• Akima, H., 1970: A new method of interpolation and smooth curve fitting based on local procedures. J. Assoc. Comput. Mech., 17 , 589602.

    • Search Google Scholar
    • Export Citation
  • Argo Data Management Team, 2004: Argo quality control manual, version 2.0b. Argo Data Management, 23 pp.

  • Argo Science Team, 2001: Argo: The global array of profiling floats. Observing the Oceans in the 21st Century, C. J. Koblinsky and N. R. Smith, Eds., GODAE Project Office, 248–258.

    • Search Google Scholar
    • Export Citation
  • Giese, B. S., , S. C. Urizar, , and N. S. Fuckar, 2002: Southern Hemisphere origins of the 1976 climate shift. Geophys. Res. Lett., 29 , 1014. doi:10.1029/2001GL013268.

    • Search Google Scholar
    • Export Citation
  • Hanawa, K., , and L. D. Talley, 2001: Mode waters. Ocean Circulation and Climate—Observing and Modelling the Global Ocean, G. Siedler, J. Church, and J. Gould, Eds., Academic Press, 373–386.

    • Search Google Scholar
    • Export Citation
  • Hautala, S. L., , and D. H. Roemmich, 1998: Subtropical mode water in the Northeast Pacific Basin. J. Geophys. Res., 103 , (C6). 1305513066.

    • Search Google Scholar
    • Export Citation
  • Hosoda, S., , T. Ohira, , and T. Nakamura, 2008: A monthly mean dataset of global oceanic temperature and salinity derived from Argo float observations. JAMSTEC Rep. Res. Develop., 8 , 4769.

    • Search Google Scholar
    • Export Citation
  • Johnson, G. C., 2006: Generation and initial evolution of a mode water θ–S anomaly. J. Phys. Oceanogr., 22 , 184195.

  • Johnson, G. C., , and M. J. McPhaden, 1999: Interior pycnocline flow from the subtropical to the equatorial Pacific Ocean. J. Phys. Oceanogr., 29 , 30733089.

    • Search Google Scholar
    • Export Citation
  • Masuzawa, J., 1969: Subtropical mode water. Deep-Sea Res., 16 , 463472.

  • Nonaka, M., , and H. Sasaki, 2007: Formation mechanism for isopycnal temperature–salinity anomalies propagating from the eastern South Pacific to the equatorial region. J. Climate, 20 , 13051315.

    • Search Google Scholar
    • Export Citation
  • Provost, C., , C. Escoffier, , K. Maamaatuaiahutapu, , A. Kartavtseff, , and V. Garcon, 1999: Subtropical mode waters in the South Atlantic Ocean. J. Geophys. Res., 104 , (C9). 2103321049.

    • Search Google Scholar
    • Export Citation
  • Roemmich, D., , and B. Cornuelle, 1992: The Subtropical Mode Water of the South Pacific Ocean. J. Phys. Oceanogr., 22 , 11781187.

  • Ruddick, B., 1983: A practical indicator of the stability of the water column to double-diffusive activity. Deep-Sea Res., 30A , 11051107.

    • Search Google Scholar
    • Export Citation
  • Schmitt, R. W., 1981: Form of the temperature–salinity relationship in the Central Water: Evidence for double-diffusive mixing. J. Phys. Oceanogr., 20 , 10151026.

    • Search Google Scholar
    • Export Citation
  • Siedler, G., , A. Kuhl, , and W. Zenk, 1987: The Madeira Mode Water. J. Phys. Oceanogr., 17 , 15611570.

  • St. Laurent, L., , and R. W. Schmitt, 1999: The contribution of salt fingers to vertical mixing in the North Atlantic Tracer Release Experiment. J. Phys. Oceanogr., 29 , 14041424.

    • Search Google Scholar
    • Export Citation
  • Suga, T., , K. Hanawa, , and Y. Toba, 1989: Subtropical mode water in the 137°E section. J. Phys. Oceanogr., 19 , 16051618.

  • Tsubouchi, T., , T. Suga, , and K. Hanawa, 2007: Three types of South Pacific subtropical mode waters: Their relation to the large-scale circulation of the South Pacific subtropical gyre and their temporal variability. J. Phys. Oceanogr., 37 , 24782490.

    • Search Google Scholar
    • Export Citation
  • Tsuchiya, M., , and L. D. Talley, 1996: Water-property distributions along an eastern Pacific hydrographic section at 135°W. J. Mar. Res., 54 , 541564.

    • Search Google Scholar
    • Export Citation
  • Wong, A. P. S., , and G. C. Johnson, 2003: South Pacific Eastern Subtropical Mode Water. J. Phys. Oceanogr., 33 , 14931509.

  • Worthington, L. V., 1959: The 18°C Water in the Sargasso Sea. Deep-Sea Res., 5 , 297305.

  • Yeager, S. G., , and W. G. Large, 2004: Late-winter generation of spiciness on subducted isopycnals. J. Phys. Oceanogr., 34 , 15281547.

  • Zhang, J., , R. W. Schmitt, , and R. X. Huang, 1998: Sensitivity of the GFDL modular ocean model to parameterization of double-diffusive processes. J. Phys. Oceanogr., 28 , 589605.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    The climatological austral spring distribution of potential vorticity on the 25.3 kg m−3 isopycnal surface, superimposed on the pressure anomaly streamfunction (contour interval of 0.5 m2 s−2) relative to 2000 dbar on the same isopycnal surface, by averaging the MOAA GPV dataset from October to December during 2000–07. The red line denotes the wintertime outcrop of the 25.3 kg m−3 isopycnal surface. The green line denotes the trajectory of the float with World Meteorological Organization (WMO) ID 3900255, which is shown in Figs. 10 and 11.

  • View in gallery

    (left) The relation between potential vorticity and density without the mixed layer at 5°–35°S, 140°–70°W from October to December (austral spring) in 2004–06. (right) A θS diagram at the depth of the minimum potential vorticity, for profiles with the potential vorticity less than 2.5 × 10−10 m−1 s−1. This diagram represents the frequency of each θS class in the same region and time period as in the left panel.

  • View in gallery

    The distribution of the properties at the core of the mode water, defined as the water with potential vorticity less than 2.5 × 10−10 m−1 s−1 and with thickness exceeding 40 m from October to December in 2006: (a) temperature, (b) salinity, (c) density, (d) thickness, (e) potential vorticity, and (f) the temperature gradient component of the potential vorticity QT. Circles (colored pluses) denote the water with (without) a Turner angle between 70° and 90° in the potential vorticity minimum layer. Red lines in (e) and (f) denote the climatological outcrop density lines of 24.5 kg m−3 (northern line) and 25.8 kg m−3 (southern line) in austral winter (July–September).

  • View in gallery

    The seasonal distribution of the minimum potential vorticity of each profile with the mode water, defined in the same manner as in Fig. 3 from July 2006 to June 2007: (a) July–September (winter), (b) October–December (spring), (c) January–March (summer), and (d) April–June (fall). Circles and colored plus signs have the same meaning as in Fig. 3.

  • View in gallery

    Same as in Fig. 4, but for the temperature gradient component of the potential vorticity at the core of the mode water. Circles and colored plus signs have the same meaning as in Fig. 3.

  • View in gallery

    The seasonal frequency of the (a) potential vorticity, (b) temperature gradient component of the potential vorticity, (c) (negative) salinity gradient component of the potential vorticity, and (d) Turner angle at the SPESTMW core in 5°–35°S, 160°–70°W from July 2006 to June 2007. All frequencies are normalized by the number of profiles in the region described above in each season.

  • View in gallery

    Same as in Fig. 6, but at the NPSTMW cores in 20°–40°N, 120°E–180° from January to December 2005.

  • View in gallery

    Turner angle profile from October to December in 2006 vs the relative depth to the SPESTMW core (green and black circles); a positive (negative) value represents the depth below (above) the SPESTMW core. The grid values within SPESTMW are plotted as green circles, and those outside of the SPESTMW are plotted as black circles. Only Tu values from 71.6° to 90° are displayed. (red line) Occurrence frequency of the depth with maximum Tu in the range between 71.6° and 90° in each profile. The number in the lower-right corner of the figure denotes the proportion of the profiles with Tu exceeding 77° to all available profiles from October to December in 2006.

  • View in gallery

    The seasonal distribution of the maximum Turner angle of each profile, with the mode water defined in the same way as in Fig. 3 from July 2006 to June 2007: (a) July–September, (b) October–December, (c) January–March, and (d) April–June. Circles and colored plus signs have the same meaning as in Fig. 3.

  • View in gallery

    Time–depth sections of the diffusive terms of (top) heat, (middle) salt, and (bottom) density computed using the salt fingering diffusive parameterization. Time–depth sections of potential temperature, salinity, and density (solid-line contour) obtained by the float with WMO ID 3900255 are superimposed. Green circles denote grids in the SPESTMW.

  • View in gallery

    Time series of (left) temperature and (right) salinity on the isopycnal surface. The circles denote values observed by the float with WMO ID 3900255, as in Fig. 10. The stars denote the values modeled by time integration of the diapycnal diffusion term of the temperature and salinity from Fig. 10.

  • View in gallery

    Time series of the density diffusive term due to salt fingering between two isopycnal surfaces within a 0.1 kg m−3 interval, from 25.0 to 25.5 kg m−3, calculated from the float with WMO ID 3900255 as in Fig. 10.

  • View in gallery

    The seasonal frequency of (left) potential temperature and (right) salinity from (a) 25.1 to 25.3 kg m−3 and (b) 25.3 to 25.5 kg m−3, within the SPESTMW in 5°–35°S, 160°–70°W from July 2005 to June 2006. All frequencies are normalized by the number of profiles in the region described above in each season.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 41 41 4
PDF Downloads 31 31 2

Structure and Modification of the South Pacific Eastern Subtropical Mode Water

View More View Less
  • 1 Ocean Climate Change Research Program, Research Institute for Global Change, Yokosuka, Japan
  • | 2 Ocean Climate Change Research Program, Research Institute for Global Change, Yokosuka, and Department of Geophysics, Graduate School of Science, Tohoku University, Sendai, Japan
© Get Permissions
Full access

Abstract

Using all available temperature and salinity profiles obtained by Argo floats from July 2004 to June 2007, this study investigated the structure and modification of the South Pacific Eastern Subtropical Mode Water (SPESTMW). Based on the observed characteristics of the vertical minima of potential vorticity over the subtropical South Pacific, SPESTMW is defined as water with potential vorticity magnitude less than 2.5 × 10−10 m−1 s−1 and thickness exceeding 40 m. It is found between 35°–5°S and 160°–70°W and has a temperature of 13°–26°C, salinity greater than 34.0, and density of 24.5–25.8 kg m−3 at its core.

This study confirmed that vertical changes in temperature and salinity tend to compensate for each other in terms of density changes, resulting in favorable salt fingering conditions, as previously reported. By analyzing many profiles of Argo data in spring immediately after the SPESTMW formation period, its temperature and salinity are vertically uniform in the formation region, but large vertical gradients of temperature and salinity are found downstream from that region, even in the SPESTMW core. Consequently, the low potential vorticity signature of SPESTMW spread much wider than its signature as a thermostad. The Argo data also captured the seasonal changes of the vertical gradients of temperature and salinity at the SPESTMW core; these gradients increased as the seasons progressed, even in the formation region. Therefore, SPESTMW is truly vertically uniform water (i.e., thermostad, halostad, and pycnostad simultaneously) only immediately after the formation period. Afterward, it is only pycnostad. This seasonal evolution is related to temperature and salinity diffusion due to salt fingering in a manner similar to the rapid modification of interannual anomalies as shown by previous research. The temperature and salinity near the SPESTMW core and lower region decreased soon after its formation.

Corresponding author address: Kanako Sato, Ocean Climate Change Research Program, Research Institute for Global Change, Japan Agency for Marine-Earth Science and Technology, 2-15 Natsushima-cho, Yokosuka 237-0061, Japan. Email: k_sato@jamstec.go.jp

Abstract

Using all available temperature and salinity profiles obtained by Argo floats from July 2004 to June 2007, this study investigated the structure and modification of the South Pacific Eastern Subtropical Mode Water (SPESTMW). Based on the observed characteristics of the vertical minima of potential vorticity over the subtropical South Pacific, SPESTMW is defined as water with potential vorticity magnitude less than 2.5 × 10−10 m−1 s−1 and thickness exceeding 40 m. It is found between 35°–5°S and 160°–70°W and has a temperature of 13°–26°C, salinity greater than 34.0, and density of 24.5–25.8 kg m−3 at its core.

This study confirmed that vertical changes in temperature and salinity tend to compensate for each other in terms of density changes, resulting in favorable salt fingering conditions, as previously reported. By analyzing many profiles of Argo data in spring immediately after the SPESTMW formation period, its temperature and salinity are vertically uniform in the formation region, but large vertical gradients of temperature and salinity are found downstream from that region, even in the SPESTMW core. Consequently, the low potential vorticity signature of SPESTMW spread much wider than its signature as a thermostad. The Argo data also captured the seasonal changes of the vertical gradients of temperature and salinity at the SPESTMW core; these gradients increased as the seasons progressed, even in the formation region. Therefore, SPESTMW is truly vertically uniform water (i.e., thermostad, halostad, and pycnostad simultaneously) only immediately after the formation period. Afterward, it is only pycnostad. This seasonal evolution is related to temperature and salinity diffusion due to salt fingering in a manner similar to the rapid modification of interannual anomalies as shown by previous research. The temperature and salinity near the SPESTMW core and lower region decreased soon after its formation.

Corresponding author address: Kanako Sato, Ocean Climate Change Research Program, Research Institute for Global Change, Japan Agency for Marine-Earth Science and Technology, 2-15 Natsushima-cho, Yokosuka 237-0061, Japan. Email: k_sato@jamstec.go.jp

1. Introduction

Mode water is characterized as a layer of nearly vertically homogeneous water found over a relatively large geographical area (Hanawa and Talley 2001). Mode waters occur above or within the top of the permanent pycnocline and are apparent through the contrast in stratification with the pycnocline waters. They are formed in the wintertime deep mixed layer and subducted into the ocean interior. Mode waters store information about the variations of atmospheric forcing and air–sea interaction in the formation region. The distribution and circulation of mode waters are also influenced by the dynamical conditions of the ocean: the activities of eddies in the formation area and distribution region, and the strength of the gyre circulation. Meanwhile, because mode waters have a huge volume, the variations in mode water properties and distribution influence the oceanic storage and transport of heat and freshwater.

Eastern subtropical mode water (ESTMW) is a type of mode water distributed in the eastern part of each subtropical gyre of the World Ocean, except in the Indian Ocean (Siedler et al. 1987; Hautala and Roemmich 1998; Provost et al. 1999; Wong and Johnson 2003). In each subtropical gyre, the density of ESTMW is generally similar to that of subtropical mode water (STMW) distributed on the equatorward side of the western boundary current extension (Worthington 1959; Masuzawa 1969; Provost et al. 1999; Roemmich and Cornuelle 1992). On the other hand, in all subtropical gyres except that in the South Pacific, the spatial extent and volume of ESTMW are much smaller than those of STMW. ESTMW in the South Pacific (SPESTMW), however, has a much larger volume and spatial extent than STMW.

The regions where potential vorticity is less than 3 × 10−10 m−1 s−1 on the eastern side of the South Pacific and on the 25.3 kg m−3 isopycnal surface, which is near the center of the density range occupied by SPESTMW (Wong and Johnson 2003), have been identified from a climatological map based on the Grid Point Value of Monthly Objective Analysis for Argo data (MOAA GPV) dataset (Hosoda et al. 2008) (Fig. 1). This global 1° grid dataset of monthly temperature and salinity distributions has been estimated from available profiles of Argo float, Triangle Trans-Ocean Buoy Network (TRITON) buoy, and conductivity–temperature–depth (CTD) casts, using the two-dimensional optimal interpolation method on pressure levels from the surface to 2000 dbar and on density levels. We found that SPESTMW spreads westward to about 140°W, and that its distribution area is wider than that of STMW (see Fig. 2 in Tsubouchi et al. 2007). It is therefore likely that SPESTMW affects the structure and circulation of the subtropical pycnocline in the South Pacific subtropical gyre on a basin scale, unlike waters in the other subtropical gyres. Moreover, SPESTMW is advected northwestward from the outcrop region after its formation. The isopycnal surface corresponding to the core of SPESTMW has a direct route to the equator (Fig. 1; e.g., Johnson and McPhaden 1999). Several studies using general circulation models found that the decadal-scale spiciness anomaly that is advected equatorward from the subtropical South Pacific to the tropical Pacific is large on this isopycnal surface (Giese et al. 2002; Yeager and Large 2004; Nonaka and Sasaki 2007). Therefore, it is important to clarify the formation process, distribution, and vertical structure of SPESTMW to understand not only the subtropical gyre variability, but also the spiciness variation on a decadal scale in the tropical Pacific.

SPESTMW was first detected and described by Tsuchiya and Talley (1996), based on the data from the World Ocean Circulation Experiment’s (WOCE) hydrographic section P17 at 135°W. Wong and Johnson (2003) described the geographic extent, water properties, vertical structure, and formation of SPESTMW using all the data available at that time from WOCE hydrographical sections, a high-resolution expendable bathythermograph (XBT) line, and several Argo profiling floats, although the spatial distribution of the Argo data was coarse. SPESTMW occupies a relatively small density range, despite its wide range of temperature and salinity values. Wong and Johnson (2003) indicated that this feature results from the density-compensating vertical and horizontal gradients of temperature and salinity. In other words, warm salty water overlies cold freshwater, and the wintertime horizontal surface water properties range from colder and fresher in the southeast to warmer and saltier in the northwest. In particular, this vertical structure of temperature (T) and salinity (S) considerably reduces the density stratification and results in a water mass with relatively low potential vorticity. On the other hand, Wong and Johnson suggested that the density-compensating structure of temperature and salinity may hasten SPESTMW destruction by double-diffusive convective mixing due to salt fingering in the lower part and base of the SPESTMW. Johnson (2006) quantitatively estimated the amount of TS variation on isopycnal surfaces due to salt fingering, using profiles observed by two Argo floats. He showed that the vertical mixing may be sufficiently vigorous to reduce the temperature–salinity anomaly on the isopycnal surface near the bottom of the SPESTMW created in the wintertime mixed layer. This could occur within 6 months after formation, by spreading of the anomaly to denser regions through diapycnal fluxes.

While previous studies have revealed a number of interesting features of SPESTMW, as described above, a full picture of SPESTMW that integrates all these features, including the seasonal evolution of its spatial extent and properties, has not yet been presented. This is because the data available to previous studies were spatially and temporally limited. For example, the vigorous vertical diffusion of the temperature–salinity anomaly described by Johnson (2006) was only analyzed using data from the few Argo floats present in the center of the formation at the time.

Since 2000, the international Argo project has deployed profiling floats over the global ocean, with a target density of one per 3° square, to monitor temperature and salinity down to a depth of 2000 dbar over 10-day intervals (Argo Science Team 2001). In the subtropical South Pacific, about 600 floats are in operation as of June 2007. These floats are also located without spatial bias. Using the observational data provided by Argo floats for 3 yr from July 2004 to June 2007, we examined the horizontal and vertical structures of SPESTMW and their seasonal changes in detail. Furthermore, we tried to elucidate processes responsible for seasonal changes in the water.

The data and analysis methods are explained in section 2. The characteristics of SPESTMW detected by Argo data are described in section 3. The vertical structure and seasonal changes of SPESTMW are examined in section 4. The seasonal changes of the TS characteristics in SPESTMW are investigated in section 5. In section 6, we try to clarify the factors controlling the seasonal changes of the vertical structure and TS characteristics in SPESTMW. Finally, the conclusions are given in section 6.

2. Data and analysis methods

We used all available temperature and salinity profiles observed by Argo floats in operation in the subtropical South Pacific (5°–40°S, 120°E–70°W) from July 2004 to June 2007. The deployment of the Argo floats started in 2000 in the South Pacific, and the number of operational floats has increased rapidly since 2004. The floats drift freely at a predetermined parking pressure (typically 1000 dbar) and ascend to the sea surface at predetermined intervals (10 days) after descending to the maximum pressure (2000 dbar). During the ascent, the floats measure temperature and conductivity at about 60–110 sampling pressure levels, using a CTD sensor module. The collected data and float location are transmitted from the surfaced floats to satellites and are made freely available within 24 h, after passing through the Argo real-time quality control procedure (Argo Data Management Team 2004).

The real-time and delayed-mode quality-controlled data used for the present analysis were obtained from the U.S. Global Ocean Data Assimilation Experiment (USGODAE) Argo Global Data Assembly Center (GDAC) in July 2007.1 We did not use the profiles of SOLO floats deployed by the Woods Hole Oceanographic Institution (WHOI) because the profiles of these floats had incorrect pressure values. One of the major problems with these data is that the profiles observed by the SOLO floats with Falmouth Scientific, Inc. (FSI), CTD may be offset upward by one or more pressure levels, resulting in a significant cold bias. More detailed information on this problem is available online (http://www-argo.ucsd.edu/Acpres_offset2.html).

While the Argo floats observe temperature and conductivity during the ascent, some floats also observe them during the descent. We used only data observed during the ascent because the position data of profiles obtained during the descent may be away from the true position. We first eliminated temperature and salinity profiles with quality flags at any measured levels that were not coded as “1” (i.e., “good” data), or which exhibited density inversion. We selected profiles with a measured shallowest (deepest) depth of less than 20 (more than 500) dbar. We also selected profiles for which the vertical sampling interval from the surface to 300 dbar was less than 30 dbar. We excluded profiles with suspicious salinity data as follows. We removed the profiles that had temperature of less than 3°C and salinity that was less than 34, or temperature of less than 8°C and salinity that was larger than 35. We also removed the profiles for which temperature and salinity considerably departed from the temperature–salinity relation of most profiles by a visual scan. Ultimately, we used 26 787 profiles containing temperature and salinity data observed from 308 floats.

Each profile of temperature and salinity was interpolated vertically onto 10-dbar grids using the Akima spline (Akima 1970) and then onto 0.01 kg m−3 grids using linear interpolation. The potential vorticity Q was calculated under the assumption that the relative vorticity was negligible and the hydrostatic approximation was valid:
i1520-0485-39-7-1700-eq1
where f, g, p, and ρ are the Coriolis parameter, gravity, pressure, and density referenced to the central value of adjacent observed pressure levels, respectively. Furthermore, to investigate the vertical structure of SPESTMW, we calculated the vertical gradients of temperature and salinity, and their contributions to the potential vorticity QT and QS:
i1520-0485-39-7-1700-eq2
where α and β are the thermal expansion coefficient and the haline contraction coefficient, respectively. To calculate the pressure deviation of temperature, salinity, and density at pressure p, the corresponding values at p + 20 and p − 20 dbar were used. While potential vorticity is negative in the Southern Hemisphere, we refer only to its absolute value hereafter.

3. Characteristics of SPESTMW

We detected SPESTMW in dramatically increasing Argo profiling float data in spring (October–December) of 2004, 2005, and 2006, as follows. First, we detected the minimum potential vorticity using the relation between density and potential vorticity in the subsurface layer below the mixed layer at 5°–35°S and 140°–70°W (Fig. 2, left). These ranges of latitude and longitude were defined to encompass the SPESTMW distribution area shown by Wong and Johnson (2003). The minimum potential vorticity less than 2.5 × 10−10 m−1 s−1 was found in the 24.5–25.8 kg m−3 layer, which was located between the high potential vorticity layers corresponding to the seasonal pycnocline (above) and the main pycnocline (below). We used the potential vorticity value of 2.5 × 10−10 m−1 s−1 as a criterion for detecting the potential vorticity minimum layer. We also added a threshold of layer thickness greater than 40 m, to take the vertical resolution of the Argo data into account. Consequently, we defined SPESTMW as vertically homogeneous water with a potential vorticity of less than 2.5 × 10−10 m−1 s−1, and with a thickness exceeding 40 m. While the criterion of 2.5 × 10−10 m−1 s−1 is less than the SPESTMW definition of Wong and Johnson (2003), who used 3.0 × 10−10 m−1 s−1, it is appropriate for detecting SPESTMW from the data used here. We refer to the vertical grid with minimum potential vorticity in SPESTMW as the core of the SPESTMW and refer to all vertical grids within the SPESTMW layer as being “within SPESTMW.”

We found that the SPESTMW core had a temperature of 13°–26°C and salinity that was larger than 34 (Fig. 2, right). The SPESTMW detected in this study had higher temperature and salinity than was found in Wong and Johnson (2003; see their Fig. 5). Although our definition of SPESTMW is slightly different from that of Wong and Johnson (2003), we captured almost the same features as they described. In particular, SPESTMW had wider temperature and salinity ranges than STMW in the South Pacific (SPSTMW) and a relatively narrower density range than expected from its range of temperatures and salinities.

The pycnostad defined above was detected over the entire subtropical South Pacific. Most of the pycnostads in the eastern side of the subtropical gyre east of 160°W have Turner angles (Tu) of between 70° and 90° (Fig. 3). Here, Tu is an index showing the contribution of the temperature vertical gradient to the density gradient with respect to the salinity vertical gradient. The Tu values between 70° and 90° represent favorable conditions for salt fingering at that depth. The distribution of pycnostads with Tu between 70° and 90° almost coincides with the climatological distribution of SPESTMW (Fig. 1). However, most pycnostads in the western side of the subtropical South Pacific have Tu lower than 70° (colored pluses in Fig. 3). They have temperatures ranging from 15° to 22°C, salinities greater than 35.5, and densities ranging from 25.0 to 26.2 kg m−3. The properties of these pycnostads are consistent with the SPSTMW described by Roemmich and Cornuelle (1992), which was detected by hydrographical data from the National Oceanic Data Center (NODC). The pycnostad with Tu between 70° and 90° appears within the East Australian Current. This is considered to be a local type of SPSTMW. Though it is interesting, we leave it to a future study because it is beyond the scope of our research purpose. We can roughly distinguish between SPESTMW and SPSTMW using Tu. We regard the pycnostads on the eastern side of the subtropical South Pacific from 160°W as SPESTMW.

In the SPESTMW formation region between the climatological wintertime isopycnal outcrop lines of 24.5 and 25.8 kg m−3, the SPESTMW is more than 100 m thick, and its properties can vary over wide ranges. The SPESTMW in the southern part of its formation region and near the coastline of the South American continent has a temperature of approximately 13°–17°C and salinity of 34–35 at its core, which are the lowest values within the SPESTMW property range. The SPESTMW has higher temperature and salinity at its core to the northwest in the formation region, with the highest values near the outcrop line of 24.5 kg m−3. This characteristic distribution of the SPESTMW core properties reflects the wintertime sea surface properties in the SPESTMW formation region (figure not shown). In this region, where the mixed layer is deeper than 100 m (figure not shown), the temperature and salinity contributions to the horizontal density change compensate each other. Their distributions reflect this, and as a result, there is a wide window of horizontally homogeneous density. As Wong and Johnson (2003) described, this is the reason why SPESTMW has wide ranges of temperatures and salinities, despite its relatively narrow range of densities.

The SPESTMW potential vorticity is low (less than 1.0 × 10−10 m−1 s−1) in the formation region (Fig. 3e). Because the vertical gradient of the temperature at the SPESTMW core is also small in the formation region, the temperature gradient component of potential vorticity QT is equivalent to the total potential vorticity (Figs. 3e and 3f). In other words, the temperature, salinity, and SPESTMW density in spring are uniformly vertical in the formation area. In the region downstream from or northwest of the formation area, however, both the potential vorticity and the temperature gradient component are larger than in the formation region (Figs. 3e and 3f). Additionally, the temperature gradient component is much larger than the total potential vorticity in the region downstream from the formation area. Salinity stratifies to compensate temperature stratification, so that the total potential vorticity is considerably lower than the temperature vertical gradient component in the region downstream from formation. As a result, the low potential vorticity signature of SPESMTW spreads much more widely than its thermostad signature.

4. Vertical structure and seasonal change of SPESTMW

As described in section 1, Wong and Johnson (2003) suggested that salt fingering effectively destroys SPESTMW. While Johnson (2006) estimated the amount of TS variation on the isopycnal surface at the base of SPESTMW due to salt fingering, we do not yet understand how fast the SPESTMW itself decays. To examine this issue, we focused on the seasonal change of the SPESTMW’s vertical structure.

First, we examined the seasonal change of the potential vorticity at the SPESTMW core, which we detected according to the definition described in section 3 (Fig. 4). We detected the vertical minimum of the potential vorticity in each profile, excluding the mixed layer. Water with low potential vorticity is left in the subsurface in spring after the formation of the seasonal pycnocline. As the season progresses, the minimum potential vorticity in each profile becomes larger in the SPESTMW formation region. Moreover, the number of profiles containing SPESTMW decreases not only in the formation region, but also in the whole of the South Pacific subtropical gyre from spring to fall. This means that the SPESTMW diminishes considerably within a couple of seasons after its formation.

The temperature gradient component of the potential vorticity at the SPESTMW core shows similar seasonal evolution to the potential vorticity: small in spring in the SPESTMW formation region, then increasing there over time (Fig. 5). In addition, the temperature gradient component increases much faster than the potential vorticity in the formation region, particularly in the southern part of the formation area (Figs. 4 and 5). This fact means that the vertical gradient of the salinity also becomes larger with the season, to compensate for the increase in the vertical gradient of the temperature, so that the associated potential vorticity increase is more moderate than expected.

Next, we looked into the vertical structure and seasonal change of SPESTMW. We used vertical gradients as indices for vertical homogeneity to understand how the water decays after its formation (Fig. 6). We counted the seasonal frequency of the appearance of SPESTMW cores from July 2006 to June 2007, excluding the mixed layer. Since the number of profiles in the South Pacific subtropical gyre increased with time during this period, all frequencies in Fig. 6 were normalized by the number of all profiles in 5°–35°S, 160°–70°W in each season, regardless of whether or not they contained a core.

As shown in Fig. 6, the occurrence rate of SPESTMW increases from winter to spring. In particular, SPESTMW with potential vorticity less than 1.5 × 10−10 m−1 s−1 increases rapidly. The SPESTMW cores with potential vorticity lower than 1.0 × 10−10 m−1 s−1 dramatically decrease from spring to summer (January–March), and those with potential vorticity higher than 1.5 × 10−10 m−1 s−1 increase. From summer to fall (April–June), the occurrence rate of SPESTMW cores decreases for all classes, but especially for classes lower than 1.5 × 10−10 m−1 s−1. The considerable decrease of SPESTMW with low potential vorticity is consistent with Fig. 4. The potential vorticity mode increases from 0.5–1.0 × 10−10 to 1.5–2.0 × 10−10 m−1 s−1 from spring to fall.

Modes of the vertical gradients of temperature and salinity, in the form of their contributions to potential vorticity, occur in the third smallest and the smallest classes, respectively, in spring, and become ever larger as the seasons progress (Figs. 6b and 6c). From spring to summer, the mode of the vertical gradient of temperature increases from 1–1.5 × 10−10 to 3.5–4 × 10−10 m−1 s−1, much faster than the potential vorticity. In addition, the mode of the salinity vertical gradient increases from 0–0.5 × 10−10 to 0.5–1 × 10−10 m−1 s−1, slower than the potential vorticity and the temperature vertical gradient. A positive vertical gradient of salinity (temperature) is a destabilizing (stabilizing) factor for the vertical stratification of density. Because the vertical gradients of both temperature and salinity increase as the seasons progress, their effects on the density stratification tend to compensate each other. As a result, the vertical gradient of the density is maintained at a lower value than what would be expected from the vertical gradient of the temperature alone. To examine the degree of this compensation and its seasonal dependency, we counted the frequency of the Turner angle: Tu = arctan(αzT/βzS) (Ruddick 1983), in which ∂z represents z derivation (Fig. 6d). Here, Tu is an indicator representing the tendency of the occurrence of double diffusion due to density stratification, compensated by the temperature and salinity stratification. The mode of Tu occurs between 70° and 75° all year, indicating that positive vertical gradients of temperature and salinity effectively compensate each other in their contributions to the density stratification. Therefore, SPESTMW is truly vertically uniform water (i.e., thermostad, halostad, and pycnostad simultaneously) only immediately after the formation period. It should be noted that for 2 yr, from October 2004 to June 2005 and from October 2005 to June 2006, the potential vorticity and vertical gradients of temperature and salinity in the SPESTMW core showed seasonal evolutions that were similar to those from October 2006 to June 2007 (figures not shown).

To emphasize the features of the SPESTMW vertical structure and seasonality, we compared them with those of the STMW in the North Pacific (NPSTMW), which is known to be a strong thermostad and pycnostad (e.g., Suga et al. 1989). We detected NPSTMW using the Argo data from January 2004 to December 2005 in the area 20°–40°N, 120°E–180°, defining it as water with potential vorticity lower than 1.5 × 10−10 m−1 s−1, density of 25.0–25.6 kg m−3, temperature of 15°–20°C, and salinity higher than 34.5.

The modes of the potential vorticity and vertical temperature gradient in the NPSTMW core increased very slightly as seasons progressed, while the mode of the salinity vertical gradient did not change at all (Fig. 7). The mode of Tu occurred around 60°, which was smaller than that of SPESTMW. The vertical structure of the salinity in NPSTMW contributed to the density stratification much less than that in SPESTMW. That is, NPSTMW was both pycnostad and thermostad throughout the year, in contrast to SPESTMW, which was pycnostad and thermostad only immediately after formation and lost its vertical homogeneity rapidly as the seasons progressed.

The fact that the mode of Tu in the SPESTMW core occurred between 70° and 75° year round (Fig. 6d) indicates that the stratification of SPESTMW is a favorable condition for salt fingering throughout the year. It also suggests that salt fingering causes the increase of in the vertical gradients of the temperature and salinity of SPESTMW as the seasons progress. We discuss this further in the next section. On the other hand, NPSTMW had Tu values of approximately 60° year round, which are smaller than those in SPESTMW. The NPSTMW did not have a favorable condition for salt fingering in any season. This may have been why the vertical gradients of the NPSTMW temperature and salinity changed much less seasonally.

5. Discussion

As described in section 4, SPESTMW is characterized as a layer of nearly vertically uniform density caused by the stratifications of temperature and salinity compensating each other, which is very different from the NPSTMW case. The SPESTMW temperature and salinity stratification is such that the Turner angle Tu is 70°–75° year round. This means that SPESTMW maintains favorable stratification for double-diffusive convection due to salt fingering. In this section, we try to elucidate the change in SPESTMW properties due to salt fingering.

Wong and Johnson (2003) calculated the horizontal extent of salt fingering in the subtropical South Pacific by inducing the density ratio from the spatially coarse data. Johnson (2006) showed the temporal change of the salt fingering tendency from only two floats. Neither successfully captured the spatial and temporal changes in the salt fingering tendency over the whole extent of the SPESTMW. Thanks to the rapid expansion of the Argo array in the South Pacific, many more data were available for our study. First, we tried to understand where salt fingering tends to occur in the vertical structure of the SPESTMW and how the layers with stratification favorable to salt fingering are distributed horizontally. Here, we adopt Tu as an indicator representing the tendency for double diffusion to occur.

It was demonstrated that most of profiles with the SPESTMW (96.71%) had Tu ranging from 71.6° to 90° from the core to the lower part of the SPESTMW. Values of Tu were frequently above 77° (in about 60% of all profiles during October–December in 2006) and sometimes approached 90° at and below the core of the water (Fig. 8), suggesting very vigorous salt fingering activity (Schmitt 1981; St. Laurent and Schmitt 1999). Most of the vertical Tu maxima in each profile occurred between the SPESTMW core and 100 m deeper than the core (red line in Fig. 8), and 85% of them occurred within a 30-m depth from the core. Moreover, Tu was close to 90° near the SPESTMW core, and the Tu maximum in each profile occurred most frequently there. Therefore, the vertical structure of SPESTMW in spring was favorable for strong salt fingering most frequently near the core, while the lower part of the SPESTMW was generally favorable.

Almost all the profiles with the SPESTMW showed Tu values larger than 70° within and beneath the SPESTMW year round (Fig. 9). The Tu values larger than 77° were confined to the SPESTMW formation region. The Tu values in the SPESTMW formation area were closest to 90° in spring and decreased from spring to fall (April–June). In addition, Tu became increasingly smaller to the northwest. It is thus suggested that the modification of SPESTMW due to salt fingering began immediately after its formation. This is a plausible explanation for why the SPESTMW did not spread far from its formation region, as suggested by Wong and Johnson (2003) and as shown in Fig. 3. It should be pointed out here that the reason why SPESTMW covers a wide area is not that SPESTMW spreads far from its formation region but that its formation region itself spreads widely.

Next, we estimated the temporal changes of temperature and salinity due to diapycnal mixing enhanced by salt fingering, to examine how much of the observed changes of SPESTMW properties can be explained by the vertical diffusion. We estimated the heat and salt diffusive terms, ∂z(KTzT) and ∂z(KSzS), respectively, according to the method used by Johnson (2006). The parameterization of diapycnal salt diffusion due to salt fingering KS was adopted and fitted to the vertical diffusion of salinity data estimated by St. Laurent and Schmitt (1999) as a function of the density ratio. We obtained KS = 2.4 × 10−4 [1 − (Rρ − 1)/(2.05 − 1)]3 + 0.1 × 10−4 m2 s−1 for 2.05 > Rρ > 1 (71° < Tu < 90°) and KS = 0.1 × 10−4 m2 s−1 for outside the range of Rρ. We adopted the parameterization of the diapycnal temperature diffusion, KT, in which KT is determined by the function, KT = (8/13)KS for 2.05 > Rρ >1 (71° < Tu < 90°), and KT = 0.1 × 10−4 m2 s−1 outside the range of Rρ.

There are some parameterizations of diapycnal diffusion due to double diffusion (e.g., Zhang et al. 1998; St. Laurent and Schmitt 1999). They were derived by microstructure data observed in a narrow region and during a short time, and are ad hoc. It is not certain how well these parameterizations represent the diapycnal diffusion in the relatively wide region of the subtropical South Pacific. Because we used data with 10-m vertical resolution, the uncertainties in the coefficients of the function were also large. Therefore, we cannot accurately discuss the amount of diffusion due to salt fingering, but we can discuss qualitatively where the changes in temperature and salinity occur.

The SPESTMW lost more heat and salt in its bottom than in its center after its formation in the next winter, despite the fact that Tu was closest to 90° at the core (Figs. 10, top and middle). This was due to the structure of the vertical gradients of temperature and salinity. The time integration of the temperature and salinity diffusive terms on isopycnal surfaces in the lower part of SPESTMW indicated decreases in temperature and salinity from the formation period to the next winter, which corresponded to the observed temperature and salinity on each isopycnal surface fairly well (Fig. 11). On the 25.2 kg m−3 isopycnal surface, which was situated near the bottom of SPESTMW, heat and salt were lost vigorously within the SPESTMW, and its temperature and salinity were lowered by about 0.3°C and 0.1, respectively, by the next winter (Fig. 11). The SPESTMW lost significant amounts of heat and salt from its formation period to the next winter at its bottom due to salt fingering.

The SPESTMW lost heat and salt at its central layers as well as at its base (Fig. 10). Heat and salt diffusive terms integrated vertically within the SPESTMW had negative values until the next winter (Fig. 11). This situation was also observed by other floats. These results suggest that salt fingering caused the decrease of temperature and salinity in the whole SPESTMW layer after its formation.

We also estimated the density diffusive term associated with diapycnal mixing due to salt fingering ∂z(ραKTzTρβKSzS) within and beneath the SPESTMW, using the diffusive coefficient of temperature and salinity calculated above (Fig. 10, bottom). The density diffusive term was largest at, and just below, the mixed layer as the mixed layer depth increases. It was also relatively large in the lower part of and just beneath the SPESTMW. The density appeared to decrease due to salt fingering in the bottom part of SPESTMW (above the 25.2 kg m−3 isopycnal surface) from November 2004 to July 2005, and it tended to increase at the 25.2–25.3 kg m−3 isopycnal surfaces just below the SPESTMW layer during the same period (Fig. 12). The bottom part of the SPESTMW may gain buoyancy, and the layer just below the SPESTMW may lose buoyancy over time.

The temperature and salinity diffusion in the SPESTMW from November 2004 to August 2005 was greater than in the following year (Fig. 10). This was due to the difference in temperature and salinity stratification between the two years. Although the float shown in Fig. 10 moved near the formation region of the SPESTMW and did not move away during the period analyzed in Fig. 10, it is not certain that this difference represents spatial variation or the interannual change of the SPESTMW. This will be clarified in future work.

6. Conclusions

We investigated the horizontal distribution, vertical structure, and seasonal evolution of SPESTMW by analyzing all available temperature and salinity profiles observed by Argo floats from July 2004 to June 2007. Examining the vertical minima of the potential vorticity over the subtropical South Pacific, we defined SPESTMW as water with a potential vorticity magnitude of less than 2.5 × 10−10 m−1 s−1 and a thickness larger than 40 m. The SPESTMW thus defined had a temperature of 13°–26°C, salinity greater than 34.0, and density of 24.5–25.8 kg m−3 at its core. Although this definition is slightly different from that of Wong and Johnson (2003), which was based on the fewer profiles available at that time, the present study captured the same general features of SPESTMW, namely, that SPESTMW has a wider temperature and salinity range than SPSTMW.

We confirmed that the vertical changes in temperature and salinity tend to compensate for each other in terms of density changes, resulting in conditions favorable to salt fingering, as Wong and Johnson (2003) mentioned. By analyzing many profiles of the Argo data over 3 yr and comparing SPESTMW with NPSTMW, which is known as a thick thermostad, we found that the temperature and salinity of SPESTMW were not as vertically uniform as those of NPSTMW of the same potential vorticity, even immediately after formation. Therefore, the low potential vorticity signature of SPESTMW spread much wider than its signature as a thermostad. The vertical gradients of temperature and salinity in the SPESTMW formation region were lowest immediately after formation, and they became larger even in the formation region as the seasons progressed. SPESTMW is truly vertically uniform water (thermostad, halostad, and pycnostad simultaneously) only in the formation region immediately after the formation period; otherwise, it is pycnostad but neither thermostad nor halostad. The analysis of numerous Argo profiles clearly captured this seasonal evolution of SPESTMW.

We also found that the seasonal evolution of the SPESTMW temperature and salinity vertical gradients was related to the diffusive flux of temperature and salinity due to salt fingering, in a manner similar to the way interannual anomalies are rapidly modified, as shown by Johnson (2006). Temperature and salinity near the SPESTMW core and its lower part began to decrease soon after formation. These changes from the SPESTMW core to the layer underneath were reasonably explained by the vertical diffusion enhanced by salt fingering.

The modes of temperature and salinity within SPESTMW detected from all profiles in each year within the layer of 25.1–25.5 kg m−3 become slightly colder and fresher than those in spring (Fig. 13), as the seasons progressed. This suggests that salt fingering may cause systematic changes in the properties of the whole SPESTMW. More detailed observations are necessary to clarify quantitatively the extent to which salt fingering contributes to the modification of the SPESTMW.

There seem marginal indications that the bottom part of SPESTMW gained buoyancy and that the layer just below the SPESTMW lost buoyancy sometimes from SPESTMW formation in the estimation of the density diffusive term associated with diapycnal mixing due to salt fingering (Fig. 10). If that is true, the layer thickness near the bottom of and under the SPESTMW would increase after its formation. In other words, there might be a possibility that salt fingering would transport low potential vorticity downward along with temperature and salinity and then affect the general circulation by redistributing potential vorticity. This would be another interesting subject for future study.

Acknowledgments

We thank the members of the Japan Agency for Marine–Earth Science and Technology (JAMSTEC) Argo group. Mr. Ohira in the JAMSTEC Argo group calculated the difference between Argo data and World Ocean Atlas 2001 data. The Argo float data used in this study were collected and made freely available by the International Argo Project and the national programs that contribute to it (information online at http://www.argo.ucsd.edu and http://argo.jcommops.org).

REFERENCES

  • Akima, H., 1970: A new method of interpolation and smooth curve fitting based on local procedures. J. Assoc. Comput. Mech., 17 , 589602.

    • Search Google Scholar
    • Export Citation
  • Argo Data Management Team, 2004: Argo quality control manual, version 2.0b. Argo Data Management, 23 pp.

  • Argo Science Team, 2001: Argo: The global array of profiling floats. Observing the Oceans in the 21st Century, C. J. Koblinsky and N. R. Smith, Eds., GODAE Project Office, 248–258.

    • Search Google Scholar
    • Export Citation
  • Giese, B. S., , S. C. Urizar, , and N. S. Fuckar, 2002: Southern Hemisphere origins of the 1976 climate shift. Geophys. Res. Lett., 29 , 1014. doi:10.1029/2001GL013268.

    • Search Google Scholar
    • Export Citation
  • Hanawa, K., , and L. D. Talley, 2001: Mode waters. Ocean Circulation and Climate—Observing and Modelling the Global Ocean, G. Siedler, J. Church, and J. Gould, Eds., Academic Press, 373–386.

    • Search Google Scholar
    • Export Citation
  • Hautala, S. L., , and D. H. Roemmich, 1998: Subtropical mode water in the Northeast Pacific Basin. J. Geophys. Res., 103 , (C6). 1305513066.

    • Search Google Scholar
    • Export Citation
  • Hosoda, S., , T. Ohira, , and T. Nakamura, 2008: A monthly mean dataset of global oceanic temperature and salinity derived from Argo float observations. JAMSTEC Rep. Res. Develop., 8 , 4769.

    • Search Google Scholar
    • Export Citation
  • Johnson, G. C., 2006: Generation and initial evolution of a mode water θ–S anomaly. J. Phys. Oceanogr., 22 , 184195.

  • Johnson, G. C., , and M. J. McPhaden, 1999: Interior pycnocline flow from the subtropical to the equatorial Pacific Ocean. J. Phys. Oceanogr., 29 , 30733089.

    • Search Google Scholar
    • Export Citation
  • Masuzawa, J., 1969: Subtropical mode water. Deep-Sea Res., 16 , 463472.

  • Nonaka, M., , and H. Sasaki, 2007: Formation mechanism for isopycnal temperature–salinity anomalies propagating from the eastern South Pacific to the equatorial region. J. Climate, 20 , 13051315.

    • Search Google Scholar
    • Export Citation
  • Provost, C., , C. Escoffier, , K. Maamaatuaiahutapu, , A. Kartavtseff, , and V. Garcon, 1999: Subtropical mode waters in the South Atlantic Ocean. J. Geophys. Res., 104 , (C9). 2103321049.

    • Search Google Scholar
    • Export Citation
  • Roemmich, D., , and B. Cornuelle, 1992: The Subtropical Mode Water of the South Pacific Ocean. J. Phys. Oceanogr., 22 , 11781187.

  • Ruddick, B., 1983: A practical indicator of the stability of the water column to double-diffusive activity. Deep-Sea Res., 30A , 11051107.

    • Search Google Scholar
    • Export Citation
  • Schmitt, R. W., 1981: Form of the temperature–salinity relationship in the Central Water: Evidence for double-diffusive mixing. J. Phys. Oceanogr., 20 , 10151026.

    • Search Google Scholar
    • Export Citation
  • Siedler, G., , A. Kuhl, , and W. Zenk, 1987: The Madeira Mode Water. J. Phys. Oceanogr., 17 , 15611570.

  • St. Laurent, L., , and R. W. Schmitt, 1999: The contribution of salt fingers to vertical mixing in the North Atlantic Tracer Release Experiment. J. Phys. Oceanogr., 29 , 14041424.

    • Search Google Scholar
    • Export Citation
  • Suga, T., , K. Hanawa, , and Y. Toba, 1989: Subtropical mode water in the 137°E section. J. Phys. Oceanogr., 19 , 16051618.

  • Tsubouchi, T., , T. Suga, , and K. Hanawa, 2007: Three types of South Pacific subtropical mode waters: Their relation to the large-scale circulation of the South Pacific subtropical gyre and their temporal variability. J. Phys. Oceanogr., 37 , 24782490.

    • Search Google Scholar
    • Export Citation
  • Tsuchiya, M., , and L. D. Talley, 1996: Water-property distributions along an eastern Pacific hydrographic section at 135°W. J. Mar. Res., 54 , 541564.

    • Search Google Scholar
    • Export Citation
  • Wong, A. P. S., , and G. C. Johnson, 2003: South Pacific Eastern Subtropical Mode Water. J. Phys. Oceanogr., 33 , 14931509.

  • Worthington, L. V., 1959: The 18°C Water in the Sargasso Sea. Deep-Sea Res., 5 , 297305.

  • Yeager, S. G., , and W. G. Large, 2004: Late-winter generation of spiciness on subducted isopycnals. J. Phys. Oceanogr., 34 , 15281547.

  • Zhang, J., , R. W. Schmitt, , and R. X. Huang, 1998: Sensitivity of the GFDL modular ocean model to parameterization of double-diffusive processes. J. Phys. Oceanogr., 28 , 589605.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

The climatological austral spring distribution of potential vorticity on the 25.3 kg m−3 isopycnal surface, superimposed on the pressure anomaly streamfunction (contour interval of 0.5 m2 s−2) relative to 2000 dbar on the same isopycnal surface, by averaging the MOAA GPV dataset from October to December during 2000–07. The red line denotes the wintertime outcrop of the 25.3 kg m−3 isopycnal surface. The green line denotes the trajectory of the float with World Meteorological Organization (WMO) ID 3900255, which is shown in Figs. 10 and 11.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 2.
Fig. 2.

(left) The relation between potential vorticity and density without the mixed layer at 5°–35°S, 140°–70°W from October to December (austral spring) in 2004–06. (right) A θS diagram at the depth of the minimum potential vorticity, for profiles with the potential vorticity less than 2.5 × 10−10 m−1 s−1. This diagram represents the frequency of each θS class in the same region and time period as in the left panel.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 3.
Fig. 3.

The distribution of the properties at the core of the mode water, defined as the water with potential vorticity less than 2.5 × 10−10 m−1 s−1 and with thickness exceeding 40 m from October to December in 2006: (a) temperature, (b) salinity, (c) density, (d) thickness, (e) potential vorticity, and (f) the temperature gradient component of the potential vorticity QT. Circles (colored pluses) denote the water with (without) a Turner angle between 70° and 90° in the potential vorticity minimum layer. Red lines in (e) and (f) denote the climatological outcrop density lines of 24.5 kg m−3 (northern line) and 25.8 kg m−3 (southern line) in austral winter (July–September).

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 4.
Fig. 4.

The seasonal distribution of the minimum potential vorticity of each profile with the mode water, defined in the same manner as in Fig. 3 from July 2006 to June 2007: (a) July–September (winter), (b) October–December (spring), (c) January–March (summer), and (d) April–June (fall). Circles and colored plus signs have the same meaning as in Fig. 3.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 5.
Fig. 5.

Same as in Fig. 4, but for the temperature gradient component of the potential vorticity at the core of the mode water. Circles and colored plus signs have the same meaning as in Fig. 3.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 6.
Fig. 6.

The seasonal frequency of the (a) potential vorticity, (b) temperature gradient component of the potential vorticity, (c) (negative) salinity gradient component of the potential vorticity, and (d) Turner angle at the SPESTMW core in 5°–35°S, 160°–70°W from July 2006 to June 2007. All frequencies are normalized by the number of profiles in the region described above in each season.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 7.
Fig. 7.

Same as in Fig. 6, but at the NPSTMW cores in 20°–40°N, 120°E–180° from January to December 2005.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 8.
Fig. 8.

Turner angle profile from October to December in 2006 vs the relative depth to the SPESTMW core (green and black circles); a positive (negative) value represents the depth below (above) the SPESTMW core. The grid values within SPESTMW are plotted as green circles, and those outside of the SPESTMW are plotted as black circles. Only Tu values from 71.6° to 90° are displayed. (red line) Occurrence frequency of the depth with maximum Tu in the range between 71.6° and 90° in each profile. The number in the lower-right corner of the figure denotes the proportion of the profiles with Tu exceeding 77° to all available profiles from October to December in 2006.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 9.
Fig. 9.

The seasonal distribution of the maximum Turner angle of each profile, with the mode water defined in the same way as in Fig. 3 from July 2006 to June 2007: (a) July–September, (b) October–December, (c) January–March, and (d) April–June. Circles and colored plus signs have the same meaning as in Fig. 3.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 10.
Fig. 10.

Time–depth sections of the diffusive terms of (top) heat, (middle) salt, and (bottom) density computed using the salt fingering diffusive parameterization. Time–depth sections of potential temperature, salinity, and density (solid-line contour) obtained by the float with WMO ID 3900255 are superimposed. Green circles denote grids in the SPESTMW.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 11.
Fig. 11.

Time series of (left) temperature and (right) salinity on the isopycnal surface. The circles denote values observed by the float with WMO ID 3900255, as in Fig. 10. The stars denote the values modeled by time integration of the diapycnal diffusion term of the temperature and salinity from Fig. 10.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 12.
Fig. 12.

Time series of the density diffusive term due to salt fingering between two isopycnal surfaces within a 0.1 kg m−3 interval, from 25.0 to 25.5 kg m−3, calculated from the float with WMO ID 3900255 as in Fig. 10.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

Fig. 13.
Fig. 13.

The seasonal frequency of (left) potential temperature and (right) salinity from (a) 25.1 to 25.3 kg m−3 and (b) 25.3 to 25.5 kg m−3, within the SPESTMW in 5°–35°S, 160°–70°W from July 2005 to June 2006. All frequencies are normalized by the number of profiles in the region described above in each season.

Citation: Journal of Physical Oceanography 39, 7; 10.1175/2008JPO3940.1

1

The real-time quality control is a very gross filter, and results with scientifically quality-controlled delayed-mode data might be different. The date when the dataset was downloaded is specified to allow future readers to determine if some problem with the data found later might affect the results.

Save