PUBLICATIONS
Journal of Geophysical Research: Oceans
RESEARCH ARTICLE
10.1002/2013JC009693
Mindanao Current/Undercurrent measured by a subsurface
mooring
Special Section:
Linlin Zhang1,2,3, Dunxin Hu1,2, Shijian Hu1,2, Fan Wang1,2, Fujun Wang1,2, and Dongliang Yuan1,2
Western Pacific Ocean Circulation and Climate
1
Key Points:
Mooring data, hydrographic data,
and model outputs suggest the
existence of MUC
The currents below 200 m show
significant intraseasonal variability of
60–80 days
Intraseasonal variability is due to
westward translating subthermocline
eddies
Correspondence to:
D. Hu,
[email protected]
Citation:
Zhang, L., D. Hu, S. Hu, F. Wang,
F. Wang, and D. Yuan (2014),
Mindanao Current/Undercurrent
measured by a subsurface mooring, J.
Geophys. Res. Oceans, 119, 3617–3628,
doi:10.1002/2013JC009693.
Received 17 DEC 2013
Accepted 22 MAY 2014
Accepted article online 24 MAY 2014
Published online 9 JUN 2014
This article was corrected on
10 JUL 2014. See the end of the full
text for details.
Institute of Oceanology, Chinese Academy of Sciences, Qingdao, China, 2Key Laboratory of Ocean Circulation and Waves,
Chinese Academy of Sciences, Qingdao, China, 3International Pacific Research Center, SOEST, University of Hawai’i at
Monoa, Honolulu, Hawaii, USA
Abstract The mean structure and variability of the currents east of Mindanao are investigated through 2
year mooring observations at about 8 N, 127 3’E from December 2010 to December 2012. The strong
southward Mindanao Current (MC) exists in the upper 600 m with a maximum mean velocity of 73 cm/s
and a standard deviation of 17 cm/s at 100 m. A northward mean flow is observed below 600 m to the
depth deeper than 1000 m, which has been called the Mindanao Undercurrent (MUC) with a maximum
mean velocity of about 10 cm/s at 950 m and a standard deviation of 19 cm/s. Further analysis with hydrographic data and an eddy-resolving model outputs also suggests this northward mean current to be the
MUC. Intraseasonal variability with a period of 60–80 days is revealed through the whole water column
from 200 m down to about 900 m. This intraseasonal variability appears to be closely related to subthermocline eddies, which translate westward and intensify near the Mindanao coast.
1. Introduction
The low-latitude western boundary currents in the north Pacific are of particular importance for ocean dynamics
and climate in view of their roles in the mass/heat partition between the tropical and subtropical gyres, and in
the heat budget of the warm pool [e.g., Lukas et al., 1996; Hu et al., 2011]. As two components of the lowlatitude western boundary currents in the tropical north Pacific, the southward Mindanao Current (MC) in the
upper layer [e.g., Nitani, 1972; Lukas, 1988; Lukas et al., 1991; Wijffels et al., 1995; Kashino et al., 2005] and the
northward Mindanao Undercurrent (MUC) in the lower layer [Hu and Cui, 1989, 1991; Hu et al., 1991; Wang and
Hu, 1998a, 1998b; Lukas et al., 1991; Qu et al., 1998, 2012] flow along the east coast of the Mindanao Island.
Many studies have been focused on the characteristics and variability of the MC with hydrographic observations, sea level time series and numerical models. The MC is a stable coastal jet with a maximum speed of
about 1 m/s [e.g., Lukas, 1988; Lukas et al., 1991; Wijffels et al., 1995]. The MC transport is believed to be
~o and Southern Oscillation (ENSO) [e.g., Kim et al., 2004; Kashino et al., 2009, 2011], and
related to the El Nin
influenced by a strong, intermittent quasi-biennial signal [e.g., Lukas, 1988; Qiu and Lukas, 1996]. Recent
studies also revealed existence of annual and semiannual signals in the MC transport [e.g., Tozuka et al.,
2002; Qu et al., 2008]. On the basis of mooring observation from October 1999 to July 2002, Kashino et al.
[2005] reported that the MC has a subsurface velocity maximum exceeding 1.3 m/s at approximately 100m
depth. They also noted that the observed interannual, seasonal, and intraseasonal variations were of comparable amplitudes, and the MC was strong during boreal summer and during the onset of El Nino.
Compared with the well-known MC, the MUC is still poorly understood. Based on three CTD transects off
Mindanao near 7.5 N, multiple northward velocity cores were found below 500 m and named MUC by Hu
and Cui [1989, 1991]. The maximum velocity of the MUC is around 20 cm/s with a transport of 8–22 Sv [Hu
et al., 1991]. This undercurrent structure was visible in the acoustic Doppler current profile (ADCP) measurements and CTD sections collected during the Western Equatorial Pacific Ocean Circulation Study (WEPOCS)
[Hacker et al., 1989; Lukas et al., 1991], showing that the MUC is 20–25 km wide and 50–75 km offshore.
They also appeared in recent shipboard ADCP sections along 7 N and 8 N [Kashino et al., 2009]. Several
analyses of hydrographic data further suggested that, although its intensity and locality vary from time to
time, the MUC is a quasi-permanent feature [e.g., Wang and Hu, 1998a, 1998b; Qu et al., 1998]. Recent studies with an eddy-resolving model also show the northward MUC within a depth range between 400 and
1000 m below the MC, though its variance is larger than the mean current [e.g., Qu et al., 2012].
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3617
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
However, based on seven shipboard ADCP transects in the
upper 350 m between 1987 and
1990, Wijffels et al. [1995] noted
that there was only a weak mean
flow offshore and below the MC,
and speculated that the MUC
might be a transient phenomenon. Obviously, their ADCP with
upper 350 m range limit could
not well reach the MUC which is
generally below 400 m. Using
lowered and shipboard ADCP
measurements from two cruises,
Firing et al. [2005] observed a
southward flow along the Mindanao coast and a northward flow
100–200 km offshore below
500 m, and suggested that they
seemed to be subthermocline
Figure 1. Location of the 8 N mooring (purple star). Color map shows the topography
eddies. Based on current meter
east of Philippines from etopo5. Contours: mean sea surface height (cm) during 1992–
2012 from AVISO. Black and yellow arrows indicate the Mindanao Current (MC) and Mindmeasurements at 700 m at
anao Undercurrent (MUC), respectively.
6 500 N/126 430 E during October
1999 and July 2002, Kashino et al.
[2005] found that intraseasonal variability of 50–100 days was dominant and there was nearly no significant
MUC at that level. They suspected that the MUC did not flow at that depth of the mooring site and it might
be present in a deeper layer or further offshore.
Despite the studies listed above, several questions remain. Is the MUC a permanent current? As many studies showed large variability in the current below the MC, what are the detailed features of this variability?
What is the source of this variability? Due to a lack of continuous direct measurements, there are still no definite answers to these questions.
As a part of the field experiment of the Northwestern Pacific Ocean Circulation and Climate Experiment
(NPOCE) program [Hu et al., 2011], a subsurface ADCP mooring was deployed at 8 N, 127 3’E east of Mindanao Island during December 2010 to December 2012. The mooring provides 2 year velocity data between 0
and 800 m and CTD measurements at some fixed depths. With these measurements, we examined the
mean structure and variability of the currents east of Mindanao in this study. The existence of the mean
northward MUC is argued to be true even with large variance, and the intraseasonal variability of the MC/
MUC is also illustrated. The source of this intraseasonal variability is further explored with an eddy resolving
model.
2. Data
2.1. ADCP Mooring
A subsurface mooring was deployed at about 8 N, 127 3’E east of Mindanao Island to measure the MC/
MUC, and the water depth is about 6100 m (Figure 1). The site of 8 N, 127 3’E was selected based on the
previously calculated geostrophic velocity field from repeated hydrographic transects, generally near the
inshore core of the MUC [Hu and Cui, 1989, 1991; Qu et al., 1998].
The observation period was from 1 December 2010 to 7 December 2012, with one time interval on 15–16
July 2011 for retrieval and redeployment of the mooring. An upward-looking and a downward-looking
Acoustic Doppler Current Profilers (75 kHz ADCP, TRDI) were mounted on the mooring at about 400 m.
There were also other instruments such as current meters and SBE 37 CTDs attached on the mooring at various depths, but we primarily focus on the 2 year ADCP measurements in this study, and CTD measurements
at about 600 m are also used. The ADCP measured velocity every 1 hour in 60 bins with a bin size of 8 m,
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3618
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 2. ADCP-measured mean velocity with standard deviation: (a) meridional, (b) zonal. Red indicates the mean velocity, and blue the
standard deviation of velocity. (c) Number of days with ADCP measurements at different depths.
and the sampling depth of the two ADCPs ranges from the sea surface to about 800 m. Due to the relatively
less buoyancy in the first observation period (December 2010 to July 2011), the main float which was
designed at 400 m sometimes got down to 800 m when the subsurface currents became stronger, therefore
the downward looking ADCP mounted on the main float could measure the currents at 1200 m sometimes
before July 2011, and at the same time the upward looking ADCP missed the currents in the upper 500 m.
The ADCP measured hourly velocity data were first interpolated vertically onto standard depths of 1 m vertical resolution, and then were daily averaged to remove tidal signals. The SBE 37 CTD at about 600 m measures the conductivity, temperature, and pressure every 10 min. The data are first calibrated onto the fixed
depth of 600 m, and then averaged to get the daily data, which are finally used in the following analysis.
2.2. OFES Output
To investigate the source of the intraseasonal variability of the currents east of Mindanao revealed by the
mooring measurements, we also used outputs from the OGCM for the Earth Simulator (OFES). Its domain
covers a near-global region extending from 75 S to 75 N, with a horizontal resolution of 0.1 both in the
longitude and latitude. The vertical resolution varies from 5 m near the surface to 330 m near the bottom,
with a total of 54 levels. The 3 day snapshot model outputs for the period 2010–2012 are used in this study.
Detailed descriptions about this model can be found in Masumoto et al. [2004] and Sasaki et al. [2008].
3. Results
3.1. Mean Currents of MC/MUC
Figure 2a shows the vertical profile of the mean meridional velocities measured by the ADCPs during the
whole mooring period. The MC is characteristic of a strong southward flow, which is located in the upper
600 m with a mean velocity exceeding 73 cm/s at a depth of about 100 m, much weaker than the one
reported by Kashino et al. [2005]. This difference may be attributed to different locations. As Lukas et al.
[1991] showed, the MC accelerates southward along the Philippine coast. The standard deviation of the MC
at 100 m is about 17 cm/s, which is much smaller than the mean flow, indicating that the MC is a very stable
current. A northward mean flow is observed below 600 m to the depth deeper than 1000 m, which has
been called the Mindanao Undercurrent (MUC). Its maximum mean velocity is around 10 cm/s at 950 m
with a standard deviation of about 19 cm/s (Figure 2a).
It should be noted that the averaging time range in obtaining the mean velocity is different at different
depths. Figure 2c displays the number of days with effective ADCP measurements at various depths. Above
780 m, the effective measurement time range at most depths is longer than 600 days, while below 780 m it
decreases with the depth, 400 days at 800 m and 200 days at 1000 m. At 950 m, there are about 250 days
of ADCP measurements. Considering that the velocity at this depth varies with a dominant period of 60–80
days as shown in the following section, the ADCP measurements provide only about 3 degrees of freedom,
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3619
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 3. Mean meridional geostrophic velocity relative to 1500 m at 8 N calculated from CTD and Argo profiles of NODC. White contour
indicates the zero-line of velocity and black line denotes the mooring site.
and the standard error of the mean current reaches 11 cm/s, therefore longer time series is needed to test
the significance of this mean current.
The mean zonal current from the ADCP measurements is shown in Figure 2b, which is eastward in the
upper 400 m with a maximum mean velocity around 5 cm/s at about 150 m. Between 400 and 1000 m, it is
westward with a mean velocity of 3 cm/s at 950 m. Then the above analysis couches that the MUC is a
northwestward subsurface flow, while the MC is southeastward in the upper layer. Similar to the meridional
counterpart, the standard deviation of the zonal velocity is also very large, about 12 cm/s at 950 m.
Many previous studies have presented the geostrophic currents along the section of 8 N based on hydrographic observations with different values of the MC/MUC during different periods [e.g., Hu and Cui, 1989,
1991; Hacker et al., 1989; Lukas et al., 1991; Kashino et al., 1996, 2009; Qu et al., 1998; Wang and Hu, 1998a,
1998b]. To obtain a relatively reasonable velocity structure of the MC/MUC, we collected all hydrographic
data available near 8 N so far from NODC, including CTD and Argo profiles, and calculated the mean geostrophic velocities in Figure 3. A strong southward MC is shown in the upper 500 m with a maximum velocity of 96 cm/s. Below the MC, a northward flow of the MUC is displayed with two cores. Obviously, the
subsurface mooring is located not exactly at, but quite near the inshore core of the MUC. The axis of the
inshore core is near 126.8 E centered at about 900 m, consistent with the mooring result, while its velocity
is about 4 cm/s, much weaker than the mooring measured. The offshore core is located at 127.7 E with a
maximum velocity of about 14 cm/s at the depth of 550 m. This multicore structure of the MUC has been
reported by many previous studies based on hydrographic sections [Hu and Cui, 1989; Wang and Hu, 1998a;
Qu et al., 1998] and model outputs [Qu et al., 2012].
It should be pointed out that the mean MUC measured by the mooring ADCP in Figure 2 is much weaker
and deeper than the previous results provided by Hu and Cui [1989] and Wang and Hu [1998a]. One apparent reason for this difference is that Figure 2 shows the mean current during the 2 year observation period
of the mooring ADCP, while previous studies were merely based on synoptic CTD sections, which were
influenced strongly by subthermocline eddies as shown in the following section. In addition, different
observation time may also contribute to the difference.
A 2 year time series of meridional velocities measured by the ADCP is shown in Figures 4. In the upper
400 m there is a strong and stable southward flow during the entire period. The maximum velocity is up to
100 cm/s in July 2011 at about 100 m. Below 400 m, there is a northward flow, which is strongly interfered
by the intraseasonal variability. Sometimes it is very strong with a velocity over 40 cm/s, but sometimes it is
weak and even reverse, which will be discussed in the next section.
3.2. Intraseasonal Variability of MC/MUC From ADCP Measurements
Power spectral density (PSD) is calculated for both meridional and zonal velocities at all depths between 0
and 800 m with the ADCP-measured data and is shown in Figure 5. Here the PSD has been normalized by
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3620
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 4. Time series of daily meridional velocity measured by the mooring ADCP during December 2010 to December 2012. Black contours indicate zero line of the velocity.
the total velocity variance at different depths. A protruding feature of the PSD is the coherent peak with a
period of 60–80 days between 200 and 800 m. In the upper 200 m, there is no peak of this period. The PSD
for the zonal currents has similar features (Figure 5b). Kashino et al. [2005] and Qu et al. [2012] also mentioned the intraseasonal variability with a broader band of 50–100 days from the current meter measurements and model outputs, respectively. The intraseasonal variability is most likely attributed to the eddy
activity in this region.
To better understand the intraseasonal variability of the currents, the ADCP-measured time series is filtered
with a 40–100 days bandpass filter at each depth (Figure 6). The filtered time series clearly demonstrates
the vertically coherent intraseasonal variability of the MC/MUC. The maximum variation takes place below
400 m with a peak-to-peak difference around 30 cm/s for the meridional component. Consistent with the
spectrum analysis, the intraseasonal variability in the upper layer is relatively weaker compared with the
deeper layer, indicating that the intraseasonal variability may be caused by subthermocline eddies, which
are nearly invisible at the sea surface [e.g., Firing et al., 2005; Qu et al., 2012]. In addition, the intraseasonal
activity appears to vary in different years, i.e., energetic in 2011 with large fluctuations and weak in 2012.
The reason for this distinct and abrupt interannual change will be explored in further studies.
Such intraseasonal variability is also captured by the CTD mounted on the mooring at about 600 m (Figure 7).
Because the main float of the mooring moved up and down strongly before July 2011 due to the strong
Figure 5. Power spectral density (PSD) of the meridional (a) and zonal (b) velocity at all depths from the mooring ADCP. The PSD has been
normalized by the total velocity variance at different depths.
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3621
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 6. Time series of 40–100 days band pass filtered meridional (a) and zonal (b) velocities from the mooring ADCP. Black contours indicate zero line of the velocity.
intraseasonal variability and the relative less buoyancy of the mooring system, the mooring CTD data may
have large interpolation error after being calibrated onto the fixed depth. Therefore, we only show the CTD
measurements during July 2011 to December 2012 (Figure 7). Both temperature and salinity measurements
exhibit large intraseasonal variations with similar peak period to that of the velocity. The correlation between
the velocity and temperature is 0.39, and that between the velocity and salinity is 0.32. The velocity anomalies
often have warm/saline or cold/fresh characteristics. A northward velocity anomaly corresponds to warm/
saline water, while a southward velocity anomaly corresponds to cold/fresh water. This probably could be
understood in the following way. Since one of the water sources of the MUC’s near-shore core is the saline
water from the southern hemisphere [Wang and Hu, 1998b], a northward velocity anomaly must be
Figure 7. (a) Vertically averaged velocity vector between 600 and 800 m during July 2011 to December 2012 from the mooring ADCP. (b)
and (c): temperature and salinity measurements at 600 m from the CTD mounted on the mooring.
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3622
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 8. Climatological mean meridional velocity along 8 N from OFES data. White contour indicates the zero line of velocity and black
line denotes the mooring site.
accompanied by warm/saline water from the south. As the North Pacific Intermediate Water with cold and
fresh water can reach this region off Mindanao from the north, a southward velocity anomaly must bring
cold/fresh water from the north.
The gross effect of these velocity anomalies on meridional heat and salinity fluxes is estimated as <v’T’>
and <v’S’> with the above time series, where v’, T’ and S’ are anomalies of the meridional velocity, temperature and salinity, respectively. The estimated northward heat flux is about 43106 W/m2, and the northward
salinity flux is about 6.731024 psu m/s. To further quantify these eddy effects, we calculated the ‘bolus’
velocity, which has been extensively used in prior studies [Marshall, 1997; Qu et al., 2012]. The ‘bolus’ velocity is defined as <v’S’>/DS, where DS represents the salinity difference between North and South Pacific
waters being converged in the western tropical Pacific. According to Qu et al. [2012], the mean DS here is
about 1631024 psu at this level. The corresponding ‘bolus’ velocity is about 0.42 m/s, much stronger than
the mean current of the MUC by a factor of 4. Therefore, subthermocline eddies here play a very important
role in the water exchange between North and South Pacific.
3.3. Intraseasonal Variability of MC/MUC From OFES
Although the ADCP data from the mooring at 8 N, 127 3’E illustrated the mean structure and intraseasonal
variability of the MC/MUC, the mooring is only one spot in space and hence insufficient for depicting the
whole picture of the MC/MUC and the source of its intraseasonal variability. Since the subthermocline intraseasonal variability is largely invisible from the sea surface, satellite altimetry is also unable to capture it.
Instead, numerical model (e.g., OFES) outputs may be used to do so. The OFES outputs have been analyzed
by a number of earlier studies. Dutrieux [2009] compared them with shipboard ADCP measurements and
suggested that the model represents reasonably well the circulation and its variability in the western tropical Pacific. Qu et al. [2012] also noted that the simulation of OFES clearly demonstrates the circulation pattern and the variability of the low-latitude western boundary currents. Even so, we still further validated this
data prior to use for this study.
3.3.1. Validation of OFES Outputs With Mooring Measurements
Figure 8 shows the mean meridional velocity along 8 N simulated by OFES. Qualitatively, the MC and MUC
are well depicted, with a formidable southward flow in the upper layer overlying weaker northward flows in
the layer of 400–1000m. This is in reasonable agreement with earlier observations [e.g., Hu and Cui, 1989,
1991; Lukas et al., 1991; Qu et al., 1998; Wang and Hu, 1998a].
To examine the ability of OFES in simulating the intraseasonal variability of the currents east of Mindanao
we compared the OFES time series of the vertical mean meridional currents between 600 and 800 m with
the mooring measurements (Figure 9a). Even though the correlation between these two time series is only
0.3, far below the 95% confidence level with about 10 degrees of freedom, both of them show obvious
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3623
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 9. (a) Vertical mean meridional velocity between 600 and 800 m at the mooring site from ADCP (red) and OFES data (blue). Mean
EKE at 605 m (b) and along 8 N (c) during 2010–2012 derived from the OFES output. The purple star in (b) and white line in (c) indicate
the mooring site.
intraseasonal variability and have roughly the same magnitude of variability around their means. Actually,
spectrum analyses by Qu et al. [2012] have suggested the OFES-simulated currents in the 200–1000 m layer
with significant intraseasonal variability of a period of 50–100 days, which is almost consistent with the
mooring measurements. In general, the OFES outputs are able to duplicate the MUC and its intraseasonal
variability, and can be used to explore the source of this intraseasonal variability.
3.3.2. Source of the MUC Intraseasonal Variability
Figures 9b and 9c show the mean EKE distribution at 605 m and along 8 N section calculated from OFES. It
can be seen that eddy activities near the Philippine coast are very strong and these eddies are amassed
between 200 and 1000m with the maximum EKE at about 500 m. These eddies are suggested to be closely
related to the subthermocline eddies reported by Firing et al. [2005] and play an important role in the intraseasonal variability of the subsurface currents east of Mindanao.
To seek the source of these subthermocline eddies, we have a time-longitude plot of the meridional current
anomalies along 8 N during 2010–2012 derived from OFES in Figure 10, which manifests a significant westward propagation feature from the east. The signals are first formed between 140 and 145 E with weak
anomalies and then get stronger when moving westward. After passing 135 E, these anomalies become
stronger and continuously move toward the Mindanao coast. It takes about 4–5 months from 135 E to the
mooring site with a speed of 6–8 cm/s, which is much smaller than the westward propagation speed
(>20 cm/s at 8 N) of the first-mode baroclinic Rossby wave [Chelton et al., 2011], but coincident well with
the mean current advection (about 6 cm/s) in this region from the OFES data. This result hints that the westward eddy translation shown in Figure 10 is probably not due to the first-mode baroclinic Rossby wave. Qiu
et al. [2013] also suggested a westward subsurface current with a speed of 1–2 cm/s in this region derived
from Argo floats. The relatively coarse resolution (0.5 ) of the data set used in their geostrophic calculation
may underestimate the strength of the current.
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3624
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 10. Time-longitude plot of the meridional velocity anomaly (color) at 605 m along 8 N during December 2010 to December 2012
from the OFES data. Black line indicates the mooring site.
Considering that those eddies are intensified west of 130 E, it is possible that there is an energy source for
subthermocline eddies near the Mindanao coast. The EKE distribution also shows locally strong eddy activity near the Mindanao coast (Figure 9). As suggested by Dutrieux [2009], most of the eddy energy might be
attributed to the mixed horizontal and vertical shear instability of subthermocline currents.
In addition, the local wind stress around the mooring site is analyzed, which indicates an intraseasonal variability around 30 days (figure not shown), inconsistent with the 60–80 days variability of the subsurface currents. That means that the intraseasonal variability is not caused by local wind. This result agrees well with
Firing et al. [2005] and Kashino et al. [2005], who suggested that the local wind forcing cannot be the energy
source of subthermocline eddies. The subthermocline eddies probably represent energy that has propagated westward from the central Pacific and/or northward along the western boundary.
Chiang and Qu [2013] proposed that subthermocline eddies east of Mindanao are from the coast of New
Guinea along a north-south dominant pathway based on the EKE analysis. The velocity vector and CTD
measurements from the 8 N mooring show a little bit of meridionally moving eddies that can be seen from
the rotation of the velocity vector (Figure 7). However, we only detected very few meridionally moving
eddies between 3 and 8 N from the OFES data. It is possible that these meridional signals are dissipated
by eddy-mean flow interactions in this region, and most of the intraseasonal signals at the mooring site
comes from the east and intensifies near the Mindanao coast.
3.4. Seasonal and Interannual Variability of MC
Beside intraseasonal variability, the 2 year time series of ADCP data also help look over the seasonal and
interannual variability of the MC. The meridional velocity averaged between 50 and 150 m shows obvious
seasonal variation of the MC accompanied by higher frequency fluctuations (Figure 11a). During the observation period, the MC is strong in summer and weak in fall. The monthly mean velocity reaches its
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3625
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 11. Vertical mean meridional velocity between 50 and 150 m (a) and 600 and 800 m (b) from the mooring ADCP measurements.
maximum of 81 cm/s in June and minimum of 56 cm/s in October. These results are closely in agreement
with the results by Kashino et al. [2005].
In addition, the MC in 2012 is weaker compared with that in 2011. Correspondingly, ENSO shifts from its
cold phase in 2011 to normal phase in 2012. Such correspondence is different from the previous result of
Kim et al. [2004] and Kashino et al. [2005] that the MC transport increased during El Nino. These differences
indicate that the interannual variation of the MC strength does not correspond to ENSO cycle all the time.
Lukas [1988] and Qiu and Lukas [1996] also suggest that interannual fluctuations of the MC are dominated
by quasi-biennial oscillations based on sea level difference between Davao and Malakal and do not always
correspond to ENSO, which seems to be in accordance with the present results.
4. Conclusion and Discussions
Based on 2 year ADCP measurements east of Mindanao Island, the mean structure and variability of the
MC/MUC are investigated. A stable southward Mindanao Current (MC) exists in the upper 600 m with the
velocity core at 100 m. A northward mean flow is observed below 600 m to the depth deeper than 1000 m,
which has been called the Mindanao Undercurrent (MUC) with a maximum mean velocity of about 10 cm/s
at 950 m and a standard deviation of 19 cm/s. Further analysis with hydrographic data and OFES outputs
also suggest this current to be the MUC. The ADCP observation reveals significant subthermocline intraseasonal variability with a period of 60–80 days below 200 m. Analysis with OFES data indicates that they are
closely related to westward translating subthermocline eddies in this region.
Through seven shipboard ADCP transects’ analysis, Wijffels et al. [1995] indicated that there was only a weak
northward mean flow offshore and below the MC, and speculated that the MUC might be a transient phenomenon. However, it, seemingly, should be noted that those ADCP transects could only measure currents
from the sea surface down to a depth of 350 m, unable to reach the MUC (its core is at about 950 m).
Based on seven SADCP sections along 7 N, Kashino et al. [2013] observed northward subsurface mean
currents below 400 m at about 127.5 E, and suggested it was a part of an anticyclonic eddy. But both
geostrophic calculations from all available hydrographic data (Figure 3) and mean velocity from OFES outputs (Figure 8) show there is a northward current west of the anticyclonic eddy, because the northward
velocity is much larger and broader than the southward one. Qu et al. [2012] also shows such kind of
northward current-MUC west of two anticyclonic eddies east of Mindanao (see their Figure 2a, Figure 3a
and Figure 4a).
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3626
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Figure 12. Meridional geostrophic velocity relative to 1500 m calculated from the in situ CTD transect along 8 N on 8 December 2012.
White contour indicates the zero velocity line and black line denotes the mooring site.
Using lowered and shipboard ADCP measurements from two cruises, Firing et al. [2005] observed a southward flow from the surface down to 2000 m along the Mindanao coast and a northward flow 100–200 km
offshore between 500 and 2000 m, and then combined with diagonal shipboard ADCP sections from the
second cruise to suggest that these flows were parts of cyclonic subthermocline eddies. This is some how
consistent with the strong intraseasonal variability of the currents below 200 m from the subsurface mooring measurements. As described above, the strong intraseasonal variability is from the westward translating
subthermocline eddies. These strong eddies heavily interfere the MUC and sometimes can make the MUC
disappear even reverse temporarily, as the case depicted by Firing et al. [2005]. Anyway, even though
strongly affected by the eddy activity, the MUC as a mean flow of about 10 cm/s is still there, as shown by
the ADCP measurements (Figure 2a), the multiyear hydrographic transects (Figure 3) and OFES outputs (Figure 8).
Acknowledgements
The authors would express their
sincere gratitude to the crew of R/V
Science 1 as well as all scientists and
technicians on board for the
deployment or retrieval of the
subsurface mooring to get these
valuable data. NODC and JAMSTEC are
also appreciated for providing the
hydrographic data and OFES data. The
authors are grateful to Eric Firing,
Tangdong Qu, and anonymous
reviewers for their insightful
comments, to Thomas Kilpatrick for his
help in editing the manuscript. This
study was supported by the Key
Project of National Natural Science
Foundation of China (grant 41330963),
the Strategic Priority Research
Program of the Chinese Academy of
Sciences (grant XDA11010101), and
the National Key Basic Research
Program of China (grants
2012CB417401 and 2013CB956202). L.
Zhang was also supported by U.S.
National Science Foundation through
grants OCE10-29704 and OCE1130050. School of Ocean and Earth
Science and Technology contribution
9127 and International Pacific
Research Center contribution IPRC1060.
ZHANG ET AL.
Notice that the current between 600 and 800 m became weak in 2012, and the currents below 800 m might
be also weak. The reason for the weakening is speculated to be related to a zonal shift of the MUC, but not
its disappearance. For instance, the northward flow could not be figured out by the mooring ADCP measurements in the upper 900 m in December 2012 (Figure 4), and then one might misunderstand it as the disappearance of the MUC, while the geostrophic velocity section calculated from the in situ CTD transect at
nearly the same time clearly shows an inshore and an offshore northward cores of the MUC, which are
away from the mooring site at this time (Figure 12).
In summary, the subsurface mooring at 8 N, 127 3’E east of Mindanao revealed some new features of the
MC/MUC. However, it is only a single point in space and unable to capture the whole picture of the MC/
MUC. As Hu and Cui [1989] mentioned, the MUC has two to three cores. The mean structure of meridional
velocity at 8 N from both geostrophic calculations (Figure 3) and OFES outputs (Figure 8) also show the
multicore structure of the MUC, while the offshore one is much stronger and shallower than the inshore
one. In this study, the mooring site seems to be located between the inshore and offshore cores, or near
the inshore one (Figures 3 and 8), therefore it may miss the major part of the MUC. For further studies on
the MC/MUC, a mooring array across the current system, especially on different cores of the MUC, is really
needed.
References
Chelton, D., M. Schlax, R. Samelson, and R. Samelson (2011), Global observations of nonlinear mesoscale eddies, Prog. Oceanogr., 91(2),
167–216.
Chiang, T., and T. Qu (2013), Subthermocline eddies in the western equatorial Pacific as shown by an eddy-resolving OGCM, J. Phys. Oceanogr., 43, 1241–1253, doi:10.1175/JPO-D-12–0187.1.
Dutrieux, P. (2009), Tropical Western Pacific currents and the origin of intraseasonal variability below the thermocline, PhD thesis, 140 pp.
Univ. of Hawaii at Manoa, Honolulu.
C 2014. American Geophysical Union. All Rights Reserved.
V
3627
Journal of Geophysical Research: Oceans
10.1002/2013JC009693
Firing, E., Y. Kashino, and P. Hacker (2005), Energetic subthermocline currents observed east of Mindanao, Deep Sea Res., Part II, 52(3-4),
605–613, doi:10.1016/j.dsr2.2004.12.007.
Hacker, P., E. Firing, R. Lukas, P. Richardson, and C. Collins (1989), Observation of the Low-latitude Western Boundary Circulation in the
Pacific during WEPOCS III, in Proceedings of Western International Meeting and Workshop on TOGA COARE, edited by J. Picaut, R. Lukas,
and T. Delcroix, pp. 135–143, Inst. Fr. de Rech. Sci. pour le Deev. en Coop., Noumeea.
Hu, D., and M. Cui (1989), The western boundary current in the far-western Pacific Ocean, in Proceedings of Western International Meeting
and Workshop on TOGA COARE, edited by J. Picaut, R. Lukas, and T. Delcroix, pp. 123–134, Inst. Fr. de Rech. Sci. pour le Deev. en Coop.,
Noumeea.
Hu, D., and M. Cui (1991), The western boundary current of the Pacific and its role in the climate, Chin. J. Oceanol. Limnol., 9(1), 1–14, doi:
10.1007/BF02849784.
Hu, D., M. Cui, T. Qu, and Y. Li (1991), A subsurface northward current off Mindanao identified by dynamic calculation, Oceanogr. Asian Marginal Seas-Elsevier Oceanogr. Ser., 54, 359–365, doi:10.1016/S0422–9894(08)70108-9.
Hu, D., et al. (2011), Northwestern Pacific Ocean Circulation and Climate Experiment (NPOCE) Science/Implementation Plan, Ocean Press,
Beijing.
Kashino, Y., M. Aoyama, T. Kawano, and N. Hendiarti (1996), The water masses between Mindanao and New Guinea, J. Geophys. Res., 101,
12,391–12,400, doi:10.1029/95JC03797.
Kashino, Y., A. Ishida, and Y. Kuroda (2005), Variability of the Mindanao Current: Mooring observation results, Geophys. Res. Lett., 32,
L18611, doi:10.1029/2005GL023880.
Kashino, Y., N. Espana, F. Syamsudin, K. Richards, T. Jensen, P. Dutrieux, and A. Ishida (2009), Observations of the North Equatorial Current,
Mindanao Current, and Kuroshio Current System during the 2006/07 El Ni~
no and 2007/08 La Ni~
na, J. Oceanogr., 65(3), 325–333, doi:
10.1007/s10872-009-0030-z.
Kashino, Y., A. Ishida, and S. Hosoda (2011), Observed ocean variability in the Mindanao Dome region, J. Phys. Oceanogr., 41(2), 287–302,
doi:10.1175/2010JPO4329.1.
Kashino, Y., A. Atmadipoera, Y. Kuroda, and Lukijanto (2013), Observed features of the Halmahera and Mindanao Eddies, J. Geophys, Res.,
118, 6543–6560, doi:10.1002/2013JC009207.
Kim, Y., T. Qu, T. Jensen, T. Miyama, H. Mitsudera, H. Kang, and A. Ishida (2004), Seasonal and interannual variations of the North Equatorial
Current bifurcation in a high-resolution OGCM, J. Geophys. Res., 109, C03040, doi:10.1029/2003JC002013.
Lukas, R. (1988), Interannual fluctuations of the Mindanao Current inferred from sea level, J. Geophys. Res., 93, 6744–6748, doi:10.1029/
JC093iC06p06744.
Lukas, R., E. Firing, P. Hacker, P. Richardson, C. Collins, R. Fine, R. Gammon (1991), Observations of the Mindanao Current during the Western Equatorial Pacific Ocean Circulation Study, J. Geophys. Res., 96, 7089–7104.
Lukas, R., T. Yamagata, and J. McCreary (1996), Pacific low-latitude western boundary currents and the Indonesian through flow, J. Geophys.
Res., 101, 12,209–12,216, doi:10.1029/96JC01204.
Marshall, D. (1997), Subduction of water masses in an eddying ocean, J. Mar. Res., 55(2), 201–222, doi:10.1357/0022240973224373.
Masumoto, Y., H. Sasaki, T. Kagimoto, N. Komori, et al. (2004), A fifty-year eddyresolving simulation of the world ocean–Preliminary outcomes of OFES (OGCM for the Earth Simulator), J. Earth Simulator, 1, 35–56.
Nitani, H. (1972), Beginning of the Kuroshio, in Kuroshio: Physical Aspects of the Japan Current, edited by H. Stommel and K. Yoshida, pp.
129–163, Univ. of Wash. Press, Seattle, Wash.
Qiu, B., and R. Lukas (1996), Seasonal and interannual variability of the North Equatorial Current, the Mindanao Current, and the Kuroshio
along the Pacific western boundary, J. Geophys. Res., 101, 12,315–12,330, doi:10.1029/95JC03204.
Qiu, B., D. Rudnick, S. Chen, and Y. Kashino (2013), Quasi-stationary North Equatorial Undercurrent jets across the tropical North Pacific
Ocean, Geophys. Res. Lett., 40, 1–5, doi:10.1002/grl.50394.
Qu, T., H. Mitsudera, and T. Yamagata (1998), On the western boundary currents in the Philippine Sea, J. Geophys. Res., 103, 7537–7548, doi:
10.1029/98JC00263.
Qu, T., J. Gan, A. Ishida, Y. Kashino, and T. Tozuka (2008), Semiannual variation in the western tropical Pacific Ocean, Geophys. Res. Lett., 35,
L16602, doi:10.1029/2008GL035058.
Qu, T., T. Chiang, C. Wu, P. Dutrieux, and D. Hu (2012), Mindanao Current/Undercurrent in an eddy-resolving GCM, J. Geophys. Res., 117,
C06026, doi:10.1029/2011JC007838.
Sasaki, H., M. Nonaka, Y. Masumoto, Y. Sasai, H. Uehara, and H. Sakuma (2008), An eddy-resolving hindcast simulation of the quasiglobal
ocean from 1950 to 2003 on the Earth Simulator, in High Resolution Numerical Modeling of the Atmosphere and Ocean, edited by K. Hamilton and W. Ohfuchi, pp. 157–185, Springer, New York, doi:10.1007/978-0-387–49791-4_10.
Tozuka, T., T. Kagimoto, Y. Masumoto, and T. Yamagata (2002), Simulated Multiscale Variations in the Western Tropical Pacific: The Mindanao Dome Revisited, J. Phys. Oceanogr., 32(5), 1338–1359, doi:10.1175/1520-0485(2002)032<1338: SMVITW>2.0.CO; 2.
Wang, F., and D. Hu (1998a), Dynamic and thermohaline properties of the Mindanao Undercurrent, part I: Dynamic structure, Chin. J. Oceanol. Limnol., 16(2), 122–127, doi:10.1007/BF02845177.
Wang, F., and D. Hu (1998b), Dynamic and thermohaline properties of the Mindanao Undercurrent, part II: Thermohaline structure, Chin. J.
Oceanol. Limnol., 16(3), 206–213, doi:10.1007/BF02848724.
Wijffels, S., E. Firing, and J. Toole (1995), The mean structure and variability of the Mindanao Current at 8 N, J. Geophys. Res., 100, 18,421–
18,435, doi:10.1029/95JC01347.
Erratum
At the authors’ request, Fan Wang, Fujun Wang, and Dongliang Yuan have been added as the fourth, fifth, and sixth authors of this article.
With the unanimous support of the remaining authors and the editor, this has since been corrected and this version may be considered
the authoritative version of record.
ZHANG ET AL.
C 2014. American Geophysical Union. All Rights Reserved.
V
3628