• Balmaseda, M. A., A. Vidard, and D. L. T. Anderson, 2008: The ECMWF ocean analysis system: ORA-S3. Mon. Wea. Rev., 136, 30183034, doi:10.1175/2008MWR2433.1.

    • Search Google Scholar
    • Export Citation
  • Book, J. W., H. Perkins, R. P. Signell, and M. Wimbush, 2007: The Adriatic Circulation Experiment winter 2002/2003 mooring data report: A case study in ADCP data processing. NRL Memo. Rep. NRL/MR/7330--07-8999, 50 pp.

  • Chelton, D. B., and M. G. Schlax, 1996: Global observations of oceanic Rossby waves. Science, 272, 234238, doi:10.1126/science.272.5259.234.

    • Search Google Scholar
    • Export Citation
  • Chiang, T.-L., and T. Qu, 2013: Subthermocline eddies in the western equatorial Pacific as shown by an eddy-resolving OGCM. J. Phys. Oceanogr., 43, 12411253, doi:10.1175/JPO-D-12-0187.1.

    • Search Google Scholar
    • Export Citation
  • Dibarboure, G., O. Lauret, F. Mertz, and V. Rosmorduc, 2009: SSALTO/DUACS user handbook: (M)SLA and (M)ADT near-real time and delayed time products. CNES Rep. CLS‐DOS‐NT-06-034, 74 pp.

  • Efron, B., 1979: Bootstrap methods: Another look at the jackknife. Ann. Stat., 7, 126, doi:10.1214/aos/1176344552.

  • Firing, E., Y. Kashino, and P. Hacker, 2005: Energetic subthermocline currents observed east of Mindanao. Deep-Sea Res., 52, 605613, doi:10.1016/j.dsr2.2004.12.007.

    • Search Google Scholar
    • Export Citation
  • Ganachaud, A., and Coauthors, 2014: The Southwest Pacific Ocean Circulation and Climate Experiment (SPICE). J. Geophys. Res. Oceans, 119, 76607686, doi:10.1002/2013JC009678.

    • Search Google Scholar
    • Export Citation
  • Gordon, A. L., and R. A. Fine, 1996: Pathways of water between the Pacific and Indian Oceans in the Indonesian Seas. Nature, 379, 146149, doi:10.1038/379146a0.

    • Search Google Scholar
    • Export Citation
  • Gu, D., and S. G. H. Philander, 1997: Interdecadal climate fluctuations that depend on exchanges between the tropics and extratropics. Science, 275, 805807, doi:10.1126/science.275.5301.805.

    • Search Google Scholar
    • Export Citation
  • Hu, D., and M. Cui, 1989: The western boundary current in the far-western Pacific Ocean. Proc. Western Pacific Int. Meeting and Workshop on TOGA COARE, ORSTOM, Nouméa, New Caledonia, 123–134.

  • Hu, D., and M. Cui, 1991: The western boundary current of the Pacific and its role in the climate. Chin. J. Oceanol. Limnol., 9, 114, doi:10.1007/BF02849784.

    • Search Google Scholar
    • Export Citation
  • Hu, D., and Coauthors, 2011: Northwestern Pacific Ocean Circulation and Climate Experiment (NPOCE) Science/Implementation Plan. China Ocean Press, 100 pp.

  • Hu, D., and Coauthors, 2013: Direct measurements of the Luzon Undercurrent. J. Phys. Oceanogr., 43, 14171425, doi:10.1175/JPO-D-12-0165.1.

    • Search Google Scholar
    • Export Citation
  • Hu, D., and Coauthors, 2015: Pacific western boundary currents and their roles in climate. Nature, 522, 299308, doi:10.1038/nature14504.

    • Search Google Scholar
    • Export Citation
  • Hu, S., and D. Hu, 2012: Heat center of the western Pacific warm pool. Chin. J. Oceanol. Liimol., 30, 169176, doi:10.1007/s00343-012-1193-9.

    • Search Google Scholar
    • Export Citation
  • Hu, S., and D. Hu, 2014: Variability of the Pacific North Equatorial Current from repeated shipboard acoustic Doppler current profiler measurements. J. Oceanogr., 70, 559571, doi:10.1007/s10872-014-0253-5.

    • Search Google Scholar
    • Export Citation
  • Jin, F. F., 1997: An equatorial ocean recharge paradigm for ENSO. Part I: Conceptual model. J. Atmos. Sci., 54, 811829, doi:10.1175/1520-0469(1997)054<0811:AEORPF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Kashino, Y., N. Espaa, F. Syamsudin, K. J. 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 Niño and 2007/08 La Niña. J. Oceanogr., 65, 325333, doi:10.1007/s10872-009-0030-z.

    • Search Google Scholar
    • Export Citation
  • Kashino, Y., A. Ishida, and S. Hosoda, 2011: Observed ocean variability in the Mindanao Dome region. J. Phys. Oceanogr., 41, 287302, doi:10.1175/2010JPO4329.1.

    • Search Google Scholar
    • Export Citation
  • Kashino, Y., I. Ueki, and H. Sasaki, 2015: Ocean variability east of Mindanao: Mooring observations at 7°N, revisited. J. Geophys. Res. Oceans, 120, 25402554, doi:10.1002/2015JC010703.

    • Search Google Scholar
    • Export Citation
  • Kim, Y. Y., T. Qu, T. Jensen, T. Miyama, H. Mitsudera, H. W. 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.

    • Search Google Scholar
    • Export Citation
  • Lukas, R., 1988: Interannual fluctuations of the Mindanao Current inferred from sea level. J. Geophys. Res., 93, 67446748, doi:10.1029/JC093iC06p06744.

    • Search Google Scholar
    • Export Citation
  • Lukas, R., E. Firing, P. Hacker, P. L. Richardson, C. A. Collins, R. Fine, and R. Gammon, 1991: Observations of the Mindanao Current during the western equatorial Pacific Ocean circulation study. J. Geophys. Res., 96, 70897104, doi:10.1029/91JC00062.

    • Search Google Scholar
    • Export Citation
  • Lukas, R., T. Yamagata, and J. P. McCreary, 1996: Pacific low-latitude western boundary currents and the Indonesian throughflow. J. Geophys. Res., 101, 12 20912 216, doi:10.1029/96JC01204.

    • Search Google Scholar
    • Export Citation
  • Masumoto, Y., and T. Yamagata, 1991: Response of the western tropical Pacific to the Asian winter monsoon: The generation of the Mindanao Dome. J. Phys. Oceanogr., 21, 13861398, doi:10.1175/1520-0485(1991)021<1386:ROTWTP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • McCreary, J. P., Jr., and P. Lu, 1994: Interaction between the subtropical and equatorial ocean circulations: The subtropical cell. J. Phys. Oceanogr., 24, 466497, doi:10.1175/1520-0485(1994)024<0466:IBTSAE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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 31512 330, doi:10.1029/95JC03204.

    • Search Google Scholar
    • Export Citation
  • Qiu, B., D. A. Koh, C. Lumpkin, and P. Flament, 1997: Existence and formation mechanism of the North Hawaiian Ridge Current. J. Phys. Oceanogr., 27, 431444, doi:10.1175/1520-0485(1997)027<0431:EAFMOT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Qiu, B., S. Chen, D. L. Rudnick, and Y. Kashino, 2015: A new paradigm for the North Pacific subthermocline low-latitude western boundary current system. J. Phys. Oceanogr., 45, 24072423, doi:10.1175/JPO-D-15-0035.1.

    • Search Google Scholar
    • Export Citation
  • Qu, T., H. Mitsudera, and T. Yamagata, 1998: On the western boundary currents in the Philippine Sea. J. Geophys. Res., 103, 75377548, doi:10.1029/98JC00263.

    • Search Google Scholar
    • Export Citation
  • Qu, T., T. L. Chiang, C. R. Wu, P. Dutrieux, and D. Hu, 2012: Mindanao Current/Undercurrent in an eddy-resolving GCM. J. Geophys. Res., 117, C06026, doi:10.1029/2011JC007838.

    • Search Google Scholar
    • Export Citation
  • Roemmich, D., and J. Gilson, 2009: The 2004–2008 mean and annual cycle of temperature, salinity, and steric height in the global ocean from the Argo program. Prog. Oceanogr., 82, 81100, doi:10.1016/j.pocean.2009.03.004.

    • Search Google Scholar
    • Export Citation
  • 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. High Resolution Numerical Modelling of the Atmosphere and Ocean, K. Hamilton and W. Ohfuchi, Eds., Springer, 157–185.

  • Smith, S. D., 1988: Coefficients for sea surface wind stress, heat flux, and wind profiles as a function of wind speed and temperature. J. Geophys. Res., 93, 15 46715 472, doi:10.1029/JC093iC12p15467.

    • Search Google Scholar
    • Export Citation
  • Sprintall, J., A. L. Gordon, A. Koch-Larrouy, T. Lee, J. T. Potemra, K. Pujiana, and S. E. Wijffels, 2014: The Indonesian seas and their role in the coupled ocean-climate system. Nat. Geosci., 7, 487492, doi:10.1038/ngeo2188.

    • Search Google Scholar
    • Export Citation
  • 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, 13381359, doi:10.1175/1520-0485(2002)032<1338:SMVITW>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Walters, R. A., and C. Heston, 1982: Removing tidal-period variations from time-series data using low-pass digital filters. J. Phys. Oceanogr., 12, 112115, doi:10.1175/1520-0485(1982)012<0112:RTPVFT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wang, F., and D. Hu, 1998: Dynamic and thermohaline properties of the Mindanao Undercurrent I. Dynamic structure. Chin. J. Oceanol. Limnol., 16, 122127, doi:10.1007/BF02845177.

    • Search Google Scholar
    • Export Citation
  • Wang, F., and D. Hu, 1999: Preliminary study on the formation mechanism of counter western boundary undercurrents below the thermocline—A conceptual model. Chin. J. Oceanol. Limnol., 17, 19, doi:10.1007/BF02842694.

    • Search Google Scholar
    • Export Citation
  • Wang, F., N. Zang, Y. Li, and D. Hu, 2015: On the subsurface countercurrents in the Philippine Sea. J. Geophys. Res. Oceans, 120, 131144, doi:10.1002/2013JC009690.

    • Search Google Scholar
    • Export Citation
  • Wang, Q., F. Zhai, F. Wang, and D. Hu, 2014: Intraseasonal variability of the subthermocline current east of Mindanao. J. Geophys. Res. Oceans, 119, 85528566, doi:10.1002/2014JC010343.

    • Search Google Scholar
    • Export Citation
  • 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 42118 435, doi:10.1029/95JC01347.

    • Search Google Scholar
    • Export 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, 36173628, doi:10.1002/2013JC009693.

    • Search Google Scholar
    • Export Citation
  • View in gallery
    Fig. 1.

    Topography (color) around the subsurface mooring (black dot) and mean current during 2011–13 in the upper 410-m layer from the ECCO2 dataset. Major surface currents, including the NEC, NECC, ITF, and the MC, are labeled.

  • View in gallery
    Fig. 2.

    (a) The 3-day low-passed daily mean meridional velocity υ (cm s−1) measured by the moored ADCPs. (b) Bootstrapped means of υ and corresponding errors at depths where the number of samples are greater than 365. Errors are defined as the standard deviations of the bootstrapped means. (c) Time series of smoothed core mean meridional velocity υMUC and (d) distribution of density of bootstrapped means of υMUC.

  • View in gallery
    Fig. 3.

    (a) Variation συ (cm s−1) of the observed currents at depths defined as the bootstrapped standard deviation of daily υ. (b) Low-passed ADCP meridional velocity υ using a Fourier filter with a cutoff period of 365 days.

  • View in gallery
    Fig. 4.

    (a) Low-passed anomalous meridional velocity υ′ (cm s−1) and (b) vertical mean of υ′ in 550–750- and 80–550-m layers. Error bars in the right panel denote the vertical variation of υ′ (standard deviations over depths).

  • View in gallery
    Fig. 5.

    (a) Hovmöller diagrams of low-passed (a) observed sea level anomaly (m) and (b) wind stress curl anomaly (Pa m−1) averaged over 4°–12°N relative to the mean fields over 1997–2013. Wind data are extracted from ERDDAP, version 1.62 Fleet Numerical Meteorology and Oceanography Center 10-m surface wind. Bottom panels are the same as upper panels [(c) SLA and (d) wind stress curl anomaly] but for the years after 2009. Black rectangle in (a) and (c) highlights the region of relative maximum signals near the coast.

  • View in gallery
    Fig. 6.

    Hovmöller diagrams of Argo temperature anomaly (°C) at depth: (a) at 200 dbar, (b) vertical mean in the 2.5–200-dbar layer, and (c) vertical mean in the 550–750-dbar layer. Bottom panels are as in the upper panels but relative to the mean during November 2010–February 2014.

  • View in gallery
    Fig. 7.

    (a) Composite AVISO SLAs (cm) and (b) ERDDAP wind stress curl anomalies (Pa m−1) over the three interannual phases (phase A: March–July of 2011; phase B: July 2011–January 2013; and phase C: January 2013–June 2014) of the observed currents.

  • View in gallery
    Fig. 8.

    Surface horizontal current anomalies during the three interannual phases from the monthly OSCAR. Cyclonic and anticyclonic current anomalies are denoted in red and blue ellipses.

  • View in gallery
    Fig. 9.

    Comparison between the first-layer thickness in the RGM (blue) and the AVISO SLA (red) at (a) the mooring point of 8°N, 127°E in the MC region and (b) 13°N, 137°E in the NEC region.

  • View in gallery
    Fig. 10.

    Comparison of the anomalous meridional current velocities υ′ at the 8°N, 127°E between RGM and observations from the ADCPs of the moorings at (a) RGM layer 1 (80–550 m for ADCP observations) and (b) layer 2 (550–750 m for ADCP observations). (c) The power spectra of the RGM transport anomalies of (left) the MC and (right) the MUC during 1992–2014. Dashed black lines in (c) indicate the 95% confidence level.

  • View in gallery
    Fig. 11.

    Comparison of Hovmöller diagrams between (a) AVISO SLA and (b) the first-layer thickness in the RGM, that is, , averaged over 4° and 12°N. Black boxes roughly denote the longitudes of maximum anomalies around the offshore part of the MC/MUC.

  • View in gallery
    Fig. 12.

    (a) Satellite-observed mean surface geostrophic current from the AVISO dataset averaged over 1993–2013 and mean horizontal currents in (b) layer 1 and (c) layer 2 of the RGM. Vectors of currents are superposed on the shaded color that indicates the meridional components of mean velocities (cm s−1). Red and blue squares denote the mooring site.

  • View in gallery
    Fig. 13.

    Anomalous volume transports (Sv) of the (a) MC and (b) MUC in various numerical experiments including FP, CP, EP, WP, FWP, and EWP runs.

  • View in gallery
    Fig. 14.

    Percentages of contributions Pi of wind forcings in different regions (FWP, EWP, CP, and EP) to the interannual variability of the (top) MC and (bottom) MUC.

  • View in gallery
    Fig. 15.

    (a) Monthly Niño-3.4 index and SLA averaged over 126°–130°E along 8°N in the MC region, and (b) their lead–lag correlation. Both the Niño-3.4 index and SLA are 13-month running means.

  • View in gallery
    Fig. 16.

    Comparison between the Niño-3.4 index and the velocity anomalies averaged over 80–550 (MC) and 550–750 m (MUC) as in Fig. 4b. The velocity anomalies are normalized by their standard deviations.

  • View in gallery
    Fig. 17.

    (a) OFES EKE and meridional velocity anomalies averaged over 5°–10°N and 126°–130°E, and the (b) lead–lag correlation between them. Both the time series are 1-yr running averaged. Positive time lag indicates EKE lags the meridional velocity anomalies.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1621 472 25
PDF Downloads 256 93 8

Interannual Variability of the Mindanao Current/Undercurrent in Direct Observations and Numerical Simulations

Shijian HuInstitute of Oceanology, and Key Laboratory of Ocean Circulation and Wave, Chinese Academy of Sciences, and Laboratory for Ocean and Climate Dynamics, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

Search for other papers by Shijian Hu in
Current site
Google Scholar
PubMed
Close
,
Dunxin HuInstitute of Oceanology, and Key Laboratory of Ocean Circulation and Wave, Chinese Academy of Sciences, and Laboratory for Ocean and Climate Dynamics, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

Search for other papers by Dunxin Hu in
Current site
Google Scholar
PubMed
Close
,
Cong GuanInstitute of Oceanology, and Key Laboratory of Ocean Circulation and Wave, Chinese Academy of Sciences, Qingdao, and University of Chinese Academy of Sciences, Beijing, China

Search for other papers by Cong Guan in
Current site
Google Scholar
PubMed
Close
,
Fan WangInstitute of Oceanology, and Key Laboratory of Ocean Circulation and Wave, Chinese Academy of Sciences, and Laboratory for Ocean and Climate Dynamics, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

Search for other papers by Fan Wang in
Current site
Google Scholar
PubMed
Close
,
Linlin ZhangInstitute of Oceanology, and Key Laboratory of Ocean Circulation and Wave, Chinese Academy of Sciences, and Laboratory for Ocean and Climate Dynamics, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

Search for other papers by Linlin Zhang in
Current site
Google Scholar
PubMed
Close
,
Fujun WangInstitute of Oceanology, and Key Laboratory of Ocean Circulation and Wave, Chinese Academy of Sciences, and Laboratory for Ocean and Climate Dynamics, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

Search for other papers by Fujun Wang in
Current site
Google Scholar
PubMed
Close
, and
Qingye WangInstitute of Oceanology, and Key Laboratory of Ocean Circulation and Wave, Chinese Academy of Sciences, and Laboratory for Ocean and Climate Dynamics, Qingdao National Laboratory for Marine Science and Technology, Qingdao, China

Search for other papers by Qingye Wang in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The interannual variability of the boundary currents east of the Mindanao Island, including the Mindanao Current/Undercurrent (MC/MUC), is investigated using moored acoustic Doppler current profiler (ADCP) measurements combined with a series of numerical experiments. The ADCP mooring system was deployed east of the Mindanao Island at 7°59′N, 127°3′E during December 2010–August 2014. Depth-dependent interannual variability is detected in the two western boundary currents: strong and lower-frequency variability dominates the upper-layer MC, while weaker and higher-frequency fluctuation controls the subsurface MUC. Throughout the duration of mooring measurements, the weakest MC was observed in June 2012, in contrast to the maximum peaks in December 2010 and June 2014, while in the deeper layer the MUC shows speed peaks circa December 2010, January 2011, April 2013, and July 2014 and valleys circa June 2011, August 2012, and November 2013. Diagnostic analysis and numerical sensitivity experiments using a 2.5-layer reduced-gravity model indicate that wind forcing in the western Pacific Ocean is a driving agent in conditioning the interannual variability of MC and MUC. Results suggest that westward-propagating Rossby waves that generate in the western Pacific Ocean (roughly 150°–180°E) are of much significance in the interannual variability of the two boundary currents. Fluctuation of Ekman pumping due to local wind stress curl anomaly in the far western Pacific Ocean (roughly 120°–150°E) also plays a role in the interannual variability of the MC. The relationship between the MC/MUC and El Niño is discussed.

Corresponding author address: Shijian Hu, Institute of Oceanology, Chinese Academy of Sciences, 7 Nanhai Road, Qingdao 266071, China. E-mail: sjhu@qdio.ac.cn

Abstract

The interannual variability of the boundary currents east of the Mindanao Island, including the Mindanao Current/Undercurrent (MC/MUC), is investigated using moored acoustic Doppler current profiler (ADCP) measurements combined with a series of numerical experiments. The ADCP mooring system was deployed east of the Mindanao Island at 7°59′N, 127°3′E during December 2010–August 2014. Depth-dependent interannual variability is detected in the two western boundary currents: strong and lower-frequency variability dominates the upper-layer MC, while weaker and higher-frequency fluctuation controls the subsurface MUC. Throughout the duration of mooring measurements, the weakest MC was observed in June 2012, in contrast to the maximum peaks in December 2010 and June 2014, while in the deeper layer the MUC shows speed peaks circa December 2010, January 2011, April 2013, and July 2014 and valleys circa June 2011, August 2012, and November 2013. Diagnostic analysis and numerical sensitivity experiments using a 2.5-layer reduced-gravity model indicate that wind forcing in the western Pacific Ocean is a driving agent in conditioning the interannual variability of MC and MUC. Results suggest that westward-propagating Rossby waves that generate in the western Pacific Ocean (roughly 150°–180°E) are of much significance in the interannual variability of the two boundary currents. Fluctuation of Ekman pumping due to local wind stress curl anomaly in the far western Pacific Ocean (roughly 120°–150°E) also plays a role in the interannual variability of the MC. The relationship between the MC/MUC and El Niño is discussed.

Corresponding author address: Shijian Hu, Institute of Oceanology, Chinese Academy of Sciences, 7 Nanhai Road, Qingdao 266071, China. E-mail: sjhu@qdio.ac.cn

1. Introduction

Low-latitude western boundary currents (LLWBCs) in the Pacific Ocean are of much importance in the tropical climate system for their role in the recharge/discharge of the El Niño–Southern Oscillation (ENSO) cycle and in regulating the western Pacific warm pool that is a key factor in the tropical Pacific Ocean (Hu and Cui 1991; Lukas et al. 1996; Jin 1997; Hu and Hu 2012; Hu et al. 2015). The Mindanao Current (MC) and the underlying Mindanao Undercurrent (MUC) are remarkable components of the LLWBCs in the Pacific Ocean (Hu and Cui 1989). They are suggested to be important in the global thermohaline circulation for their significant contribution to the Indonesian Throughflow (ITF) that connects different ocean basins (e.g., Gordon and Fine 1996; Sprintall et al. 2014). The MC is also a crucial pathway between the midlatitudes and equatorial region and influences the interdecadal climate variability (e.g., Gu and Philander 1997). The significance of LLWBCs in the Pacific Ocean has led to international research efforts such as the Northwestern Pacific Ocean Circulation and Climate Experiment (NPOCE; Hu et al. 2011) and Southwest Pacific Ocean Circulation and Climate Experiment (SPICE; Ganachaud et al. 2014).

The MC and MUC are of particular interest to the oceanography community for their spatial complexity (e.g., Hu and Cui 1989, 1991; Wijffels et al. 1995; Wang et al. 2015) and multi-time-scale variabilities (e.g., Lukas 1988; Qiu and Lukas 1996; Tozuka et al. 2002; Kashino et al. 2005, 2009, 2011; Zhang et al. 2014). On an interannual time scale, Lukas (1988) found that fluctuations of the MC transport are at least 50% of the mean transport, with a period of 2 yr, but have no apparent relationship to the strength of ENSO, on the basis of sea level observations at the islands of Mindanao in the Philippines and Malakal in Palau. Using hydrographic observations from eight cruises, Wijffels et al. (1995) suggested that the MC is a stable coastal jet with a maximum speed of 1 m s−1 and speculated that the variation of MC is related to the eddy or meanderlike anomalous circulation. Qiu and Lukas (1996) pointed out that the interannual variability of the MC is influenced by both the ENSO wind on a time scale of 3–7 yr and the quasi-biennial wind that is confined to the tropical gyre of the North Pacific. Kashino et al. (2005) examined the variability of the MC using mooring observations conducted east of the Mindanao Island and suggested that the MC was enhanced during the onset of 2002 El Niño.

With regard to the MUC, direct observations suggest that the MUC around 8°N is probably below 600 m to the depth deeper than 1000 m and features significant intraseasonal variability (Wang et al. 2014; Zhang et al. 2014; Qiu et al. 2015; Wang et al. 2015). It seems that the MUC is a relatively weak northward continuous mean flow with a mean speed of several centimeters per second, possesses a double-core structure [a major inshore core and a secondary offshore core (e.g., Qiu et al. 2015)], and probably is related to the subthermocline eddies. But debate persists as to if it is a transient or permanent current (Hu and Cui 1989, 1991; Lukas et al. 1991; Wijffels et al. 1995; Wang and Hu 1998; Qu et al. 1998; Firing et al. 2005; Kashino et al. 2005; Qu et al. 2012; Kashino et al. 2015). This debate might be due to the absence of enough observations of the MUC, especially the shortage in the depths of measurements. For example, Kashino et al. (2013, p. 1) suggested that “the stationary northward undercurrent, the Mindanao Undercurrent, was also not found at 7°N east of Mindanao” on the basis of observations shallower than 600 m (thus shallower than the MUC layer). Kashino et al. (2015, p. 1) proposed that the “Mindanao Undercurrent was not confirmed at the 7°N line,” but the measurements in that paper are confined to the upper 350-m layer and depths of 560, 960, and 1460 m, which expectedly might miss the MUC core. Lack of enough observations gives rise to some studies using model outputs (e.g., Kashino et al. 2015), but unfortunately few of these simulations can be validated by observations. Recently, Qiu et al. (2015) pointed out that a time-mean MUC is observed from 6° to 13°N and extends from the about 400- to 1200-m layer at 6°N to the about 800- to 1200-m layer at about 12°N, on the basis of 14-yr Argo float profiling data (obviously with depth up to 2000 m) from 2001 to 2014. The study by Qiu et al. (2015) together with other previous observations seems to confirm that the MUC is a mean flow along the east coast of Mindanao Island (e.g., Zhang et al. 2014; Hu et al. 2015).

Previous studies have tremendously facilitated our understanding of the structure and variability of these currents. However, in a situation of a lack of sustained measurements of these boundary currents, the feature and mechanism of the interannual variability of MC are still controversial. For example, we do not know much about the vertical feature of the interannual variability, and it is unclear yet what the relative contributions are from remote forcing and local wind forcing. Geostrophic calculation in the MC/MUC region is possibly of nonignorable error, and thus direct observations of the undercurrents in this region are extremely needed. The interannual variability of MUC in the direct observations is even nearly unknown yet.

Here, we present the direct observations of the MC and MUC from December 2010 to August 2014 with emphasis on an interannual time scale. Spatial and temporal features of the mean current and interannual variability will be described in sections 2 and 3. Dynamic mechanisms about the MC/MUC interannual variability will be explored in sections 4 and 5 by combining dynamic diagnostic analysis and sensitivity numerical experiments using a 2.5-layer reduced-gravity model. Section 6 will discuss the relationships between the currents and ENSO and between currents and interannual modulation of mesoscale subthermocline eddies. Results will be summarized in section 7.

2. Observations and mean structure

To monitor the boundary currents east of Mindanao Island, a subsurface mooring has been sequentially deployed and recovered at 7°59′N, 127°3′E since December 2010 (Fig. 1), as a part of the observation project of the NPOCE. The mean horizontal structure of upper-layer (upper 410 m) currents, including the North Equatorial Current (NEC), North Equatorial Countercurrent (NECC), ITF, and MC, is presented in Fig. 1 using the Estimating the Circulation and Climate of the Ocean, Phase II (ECCO2), dataset over 2011–2013. According to Qiu et al. (2015), the inshore core of the mean MUC is much stronger than the offshore core, and the MC core and the inshore core of the MUC at 8°N are approximately located at the same longitude, that is, between about 127° and 127.5°E. As shown in Fig. 1, the mooring was located at the principal axis of the major cores of the MC and MUC.

Fig. 1.
Fig. 1.

Topography (color) around the subsurface mooring (black dot) and mean current during 2011–13 in the upper 410-m layer from the ECCO2 dataset. Major surface currents, including the NEC, NECC, ITF, and the MC, are labeled.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

The first mooring was deployed in December 2010 and recovered in July 2011, the second one was deployed in July 2011 and recovered in December 2012, and the third mooring was deployed in December 2012 and recovered in August 2014. Each mooring was equipped with an upward-looking and a downward-looking 75-kHz Teledyne RD Instruments (TRDI) acoustic Doppler current profilers (ADCPs) at about 400-m depth. The velocity accuracy is 1% of the current magnitude ±5 mm s−1. The raw data are quality controlled using percent good quality control, internal ADCP quality control, and correlation quality control (Book et al. 2007). Vertical movement of the instruments wADCP is estimated using the records of the pressure sensors mounted in the ADCPs. Result shows that wADCP is O(10−4) m s−1, which is much less than the vertical or horizontal velocities of the seawater. Thus, the errors in the horizontal velocities induced by the up–down motion of the buoy are not taken into consideration in the present paper. Then daily meridional and zonal components of current velocity υ and u are obtained from the quality-controlled hourly records and further processed according to our documented methods (Hu et al. 2013). All the time series are low-pass filtered with a cutoff period of 3 days to remove the tidal effect.

Figure 2a presents the daily mean υ observed by the moored ADCPs. Northward current in the subthermocline layer was observed during most of the time. The maximum instantaneous meridional speeds of the northward and southward currents exceed 48 and 140 cm s−1, respectively. To obtain approximate means of the observed currents, we apply the bootstrap method to the daily mean meridional velocities with a trial number of 300 at depths where the numbers of samples are greater than 365 (Efron 1979). Figure 2b presents the bootstrapped means of daily υ and corresponding errors. It shows that maximum mean northward and southward speeds are about 4 and 78 cm s−1 and statistically significant relative to the errors, though the observed mean undercurrent is weak as expected. Note that the mean undercurrent is enhanced with the increase of depth below 600 m; it is very likely stronger at deeper layers (e.g., Qiu et al. 2015). The observed mean surface layer current should be the MC, while the mean subsurface current should be a part of the MUC, as mentioned in previous studies of the boundary currents in this region (Hu and Cui 1989, 1991; Lukas et al. 1991; Kim et al. 2004; Qu et al. 2012; Wang et al. 2014; Zhang et al. 2014; Wang et al. 2015; Qiu et al. 2015).

Fig. 2.
Fig. 2.

(a) The 3-day low-passed daily mean meridional velocity υ (cm s−1) measured by the moored ADCPs. (b) Bootstrapped means of υ and corresponding errors at depths where the number of samples are greater than 365. Errors are defined as the standard deviations of the bootstrapped means. (c) Time series of smoothed core mean meridional velocity υMUC and (d) distribution of density of bootstrapped means of υMUC.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

To examine the velocity of the subsurface MUC, we plot in Fig. 2c the core-mean (i.e., vertical average of the MUC speed) northward velocity υMUC, which is smoothed to remove high-frequency variability (mainly including intraseasonal and seasonal signals). Fluctuation of υMUC in Fig. 2c is induced by both the natural variability of the MUC and the variation of depths of the ADCPs, and the maximum velocity of the MUC is probably larger than the velocity in Fig. 2c. The υMUC has been positive in the entire duration of the mooring observation, suggesting that a steady mean current does exist below the MC at this place. The bootstrap method is applied to the υMUC to calculate its mean with trial number of 1000. Figure 2d presents the density of bootstrapped means of υMUC (i.e., ) and indicates that the is about 6 cm s−1.

In addition, variation συ of the observed currents, defined as the standard deviation of daily υ, is further examined by applying the bootstrap method. The συ shown in Fig. 3a covers time scales from daily to interannual scales, say synoptic, intraseasonal, seasonal, and interannual variabilities. For the subsurface layer, the variation is about 12 cm s−1, which is about twice the . Of the total variations of υ for the MC and MUC in Fig. 3, about 33% and 14%, respectively, are induced by interannual variability (estimated by comparing with the συ of 1-yr low-pass filtered υ). Figure 3b presents the low-pass filtered υ using a Fourier filter (Walters and Heston 1982) with a cutoff period of 365 days to remove the mesoscale to small-scale effects. In spite of the strong variability as in Fig. 3a, the mean flows (i.e., the MC and MUC) were observed by the moored ADCPs in most (about 89%) of the duration. The MUC is roughly below 600 m with several centimeters per second, as we mentioned in Fig. 2, but apparently the mooring has missed the deeper part of the undercurrent. This result implies that measurements at 8°N shallower than 600 m naturally cannot capture the MUC.

Fig. 3.
Fig. 3.

(a) Variation συ (cm s−1) of the observed currents at depths defined as the bootstrapped standard deviation of daily υ. (b) Low-passed ADCP meridional velocity υ using a Fourier filter with a cutoff period of 365 days.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

3. Observed interannual variability

The interannual variations of boundary currents are significant during 2010–2014, though the duration of the ADCP measurements is very limited. The MC is stronger during December 2010–August 2011 and August 2013–August 2014 than October 2011–January 2013 (Fig. 2a). To illustrate the interannual features, we extract the interannual series of daily meridional velocity anomalies relative to the mean velocities over the observation duration by low-pass filtering with a cutoff period of 365 days. As shown in Fig. 4a, there are three interannual phases of υ′ in the upper 550-m layer (i.e., the MC layer). Positive υ′ was significant during the second half of 2011 and 2012 (phase B for simplification), but in other periods (phase A during the first half-year of 2011 and phase C in 2013–2014) negative υ′ was dominant. The amplitude of interannual MC exceeds 15 cm s−1 with an approximate 3-yr period, though it is difficult to assess the statistical significance. Higher-frequency but weaker interannual variability exists in the ocean below 550 m (i.e., the MUC layer) in contrast to the interannual variability of MC (Figs. 4a,b). The amplitude of interannual variability of MUC is no more than 5 cm s−1 with an approximate period of 1.2 yr. We averaged υ′ vertically over the 80–550-m layer and the 550–750-m layer to present the respective interannual features in the upper- and deeper-layer currents, respectively (Figs. 4b). Over the observation duration, the weakest MC was observed in June 2012, in contrast to the maximum peaks in December 2010 and June 2014. In the deeper layer, the MUC shows speed peaks around December 2010, January 2011, April 2013, and July 2014 and valleys in June 2011, August 2012, and November 2013.

Fig. 4.
Fig. 4.

(a) Low-passed anomalous meridional velocity υ′ (cm s−1) and (b) vertical mean of υ′ in 550–750- and 80–550-m layers. Error bars in the right panel denote the vertical variation of υ′ (standard deviations over depths).

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

4. Related dynamics

Previous studies suggest that the MC and MUC are approximately geostrophic flows (e.g., Qu et al. 2012; Wang and Hu 1999). An eastward-directed cross-shelf pressure gradient is set up in the upper layer and balances the westward Coriolis force associated with the southward-flowing MC, while a westward-directed cross-shelf pressure gradient is built up in the subthermocline layer and balances the eastward Coriolis force associated with the northward-flowing MUC. Thus, local wind forcing and Rossby waves forced by remote winds, which are major dynamical processes that affect the pressure structure, might play an important role in controlling the interannual variability of both the MC and MUC.

Figure 5a presents the Hovmöller diagram of sea level anomaly (SLA) averaged over 4°–12°N. Here, the SLA data are provided by the Data Unification and Altimeter Combination System (DUACS) and distributed by Archiving, Validation, and Interpretation of Satellite Oceanographic Data (AVISO)/Centre National d’Etudes Spatiales (Dibarboure et al. 2009). High-frequency variability (periods less than 1 yr) are excluded by applying a 13-month running mean to the monthly SLA. Correspondingly, Fig. 5b presents the Hovmöller diagram of wind stress curl anomaly to compare with SLA, where the wind stress curl data are provided by the Environmental Research Division Data Access Program (ERDDAP) data server at the National Oceanic and Atmospheric Administration (NOAA) (Smith 1988). The difference of SLA is clear between the western and central-eastern Pacific Ocean. As shown in Fig. 5a, interannual anomalies generated in the eastern to central Pacific Ocean propagate westward to the central Pacific Ocean and dominate the interannual variability of SLA in the eastern to central Pacific Ocean. But in the western Pacific Ocean, in particular the far western Pacific Ocean centered at 140°E, the wave signals generally originate in the central to western Pacific Ocean most of the time. Interannual variability of SLA in the western Pacific Ocean seems to be composed of local, low-frequency variation and relatively higher-frequency propagating variation from the western-central Pacific Ocean and is much different from that in the eastern Pacific Ocean. Power spectra of SLA at 8°N shows that the far western Pacific Ocean (120°–150°E) possesses an interannual variability with a period of about 3 yr and is consistent with the MC feature. The eastern part of the western Pacific Ocean (150°E–180°) is an origin of the Rossby waves (Fig. 5a) and is characterized by interannual variability with a peak of period between 1 and 2 yr that is in agreement with the observed MUC feature (figure not shown). Comparison indicates that the SLA variation is in agreement with the fluctuation of wind stress curl phase to phase; SLAs were positive during July 1998–July 2001 and July 2007–July 2013, corresponding to negative wind stress curl anomalies, and negative during July 2001–July 2007 when wind stress curl anomalies were positive. The strongest wind stress curl anomalies are presented in the western Pacific Ocean. The above features imply that the western Pacific Ocean is possibly very important in the interannual variability of the western boundary currents like the MC.

Fig. 5.
Fig. 5.

(a) Hovmöller diagrams of low-passed (a) observed sea level anomaly (m) and (b) wind stress curl anomaly (Pa m−1) averaged over 4°–12°N relative to the mean fields over 1997–2013. Wind data are extracted from ERDDAP, version 1.62 Fleet Numerical Meteorology and Oceanography Center 10-m surface wind. Bottom panels are the same as upper panels [(c) SLA and (d) wind stress curl anomaly] but for the years after 2009. Black rectangle in (a) and (c) highlights the region of relative maximum signals near the coast.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

Rossby waves forced by wind forcing in the western Pacific Ocean propagate westward until the Philippine Sea, but at different depths the propagation is different. To examine the vertical feature of the wave propagation, we present in Fig. 6 a Hovmöller diagram of temperature anomalies at depths along 8°N using gridded Argo observations since 2004 (Roemmich and Gilson 2009). For the upper layer (at 200 dbar and 2.5–200-dbar mean), the fluctuation is strong and the propagation is very clear. But the anomalies in the deeper layer (at 650 dbar) are relatively small, as expected. A notable characteristic is that whether the upper layer or the deeper layer, the Rossby waves are generally generated at the region west of the date line except for a few special years like 2008 and 2010. The temperature phases over the mooring duration are consistent with the interannual variation of the MC: cooling and upwelling during phases A and C but relative warming during phase B. It should be noted that the deeper-layer temperature variability is different from the upper layer in the far western Pacific Ocean, that is, west of about 150°E: it is of higher frequency but weaker amplitude. This difference might explain why the MC shows stronger and lower frequency than the MUC. This is reasonable because the local wind forcing in the far western Pacific Ocean gives rise to great upper-layer influence on the signals propagating from the eastern part of the western Pacific Ocean (150°E–180°). Disparity of the temperature anomaly in Fig. 6 and MUC velocity in Fig. 4 also exists, suggesting that the interannual variability of the MUC is also modulated by other processes besides the Rossby waves.

Fig. 6.
Fig. 6.

Hovmöller diagrams of Argo temperature anomaly (°C) at depth: (a) at 200 dbar, (b) vertical mean in the 2.5–200-dbar layer, and (c) vertical mean in the 550–750-dbar layer. Bottom panels are as in the upper panels but relative to the mean during November 2010–February 2014.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

Rossby waves triggered by remote wind forcing get enhanced toward the west as shown in Fig. 5a. It should be noted that almost all the Rossby wave signals reach a zonal maximum (minimum for negative phases) around 130°E, but in contrast the signal is very small in the region between the coast and the maximum (minimum for negative phases). Because the MC and MUC are approximately geostrophic flows, this feature implies that a downwelling Rossby wave (positive SLA) will cancel the eastward pressure gradient, reduce the southward-flowing MC, but enhance the northward-flowing MUC and vice versa for the response of the MC and MUC to upwelling Rossby waves. Recently, Qiu et al. (2015) suggested that the MUC are related to baroclinic instability of the overlying wind-driven western boundary currents. Thus, beside the direct influence of Rossby waves, the interannual variability of the MUC might also be influenced by the Rossby waves through modulating the baroclinic instability in the MC.

To examine the role of local wind forcing, we present the composite maps of AVISO SLAs and ERDDAP wind stress curl anomaly during phases A, B, and C. As shown in Fig. 7a, SLAs were negative during phases A and C but positive in phase B, suggesting that anomalous upwelling was significant east of Mindanao during phases A and C, while anomalous downwelling governed the current anomaly. To examine the horizontal current pattern during these phases, we extract the surface current anomalies from monthly Ocean Surface Currents Analyses–Real Time (OSCAR) datasets. As a result, surface currents show cyclonic current anomalies and southward current anomalies near the western boundary during negative phases of the MC, and on the contrary, anticyclonic current anomaly and northward anomaly are distinct during positive phases of MC (Fig. 8). The composite wind stress curl anomalies in three MC phases are consistent with the corresponding SLAs in the south Philippine Sea. Positive wind stress curl anomalies in phases A and C lead to local Ekman pumping, cyclonic current anomalies surrounding the wind stress curl anomalies, and thus northward current anomaly close to the western boundary. Meanwhile, negative wind stress curl anomaly gives rise to anomalous downwelling in the Mindanao Dome (Masumoto and Yamagata 1991) and anticyclonic current anomaly and thus causes southward current anomaly near the western boundary and interannual intensification of MC (Figs. 5, 6).

Fig. 7.
Fig. 7.

(a) Composite AVISO SLAs (cm) and (b) ERDDAP wind stress curl anomalies (Pa m−1) over the three interannual phases (phase A: March–July of 2011; phase B: July 2011–January 2013; and phase C: January 2013–June 2014) of the observed currents.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

Fig. 8.
Fig. 8.

Surface horizontal current anomalies during the three interannual phases from the monthly OSCAR. Cyclonic and anticyclonic current anomalies are denoted in red and blue ellipses.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

5. Sensitivity experiments

To further understand the variability of MC and MUC, we adopt in this section a 2.5-layer (the third layer is assumed to be inert) reduced-gravity model (RGM) built by Qiu et al. (1997). On the basis of this RGM, several sensitivity numerical experiments are conducted. A 2.5-layer RGM is found to be suitable to study the wind-driven circulations and oceanic adjustment by the Rossby waves (e.g., McCreary and Lu 1994; Qiu et al. 1997).

Equations that govern the upper two layers in the RGM and basic model sets are the same as those described by Qiu et al. (1997), except for the following changes. First, the model domain covers a closed region from the tropical South Pacific Ocean to subtropical North Pacific Ocean (20°S–40°N, 120°E–70°W) with a horizontal resolution of 0.25° × 0.25°. Second, the mean thicknesses for the upper two layers are chosen according to the vertical structure of the MC/MUC system, and we require that the RGM can reproduce the upper MC and the subsurface MUC and their interannual variability. As shown in Figs. 2 and 4, because the MC core is mainly shallower than 400-m and interannual variability in the upper 400-m layer is distinct from the deeper layer, the initial thickness of the first layer is chosen to be 400 m. Figure 2 also indicates that the subsurface current underlying the MC is possibly as deep as 1000 m; thus, the initial thickness of the second layer is set to be 600 m. Third, the horizontal eddy viscosity coefficient is 700 m2 s−1 in the whole model domain but increases linearly to 1500 m2 s−1 near the model boundaries to suppress the exaggerated instabilities. Finally, the RGM in the present paper is spun up from rest by the climatological wind stress from the European Centre for Medium-Range Weather Forecasts (ECMWF) Ocean Reanalysis, system 3 (ORA-S3; Balmaseda et al. 2008), wind stress data (1992–2011) and ERDDAP wind stress data (2012–2014) for 30 yr to reach a quasi-steady state. We extract the monthly wind stress data (1992–2011) from the ECMWF ORA-S3 and ERDDAP (2012–14) and remove the seasonal variation in the wind field to focus on the interannual variability. The RGM ocean circulation is then driven by these monthly wind stress data, namely, FP run.

At first, model results are validated by observations from satellites and mooring ADCPs (Figs. 911). The anomalous thickness of the first-layer is compared with the observational SLAs from AVISO at two points: 8°N, 127°E in the MC/MUC region and 13°N, 137°E in the NEC region (Fig. 9). Correlation coefficients between SLAs and in Fig. 9 are 0.9 at 8°N, 127°E and 0.9 at 13°N, 137°E, indicating that the RGM outputs are consistent with the satellite observations.

Fig. 9.
Fig. 9.

Comparison between the first-layer thickness in the RGM (blue) and the AVISO SLA (red) at (a) the mooring point of 8°N, 127°E in the MC region and (b) 13°N, 137°E in the NEC region.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

Fig. 10.
Fig. 10.

Comparison of the anomalous meridional current velocities υ′ at the 8°N, 127°E between RGM and observations from the ADCPs of the moorings at (a) RGM layer 1 (80–550 m for ADCP observations) and (b) layer 2 (550–750 m for ADCP observations). (c) The power spectra of the RGM transport anomalies of (left) the MC and (right) the MUC during 1992–2014. Dashed black lines in (c) indicate the 95% confidence level.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

Fig. 11.
Fig. 11.

Comparison of Hovmöller diagrams between (a) AVISO SLA and (b) the first-layer thickness in the RGM, that is, , averaged over 4° and 12°N. Black boxes roughly denote the longitudes of maximum anomalies around the offshore part of the MC/MUC.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

The RGM current is also compared with the observations from the ADCPs of the moorings at the 8°N, 127°E. Figure 10 presents the anomalous meridional current velocities in the RGM layers 1 and 2 and ADCP observations averaged over 80–550 and 550–750 m. Correlation coefficients between RGM and ADCP observations are 0.8 in the upper layer and 0.5 in the second layer and are both statistically significant at the 98% confidence level.

Transports of the MC and MUC are calculated along 8°N as below:
e1
e2
where υ1 and υ2 are meridional velocities in the first and second RGM layers, and H1 and H2 are the thicknesses of the first and second RGM layers.

Power spectra of the MC and MUC transport anomalies are calculated using the monthly RGM outputs during 1992–2014. As shown in Fig. 10c, both the MC and MUC have significant seasonal variability. On the interannual time scales, the spectral peaks of the MC transport are located at 1–2 and 2–3 yr, and the latter near the 3-yr period is much stronger. In contrast, the significant interannual period of the MUC transport is about 1–2 yr and less than the MC periods. The spectral feature in Fig. 10c is consistent with the moored ADCP observations as described in section 3.

The RGM outputs are further validated in terms of wave propagation by comparing the Hovmöller diagram of the first-layer thickness anomaly of the RGM with that of AVISO SLA along the zonal band between 4° and 12°N (Fig. 11). Results show that the simulated presents significant westward propagation and is in good agreement with the observed SLA phase by phase. Based on Radon transform calculation (Chelton and Schlax 1996), the phase speeds of and SLA in Fig. 11 are about −40 cm s−1, which is close to the baroclinic Rossby waves at the latitude of about 8°N and in good agreement with the results by Chelton and Schlax (1996). Both the SLA and are increased with the westward propagation of the Rossby waves and reach a peak in the eastern part of the MC/MUC (roughly 127°–130°E), but they are very small near the Mindanao coast. As we discussed above, this characteristic gives rise to significant influence of Rossby waves on the MC/MUC by modulating the zonal pressure gradient in the two currents.

Figure 12 presents the temporal mean field of the RGM currents in the first and second layers and compares with satellite observed mean surface geostrophic current from the AVISO dataset averaged over 1993–2013. The RGM layer 1 shows a similar horizontal currents pattern to the AVISO surface geostrophic current. Disparity between the RGM layer 1 and AVISO surface currents might be due to the difference of depths of the two datasets and the ideal topography of the RGM. Both the MC and MUC are well reproduced by the RGM in terms of spatial structure, path, and amplitude. In the second layer, although horizontal currents possess strong mesoscale eddy activities, undercurrents, the MUC for instance, are also very distinct from eddies. The RGM MUC originates at about 5.5°N and flows northward until 14°N. The RGM currents are consistent with the geostrophic currents derived from 14-yr Argo float profiling data presented by Qiu et al. (2015). In particular, Qiu et al. (2015) pointed out that the MUC was observed from 6° to 13°N.

Fig. 12.
Fig. 12.

(a) Satellite-observed mean surface geostrophic current from the AVISO dataset averaged over 1993–2013 and mean horizontal currents in (b) layer 1 and (c) layer 2 of the RGM. Vectors of currents are superposed on the shaded color that indicates the meridional components of mean velocities (cm s−1). Red and blue squares denote the mooring site.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

Sensitivity experiments include five parts: the eastern Pacific (EP) run, the central Pacific (CP) run, the western Pacific (WP) run, the eastern part of the western Pacific (EWP) run, and the far western Pacific (FWP) run. All the experiments utilize a unified model set as introduced above but forced by different wind fields. In the EP run, interannual variability of wind stress is removed except the eastern Pacific Ocean (130°–70°W). Similarly, for the CP, WP, EWP, and FWP runs, interannual variability is permitted only in the central Pacific (180°–130°W), the western Pacific (120°E–180°), the eastern part of western Pacific (150°E–180°), and the far western Pacific (120°–150°E) Ocean, respectively.

MC and MUC transports in the control run FP are about 27 Sverdrups (Sv; 1 Sv ≡ 106 m3 s−1) and 5 Sv and are consistent with previous studies (e.g., Hu and Cui 1991; Lukas et al. 1991; Wang and Hu 1998; Qu et al. 2012). For example, Qu et al. (2012) reported that the MC and MUC are 23.9 and 3.8 Sv. Transport anomalies of the MC and MUC in various sensitivity experiments are presented in the Fig. 13. Among the three parts of Pacific regions, WP contributes the most relative to EP and CP for both the MC and MUC, and in the western Pacific, EWP is more important than the FWP region. Table 1 shows some statistics in terms of the MC and MUC transports in Fig. 13. Correlation coefficients between the FP and WP are 0.8 for the MC and 0.7 for the MUC. But for the MC, the FP–FWP correlation coefficient (0.6) is also significant. To quantify the contributions to the total MC/MUC transport variability, we define the variations Si in the i run as
e3
where i = FWP, EWP, CP, and EP; CORRi is correlation coefficient between the FP run and the i run; and STDi is the standard deviation of the i run. The relative contribution is then calculated as below:
e4
Fig. 13.
Fig. 13.

Anomalous volume transports (Sv) of the (a) MC and (b) MUC in various numerical experiments including FP, CP, EP, WP, FWP, and EWP runs.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

Table 1.

Statistics of the MC and MUC transports (Sv) in the sensitivity experiments. CORR is the correlation coefficients between sensitivity experiments and control run FP. Mean and std dev are mean transports and corresponding standard deviations.

Table 1.

As shown in Fig. 14, the WP wind (FWP plus EWP) accounts for (72% ± 3%) of the MUC variation and (54% ± 2%) of the MC variation, suggesting that the interannual variability of both the MC and MUC are largely controlled by the WP wind forcing. Meanwhile, the EWP wind forcing contributes (29% ± 1%) to the MC variability and (47% ± 2%) to the MUC variability, supporting the results discussed above: Rossby waves generate at the EWP, propagate westward, and contribute to the interannual variability of the MC and MUC in the FWP. Considering that the MUC are related to the baroclinic instability of the overlying wind-driven western boundary currents (Qiu et al. 2015), it is possible that the baroclinic Rossby waves induced by the EWP wind forcing also influence the MUC through modulating the baroclinic instability in the MC. But this hypothesis needs to be confirmed by more subthermocline observations. For the upper-layer MC, the FWP wind is also of significance, pointing to the importance of local Ekman pumping in the FWP as discussed in section 4.

Fig. 14.
Fig. 14.

Percentages of contributions Pi of wind forcings in different regions (FWP, EWP, CP, and EP) to the interannual variability of the (top) MC and (bottom) MUC.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

6. Discussion

Interannual variability of ocean circulations in the western Pacific Ocean are commonly related to the ENSO cycle (e.g., Qiu and Lukas 1996; Kashino et al. 2009; Kashino et al. 2011; Hu and Hu 2014; Hu et al. 2015). During El Niño development, the MC is suggested to be enhanced, transport more cool water equatorward, and contribute to the discharge of the ENSO cycle (Kashino et al. 2009; Hu et al. 2015). Figure 15 depicts the Niño-3.4 index and the SLA averaged over 126°–130°E along 8°N in the MC region as well as their lead–lag correlation. Result indicates that the SLA varies out of phase with the Niño-3.4 index, and the latter leads the former by about 1 month, suggesting a fast response of the SLA in the MC region to the Niño-3.4 index.

Fig. 15.
Fig. 15.

(a) Monthly Niño-3.4 index and SLA averaged over 126°–130°E along 8°N in the MC region, and (b) their lead–lag correlation. Both the Niño-3.4 index and SLA are 13-month running means.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

During the mooring observations, two La Niña events occurred during December 2010–April 2011 and August 2011–March 2012, respectively. Here, the observed velocity anomalies of the MC and MUC in Fig. 4b are normalized and compared with the Niño-3.4 index (Fig. 16). During the two La Niña events, the MUC and MC showed no abnormal variations. The MUC velocity and Niño-3.4 index shows a correlation with a correlation coefficient of −0.33, but no significant relation is found between the MC velocity and Niño-3.4 index. This is consistent with the result by Lukas (1988) and also probably because the ENSO signal during the present mooring observation is not strong enough.

Fig. 16.
Fig. 16.

Comparison between the Niño-3.4 index and the velocity anomalies averaged over 80–550 (MC) and 550–750 m (MUC) as in Fig. 4b. The velocity anomalies are normalized by their standard deviations.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

Previous studies also suggest that subthermocline eddies exist in the western Pacific Ocean below the upper layer and influence the subsurface currents (e.g., Qu et al. 2012; Chiang and Qu 2013). Figure 17 presents meridional velocity anomaly and eddy kinetic energy (EKE) averaged over 5°–10°N and 126°–130°E at 605 m. EKE is defined as
e5
where u′ and υ′ are 3-day zonal and meridional velocity anomalies relative to the temporal mean over 1980–2011 from the Ocean General Circulation Model for the Earth Simulator (OFES) output, which is eddy resolving with a horizontal resolution of 0.1° × 0.1° (Sasaki et al. 2008). It seems that the subthermocline EKE possesses significant interannual fluctuations and a close relationship with (Fig. 17), and statistically the latter leads the EKE by about 3 months. This indicates that the interannual variability of the background mean currents probably play an important role in the interannual variation of the subthermocline eddies. But the interaction between the mean circulation and subthermocline eddies needs further studies. As suggested by Qiu et al. (2015), the broad-scale subthermocline boundary flows (including the MUC) are related to the baroclinic instability of the overlying wind-driven western boundary currents. Kashino et al. (2011) pointed out that the interannual variability of temperature and heat content at 8°N (130° and 137°E) are different from that at 5°N, 137°E. For the MC and MUC, the latitude dependence of variability on such a time scale probably also exists but remains to be addressed.
Fig. 17.
Fig. 17.

(a) OFES EKE and meridional velocity anomalies averaged over 5°–10°N and 126°–130°E, and the (b) lead–lag correlation between them. Both the time series are 1-yr running averaged. Positive time lag indicates EKE lags the meridional velocity anomalies.

Citation: Journal of Physical Oceanography 46, 2; 10.1175/JPO-D-15-0092.1

In addition, two serious facts should be noted: First, the depth of MUC is roughly about 600 m to deeper than 800 m (e.g., Qiu et al. 2015). This implies that observations shallower than 600 m naturally miss the MUC core. Deductions of the MUC based on this kind of observations are probably misleading to readers and to a certain extent lead to the dispute on the existence of the MUC. Second, the MUC is referred to be a mean flow and an undercurrent. A mean flow like MUC typically has a very different time scale from mesoscale to small-scale processes, as suggested by Qiu et al. (2015).

7. Conclusions

On the basis of nearly 4-yr ADCP measurements from subsurface moorings in the south Philippine Sea, we investigate the interannual variability of the observed MC and MUC during December 2010–August 2014. Both the MC and MUC are characterized by significant interannual variability but different frequencies and amplitudes. By combining diagnostic analysis and numerical sensitivity experiments, we conclude that the wind forcing over the western Pacific Ocean acts as a driving agent in conditioning the interannual variability of MC and MUC. Westward-propagating Rossby waves generated by the EWP wind forcing might play an essential role in the interannual variability of the MC/MUC by modulating the zonal pressure gradient in the two boundary currents and changing the baroclinic instability in the MC. For the upper-layer MC, the fluctuation of local wind stress curl over the FWP in the south Philippine Sea is also of much importance. Relationships between the MC/MUC and ENSO cycle and between the MC/MUC and subthermocline eddy activities are also discussed.

Some important issues remain. The spatial structure and path of the MUC cannot be observed by a single mooring. The vertical range of the ADCP observation in the present paper is not long enough to capture the entire MUC. Interaction between the mean currents and mesoscale eddies, especially in the subthermocline layer, is not clear yet because of the lack of field observations in the subthermocline ocean there. These issues point to the significance of extending and enhancing the mooring observations of the western boundary currents in this area.

Acknowledgments

The authors express their sincere gratitude to the crews of R/V Science 1 and R/V Science and all scientists and technicians onboard for the deployment and retrieval of these subsurface moorings. We are gratefully indebted to Bo Qiu for his generous support in terms of the RGM and insightful suggestions. Comments and constructive suggestions from two anonymous reviewers and Gregory Foltz are of much help. We thank Janet Sprintall and Jinbo Wang for their very valuable comments on how to improve the manuscript. Discussion with Ru Chen and Junqiao Feng was helpful. The ECMWF ORA-S3 wind stress data are provided by the European Centre for Medium-Range Weather Forecasts. The ERDDAP wind stress data are provided by the ERDDAP data server at NOAA (http://coastwatch.pfeg.noaa.gov/erddap/griddap/erdQAstressmday.html). The ECCO2 dataset over 2011–13 is from http://ecco2.jpl.nasa.gov/. The OSCAR data were obtained from http://www.oscar.noaa.gov. The SLA data are distributed by the AVISO can be found online (at http://www.aviso.oceanobs.com/duacs/). RG Argo dataset is provided by the Scripps Institution of Oceanography (http://sio-argo.ucsd.edu/RG_Climatology.html). The OFES outputs are provided by the JAMSTEC (http://www.jamstec.go.jp/esc/research/AtmOcn/product/ofes.html). The mooring data are available at the NPOCE website (http://npoce.qdio.ac.cn/moored).This work was supported by the National Natural Science Foundation of China (Grants 41406016 and 41330963), the NSFC-Shandong Joint Fund for Marine Science Research Centers (Grant U1406401), the CAS Strategic Priority Research Program (Grant XDA11010101), and the National Key Basic Research Program of China (Program 973, Grant 2013CB956202).

REFERENCES

  • Balmaseda, M. A., A. Vidard, and D. L. T. Anderson, 2008: The ECMWF ocean analysis system: ORA-S3. Mon. Wea. Rev., 136, 30183034, doi:10.1175/2008MWR2433.1.

    • Search Google Scholar
    • Export Citation
  • Book, J. W., H. Perkins, R. P. Signell, and M. Wimbush, 2007: The Adriatic Circulation Experiment winter 2002/2003 mooring data report: A case study in ADCP data processing. NRL Memo. Rep. NRL/MR/7330--07-8999, 50 pp.

  • Chelton, D. B., and M. G. Schlax, 1996: Global observations of oceanic Rossby waves. Science, 272, 234238, doi:10.1126/science.272.5259.234.

    • Search Google Scholar
    • Export Citation
  • Chiang, T.-L., and T. Qu, 2013: Subthermocline eddies in the western equatorial Pacific as shown by an eddy-resolving OGCM. J. Phys. Oceanogr., 43, 12411253, doi:10.1175/JPO-D-12-0187.1.

    • Search Google Scholar
    • Export Citation
  • Dibarboure, G., O. Lauret, F. Mertz, and V. Rosmorduc, 2009: SSALTO/DUACS user handbook: (M)SLA and (M)ADT near-real time and delayed time products. CNES Rep. CLS‐DOS‐NT-06-034, 74 pp.

  • Efron, B., 1979: Bootstrap methods: Another look at the jackknife. Ann. Stat., 7, 126, doi:10.1214/aos/1176344552.

  • Firing, E., Y. Kashino, and P. Hacker, 2005: Energetic subthermocline currents observed east of Mindanao. Deep-Sea Res., 52, 605613, doi:10.1016/j.dsr2.2004.12.007.

    • Search Google Scholar
    • Export Citation
  • Ganachaud, A., and Coauthors, 2014: The Southwest Pacific Ocean Circulation and Climate Experiment (SPICE). J. Geophys. Res. Oceans, 119, 76607686, doi:10.1002/2013JC009678.

    • Search Google Scholar
    • Export Citation
  • Gordon, A. L., and R. A. Fine, 1996: Pathways of water between the Pacific and Indian Oceans in the Indonesian Seas. Nature, 379, 146149, doi:10.1038/379146a0.

    • Search Google Scholar
    • Export Citation
  • Gu, D., and S. G. H. Philander, 1997: Interdecadal climate fluctuations that depend on exchanges between the tropics and extratropics. Science, 275, 805807, doi:10.1126/science.275.5301.805.

    • Search Google Scholar
    • Export Citation
  • Hu, D., and M. Cui, 1989: The western boundary current in the far-western Pacific Ocean. Proc. Western Pacific Int. Meeting and Workshop on TOGA COARE, ORSTOM, Nouméa, New Caledonia, 123–134.

  • Hu, D., and M. Cui, 1991: The western boundary current of the Pacific and its role in the climate. Chin. J. Oceanol. Limnol., 9, 114, doi:10.1007/BF02849784.

    • Search Google Scholar
    • Export Citation
  • Hu, D., and Coauthors, 2011: Northwestern Pacific Ocean Circulation and Climate Experiment (NPOCE) Science/Implementation Plan. China Ocean Press, 100 pp.

  • Hu, D., and Coauthors, 2013: Direct measurements of the Luzon Undercurrent. J. Phys. Oceanogr., 43, 14171425, doi:10.1175/JPO-D-12-0165.1.

    • Search Google Scholar
    • Export Citation
  • Hu, D., and Coauthors, 2015: Pacific western boundary currents and their roles in climate. Nature, 522, 299308, doi:10.1038/nature14504.

    • Search Google Scholar
    • Export Citation
  • Hu, S., and D. Hu, 2012: Heat center of the western Pacific warm pool. Chin. J. Oceanol. Liimol., 30, 169176, doi:10.1007/s00343-012-1193-9.

    • Search Google Scholar
    • Export Citation
  • Hu, S., and D. Hu, 2014: Variability of the Pacific North Equatorial Current from repeated shipboard acoustic Doppler current profiler measurements. J. Oceanogr., 70, 559571, doi:10.1007/s10872-014-0253-5.

    • Search Google Scholar
    • Export Citation
  • Jin, F. F., 1997: An equatorial ocean recharge paradigm for ENSO. Part I: Conceptual model. J. Atmos. Sci., 54, 811829, doi:10.1175/1520-0469(1997)054<0811:AEORPF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Kashino, Y., N. Espaa, F. Syamsudin, K. J. 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 Niño and 2007/08 La Niña. J. Oceanogr., 65, 325333, doi:10.1007/s10872-009-0030-z.

    • Search Google Scholar
    • Export Citation
  • Kashino, Y., A. Ishida, and S. Hosoda, 2011: Observed ocean variability in the Mindanao Dome region. J. Phys. Oceanogr., 41, 287302, doi:10.1175/2010JPO4329.1.

    • Search Google Scholar
    • Export Citation
  • Kashino, Y., I. Ueki, and H. Sasaki, 2015: Ocean variability east of Mindanao: Mooring observations at 7°N, revisited. J. Geophys. Res. Oceans, 120, 25402554, doi:10.1002/2015JC010703.

    • Search Google Scholar
    • Export Citation
  • Kim, Y. Y., T. Qu, T. Jensen, T. Miyama, H. Mitsudera, H. W. 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.

    • Search Google Scholar
    • Export Citation
  • Lukas, R., 1988: Interannual fluctuations of the Mindanao Current inferred from sea level. J. Geophys. Res., 93, 67446748, doi:10.1029/JC093iC06p06744.

    • Search Google Scholar
    • Export Citation
  • Lukas, R., E. Firing, P. Hacker, P. L. Richardson, C. A. Collins, R. Fine, and R. Gammon, 1991: Observations of the Mindanao Current during the western equatorial Pacific Ocean circulation study. J. Geophys. Res., 96, 70897104, doi:10.1029/91JC00062.

    • Search Google Scholar
    • Export Citation
  • Lukas, R., T. Yamagata, and J. P. McCreary, 1996: Pacific low-latitude western boundary currents and the Indonesian throughflow. J. Geophys. Res., 101, 12 20912 216, doi:10.1029/96JC01204.

    • Search Google Scholar
    • Export Citation
  • Masumoto, Y., and T. Yamagata, 1991: Response of the western tropical Pacific to the Asian winter monsoon: The generation of the Mindanao Dome. J. Phys. Oceanogr., 21, 13861398, doi:10.1175/1520-0485(1991)021<1386:ROTWTP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • McCreary, J. P., Jr., and P. Lu, 1994: Interaction between the subtropical and equatorial ocean circulations: The subtropical cell. J. Phys. Oceanogr., 24, 466497, doi:10.1175/1520-0485(1994)024<0466:IBTSAE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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 31512 330, doi:10.1029/95JC03204.

    • Search Google Scholar
    • Export Citation
  • Qiu, B., D. A. Koh, C. Lumpkin, and P. Flament, 1997: Existence and formation mechanism of the North Hawaiian Ridge Current. J. Phys. Oceanogr., 27, 431444, doi:10.1175/1520-0485(1997)027<0431:EAFMOT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Qiu, B., S. Chen, D. L. Rudnick, and Y. Kashino, 2015: A new paradigm for the North Pacific subthermocline low-latitude western boundary current system. J. Phys. Oceanogr., 45, 24072423, doi:10.1175/JPO-D-15-0035.1.

    • Search Google Scholar
    • Export Citation
  • Qu, T., H. Mitsudera, and T. Yamagata, 1998: On the western boundary currents in the Philippine Sea. J. Geophys. Res., 103, 75377548, doi:10.1029/98JC00263.

    • Search Google Scholar
    • Export Citation
  • Qu, T., T. L. Chiang, C. R. Wu, P. Dutrieux, and D. Hu, 2012: Mindanao Current/Undercurrent in an eddy-resolving GCM. J. Geophys. Res., 117, C06026, doi:10.1029/2011JC007838.

    • Search Google Scholar
    • Export Citation
  • Roemmich, D., and J. Gilson, 2009: The 2004–2008 mean and annual cycle of temperature, salinity, and steric height in the global ocean from the Argo program. Prog. Oceanogr., 82, 81100, doi:10.1016/j.pocean.2009.03.004.

    • Search Google Scholar
    • Export Citation
  • 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. High Resolution Numerical Modelling of the Atmosphere and Ocean, K. Hamilton and W. Ohfuchi, Eds., Springer, 157–185.

  • Smith, S. D., 1988: Coefficients for sea surface wind stress, heat flux, and wind profiles as a function of wind speed and temperature. J. Geophys. Res., 93, 15 46715 472, doi:10.1029/JC093iC12p15467.

    • Search Google Scholar
    • Export Citation
  • Sprintall, J., A. L. Gordon, A. Koch-Larrouy, T. Lee, J. T. Potemra, K. Pujiana, and S. E. Wijffels, 2014: The Indonesian seas and their role in the coupled ocean-climate system. Nat. Geosci., 7, 487492, doi:10.1038/ngeo2188.

    • Search Google Scholar
    • Export Citation
  • 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, 13381359, doi:10.1175/1520-0485(2002)032<1338:SMVITW>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Walters, R. A., and C. Heston, 1982: Removing tidal-period variations from time-series data using low-pass digital filters. J. Phys. Oceanogr., 12, 112115, doi:10.1175/1520-0485(1982)012<0112:RTPVFT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wang, F., and D. Hu, 1998: Dynamic and thermohaline properties of the Mindanao Undercurrent I. Dynamic structure. Chin. J. Oceanol. Limnol., 16, 122127, doi:10.1007/BF02845177.

    • Search Google Scholar
    • Export Citation
  • Wang, F., and D. Hu, 1999: Preliminary study on the formation mechanism of counter western boundary undercurrents below the thermocline—A conceptual model. Chin. J. Oceanol. Limnol., 17, 19, doi:10.1007/BF02842694.

    • Search Google Scholar
    • Export Citation
  • Wang, F., N. Zang, Y. Li, and D. Hu, 2015: On the subsurface countercurrents in the Philippine Sea. J. Geophys. Res. Oceans, 120, 131144, doi:10.1002/2013JC009690.

    • Search Google Scholar
    • Export Citation
  • Wang, Q., F. Zhai, F. Wang, and D. Hu, 2014: Intraseasonal variability of the subthermocline current east of Mindanao. J. Geophys. Res. Oceans, 119, 85528566, doi:10.1002/2014JC010343.

    • Search Google Scholar
    • Export Citation
  • 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 42118 435, doi:10.1029/95JC01347.

    • Search Google Scholar
    • Export 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, 36173628, doi:10.1002/2013JC009693.

    • Search Google Scholar
    • Export Citation
Save