592–606 (2007).
72. Taguchi, B., Nakamura, H., Nonaka, M. & Xie, S.-P. Influences of the Kuroshio/
Oyashio Extensions on air–sea heat exchanges and storm-track activity as
revealed in regional atmospheric model simulations for the 2003/04 cold
season. J. Clim. 22, 6536–6560 (2009).
73. Taguchi, B., Xie, S.-P., Mitsudera, H. & Kubokawa, A. Response of the Kuroshio
Extension to Rossby waves associated with the 1970s climate regime shift in a
high-resolution ocean model. J. Clim. 18, 2979–2995 (2005).
74. Liu, Z. & Alexander, M. Atmospheric bridge, oceanic tunnel and global climatic
teleconnections. Rev. Geophys. 45, RG2005 (2007).
75. Tokinaga, H. et al. Atmospheric sounding over the winter Kuroshio Extension:
effect of surface stability on atmospheric boundary layer structure. Geophys.Res.
Lett.
33, L04703 (2006).
76. Norris, J. R. & Leovy, C. B. Interannual variability in stratiform cloudiness and sea
surface temperature. J. Clim. 7, 1915–1925 (1994).
77. Klein, S. A., Hartmann, D. L. & Norris, J. R. On the relationships among low-cloud
structure, sea surface temperature and atmospheric circulation in the
summertime northeast Pacific. J. Clim. 8, 1140–1155 (1995).
78. Park, S., Alexander, M. A. & Deser, C. The impact of cloud radiative feedback,
remote ENSO forcing, and entrainment on the persistence of North Pacific sea
surface temperature anomalies. J. Clim. 19, 6243–6261 (2006).
79. Alexander, M. A. & Deser, C. A mechanism for the recurrence of wintertime
midlatitude SST anomalies. J. Phys. Oceanogr. 25, 122–137 (1995).
80. Sugimoto, S. & Hanawa, K. Remote reemergence areas of winter sea surface
temperature anomalies in the North Pacific. Geophys. Res. Lett. 32, L01606
(2005).
81. Nakamura, H. & Yamagata, T. in Beyond El Nino: Decadal and Interdecadal Climate
Variability (ed. Navarra, A.) 49–72 (Springer, 1999).
82. Sprintall, J., Roemmich, D., Stanton, B. & Bailey, R. Regional climate variability
and ocean heat transport in the southwest Pacific Ocean. J. Geophys. Res. 100,
15865–15871 (1995).
83. Shi, G., Ribbe, J., Cai, W. & Cowan, T. An interpretation of Australian rainfall
projections. Geophys. Res. Lett. 35, L02702 (2008).
84. Godfrey, S. The effect of the Indonesian Throughflow on ocean circulation and
heat exchange with the atmosphere: A review. J. Geophys. Res. Oceans 101,
12217–12237 (1996).
85. Vranes, K., Gordon, A. L. & Field, A. The heat transport of the Indonesian
throughflow and implications for the Indian Ocean heat budget. Deep Sea Res.
Part II 49, 1391–1410 (2002).
86. Gordon, A. L. Oceanography of the Indonesian Seas and their throughflow.
Oceanography (Wash. D.C.) 18, 14–27 (2005).
87. Song, Q., Gordon, A. L. & Visbeck, M. Spreading of the Indonesian throughflow in
the Indian Ocean. J. Phys. Oceanogr. 34, 772–792 (2004).
88. Wu, L. et al. Enhanced warming over the global subtropical western boundary
current. Nature Clim. Change 2, 161–166 (2012).
Demonstrates synchronized enhanced warming along subtropical WBCs and
emphasizes the important roles played by the WBCs in global climate change.
89. Zhai, F., Hu, D., Wang, Q. & Wang, F. Long-term trend of Pacific South Equatorial
Current bifurcation. Geophys. Res. Lett. 41, 3172–3180 (2014).
90. Chen, Z. & Wu, L. Long-term change of the Pacific North Equatorial Current
bifurcation in SODA. J. Geophys. Res. 117, C06016 (2012).
REVIEW RESEARCH
G
2015 Macmillan Publishers Limited. All rights reserved
18 JUNE 2015 | VOL 522 | NATURE | 307
91. Cai, W. & Cowan, T. Trends in Southern Hemisphere circulation in IPCC AR4
models over 1950–1999: ozone-depletion vs greenhouse forcing. J. Clim. 20,
681–693 (2007).
92. Cai, W., Shi, G., Cowan, T., Bi, D. & Ribbe, J. The response of the Southern Annular
Mode, the East Australian Current, and the southern mid-latitude ocean
circulation to global warming. Geophys. Res. Lett. 32, L23706 (2005).
93. Thompson, D. W. J. et al. Signatures of the Antarctic ozone hole in Southern
Hemisphere surface climate change. Nature Geosci. 4, 741–749 (2011).
94. Wang, G., Cai, W. & Purich, A. Trends in Southern Hemisphere wind driven
circulation in CMIP5 models over the 21st century: ozone recovery versus
greenhouse forcing. J. Geophys. Res. Oceans 119, 2974–2986 (2014).
95. Ganachaud, A. et al. Projected changes in the tropical Pacific Ocean of
importance to tuna fisheries. Clim. Change 119, 163–179 (2013).
Presents global warming projections showing a shift in the Pacific winds and
surface temperatures that affect oceanic currentsand vertical ocean structure,
enhancing stratification and increasing the extent of the warm pool with
consequences for tuna habitat.
96. Sen Gupta, A., Ganachaud, A., McGregor, S., Brown, J. N. & Muir, L. Drivers of the
projected changes to the Pacific Ocean equatorial circulation. Geophys. Res. Lett.
39, L09605 (2012).
97. Grenier, M., Jeandel, C. & Cravatte, S. From the subtropics to the equator in the
Southwest Pacific: continental material fluxes quantified using neodymium data
along modelled thermocline water pathways. J. Geophys. Res. Oceans 119,
3948–3966 (2014).
98. Gregory, J. M. et al. A model intercomparison of changes in the Atlantic
thermohaline circulation in response to increasing atmospheric CO
2
concentration. Geophys. Res. Lett. 32, L12703 (2005).
99. Li, W., Li, L., Ting, M. & Liu, Y. Intensification of Northern Hemisphere subtropical
highs in a warming climate. Nature Geosci. 5, 830–834 (2012).
100. Oliver, E. C. J. & Holbrook, N. J. Extending our understanding of South Pacific gyre
‘‘spin-up’’: modeling the East Australian Current in a future climate. J. Geophys.
Res. Oceans 119, 2788–2805 (2014).
Shows that an increase in the EAC extension of 4–5 Sv by 2060 simulated by a
high-resolution model closely matches results from a low-resolution climate
model.
Acknowledgements W.C. and G.W. are supported by the Australian Climate Change
Science Program, a CSIRO Office of the Chief Executive Science Leader award, and
CSIRO Office of the Chief Executive postdoctoral awards. L.W., Z.C. and X.L. are
supported by projects (41130859,41490640, 41306001) of the National Science
Foundation of China (NSFC), and a project (2013CB956200) of the National Basic
Research Program of China (MOST). D.H. is supported by CAS Program XDA
11010101, and NSFC Grants 41330963 and 41421005. S.H. is supported by NSFC
Grant41406016. Q.W.issupported by MOST Grant 2013CB956202.F.W.issupported
by MOST Grant 2012CB417401 and NSFC/Shangdong Grant U1406401, A.G. is
supported by CNRS/INSU/LEFE project MoorSPICE. This is PMEL Contribution
Number 4207, and Lamont-Doherty Earth Observatory Contribution Number 7875.
J.S. is supported by the National Aeronautics and Space Administration (NASA) under
award no. NNX13AO38G. Y.K. is supported by the Tropical Ocean Climate Study of
Japan Agency for Marine-Earth Science and Technology. This is a contribution to the
CLIVAR SPICE and NPOCE programmes. We thank A. Purich and T. Cowan for their
comments before submission. We acknowledge the World Climate Research
Programme’s Working Group on Coupled Modelling, and we thank the climate
modelling groups for producing and making available their model output. The US
Department of Energy’s Program for Climate Model Diagnosis and Intercomparison
provides coordinating support and led the development of software infrastructure in
partnership with the Global Organization for Earth System Science Portals.
Author Contributions D.H., L.W. and W.C. conceived the study. L.W., W.C. and D.H.
determined the scope. W.C. wrote the draft of the paper and finalized the manuscript
with help from G.W. A.S.G. conducted model output analysis for future projections and
plotted Fig. 5. A.G., A.S.G. and W.C. constructed the schematic of Figs 1 and 3. Z.C.
generated Fig. 4. All authors contributed to interpreting results, discussion of the
associated dynamics and improvement of this paper.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Readers are welcome to comment on the online version of the paper. Correspondence
and requests for materials should be addressed to W.C. (wenju.cai@csiro.au) or L.W.
(lxwu@ouc.edu.cn).
RESEARCH REVIEW
G
2015 Macmillan Publishers Limited. All rights reserved
308 | NATURE | VOL 522 | 18 JUNE 2015
