Volume 40, Issue 12 p. 3102-3105
Regular Article
Free Access

Marked freshening of North Pacific subtropical mode water in 2009 and 2010: Influence of freshwater supply in the 2008 warm season

Shusaku Sugimoto

Corresponding Author

Shusaku Sugimoto

Department of Geophysics, Graduate School of Science, Tohoku University, Sendai, Japan

Corresponding author: S. Sugimoto, Department of Geophysics, Graduate School of Science, Tohoku University, 6-3 Aramaki-aza-Aoba, Aoba-ku, Sendai, Miyagi, 980-8578, Japan. ([email protected])Search for more papers by this author
Nobuto Takahashi

Nobuto Takahashi

School of Food, Agricultural and Environmental Sciences, Miyagi University, Sendai, Japan

Search for more papers by this author
Kimio Hanawa

Kimio Hanawa

Department of Geophysics, Graduate School of Science, Tohoku University, Sendai, Japan

Search for more papers by this author
First published: 31 May 2013
Citations: 16

Abstract

[1] Changes in salinity of the North Pacific subtropical mode water (STMW) were investigated using profiles for 2003–2011. In 2009 and 2010, the STMW freshened markedly (S < 34.7 psu), with salinity 0.1 psu lower than other years. Freshening in 2009 was caused by excess rainfall in the 2008 warm season, which was related to an increase in the number of low pressures passing over the STMW formation region associated with a southeastward shrinkage of the summertime North Pacific subtropical high. The freshening signal persisted under the seasonal pycnocline during the 2009 warm season. This resulted in the freshening of STMW in 2010.

Key Points

  • We investigated salinity changes using Argo float
  • The STMW freshened markedly in 2009 and 2010
  • The freshening was caused by excess rainfall in the 2008 warm season

1 Introduction

[2] The North Pacific subtropical mode water (STMW), characterized by low potential vorticity (PV), is distributed widely in the northwestern part of the subtropical gyre [Masuzawa, 1969]. The STMW has a temperature of 16°–20°C and salinity of about 34.8 psu (see reviews by Hanawa and Talley [2001] and Oka and Qiu [2012]). Numerous authors have investigated STMW temperature variations [e.g., Hanawa and Kamada, 2001; Oka, 2009]. However, changes in salinity have not yet been clarified because of a lack of data.

[3] The availability of salinity profiles has increased dramatically year by year since 2000, when the international Argo project [Argo Science Team, 2001] started. These data can provide a new perspective on STMW salinity. In this study, we investigate changes in salinity and then attempt to assess the role of freshwater exchanges between the atmosphere and the ocean in determining salinity by using a salt budget analysis. The remainder of this paper is organized as follows. Section 2 outlines the data sets and definitions. Section 3 investigates the STMW salinity and then explores the role of freshwater exchanges. Section 4 describes our summary and concluding remarks.

2 Data Sets and Definitions

[4] We use temperature (T) and salinity (S) data archived in the World Ocean Database 2009 (WOD09) [Boyer et al., 2009] and at the Japan Oceanographic Data Center (JODC; http://www.jodc.go.jp) and T-S profiles from Argo floats [Oka et al., 2007] for 2003–2011. To capture the STMW adequately, we only use the profiles with maximum depth greater than 450 dbar. To control data quality, we first remove profiles duplicated in the different data sources. For each profile, the measured T-S data are compared with all values measured in the same month within a 1° × 1° box that centers on the observation point: data are excluded if they fall outside of three standard deviations of the mean. Profiles with large T inversions (dT/dz < −0.1°C dbar−1) from the surface to 450 dbar are also removed. After quality control, T-S profiles are interpolated vertically onto a 1 dbar interval using the Akima [1970] scheme. Then θ and σθ are calculated. The PV (Q) is defined as follows [Talley, 1988]:
urn:x-wiley:00948276:media:grl50600:grl50600-math-0001(1)
where f is the Coriolis parameter and ρ is the water density. In equation (1), the relative vorticity is neglected because it is much smaller than the planetary vorticity, except near vicinity of the Kuroshio/Kuroshio Extension (KE) axis [Qiu et al., 2006]. The STMW is detected as a low-Q layer of less than 2.0 × 10−10 m−1 s−1, with θ of 16°–19.5°C [Oka, 2009] and thickness greater than 100 dbar. A local vertical minimum of Q is identified from each profile and then labeled as a core of the STMW.

[5] The mixed layer depth (MLD) is defined as the depth at which σθ increases by 0.1 kg m−3 from the 10 dbar depth value [Oka, 2009]. Properties in the mixed layer (ML) are regarded as values at a 10 dbar depth. In this study, we investigate properties of the STMW and ML in a region south of the Kuroshio/KE: we select the θ = 16°C isotherm at 200 dbar as the Kuroshio/KE axis rather than θ = 15°C, which is normally regarded as a good indicator of the Kuroshio/KE axis [Kawai, 1972]. This is to ensure there is no risk of identifying subarctic profiles as coming from north of the Kuroshio/KE. Figure 1 represents the number of profiles per month used for this study.

Details are in the caption following the image
Number of θ-S profiles for each month in the northwestern part of the subtropical gyre (125°E–180°E, 20°N–36°N) (gray bars). The black line represents the number of profiles in the STMW formation region (141°E–155°E, 30°N–35°N).

[6] We use precipitation data from the Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP) [Xie and Arkin, 1997] and evaporation data and geopotential height data at 850 hPa (Z850) from the Japanese 25 year reanalysis/Japan Meteorological Agency climate data assimilation system (JRA25/JCDAS) [Onogi et al., 2007].

3 Results

[7] We investigate STMW properties in spring (April–June), for which the water properties are preserved at the time of formation. In 2009 and 2010, the STMW shows a marked freshening (core S < 34.7 psu), with salinity 0.1 psu lower than other years (Figure 2a). Such lower S has not been reported before [e.g., Hanawa and Talley, 2001]. The low-S core is distributed widely in the northwestern part of the subtropical gyre (Figures 2b and 2c). We examine the ML salinity (MLS) in the cold season (February–March) because the STMW is formed in the deep winter ML [Suga and Hanawa, 1990]. In 2009 and 2010, the lower MLS (S < 34.7 psu) is formed (Figure 2d) at locations mostly in an area north of 30°N (Figures 2e and 2f). These locations correspond closely to the STMW formation region [Hanawa and Yoritaka, 2001]. It is apparent that the fresh STMW results from the MLS in the cold season.

Details are in the caption following the image
(a) θ-S diagram of the STMW core in spring (April–June). Blue and red dots represent core properties in 2009 and 2010, respectively, and gray dots are those in other years. (b) Distributions of STMW core in spring 2009: core S < 34.7 psu (orange circles) and core S > 34.7 psu (green circles). Gray symbols represent observations without STMW. Contours (20 cm interval) indicate the sea surface height from the satellite-derived altimetry data set of the Archiving Validation and Interpretation of Satellite Oceanographic data (AVISO: http://www.aviso.com). (c) As in Figure 2b, but for spring 2010. (d) As in Figure 2a, but for ML properties in the cold season (February–March); MLD > 150 dbar. (e and f) As in Figures 2b and 2c, but for ML properties (MLD > 150 dbar); MLS < 34.7 psu (orange circles) and MLS > 34.7 psu (green circles). Gray symbols represent observations with shallow MLD (<150 dbar). The black rectangular area indicates the STMW formation region (141°E–155°E, 30°N–35°N).
[8] To investigate the causes of MLS freshening in the 2009 and 2010 cold seasons, we prepared a monthly time series of MLS averaged within the STMW formation region (141°E–155°E, 30°N–35°N) (Figure 3). The MLS exhibits a seasonal cycle: low values in the warm season and high values in the cold season. Interestingly, marked freshening of MLS (S < 34.2 psu) is observed in the 2008 warm season, half a year before the 2009 cold season; this is the lowest salinity throughout the analysis period. It is therefore expected that the freshening in the 2009 and 2010 cold seasons results from the MLS in the previous warm season. We examined the relationship between the MLS in the cold season and that in the previous warm season (May–October) by performing a correlation analysis, yielding a high correlation coefficient (R = 0.85). We evaluate the extent to which air-sea freshwater exchanges can explain the MLS in the warm season by performing a simple salt budget analysis as follows:
urn:x-wiley:00948276:media:grl50600:grl50600-math-0002(2)
where E is the monthly evaporation, P is the monthly precipitation, S is the monthly MLS (Figure 3), h represents the monthly MLD averaged within the STMW formation region, and ε is a residual term that includes advection and entrainment terms. The terms on the left and right sides in equation (2) represent the time rate of changes in MLS and air-sea freshwater exchange, respectively.
Details are in the caption following the image
Monthly time series of MLS averaged within the STMW formation region (141°E–155°E, 30°N–35°N). Error bars denote one standard error.

[9] The time rate of changes in MLS (Figure 4a) shows the largest negative value in 2008, and the time series closely resembles the air-sea freshwater exchange term (red line in Figure 4a); a significant correlation coefficient is obtained (R = 0.77). The freshwater exchange is primarily responsible for the determination of MLS. To provide further evidence for the role of freshwater exchange, we estimated the respective behavior of Ekman advection, geostrophic advection, and entrainment terms regarded as a residual term in equation (2) by performing a salt budget analysis of Ren and Riser [2009], using the monthly gridded θ-S data set produced by Hosoda et al. [2008]. Results (not shown here) showed that variations in the three terms listed above were quite small compared with the freshwater exchange term. This indicates that the MLS in the warm season is determined predominantly by air-sea freshwater exchange.

Details are in the caption following the image
(a) Time rate of changes in MLS in the STMW formation region during the warm season, defined as MLS in October minus MLS in April (bars). The red line represents the air-sea freshwater exchange term in equation (2). (b) E-P flux averaged within the STMW formation region in the warm season. The dashed line denotes the mean value. (c) E-P flux anomaly in the 2008 warm season [mm day−1] (shading). Contours represent the E-P flux climatology in the warm season (2003–2011) (contour interval of 1 mm day−1; zero contours are omitted). Black rectangular area indicates the STMW formation region. (d) As in Figure 4c, but for Z850. (e) Number of low pressures passing over the STMW formation region in the warm season, as detected from the 6-hourly JRA25/JCDAS data set using the method of Mizuta et al. [2011].

[10] We investigate the causes of the low MLS in the warm season. The 2008 warm season had an excess freshwater supply, highest recorded in the CMAP and JRA25/JCDAS data sets, which start in 1979 (Figure 4b). Figure 4c displays the E-P flux anomaly for 2008. The rainfall region moved southward by about 10° in latitude compared with the climatology, and this resulted in enhanced freshwater supply to the STMW region. The spatial pattern of the E-P flux anomaly is similar to the Z850 anomaly (Figure 4d). It can be pointed out that the increase in rainfall is caused by changes in the tracks of low pressures associated with a southeastward shrinkage of the summertime North Pacific subtropical high. In fact, the number of low pressures passing over the STMW region in 2008 was the highest of the analysis period (Figure 4e). On the other hand, no such marked freshwater supply existed in the 2009 warm season (Figure 4b), which gives rise to the following question: Why was the MLS anomalously fresh in the 2010 cold season? It appears from Figure 5 that the fresher signal that formed in the 2008 warm season persisted under the seasonal pycnocline during the 2009 warm season and so was able to influence the MLS in the following cold season (2010 cold season).

Details are in the caption following the image
Monthly time-depth section of salinity averaged within the STMW formation region (141°E–155°E, 30°N–35°N) (contours with an interval of 0.1 psu: green and light blue lines represent 34.7 and 34.6 psu, respectively). Shading denotes the salinity anomaly. The thick black line represents the MLD.

4 Summary and Concluding Remarks

[11] We have investigated changes in STMW salinity using vertical profiles for 2003–2011. In 2009 and 2010, the STMW was markedly fresher (core S < 34.7 psu), with salinity 0.1 psu lower than other years. The freshening in 2009 was caused by the excess rainfall in the 2008 warm season, which was related to an increase in the number of low pressures passing over the STMW formation region associated with a southeastward shrinkage of the summertime North Pacific subtropical high. The fresh signal appeared to persist under the seasonal pycnocline during the 2009 warm season and to influence the MLS in the following cold season. It can be pointed out that the freshening of STMW in 2010 resulted from the lower salinity in the subsurface layer.

[12] The freshening signal in the 2010 cold season also tends to be preserved under the shallow summer ML (Figure 5), and then the fresh anomaly appears to recur in the ML in the 2011 cold season as the ML deepened. Kwon and Riser [2004] pointed out that changes in properties of the North Atlantic subtropical mode water reflect atmospheric forcing over the previous 4–5 years. In future work, the role of STMW as an interannual-scale memory should be clarified by producing a long-term salinity data set. To understand STMW persistence, it is imperative to assess quantitatively the influence of the subsurface ocean current field.

Acknowledgments

[13] The authors thank members of the Physical Oceanography Group at Tohoku University for their useful discussions. The first author (S.S.) was partly supported by the Grant-in-Aid for Young Scientists (B) (23740348) from the Japan Society for the Promotion of Science and by the Grant-in-Aid for Scientific Research on Innovative Areas (25106702, “A ‘hot spot’ in the climate system: Extra-tropical air-sea interaction under the East Asian monsoon system”) from the Ministry of Education, Culture, Sports, Science and Technology. Comments from two anonymous reviewers improved the manuscript.

[14] The Editor thanks an anonymous reviewer for assisting in the evaluation of this paper.